Regulation of tyrosine hydroxylase activity


At first, it was thought that control of TH activity depended on inhibition by its end-product, noradrenaline, which competes with the binding of co-factor. According to this scheme, release of noradrenaline would diminish end-product inhibition of the enzyme and so ensure that synthesis is increased to replenish the stores. When the neurons are quiescent, the opposite would occur: i.e. intraneuronal accumulation of noradrenaline would automatically blunt synthesis. Much evidence was deemed to support this view. For instance, when metabolic breakdown of cytoplasmic noradrenaline was prevented by treatment with an inhibitor of the enzyme, monoamine oxidase (MAO: see below), the rate of synthesis of [3H]noradrenaline from [3H]tyrosine was markedly reduced (Neff and Costa 1966).

However, as early as the 1970s, it was obvious that end-product inhibition of TH could not be the main factor regulating the rate of noradrenaline synthesis. Clearly, the hydroxylation of tyrosine takes place in the cytoplasm and so it must be cytoplasmic noradrenaline that governs enzyme activity. Yet, it is vesicle-bound transmitter that undergoes impulse-evoked release from the neuron. Also, when neurons are releasing noradrenaline, its reuptake from the synapse is increased and, even though some of this transmitter ends up in the vesicles, or is metabolised by MAO, there should be a transient increase in the concentration of cytoplasmic noradrenaline which would increase end-product inhibition of TH.

To overcome these difficulties it was suggested that there was a small 'strategic pool' of cytoplasmic noradrenaline that inhibited the activity of TH. Nevertheless, even this small pool was eventually ruled out as a regulator of TH. This followed in vitro experiments investigating the effects of addition of reduced pterin co-factor on the activity of the enzyme derived from the vas deferens. It was predicted that the activity of enzyme derived from control tissues would be increased by the addition of co-factor in vitro whereas that enzyme derived from stimulated tissues should not increase because the TH would already have been maximally activated by endogenous co-factor during nerve stimulation. In fact, co-factor increased noradrenaline synthesis in both instances, suggesting that noradrenaline synthesis depended primarily on factors that directly activate TH, rather than on removal of end-product inhibition. The extent to which end-product inhibition of TH contributes to the regulation of its activity under physiologically relevant (e.g. drug-free) conditions remains uncertain.

The first clue to the processes which normally regulate TH activity came from experiments showing that electrical stimulation of sympathetic neurons increased the affinity of this enzyme for its co-factor and reduced its affinity for noradrenaline (for detailed reviews of this topic see Zigmond, Schwarzschild and Rittenhouse 1989; Fillenz 1993; Kaufman 1995; Kumar and Vrana 1996). Several lines of investigation showed that activation of TH was in fact paralleled by its phosphorylation and it was this process that accounted for the changes in the enzyme's kinetics (Table 8.2).

Table 8.2 The effects of phosphorylation of tyrosine hydroxylase on enzyme kinetics (based on Kaufman 1995)

Km for co-factor

Y max

cAMP-dependent protein kinase (PKA) (pH 6.0)


No change

cAMP-dependent protein kinase (PKA) (pH 7.0-7.4)



Ca2+-calmodulin dependent protein kinase (CAM-PK II)

No change


Ca2+-phospholipid dependent protein kinase (PKC) (pH 7.0)


No change

cGMP-dependent protein kinase (PKG) (pH 6.0)



cAMP-independent protein kinase

No change


It is now known that such phosphorylation is activated by several protein kinases, including Ca2+/phospholipid-dependent protein kinase (PKC), which reduces its Km for co-factor, and cGMP-dependent protein kinase. These factors phosphorylate different sites on the enzyme, although some are shared by different kinases. In rat TH, serine residues, Ser8, Ser19, Ser31 and Ser40, have been identified as targets and Ser40 seems to be a common target for all the kinases. It is thought that this site produces a conformational change in the enzyme that reduces its affinity for catecholamines. All these regulatory sites reside on the N-terminus of the enzyme, whereas it is the C-terminus that comprises the catalytic site. In addition to all these changes, phosphorylation of the enzyme changes the pH optimum for maximal enzyme activity and so the kinetics of this enzyme depend on the pH of the incubation medium to some extent.

