Figure 20.4 The chemical structure of mianserin and maprotiline antidepressants is mirtazepine, which is an analogue of mianserin but with fewer pronounced side-effects.
SELECTIVE NORADRENALINE REUPTAKE INHIBITORS (NARIs)
Once it was realised that the adverse effects of the TCAs were due to their interactions with transmitter receptors, rather than their inhibition of noradrenaline reuptake, one objective in the development of novel drug treatments was to produce a 'clean' and selective noradrenaline reuptake inhibitor. The first of these was maprotiline, a bridged tricyclic agent (Fig. 20.4) which has a four hundred and fifty-fold selectivity for inhibition of noradrenaline versus 5-HT uptake in vitro. Although it has little antimuscarinic activity, its antidepressant activity is compromised because it is highly sedative, probably because of its appreciable Hi-receptor antagonism, and it is also an ^-adrenoceptor antagonist. Viloxazine, an oxazine derivative of propranolol, is a bicyclic agent which similarly inhibits noradrenaline uptake more than that of 5-HT (hundred-fold selectivity in vitro) but which has little anticholinergic or antihistaminic activity. The latest NARI to be recruited in the clinic (1997) is another bicyclic anti-depressant, reboxetine (Dostert, Benedetti and Poggesi 1997).
SELECTIVE SEROTONIN REUPTAKE INHIBITORS (SSRIs)
An alternative strategy was to develop drugs that are selective inhibitors of 5-HT reuptake but which, because they are chemically unrelated to the TCAs, would be unlikely to share their side-effects. This direction of research was prompted by the finding, in the late 1960s, that imipramine inhibited 5-HT reuptake, as well as that of noradrenaline, and was reinforced by the evidence that the TCA, clomipramine, was a preferential 5-HT reuptake inhibitor. The first selective serotonin reuptake inhibitor, zimelidine, was tested in the clinic in 1971 but, although it proved to be an effective antidepressant, it was subsequently withdrawn because it could apparently induce the serious neurological disorder, Guillain-Barre syndrome. Nevertheless, other SSRIs quickly followed and five agents are currently available in the UK: fluvoxamine, fluoxetine, paroxetine, sertraline and citalopram (Fig. 20.5).
The SSRIs are all chemically unrelated but their benefits and adverse effects are broadly similar. Their efficacy in depression is not superior to that of the TCAs but their side-effects (nausea, agitation, akathisia and sexual dysfunction), although sometimes problematic, are not life-threatening. They are also considerably safer
than the TCAs in overdose, but the excessive activation of serotonergic systems can culminate in the 5-HT syndrome, a life-threatening delirium. All SSRIs have other clinical applications, such as in treatment of bulimia, anxiety disorders (e.g. obsessive compulsive disorder, panic disorder, social phobia) and Seasonal Affective disorder.
Paroxetine is the most potent inhibitor of 5-HT reuptake but, in terms of distinguishing one compound from another, their preferential selectivity for inhibition of 5-HT rather than noradrenaline reuptake is the key criterion. Citalopram is by far the most selective in vitro (1500-3000-fold) and fluoxetine, the most frequently prescribed SSRI in the UK, is the least selective of all these agents (see Stanford 1999). In fact, it is worth questioning whether fluoxetine is a true SSRI at all.
The most reliable estimates of the selectivity of fluoxetine for inhibition of 5-HT, versus noradrenaline reuptake, put this at twenty-fold, with a high K for noradrenaline uptake of between 1 and 10 ^M. However, its active metabolite, norfluoxetine, is an even more effective inhibitor of noradrenaline uptake (Ki: 0.1 p,M). After chronic administration, the concentration of fluoxetine in the plasma of patients is between 0.5 and 1.5 p,M and is thought to be even higher in the brain. Thus, even accounting for pharmacokinetic factors, such as protein binding, the brain concentrations of fluoxetine and norfluoxetine could well be high enough to inhibit noradrenaline reuptake. Similarly, the plasma concentration of citalopram (285 nM) after chronic administration of the recommended therapeutic dose is about a hundred times greater than its Ki for inhibition of 5-HT uptake (1-10 nM), and its corresponding brain concentration is ten-fold greater still. With a Ki for inhibition of noradrenaline uptake of 4 ^M, even this drug, the most selective of all the SSRIs, could still express this inhibition in patients.
It is perhaps not surprising that, even after taking into account pharmacokinetic differences between these drugs, the therapeutic doses of the SSRIs do not parallel their Ki for inhibition of 5-HT reuptake. For instance, citalopram is about a thousand times more selective than fluoxetine for inhibition of 5-HT uptake, and yet their clinically effective doses are similar. In short, not only is their selectivity for the 5-HT transporter in vitro a poor predictor of their efficacy in vivo but it has to be questioned whether any of these compounds actually work by blocking 5-HT uptake alone.
