Dual action considerations and receptor affinity data

One of the things that my research has clarified is the relationship between the receptor affinities measured in experimental test-tube (in vitro) assays and the actual effect of drugs in humans %911%93. This is a vital consideration for two reasons:

1) Because the potency of many of these drugs varies by several orders of magnitude (see table Receptor Affinity data below)

2) There is not a simple or direct relationship between receptor affinity and effect

So, one key question is, how potent do these drugs have to be before they have significant effects?

I have been using treatments that produce dual SNRI actions for many years and have no doubt that they are of superior effectiveness compared to other currently available drugs that are called ‘SNRIs’. The evidence and opinion supporting that position is getting stronger gradually.

I would also note that the evidence is coming from clinical observation not randomised controlled trials. That makes it the same as almost every other advance in psychiatry over the last five decades, none of which have emanated from the results of controlled trials — think of, chlorpromazine, lithium, MAOIs, sodium valproate, imipramine, clomipramine, clozapine, ketamine, and more.

All from clinical observation — not RCTs.

Dual action strategies possible in the past have been Clomipramine (the superiority of which was strongly supported by the DUAG studies %912-5%93, and since about 1990, combinations like sertraline + nortriptyline.

Fixed Relative Potency

There is one major disadvantage of using a single drug that supposedly has dual action on both serotonin and noradrenaline reuptake, i.e. a so-called SNRI: it is that the ratio of the potency for effecting these two pathways is fixed, and not ideal because most such drugs affect one pathway much more than the other.

Needless to say the same logic applies to the attempted development of ‘triple reuptake inhibitors’ (%916%93) which in my opinion is an extraordinary example of a triumph of hope over experience.

A useful way of picturing this is to think of a four-wheel drive vehicle that puts 90% of the power to the ground through one set of wheels, and only 10% through the other; that is fixed and cannot be varied to suit different conditions. That is clearly a dumb idea and a sub-standard system. No-one would buy a four-wheel-drive vehicle that did not have intelligent control systems that varied the power to each wheel, depending on the traction available. This analogy makes it clear that using two different drugs (to achieve SNRI) allows the dose of each one to be varied according to the characteristics and severity of the illness, in a way which is definitely more flexible, and probably more effective, than a single drug with a fixed ratio of effects.

It is logical to make an initial assumption (in the absence of basic science knowledge that would indicate otherwise) that the potencies for each pathway need to be similar, or there would be an excessive effect on one pathway and an inadequate effect on the other. How similar is similar? Until evidence to the contrary is available it is logical to assume that a ratio of 1:1 is optimal. Reference to the table indicates that the new SNRI drugs do not come near this figure, and that is a consequential departure from the ideal. But, on paper, some of the old TCAs are closer to 1:1 — however, only in the case of clomipramine does that make them a practical proposition for treatment in humans, because of other pharmacological actions and toxicity. But that is only part of the story.

There is substantive evidence (from my research and reviews) that clinical effectiveness, at least for serotonin pathways, requires a drug with low single figure SRI potency, i.e. a Ki of less than 10 nM (21) — indeed, probably closer to 1 or less (CMI is around 0.2), rather than 10. If the same applies to noradrenaline reuptake, which on present evidence would have to be considered a reasonable proposition, then it is evident that the newer drugs are unlikely to be sufficiently potent to produce optimal effects (see table).

If it is not possible to obtain therapeutic levels of clomipramine without unacceptable side-effects, then this suggests an alternative strategy for achieving effective dual action is use two different drugs at the same time (e.g. sertraline + nortriptyline). This strategy is simple and safe (I have used it in many hundreds of cases) providing a few basics about drug interactions are understood.

Tyramine Pressor Response

In the past various methodologies have been applied to attempt to find an answer to this question of the minimum required potency for optimum NRI efficacy. It is possible to test NRIs for their ability to modify the tyramine pressor response. The amine, tyramine, when ingested in larger quantities, raises the blood pressure. The usual methodology is to give increasing doses, either by mouth, or intravenously, until the systolic blood pressure increases by 30 mm of mercury. The dose required to achieve this is referred to as the TYR30. The mechanism of this reaction is via tyramine releasing the body’s own noradrenaline from the sympathetic nerve endings. To do this it must first be taken up into the pre-synaptic nerve ending. This occurs via the noradrenaline transporter %91reuptake inhibition%93.

TCAs inhibit the noradrenaline transporter — but with potencies that vary by a thousandfold. The more potent ones prevent powerfully inhibit the pressor response. There is a great deal of work, over several decades, relating to the capacity of various drugs to do this. It is now possible to correlate that with the latest human cloned receptor data.