In the periphery, some of the primary triggers for these processes have been identified. Acetylcholine seems to be one such factor because stimulation of preganglionic nerves in vivo increases enzyme activity. However, nicotinic and muscarinic receptor antagonists do not completely prevent this increase. The residual activation is attributed to peptides of the secretin-glucagon subgroup, including VIP and secretin; both these peptides activate cAMP synthesis. Purinergic transmitters could also be involved.

Finally, recent findings suggest that, in humans, four different mRNAs for TH are produced from a single gene. The translation products of these mRNAs differ in their amino-acid sequence in the N-terminal domain, rather than the catalytic C-terminus, and are likely to differ in their kinetics and susceptibility to different protein kinases. Regional differences in the distribution of these enzyme isoforms suggest that they might differ functionally, a possibility that is being explored currently.


The first reports that TH activity could be altered without changes in its kinetics came from studies of the adrenal medulla of rats in which catecholamine release was stimulated by exposure of rats to a cold environment. The increase in enzyme activity was prevented by protein synthesis inhibitors, suggesting that it was due to an increase in TH gene transcription rather than activation of existing enzyme. Since then, physiological and pharmacological stimuli that increase demand on the transmitter store have consistently been shown to trigger such induction of TH enzyme. Increased TH protein has also been detected in noradrenergic cell bodies of sympathetic ganglia and the locus coeruleus. At all these sites, as in the adrenal medulla, the increase is evident after about 24 h. However, changes in the terminals take several days to appear, presumably because of the time required for axoplasmic transport of the enzyme.

The signal for increased synthesis of TH protein in the adrenal gland certainly depends on an intact cholinergic innervation. Moreover, in the denervated gland, the increase induced by perfusion with exogenous acetylcholine is prevented by nicotinic antagonists. However, nicotinic antagonists do not completely prevent the increase in glands with an intact cholinergic innervation. These findings suggest that activation of nicotinic receptors by ACh is normally only partly responsible for the increase. Other factors now known to regulate TH gene transcription include glucocorticoids and nerve growth factor (NGF). Although details are far from clear, protein kinases (especially PKA), diacyl glycerol and Ca2+ are all thought to be crucial intracellular messengers for increased gene transcription. It should also be borne in mind that enzyme induction is not limited to TH: the same stimuli also increase DpH synthesis but less is known about factors mediating this process.


In common with other classical transmitters, noradrenaline is stored in vesicles that accumulate in the terminal varicosities. This was first shown by experiments that combined sucrose density-gradient centrifugation of tissue homogenates (see Fig. 4.3) with electron microscopy and assay of the noradrenaline content of the different layers of the gradient. These studies confirmed that the noradrenaline-rich layers of the gradient coincided with those layers in which the vesicles were clustered. This suggested that the vesicles were the major storage site for noradrenaline within the nerve terminals. Further studies examined the effects of ligation or cooling the axons of sympathetic neurons for several days. Electron micrographs of the zone around the obstruction showed that the vesicles accumulated on the side nearest the cell body, confirming that they were assembled in the cell body and transported to the terminals by anterograde axoplasmic transport. The life cycle of these vesicles was discussed in more detail in Chapter 4.

The concentration of noradrenaline in the vesicles is thought to be in the region of 0.1-0.2 M and it is estimated that there is a concentration gradient, in the order of 104106-fold, driving the transmitter out of the vesicles towards the cytoplasm. The vesicular compartmentalisation of noradrenaline is made possible by its active uptake on vesicular monoamine transporters (VMATs) and its subsequent binding, in an osmotically inert matrix, within the vesicles. One obvious function of these transporters is thus to protect and conserve the releasable vesicular pool of transmitter. However, it is thought that they also protect neurons from potentially toxic effects of an excess of cytoplasmic noradrenaline and also maintain a concentration gradient favouring noradrenaline reuptake from the synapse (see below).

Uptake of noradrenaline into the vesicles depends on an electrochemical gradient driven by an excess of protons inside the vesicle core. This gradient is maintained by an ATP-dependent vesicular H+-triphosphatase. Uptake of one molecule of noradrenaline into the vesicle by the transporter is balanced by the counter-transport of two H+ ions (reviewed by Schuldiner 1998). It is thought that either binding or translocation of one H+ ion increases the affinity of the transporter for noradrenaline and that binding of the second H+ actually triggers its translocation.

