So far we’ve looked at some human and some molecular research into Salvia divinorum. But a huge chunk of biology research involves animals; here I’ll look at just a small selection of the papers that are using animals to investigate the unique effects of Salvia.

Before I go on, I’d like to make it clear that animals in modern research very rarely suffer any pain or discomfort, and are treated with immense care. Without animal research, our understanding and treatment of devastating diseases like Alzheimer’s would be decades behind where it is now. If you want to learn the facts about modern animal research, go to The research we are looking at today is NOT the kind of research done by legal highs manufacturers before they bring their products to market; this is research that furthers medical science due to the unique and scientifically fascinating properties of Salvia and the importance of dosage.

We’ll look at two papers that use animals to learn about Salvia. Firstly, a paper that uses mice to argue that Salvia is a very potent and dysphoric drug (something that causes negative feelings, the opposite of euphoria). Then we’ll look at a paper that uses zebrafish to challenge those claims. Both of them together tell us some interesting things about Salvia!

Zhang et al (2005)

Previous studies with rodents had shown that kappa-opioid receptor (KOR) agonists caused place aversion; that is, rodents didn’t like returning to the place where they were given the drug. Since Salvinorin A, the main psychedelic compound of Salvia, is a naturally occurring KOR agonist (you can find out more about that in our previous blog post), Zhang et al decided to see if it had similar effects to the previously studied KOR agonists.

(A C57BL/6J mouse, a common experimental mouse strain, as used in Zhang et al (2005). Picture credit The Jackson Laboratory.) 

Their study is in two parts; firstly, the authors wanted to see if dopamine levels changed after Salvinorin A injections. Dopamine is a neurotransmitter involved in learning and reward, and has been implicated in mechanisms of addiction; dopamine levels are greatly increased during cocaine use, for example.

Secondly, the authors looked at Salvinorin A and Conditioned Place Aversion (CPA). If Salvinorin A makes mice not want to return to the place where they received Salvinorin A, it’s called CPA, and it might mean that Salvia is causing very negative experiences in these mice.

Salvinorin A’s effect on Dopamine

The first part of the study, looking at dopamine levels, was achieved by means of a dopamine probe inserted into an area of the mouse brain known for its high dopamine levels (fig. 1). While the mice were free to move around, they were injected with several possible concentrations of Salvinorin A (0, 0.32, 1 or 3.2 mg per kg bodyweight), and their dopamine levels were monitored for up to ten hours after injection. The authors found that dopamine levels were significantly lowered after injection with the two highest doses of Salvinorin A. At the very highest dose, this lowering of dopamine levels remained for ten hours, only returning to normal the next day. This would suggest, at the very least, that Salvinorin A is not addictive through the common mechanism of increasing dopamine levels.

Figure 1: Dopamine probe inserted into the mouse caudate putamen. (A) shows the cannula inserted into the mouse head, allowing a dopamine probe to be inserted into the mouse brain. (B) shows a cross-section of the mouse brain, with the probe entering the caudate putamen, an area with high baseline dopamine levels.

Salvinorin A’s effect on place preference

The second part of the study, dealing with CPA, took place in a specially designed setup (fig. 2). Mice were introduced into the gray compartment, and allowed to roam throughout the other two compartments; one black, one white. The mice were allowed to wander for 30 minutes, and the amount of time they spent in each room was recorded. This ‘baseline preference’ phase is important, since some eccentric mice may have a preference for one room before even being given any drug, and this needs to be accounted for during later analysis. After checking this baseline, the mouse were injected with a certain concentration of Salvinorin A (0, 1 or 3.2 mg/kg bodyweight) in either the black or the white room, and kept in that room for 30 minutes. This ‘conditioning’ was repeated in the same room several times, so the mice would start to associate one room with the effects of the drug. Then the mice were allowed free access to the entire area to see if their preference for the rooms had changed, now that they associate one of the rooms with the drug.. The time they spent in each room was again recorded. If they spent more time in the room they were given the drug, it was called Conditioned Place Preference (CPP, i.e., they liked it). If they avoided their drug room, it’s CPA, and they didn’t like it.