The table below contains a range of values for the receptor affinity measurements. This is because it is not a precise process and the in vitro (test tube) assays that produce these measurements are subject to various experimental variations. Therefore, when values are compared between different laboratories employing essentially the same technique there may be a tenfold variation in the affinity data produced. This needs to be taken account of when comparisons are made, and it is frequently not valid to try to make precise comparisons. The data must be interpreted in the light of other data and the physicochemical properties of the drug, so that a best guess can be made as to what the data indicates. When reading different papers attention must be given to whether the data derives from human cloned receptor assays, or some other technique.

HCR data is the benchmark and the data here are HCR data unless otherwise indicated.

Decades of experience with tricyclic antidepressants indicates that those with weak reuptake capacity are ineffective as antidepressants, viz. trimipramine, doxepin etc. However, it may be noted that this hypothesis has never been rigorously tested using up-to-date methodology and rating scales. That should have been done. If that test resulted in a demonstration that weaker NRI potency was less effective than strong, that would in itself be evidence for a role for noradrenaline in promoting antidepressant efficacy.

I emphasize this point particularly because the NRIs, like reboxetine, do not appear to be very effective ‘antidepressants’ %917%93 — but see other discussions relating to the imprecision and inappropriateness of such comparisons. That accords with my experience (e.g. that nortriptyline is effective when added to sertraline, but doxepin is not — except as a sedative). But see also %918%93. I never used reboxetine by itself is an antidepressant because I regarded it as insufficiently effective. I think it did augment the effect of sertraline in combination; but I would note that patients often said they preferred nortriptyline.

Note also that atomoxetine (a selective NRI like reboxetine) was strategically marketed by Lilly as an ADH drug because it ‘failed’ in AD trials (like duloxetine and sibutramine).

Indeed it seems that development of NRIs has ceased, partly because of the unpopularity and perceived low efficacy of reboxetine (despite the usual meta-analyses that find apparent equal effectiveness with SSRIs %918%93).

Various methodologies have been applied to attempt to answer the question of how potent these drugs need to be (in terms of receptor affinities) before they have significant antidepressant actions.

Table Human cloned receptor (HCR) data relating to ‘dual action’

Reuptake inhibitor affinities Ki nM (modified PKG March 2020)

SSRIs (for comparison)>10000.1-20~ 1:1000
AmitriptylineN/AAve~60Ave ~40~1:1.5
ImipramineN/A20 – 1421.3 – 20~1:2
Desipramine+++0.63 – 8.622 -180
Duloxetine**02-50 (~6)0.8-3.7~10:1
SibutramineNo HCR dataNo HCR data 

Table Legend TYR30: tyramine pressor response +++ potent effect (i.e. strong NRI) ++ moderate + weak 0 no significant effect. NA/5-HT: ratio of potency NA vs 5-HT. Note that for tricyclics, such as amitriptyline that have active metabolites (nortriptyline), the TYR30 of the parent compound is irrelevant.

See PDSP database for recent values

What affinity at the NAT constitutes a clinically useful effec? The pressor response to tyramine, the ‘TYR30’ test, provides an in vivo index of peripheral NRI potency. The first direct comparison between the posited SNRI, venlafaxine, and a TCA, desipramine has only recently been done %919%93.

NRIs inhibit tyramine uptake and thus lessen or abolish the response, potent NRIs (reboxetine, desipramine and nortriptyline block this response strongly %9110, 11%93).

The effect of TCAs on the TYR30 is proportional to their relative NRI potency, only affinities of nortriptyline, or greater, are associated with marked attenuation (see table). Duloxetine and venlafaxine have little or no effect on TYR30 even at supra-therapeutic doses. There are no known data for sibutramine or milnacipran. Duloxetine, at 60 mg daily attenuates TYR30 very slightly %9112%93, even at 240 mg per day it is still weak %9113%93. Chalon et al found no effect at 80 mg %9114%93. Venlafaxine only marginally lessens the pressor response at the usual maximum dose of 375 mg daily %9115, 16%93. A complete set of data comparing all the TCAs etc under the same conditions would be revealing, but the present evidence indicates that drugs with an NRI potency less than nortriptyline are sub-optimally effective. Dothiepin (dosulepine), doxepin, mianserin and mirtazapine (6-aza-mianserin) are weak NRIs and have no effect on TYR30 %9117-19%93.

There is no special evidence suggesting the cerebral drug levels are much different between drugs (although that is definitely possible, see e.g. %9120%93), so the provisional supposition that the central effects mirror the TYR30 data seems reasonable.