Reserpine irreversibly inhibits the triphosphatase that maintains the proton gradient and so it depletes neurons of their vesicular store of transmitter. This explains why restoration of normal neuronal function rests on delivery of new vesicles from the cell bodies. Some amphetamine derivatives, including methylenedioxymethamphetamine (MDMA), are also substrates for the transporter and, as a result, competitively inhibit noradrenaline uptake. Another way of inhibiting the transporter is by dissipation of the pH gradient across the vesicular membrane: ^-chloroamphetamine is thought to act in this way.

Much of the early work on these transporters was carried out on the chromaffin granules of the bovine adrenal medulla. These studies revealed the transporter to be a polypeptide of 80kDa. However, two VMATs have now been characterised and these are the products of different genes. Evidence suggests that both have 12

Figure 8.6 Schematic diagram of the proposed structure of the vesicular monoamine transporter. There are 12 transmembrane segments with both the N- and C-termini projecting towards the neuronal cytosol. (Based on Schuldiner 1998)

Neuronal cytosol

Figure 8.6 Schematic diagram of the proposed structure of the vesicular monoamine transporter. There are 12 transmembrane segments with both the N- and C-termini projecting towards the neuronal cytosol. (Based on Schuldiner 1998)

transmembrane-spanning domains (TMDs) with a large hydrophobic loop, facing the vesicular core, between TMD1 and TMD2. Both the N- and C-termini project towards the neuronal cytosol (Fig. 8.6). There are species differences, but VMAT1 and VMAT2 differ in their distribution. In fact, the expression of these proteins in individual cells might be mutually exclusive. They also differ in their sensitivity to the reversible uptake inhibitor, tetrabenazine, and their affinity for substrates such as amphetamine and histamine. Only VMAT2 binds histamine and tetrabenazine and this protein consistently binds amines with a higher affinity than does VMAT1. In the rat, VMAT1 is found in non-neuronal tissue, including the adrenal medulla, whereas VMAT2 is found in neurons, only. In other species, the distribution is not so distinct, with mRNA for VMAT2 being reported in the adrenal medulla as well as the brain.


Studies of release of noradrenaline from sympathetic neurons provided the first convincing evidence that impulse (Ca2+)-dependent release of any transmitter depended on vesicular exocytosis. Landmark studies carried out in the 1960s, using the perfused cat spleen preparation, showed that stimulation of the splenic nerve not only led to the detection of noradrenaline in the effluent perfusate but the vesicular enzyme, DpH, was also present. As mentioned above, this enzyme is found only within the noradrenaline storage vesicles and so its appearance along with noradrenaline indicated that both these factors were released from the vesicles. By contrast, there was no sign in the perfusate of any lactate dehydrogenase, an enzyme that is found only in the cell cytosol. The processes by which neuronal excitation increases transmitter release were described in Chapter 4.

While the amount of noradrenaline released from the terminals can be increased by nerve stimulation, it can be increased much more by drugs, like phenoxybenzamine, which block presynaptic a-adrenoceptors. These receptors are normally activated by increased noradrenaline in the synapse and trigger a feedback cascade, mediated by second messengers, which blunts further release of noradrenaline. These presynaptic autoreceptors play an important part in ensuring that transmitter stores are conserved and preventing excessive stimulation of the postsynaptic cells.

Pharmacological characterisation of this receptor revealed that it was unlike classic a-adrenoceptors found on smooth muscle. In particular, receptors modulating noradrenaline release have a higher affinity for the agonist, clonidine, and the antagonist, yohimbine. This distinctive pharmacology led to the subdivision of a-adrenoceptors into the a1- and the a2-subtypes. Although the latter is the subtype responsible for feedback inhibition of noradrenaline release, the majority of a2-adrenoceptors are actually found postsynaptically in some brain regions. There is still some debate over the identity of the subtype of a2-adrenoceptors responsible for feedback inhibition of transmitter release. However, most studies agree that the a2A/D-subtype has the major role, although the a2B- and a2C-subtypes might contribute to this action. Species differences in the relative contributions of these different receptors are also possible.