Figure 2: Conditioned Place Preference setup in Zhang et al (2005). Mice were placed in the grey area, and their baseline preference for the black or white rooms recorded (A). Mice were then injected with a drug dose and kept in one room over several sessions (B). Mice were then placed in the grey room again (C) and the time spent in each room recorded.

The authors found that the mice spent less time in the room in which they had previously been given Salvinorin A, compared to mice given no drug (fig. 3). Although the difference is not large (a couple of minutes difference out of 30 minutes of observation), Salvinorin A induces significant CPA. This suggests that at doses of 1 and 3.2 mg/kg Salvinorin A, the drug causes a negative (or at least unusual) experience in mice..

Figure 3: Salvinorin A causes Conditioned Place Aversion in mice. The difference in time mice spent in a room before and after associating that room with a dose of Salvinorin A. Mice spent significantly less time in the room in which they were given Salvinorin A compared to mice given no drug. Adapted from Zhang et al (2005).

Salvinorin A and the KOR

As examined in the previous blog post, Salvinorin A binds strongly to KORs, and this is the main mechanism of its psychedelic action. In this study, the authors wanted to confirm this; so in all the experiments above, they added an extra condition where mice were given Salvinorin A and a KOR inhibitor called nor-BNI. They found that if the nor-BNI was around, Salvinorin A didn’t manage to have any effect; either on dopamine levels or place aversion. This just confirms, in a behavioural setting, our theory that the KOR is how Salvia produces its psychedelic effects.

Author’s conclusions

The authors suggest that the lowering of dopamine caused by Salvinorin A could be part of the reason that mice appear to dislike it. They also suggest that Salvinorin A might have ‘limited therapeutic potential’ because of its apparent negative effects… but let’s have a look at those doses again; 1 and 3.2 mg/kg Salvinorin A? Compare that to the levels of Salvinorin A required to induce a psychedelic effect in humans: between 3 and 7.5 μg/kg. That’s a 1000 fold difference! No wonder the mice are not enjoying their experience; their bodies are being flooded with the world’s most potent natural hallucinogen! So, Braida et al to the rescue…

Braida et al (2007)

Straight away, these guys emphasise the relevance of dose. If we want to make any kind of conclusions about Salvinorin A’s effects in humans, we want to be administering doses covering a wide spectrum of human experience. So Braida et al start with doses much more similar to those effective in humans, but they also look at relatively high doses.

Braida et al use zebrafish as their model animal. You might think zebrafish sound like a bad model of humans, especially compared to mice; but actually, the zebrafish nervous system is very similar to the mammalian one in several ways. In fact, the zebrafish is a good model for the study of the mechanisms of addiction in humans; several successful studies into the addictive properties of cocaine and amphetamine have proven the usefulness of the zebrafish model. 

(A zebrafish as used in Braida et al (2007). Picture from Wikimedia)

Salvinorin A and place preference

In this study, the authors look at place preference in a slightly different set up (fig. 4). The fish are in a tank with two sides; one normal, one with two black dots on the floor, so the fish can tell the sides apart. A perforated barrier allows the fish to move between the two sides. The fish are left to swim around for 15 minutes, and the experimenters measure a baseline preference; which side the fish naturally likes. Overly active, inactive or troublemaker fish* are rejected at this point. The fish are then injected with Salvinorin A (0, 0.2, 0.5, 1, 80 μg/kg) and kept on one side of the tank for 30 minutes. One day later, they’re given free access to the tank for 15 minutes, and the amount of time they spend in each side is recorded.

* Troublemaker fish: “The fish that displayed abnormal behavior in the apparatus, such as deficient or excessive swimming or a baseline preference greater than 70%”

Figure 4: Conditioned Place Preference setup in Braida et al (2007). Zebrafish are introduced to a tank with two sides of different designs, separated by a perforated barrier. Picture from Darland & Dowling (2001).

At doses of 0.2 and 0.5 μg/kg, Salvinorin A actually causes CPP in zebrafish, meaning they spend more time on their drug side… they seem to like it at these low doses (fig. 5). At these doses, Salvinorin A induces a place preference comparable to cocaine, the authors state. However, at the high dose of 80 μg/kg, Salvinorin A produces strong CPA; the fish don’t like these high doses, in agreement with Zhang et al’s findings in mice.