In my opinion the two telling bits of data are the SRI potency difference between imipramine and clomipramine and the TYR30 difference between nortriptyline and desipramine when compared to duloxetine and venlafaxine. It is well established that clomipramine is the only tricyclic antidepressant (TCA) effective for obsessive compulsive disorder (and cataplexy) and that imipramine and amitriptyline are ineffective. It is unlikely that this is accounted for by anything other than their different SRI potency, e.g. 5-HT2A affinity (see also %911%93). Furthermore, we also know that there is a congruent difference in the ability of amitriptyline & imipramine & clomipramine to precipitate serotonin toxicity. The above receptor affinity data clearly indicate that (at least for TCAs) potencies of less than one (i.e. Ki < 1 nM) are required for good clinical effectiveness. It is therefore reasonable to suggest that in designing a dual action strategy it would be ideal to aim for those kinds of potencies in both pathways. Neither venlafaxine nor duloxetine come anywhere near that. Clomipramine does, as do combinations like sertraline + nortriptyline, or sertraline + reboxetine.

But who is going to advertise or promote such a combination when the two different drugs are made by different companies?

It is appropriate to continue to regard clomipramine as the gold standard SNRI antidepressant reference compound because the newer ‘dual action’ drugs just do not measure up.

However, combinations may achieve the same result with less side effects. In my opinion, and extensive experience (of >1,000 patients), sertraline + nortriptyline probably works a bit better than sertraline + reboxetine (a highly selective NRI), so maybe 5-HT2A receptors, or some other property of TCAs, does make a difference.

It is possible that adding an atypical neuroleptic for the treatment of refractory depression may be emulating the 5-HT2A antagonism that is missing with selective serotonin reuptake inhibitors (SSRIs). Which brings us back ‘full circle’ to clomipramine, as I suggested many years ago %9121%93, in a piece titled after Alphonse Karr, who neatly reiterated a couple of centuries ago ‘Plus ça change, plus c’est la même chose’ %9122%93.

Another little diversion is appropriate. Long ago van Praag suggested (as cited my publication above) that depression could be better considered as a group of psycho-pathological changes, each represented by different neuro-pharmacological alterations and to different aspects of the syndrome. For instance, he specifically suggested that psychomotor retardation was related to dopamine %9123-25%93. I must say, it always seemed to me that his ideas obviously had considerable merit, and I have been surprised how little attention they appear to have attracted (until recently — see my update re DA). Van Praag certainly moulded my thinking on multi-aminergic strategies (i.e. SNRI & DA strategies) decades ago %9121%93.

Indeed, the sensible initial assumption concerning the underlying abnormalities in depressive illness would have to be that it is not a unitary entity, it is not represented by a single discrete abnormality.

It is most unlikely to be effectively treated with a drug that has a specific action only one pathway (like SSRIs).

Comparisons and meta-analysis concerning dual action drugs have major problems. The drugs considered probably do not have true dual action; the severity of illness of the patients considered may not be great enough, e.g. to show differences, either between drugs, or between drugs and placebo %9126%93. Most meta-analysis is flawed anyway (garbage in, garbage out) and there is a ‘pot pourri’ of studies from which to select ones favoured result. It is clear from the fuss concerning the ineffectiveness of the SSRIs, that clear differences between different drugs and between drugs and placebo are only evident when one considers severely depressed patients (so that rules out almost all drug company sponsored trials).

A little history is relevant as always. The first dual action antidepressant to be trialled, at least that I have found information about, was in the early 1980s. A drug that was with us for a while, sibutramine (Reductil etc, google it). The initial depression trials were a failure and were lost; it resurrected itself in the market as a diet drug, then was withdrawn. Most psycho-pharmacologists I have spoken to do not even know about sibutramine.


1. Gillman, P.K., A review of serotonin toxicity data: implications for the mechanisms of antidepressant drug action. Biological Psychiatry, 2006. 59(11): p. 1046-51.

2. DUAG, Citalopram: clinical effect profile in comparison with clomipramine. A controlled multicenter study. Danish University Antidepressant Group. Psychopharmacology, 1986. 90(1): p. 131-8.

3. DUAG, Moclobemide: a reversible MAO-A-inhibitor showing weaker antidepressant effect than clomipramine in a controlled multicenter study. Danish University Antidepressant Group. Journal of Affective Disorders, 1993. 28(2): p. 105-16.

4. DUAG, Paroxetine: a selective serotonin reuptake inhibitor showing better tolerance, but weaker antidepressant effect than clomipramine in a controlled multicenter study. Danish University Antidepressant Group. J Affect Disord, 1990. 18(4): p. 289-99.

5. Vestergaard, P., et al., Therapeutic potentials of recently introduced antidepressants. Danish University Antidepressant Group. Psychopharmacology Series, 1993. 10: p. 190-8.