It is a2A-adrenoceptors that are found on cell bodies of noradrenergic neurons in the locus coeruleus. These receptors are activated by noradrenaline released from branches ('recurrent collaterals') of noradrenergic neurons projecting from the locus coeruleus and inhibit neuronal firing (Cederbaum and Aghajanian 1976). a2-Adrenoceptors in the brain thus depress noradrenaline release through two distinct processes: inhibition of the release process following activation of terminal autoreceptors and depression of neuronal firing following activation of receptors on the cell bodies.

The exact process(es) by which a2-adrenoceptors blunt release of transmitter from the terminals is still controversial but a reduction in the synthesis of the second messenger, cAMP, contributes to this process. a2-Adrenoceptors are negatively coupled to adenylyl cyclase, through a Pertussis toxin-sensitive Gi-like protein, and so their activation will reduce the cAMP production which is vital for several stages of the chemical cascade that culminates in vesicular exocytosis (see Chapter 4). The reduction in cAMP also indirectly reduces Ca2+ influx into the terminal and increases K+ conductance, thereby reducing neuronal excitability (reviewed by Starke 1987). Whichever of these releasecontrolling processes predominates is uncertain but it is likely that their relative importance depends on the type (or location) of the neuron.

a2-Adrenoceptors are not the only receptors to modulate noradrenaline release. In the periphery and CNS (Murugaiah and O'Donnell 1995) activation of presynaptic ^-adrenoceptors has the opposite effect: i.e. it augments release of noradrenaline. The increase in cAMP production resulting from activation of these receptors is an obvious explanation for how this might occur. The precise role of these receptors in regulation of noradrenaline release in vivo is uncertain because noradrenaline has a relatively low affinity for these receptors. However, one suggestion is that, in the periphery, they are preferentially activated by circulating adrenaline which has a relatively high affinity for these receptors. This activation could enable circulating adrenaline to augment neuronal release of noradrenaline and thereby effect a functional link between these different elements of the sympathoadrenal system. However, the extent to which this actually happens is uncertain as is a physiological role for ^-adrenoceptors in regulation of nor-adrenaline release in the brain.

Noradrenaline release might also be modulated by receptors on noradrenergic nerve terminals that are activated by other neurotransmitters ('heteroceptors'). Unfortunately, most studies of this type of modulation have been carried out in tissue slices and so it is not possible to rule out the possibility that 'heteroceptors' are in fact part of a polysynaptic loop and that they influence noradrenaline release only indirectly. Nevertheless, there is some evidence from studies of hippocampal synaptosomes that activation of muscarinic, GABAb or adenosine (A1) receptors depresses noradrenaline release while activation of GABAA receptors increases it.

A further possible mechanism, that would enable different types of neurons to modify noradrenaline release, is suggested by recent in vitro studies of brain slices. These have revealed that noradrenaline release is increased when the slices are superfused with a solution containing GABA. This release is prevented by an inhibitor of GABA uptake but unaffected by the presence of GABAA receptor antagonists, such as bicuculline. There is no doubt that this form of release depends on vesicular exocytosis because it is Ca2+-dependent, sensitive to tetrodotoxin and, like impulse-dependent release, it is attenuated by a2-adrenoceptor agonists (see above). Since uptake of GABA by GABA transporters on noradrenergic nerve terminals ('heterocarriers') involves co-transport of Na+ ions into the terminal (Fassio et al. 1996) it is possible that this uptake increases Na+ influx enough to depolarise the terminals and trigger exocytotic release of noradrenaline. The extent to which this process occurs under normal physiological conditions in vivo remains to be seen.


In common with other monoamines, the actions of released noradrenaline are terminated by its rapid reuptake from the synaptic cleft. This uptake process relies on membrane-bound noradrenaline transporters which are glycoproteins closely related ionic dependence, affinity for inhibited 8 substrate extrac&ltular extrac&ltular affinity for substrate. Vmax, stereoselectivity uptake inhibitors, translocation

Figure 8.7 Schematic diagram of the proposed structure of the noradrenaline neuronal transporter showing the 12 transmembrane, hydrophobic domains with the N- and C-termini projecting towards the cell cytoplasm. Binding domains for specific ligands are thought to be within regions indicated by the solid bars. (From Stanford 1999, reproduced with permission)