Figure 5: Salvinorin A induces both CPP and CPA depending on dose. The room preference of the zebrafish is compared to their preference after associating one of the rooms with a Salvinorin A dose. Adapted from Braida et al (2007)

Salvinorin A changes swimming behavior

The authors also looked at how the fish changed their swimming behaviour after being given doses of Salvinorin A (0, 0.1, 0.2, 0.5, 1, 5 or 10 μg/kg). They found that fish swam faster and more frenetically than normal after being given low doses, but at the highest two doses they swam more slowly than normal (fig. 6). This again points out the spectrum of Salvinorin A effects; at low doses it seems to be very different to high doses.

Figure 6: Salvinorin A has dose-dependent effects on swimming. Zebrafish were dosed with Salvinorin A and the effect on their swimming was recorded; a low score indicates very slow, listless swimming, while a high score indicates frenetic, fast swimming. Swimming significantly different from controls at 0.1, 0.2, 5 and 10 μg/kg. Adapted from Braida et al (2007).

Salvinorin A and cannabinoid receptors

In a similar way that Zhang et al used a KOR antagonist to show that Salvinorin A works through KORs, Braida et al used a cannabinoid receptor (CB1) antagonist to show that Salvinorin A actually works through cannabinoid receptors too. Without CB1 receptors, Salvinorin A can’t achieve its effects on place preference or swimming. This doesn’t mean that Salvinorin A directly activates cannabinoid receptors; more likely cannabinoid receptors are being activated downstream of a chain of processes occurring after KOR activation. In fact, it’s been shown that dopamine release is directly modulated by CB1 receptors (Lam et al, 2006), so it’s possible that all three are part of a cascade of events after Salvinorin A administration. 

Author’s conclusions

The authors have shown that at lower, more human-relevant concentrations, Salvinorin A can actually have positive effects. The authors suggest that Salvinorin A may have potential to be abused at low concentrations; but so far no documented case of Salvia abuse is known. The authors have confirmed that at high concentrations, just as in Zhang et al, Salvinorin A mimics synthetic KOR agonists by causing CPA (animals don’t like it!). 

Is Salvinorin A still special?

You might be wondering why, if Salvinorin A is just the same as other KOR agonists, are scientists still interested in it? Well in another paper, Wang et al (2005) compared Salvinorin A to several synthetic KOR agonists, and found that Salvinorin A was unique in several ways. Salvinorin A did not cause KORs to ‘internalize’ (be taken back in by the cell, usually in response to them being overly-activated) anywhere near as much as the other KOR agonists. Salvinorin A also caused a much lower down-regulation of KORs (the cell making fewer of the receptors) compared to its synthetic pals. Additionally, whilst synthetic KOR agonists are designed to specifically activate only one receptor, Salvinorin A may activate many other receptors. There are thousands of receptors in the brain, and only a handful have been analysed for Salvinorin A activity; so Salvinorin A is still a special molecule, and still an important avenue of research for neuroscientists. 


Braida D, Limonta V, Pegorini S, Zani A, Guerini-Roxxo C, Gori E & Sala M (2007) Hallucinatory and rewarding effect of Salvinorin A in zebrafish: kappa-opioid and CB1-cannabinoid receptor involvement. Psychopharmacology, 190:441-448

Darland T & Dowling JE (2001) Behavioural screening for cocaine sensitivity in mutagenized zebrafish. PNAS, 98(20):11691-11696

Lam CS, Rastegar S & Strahle U (2006) Distribution of cannabinoid receptor 1 in the CNS of zebrafish. Neuroscience 138:83-95

Wang Y, Tang K, Inan S, Siebert D, Holzgrabe U, Lee DYW, Huang P, Li JG, Cowan A, Liu-Chen LY (2005) Comparison of pharmacological activities of three distinct kappa ligands on kappa-opioid receptors in vitro and their antipruritic and antinociceptive activites in vivo. JPET, 312:220-230

Zhang Y, Butelman ER, Schlussman SD, Ho A & Kreek MJ (2005) Effects of the plant-derived hallucinogen salvinorin A on basal dopamine levels in the caudate putamen and in a conditioned place aversion assay in mice: agonist actions at kappa opioid receptors. Psychopharmacology, 179:551-558