Why do you think that tianeptine "counteracts" its respiratory depression, as opposed to just not having a strong effect on respiration? There are plenty of opioids that do not produce significant respiratory depression. For example, partial agonists like buprenorphine. Agonists that are selective for mu-1 (a MOR subtype produced by post-transcriptional modification) are also believed to produce relatively little respiratory depression.
So little is known about the opioid activity of tianeptine that it is premature to view it as a full agonist that should produce respiratory depression. The effects it produces on opioid receptors would appear to be moderate, at best, based on reports from people who have taken high doses.
You're right, perhaps that's premature. However, given that the study he linked above demonstrates that tianeptine also prevents respiratory depression induced by morphine, it seems fairly logical to put Occam's Razor to work here and assume that this effect would minimize any respiratory depression from tianeptine, and therefore it's impossible to conclude its selectivity for mu subtypes.
However, this article does provide some interesting insight:
http://jpp.krakow.pl/journal/archive/03_02/articles/09_article.html
Particularly this section:
The effect of an acute and repeated treatment with TIA or FLU on the levels of Met-Enk in the studied tissues and plasma of the rat are summarized in Table 1 .
Single administration of TIA resulted in an increase in the level of Met-Enk in pituitary gland (both intermediate and anterior lobes),while in the hypothalamus, adrenals,plasma and striatum,the decrease in the level of Met-Enk was observed.In addition,single administration of FLU induced the decrease in the level of Met-Enk in the hypothalamus,adrenals,plasma and striatum.
On the other hand,both TIA and FLU,administered repeatedly,decreased the level of Met-Enk in striatum and hypothalamus,but in the hippocampus as well as in the neurointermediate lobe of pituitary,the increase in the level of Met-Enk was observed after repeated administration of both TIA or FLU.
In the anterior lobe of pituitary,adrenal glans and plasma,the effects of repeated administration of TIA or FLU were differentiated,i.e.FLU did not induce any significant changes in the level of Met-Enk,while an increase in the level of this peptide in an anterior lobe of pituitary and adrenals,and a decrease in this parameter in plasma,was observed following repeated administration of TIA.
No significant changes were observed in the level of PENK mRNA following acute or repeated administration of TIA or FLU in the rat brain,pituitary or adrenal glands (Table 2 ).
Now this is comparatively ancient research, and predates all the recent talk about tianeptine having a direct mu or delta opioid affinity, but it differentiates it from fluoxetine (which also changes opioid peptide expression) by its ability to increase Met-enkephalin in the pituary after single administration (though it decreases it in adrenals, like fluoxetine), and after repeated administration in both the pituitary and adrenals. This would imply some interesting selectivity to its effects. The authors would have assumed at the time that this effect, like fluoxetine's, is downstream of some regulation of serotonin (they still characterized it as an SRE).
However, we have research that indicates it's affinity for mu and delta:
http://www.nature.com/tp/journal/v4/n7/full/tp201430a.html
And we've also failed to find any direct binding at monoamine or glutamate receptors. It appears to increase or decrease NMDA EPSPs depending on brain region, and it increases the phosphorylation state of AMPA1, at the CAMKII site Ser831 and at the PKA site Ser845. But as far as I know, no binding event has been found to explain any of this. It doesn't bind receptors or enter transporters.
There's obviously something funny about the stuff--and I'm willing to believe that it might bind selectively to non-respiratory depressive mu sites, but given that its interaction with AMPA is obvious--even if the exact point at which it interacts is still a mystery--and given that AMPAkines are known respiratory stimulants, and finally given that tianeptine is a demonstrated respiratory stimulant, and meanwhile that other selective/partial agonist opioids generally aren't...
Are you suggesting that tianeptine's NMDA/AMPA modulating effects are purely mediated by its activation of specific MOR?
I've read about the extensive cross-talk between opioid and AMPA and NMDA, like here:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4034717/
And while fascinating, it seems contradictory to the the hypothesis that an opioid agonist could directly cause respiratory stimulation. If some MOR are ultimately stimulating via AMPA cross-talk, but other MOR (and DOR, which are often heterodimerized with mOR) are depressive, then how do you suggest that tianeptine, with its relatively unimpressive binding values, could displace a stronger, less selective, opiate like morphine, which is blanketing receptors?