Figure 8.7 Schematic diagram of the proposed structure of the noradrenaline neuronal transporter showing the 12 transmembrane, hydrophobic domains with the N- and C-termini projecting towards the cell cytoplasm. Binding domains for specific ligands are thought to be within regions indicated by the solid bars. (From Stanford 1999, reproduced with permission)

affinity for substrate. Vmax, stereoselectivity uptake inhibitors, translocation to the transporters for several other transmitters (e.g. GABA and 5-HT). All these transporters have 12 hydrophobic transmembrane domains (TMDs), a large hydrophyllic loop between TM3 and TM4, and intracellular N- and C-termini. The hypothetical structure of the noradrenaline transporter is illustrated in Fig. 8.7. Because co-transport of both Cl_ and Na+ is required for the uptake of noradrenaline, this is regarded as one of the family of Na+/Cl~ transporters.

Exactly how this transporter carries noradrenaline across the neuronal membrane is not known but one popular model proposes that it can exist in two interchangeable states. Binding of Na+ and noradrenaline to a domain on its extracellular surface could trigger a conformation change that results in the sequential opening of outer and inner channel 'gates' on the transporter. This process enables the translocation of noradrenaline from the extracellular space towards the neuronal cytosol.

So far, only one noradrenaline transporter has been cloned. Point-mutation and splicing studies indicate that different zones of the transporter determine its substrate affinity and selectivity, ionic dependence, Fmax, and the binding site for uptake inhibitors such as desipramine (Povlock and Amara 1997). Because the cloned transporter is a target for the reuptake inhibitor, desipramine, it is thought to reflect the native transporter in the brain and peripheral tissues. However, in the periphery, two native reuptake processes (neuronal uptake, 'uptake1' and extraneuronal uptake, 'uptake2') have been recognised for over 30 years and recently, a third, desipramine-insensitive uptake site has been found in hepatocytes. These are quite distinct uptake mechanisms because they have different substrate affinities and antagonist sensitivities. As yet, few studies have investigated the possibility that more than one uptake process exists in the brain but since two mRNAs for noradrenaline transporters have been isolated from brain tissue (Pacholczyk, Blakely and Amara 1991) there could be more than one transcription factor. Also, the so-called 'extraneuronal transporter' for noradrenaline, responsible for 'uptake 2', has recently been found on glial cells in the brain (Russ et a/. 1996). At the very least, intracellular messengers could modify substrate affinity of the transporter, by causing its phosphorylation or glycosylation (Bonisch, Hammermann and Bruss 1998), and so markedly affect its function. Whether or not there are different gene products, splice variants, or posttranslational changes, it has been suggested that abnormal distributions of functionally distinctive noradrena-line transporters could underlie some psychiatric and neurological disorders.


After reuptake into the cytosol, some noradrenaline may be taken up into the storage vesicles by the vesicular transporter and stored in the vesicles for subsequent release (see above). However, it is thought that the majority is broken down within the cytosol of the nerve terminal by monoamine oxidase (MAO; EC1.4.3.4). A second degradative enzyme, catechol-O-methyl transferase (COMT; EC2.1.1.6), is found mostly in nonneuronal tissues, such as smooth muscle, endothelial cells or glia. The metabolic pathway for noradrenaline follows a complex sequence of alternatives because the metabolic product of each of these enzymes can act as a substrate for the other (Fig 8.8). This could enable one of these enzymes to compensate for a deficiency in the other to some extent.

MAO is bound to the outer membrane of mitochondria and is responsible for the oxidative deamination of noradrenaline. There are two isoforms of this enzyme, MAO-A

Metabolism Noradrenaline

Figure 8.8 The metabolic pathway(s) for noradrenaline. MAO is responsible for the oxidative deamination of noradrenaline derivatives while COMT O-methylates noradrenaline. Most intraneuronal metabolism involves MAO while COMT is mainly found extraneuronally. However, both these enzymes can act on each other's products, yielding a complex cocktail of metabolites. The reasons for this complex network of metabolites are not known