Of course, it has to be borne in mind that there are functional interactions between serotonergic and noradrenergic neurons in the CNS. Indeed, intracerebral microdialysis studies in rats have confirmed that, with the exception of fluvoxamine, all SSRIs increase the concentration of extracellular noradrenaline whether they are given systemically, or by local intracranial infusion. Such an increase could result from activation of 5-HT heteroceptors on noradrenergic neurons. There is plenty of evidence that activation of 5-HT2, and possibly 5-HT3 receptors, in the terminal field increases noradrenaline release. There is also evidence that activation of presynaptic 5-HT1A, and possibly 5-HT2, receptors, increases the activity of noradrenergic neurons in the locus coeruleus. The complex interactions between serotonergic and noradrenergic neurons that could mediate SSRI-induced changes in noradrenaline release are discussed in more detail in Stanford (1999).
Because the SSRIs are derived from different chemical groups, their receptor interactions vary from compound to compound but, apart from paroxetine, none of them shows any appreciable binding to muscarinic receptors, a prime objective of their development. However, compared with other SSRIs, fluoxetine binds with moderately high affinity to human 5-HT2A (Ki: 280 nM) and 5-HT2C receptors (Ki: 55 nM); sertraline is a relatively potent ligand for ^-adrenoceptors, ^-adrenoceptors and D1 receptors and citalopram shows appreciable binding to 5-HT1A, ^-adrenoceptors and H1 receptors (Table 20.6; Stanford 1996). The extent to which any of these receptor interactions affects the efficacy of these compounds is not known.
SEROTONIN AND NORADRENALINE REUPTAKE INHIBITORS (SNRIs)
Over the last ten years or so, the emphasis on selectivity of action has waned. This is because the relative importance of blocking noradrenaline and 5-HT reuptake remains uncertain and it is possible that it could be beneficial to block both. Some drugs that act in this way have already been developed. It is hoped that this approach might increase the response rate of patients who are resistant to more selective drug treatments and even reduce the therapeutic lag that dogs their predecessors. As yet, there is not enough information on these compounds to know whether or not this has turned out to be the case.
One of these compounds, venlafaxine (licensed in the UK in 1996), is regarded as an inhibitor of both 5-HT and noradrenaline reuptake but this is based on its actions in vitro. At low doses in vivo, it is a more potent inhibitor of 5-HT (Ki: 39 nM) than noradrenaline reuptake (Ki: 210 nM). Moreover, its active metabolite, O-demethylven-lafaxine, is a weaker inhibitor of NA reuptake, and has a longer half-life, than its parent compound. However, at high doses, venlafaxine inhibits reuptake of both these mono-amines but has negligible activity at muscarinic, H1-receptors or ^-adrenoceptors and to
Table 20.6 Rank order of affinity for receptor binding of the SSRIs
Muscarinic Hi ffi c2
sert zimel parox > sert citai fluvox > sert > fluox > citai zimel > citai. > fluox ^ parox > fluvox sert fluox fluox fluox > citai zimel > fluox parox sert = citai > fluvox fluox > fluvox ^ citai > parox fluox > citai > fluvox > sert sert ^ citai > fluvox sert > cital^>fluvox sert > fluox > parox zimel > fluvox parox ^ sert »fluvox fluox > parox
cital = citalopram; fluox = fluoxetine; fluvox = fluvoxamine; parox = paroxetine; serf = sertraline; zimel = zimelidine. Sequences derived from within study A'; s or A s so lacks the problematic side-effects of the TCAs. Milnacipran is another agent in this group, and has only a two-fold preference for noradrenaline versus 5-HT reuptake inhibition. Finally, another SNRI, sibutramine, has been found to induce weight loss, for reasons that are not fully understood, and it is licensed for use as an anti-obesity agent rather than as an antidepressant.
A final group of antidepressants targets both uptake and release of monoamines. These are triazolopyridine derivatives and include trazodone and the more recent addition, nefazodone. Trazodone is a weak inhibitor of 5-HT uptake but shows appreciable binding to 5-HTia receptors, ^-adrenoceptors and Hi-receptors and so shares some of the disadvantages of the TCAs. It is also a 5-HT2A/2C receptor antagonist and an a2-adrenoceptor antagonist, an action that is thought to contribute to its antidepressant actions. A related compound that has recently been introduced into the clinic is nefazodone. This is another weak 5-HT reuptake inhibitor with 5-HT2A antagonist effects but it also inhibits uptake of noradrenaline to some extent. It has a lower affinity for the receptors that are responsible for the unwanted side-effects of trazodone, in particular ^-adrenoceptors and muscarinic receptors.
Both these compounds have several metabolites and one of these, albeit constituting only 1% of the total, is m-chlorophenylpiperazine (mCPP). This is a 5-HT2C-receptor agonist/5-HT2A antagonist and has been suggested to contribute to the antidepressant effects of these compounds. In fact, 5-HT2C receptor agonists are currently being explored as potential antidepressants. This is interesting because mCPP induces anxiety in humans (Rotzinger et al. 1999) and trazodone is contraindicated in the treatment of patients experiencing depression with panic attacks. The enzyme responsible for this metabolic product, CYP2D6, shows genetic polymorphism and so it is possible that the accumulation of mCPP is more problematic in some individuals than others.
Ultimately, agonist drugs that directly activate monoamine receptors would appear to be a logical development in this field. Unfortunately, the peripheral side-effects of such compounds could well limit their acceptability even if we were to discover what subset of receptors to target.
Was this article helpful?