6. Subbaiah, M.A.M., Triple Reuptake Inhibitors as Potential Therapeutics for Depression and Other Disorders: Design Paradigm and Developmental Challenges. J Med Chem, 2018. 61(6): p. 2133-2165.

7. Eyding, D., et al., Reboxetine for acute treatment of major depression: systematic review and meta-analysis of published and unpublished placebo and selective serotonin reuptake inhibitor controlled trials. BMJ, 2010. 341: p. c4737.

8. Papakostas, G.I., et al., A meta-analysis of clinical trials comparing reboxetine, a norepinephrine reuptake inhibitor, with selective serotonin reuptake inhibitors for the treatment of major depressive disorder. Eur Neuropsychopharmacol, 2008. 18(2): p. 122-7.

9. Blier, P., et al., Effects of different doses of venlafaxine on serotonin and norepinephrine reuptake in healthy volunteers. International Journal of Neuropsychopharmacology, 2007. 10(1): p. 41-50.

10. Dostert, P., et al., Reboxetine prevents the tranylcypromine-induced increase in tyramine levels in rat heart. Journal of Neural Transmission, 1994. 41: p. 149-53.

11. Doggrell, S.A. and G.N. Woodruff, Effects of antidepressant drugs on noradrenaline accumulation and contractile responses in the rat anococcygeus muscle. British Journal of Pharmacology, 1977. 59(3): p. 403-9.

12. Turcotte, J.E., et al., Assessment of the serotonin and norepinephrine reuptake blocking properties of duloxetine in healthy subjects. Neuropsychopharmacology, 2001. 24(5): p. 511-21.

13. Vincent, S., et al., Clinical assessment of norepinephrine transporter blockade through biochemical and pharmacological profiles. Circulation, 2004. 109(25): p. 3202-7.

14. Chalon, S.A., et al., Duloxetine increases serotonin and norepinephrine availability in healthy subjects: a double-blind, controlled study. Neuropsychopharmacology, 2003. 28(9): p. 1685.

15. Harvey, A.T., R.L. Rudolph, and S.H. Preskorn, Evidence of the dual mechanisms of action of venlafaxine. Archives of General Psychiatry, 2000. 57(5): p. 503-9.

16. Debonnel, G., et al., Differential physiological effects of a low dose and high doses of venlafaxine in major depression. International Journal of Neuropsychopharmacology, 2007. 10(1): p. 51-61.

17. Coppen, A., et al., Effect of mianserin hydrochloride on peripheral uptake mechanisms for noradrenaline and 5-hydroxytryptamine in man. British Journal of Clinical Pharmacology, 1978. 5 Suppl 1: p. 13S-17S.

18. Pare, C.M., et al., Will amitriptyline prevent the “cheese” reaction of monoamine-oxidase inhibitors? Lancet, 1982. 2(8291): p. 183-6.

19. Ghose, K., et al., Studies of the interaction of desmethylimipramine with tyramine in man after a single oral dose, and its correlation with plasma concentration. Br J Clin Pharmacol, 1976. 3(2): p. 334-7.

20. Schomburg, R., et al., Doxepin concentrations in plasma and cerebrospinal fluid. J Neural Transm, 2011. 118(4): p. 641-5.

21. Gillman, P.K., Plus ça change, plus c’est la même chose. Pharmabulletin, 1994. 18(1): p. 10-11.

22. Karr, A., ‘Les Guelphes’. 1849.

23. van Praag, H., et al., Monoamines and Abnormal Behaviour. A Multi-Aminergic Perspective. British Journal of Psychiatry, 1990. 157: p. 723-734.

24. van Praag, H.M., Diagnosis, the rate-limiting factor of biological depression research. Neuropsychobiology, 1993. 28(4): p. 197-206.

25. van Praag, H.M., The diagnosis of depression in disorder. Aust N Z J Psychiatry, 1998. 32(6): p. 767-72; discussion 773-7.

26. Carroll, B.J., Bringing back melancholia. Bipolar Disord, 2012. 14(1): p. 1-5.



Other Resources

Share This Page

Consider Donating to PsychoTropical

PsychoTropical is funded solely through generous donations, which has enabled extensive development and improvement of all associated activities. Many people who follow the advice on the website will save enormously on doctors, treatment costs, hospitalization, etc. which in some cases will amount to many thousands of dollars, even tens of thousands — never mind all the reduction in suffering and the resultant destruction of family, work, social, and leisure capability. A donation of $100, or $500, is little compared to those savings. Some less-advantaged people feel that the little they can give is so small it won’t make a difference – but five dollars monthly helps: so, do not think that a little donation is not useful.

– Dr Ken Gillman

Dr Ken Gillman