Figure 8.8 The metabolic pathway(s) for noradrenaline. MAO is responsible for the oxidative deamination of noradrenaline derivatives while COMT O-methylates noradrenaline. Most intraneuronal metabolism involves MAO while COMT is mainly found extraneuronally. However, both these enzymes can act on each other's products, yielding a complex cocktail of metabolites. The reasons for this complex network of metabolites are not known and MAO-B, which hybridise to different cDNAs and are encoded by different genes on the X chromosome. MAO-A is the more important in vivo because it preferentially metabolises noradrenaline. However, in vitro, MAO-B will metabolise noradrenaline at high substrate concentrations. MAO probably also has an important role in development: a genetic deficiency of MAO-A causes some mental retardation and a tendency to bouts of aggression. MAO-B deficiency has no overt effects in the phenotype but a deficiency of both enzymes causes severe mental retardation and behavioural problems (Lenders et al. 1996). Of course, some of these abnormalities could be due to disruption of the metabolism of other monoamines, such as tyramine, which are also substrates for MAO.

Certainly, such a complex system for metabolism of noradrenaline (which is shared with the other catecholamines) strongly suggests that its function extends beyond that of merely destroying transmitter sequestered from the synapse. However, as yet, little is known about the regulation of this pathway and any influence it might have on noradrenergic transmission. One crucial, additional role for MAO appears to be the regulation of the intraneuronal stores of noradrenaline. Its predominantly intraneuronal location would suggest that its primary function is to ensure that there is always a low concentration of cytoplasmic noradrenaline. What can happen when the concentration of cytosplasmic noradrenaline is increased is illustrated by amphetamine. This drug causes a rise in the cytoplasmic noradrenaline and results in increased binding of this transmitter to the cytoplasmic side of the transporter which then carries it out of the neuron. Importantly, this form of noradrenaline release ('retrotransport') is independent of neuronal activation or intracellular Ca2+.

By maintaining low concentrations of cytoplasmic noradrenaline, MAO will also regulate the vesicular (releasable) pool of transmitter. When this enzyme is inhibited, the amount of noradrenaline held in the vesicles is greatly increased and there is an increase in transmitter release. It is this action which is thought to underlie the therapeutic effects of an important group of antidepressant drugs, the MAO inhibitors (MAOIs) which are discussed in Chapter 20.

Because of their lack of selectivity and their irreversible inhibition of MAO, the first MAOIs to be developed presented a high risk of adverse interactions with dietary tyramine (see Chapter 20). However, more recently, drugs which are selective for and, more importantly, reversible inhibitors of MAO-A (RIMAs) have been developed (e.g. moclobemide). These drugs are proving to be highly effective antidepressants which avoid the need for a tyramine-free diet.

Further interest in MAO has been aroused as a result of recent research on drugs with an imidazole or imidazoline nucleus (Fig. 8.9). Although many of these compounds are potent and selective a2-adrenoceptor ligands (e.g. the agonist, clonidine, or the antagonist, idazoxan), not all the binding of these compounds is explained by their high affinity for ^-adrenoceptors. It is now known that many of these drugs have their own binding sites that are now classified as imidazoline (I-) receptors. One of these, the so-called I2-receptor, has been found on MAO-B but there is general agreement that the I2-receptor is not the same as the catalytic site on the MAO enzyme. Instead, it is thought that the I2-receptor is an allosteric modulator of the catalytic site on MAO which, when activated, reduces enzyme activity. So far, the function of this



Atipamezole Cimetidine Histamine Medetomidine



Monoxidine p-Aminoclonidine


Figure 8.9 The chemical structure of imidazole and imidazoline, together with some well-known derivatives receptor is unknown but it has been suggested that a dysfunction of I2-receptors could contribute to neurodegenerative disorders such as dementia and Parkinson's disease.

There is also some evidence for subtypes of COMT but this has not yet been exploited pharmacologically. Certainly, the majority of COMT is found as soluble enzyme in the cell cytosol but a small proportion of neuronal enzyme appears to be membrane bound. The functional distinction between these different sources of COMT is unknown. COMT inhibitors also exist (e.g. pyrogallol), mostly as catechol derivatives, but so far, most have proved to be highly toxic. Only recently have drugs been developed which are selective for COMT; one of these agents, tolcapone, is used currently in treatment of Parkinson's disease (see Chapter 15).

Continue reading here: Noradrenergic Receptors

Was this article helpful?

0 0