Salvinorin A: a unique painkiller?

Salvinorin A: a unique painkiller?

In my previous blog posts I’ve talked about various potential applications of Salvia, such as an antidepressant or a treatment for addiction. Another large area of research into this unique plant is focussed on investigating the analgesic (painkilling) properties of Salvinorin A, the main psychedelic constituent of Salvia.

A common issue with modern painkillers is the fact that they are often very addictive. Many effective painkillers, such as morphine or codeine, are mu-opioid receptor (MOR) activators. Addiction can result from improper use of opioids like morphine. An ideal painkiller would relieve pain without causing this addiction; this is where the potential of Salvinorin A comes in.

Salvinorin A activates kappa-opioid receptors (KORs), a receptor in the same family as the MORs. As I’ve discussed previously, KOR activation can lead to the regulation of addiction, and it’s thought that one of the receptor’s natural roles is to try and control addiction by reducing the activity of MORs. However, if MOR activation reduces pain, doesn’t this mean that KOR activation will increase pain? You would think so; however, as a perfect example of how little we understand about neuroscience, KOR activators sometimes have analgesic properties. Ketamine and dynorphin are two such examples.

It’s not as simple as switching from MOR activators to KOR activators, however. KOR activators have their own problems when used as painkillers. They often cause unpleasant feelings, including nausea and confusion. As you may be aware, the potent KOR activator Salvinorin A also induces very strong psychedelic effects! 

Firstly, I’m going to look at a paper that examines the analgesic properties of Salvinorin A, and presents some of the issues that may arise from using Salvinorin A as a painkiller. Then we’ll look at the remaining challenges scientists face before Salvinorin A research can lead to an effective painkilling drug, and how these challenges are already being met.

DISCLAIMER: The typical methods used in modern research to measure levels of pain in animals are extremely humane. Animals are only made to feel small levels of discomfort, rather than genuinely painful stimuli. In this way scientists measure when animals begin to sense pain, at its very lowest levels. Although this isn’t a true measure of pain, since extreme pain will have different properties to low-level discomfort, it keeps the suffering of animals to a minimum.

McCurdy et al (2006)

A note on statistical significance: up until this point in these series of blog posts I’ve avoided using the term ‘significance’ when I’ve been talking about experimental results. I wanted to avoid confusion by getting overly technical/boring,  but there’s a point where results get more confusing if I don’t mention it. For example, why should you believe me when I tell you that one bar on a graph is higher than another, if it doesn’t look much higher at all? Statistical significance means that the authors of a study have performed some mathematical tests on their results, which has told them that differences in their data are unlikely to be just a fluke. If you see a bar on a graph with an asterisk above it, it means it’s significantly different from the other bars. If there is no asterisk, it means the maths says there’s a chance it’s a fluke, and you shouldn’t necessarily believe there’s any difference.

Tail-flick test

McCurdy and colleagues begin their investigation of the anti-nociceptive (nociception is the detection of harmful stimuli; a the physical component of “pain”) properties of Salvinorin A with a tail-flick test. In this test, a small light is shone on the tail of a mouse, and the mouse flicks its tail away when it can feel the heat from the light – this is considered an indicator of nociception. The anti-nociceptive properties of a drug can be assessed by measuring the time it takes for the mouse to remove its tail away from the light. The time it takes for the mouse to remove its tail from the light is considered an indicator of nociception. The light is kept on for a maximum of ten seconds to minimise the chance of causing harm to the animal. To determine the effect of Salvinorin A on nociception, a dose of either 0.5, 1, 2 or 4 mg/kg Salvinorin A was injected into the mouse either 10, 20 or 30 minutes before undergoing the tail-flick test. As a control, animals were injected with a saline solution at the same time-points and given the tail-flick test in the same way. The results show that if Salvinorin A is injected 10 minutes before the test, the mice move their tails away significantly slower than control mice at doses of 1, 2 and 4 mg/kg of Salvinorin A (figure 1). If Salvinorin A is injected 20 minutes before the test, significant differences are only seen at doses of 2 and 4 mg/kg (data not shown). At the 30 minute time-point, no significant differences are found (data not shown). This means that at relatively high doses, and for only a short period of time after administration, Salvinorin A makes mice respond slower to heat. This suggests that Salvinorin A may be acting as an anti-nociceptive for heat stimuli.

Tail-flick response time and Salvinorin A

Figure 1: Tail-flick response time and Salvinorin A.Mice were administered Salvinorin A at various doses ten minutes before having a light shone on their tails. The average time until the mice moved their tails away from the light is shown for each dose of Salvinorin A. Asterisks indicate significance compared to control (Salvinorin A dose of 0 mg/kg) at p<0.05. Adapted from McCurdy et al (2006).

Hotplate test

Another method of testing heat-related nociception is the hotplate test. Exactly as you’d expect, mice are placed on a warm plate, which slowly increases in temperature from 40°C to 52°C (where it is held for a maximum of 45 seconds to avoid causing harm to the animal). The experimenters record the time it takes for the mouse to start licking its paws or jump off the plate, indicating discomfort. In this test, doses of either 0.5, 1 or 2 mg/kg Salvinorin A were administered, always ten minutes before the test. The results show that only at the dose of 1mg/kg does Salvinorin A significantly increase the reaction times of the mice (figure 2). This confirms the previous finding in the tail-flick test that Salvinorin A may reduce the sensation of heat.

Hotplate response time and Salvinorin A

Figure 2: Hotplate response time and Salvinorin A.Mice were administered Salvinorin A at various doses ten minutes before being placed on a warm plate that steadily increased in temperature. The average time taken until the mice respond to the heat is shown for each dose of Salvinorin A. Asterisk indicates significance compared to control (Salvinorin A dose of 0 mg/kg) at p<0.05. Adapted from McCurdy et al (2006).

Abdominal constriction test

The final method the authors use for measuring nociception is this unpleasant sounding one. A weak solution of acetic acid (vinegar) is injected under the abdominal skin of the mouse; this makes the mouse writhe around in discomfort, which experimenters record for 30 minutes. This test may sound cruel, but the solution injected into the mice is so weak that up to 20% of mice don’t respond at all to the injection, and the mice that do writhe around are only experiencing discomfort. In this test, mice were again injected with doses of 0.5, 1 or 2 mg/kg of Salvinorin A, always five minutes before the test. The results show that for the first fifteen minutes of the test, all three Salvinorin A doses significantly reduced the amount of writhing in the mice compared to controls (figure 3).

Abdominal constriction test and Salvinorin A

Figure 3: Abdominal constriction test and Salvinorin A. Mice were administered Salvinorin A at various doses five minutes before being injected with a weak acidic solution under their abdomen. The number of writhing responses between minutes 6-10 of the test is shown for each Salvinorin A dose. Asterisks indicate significance compared to control (Salvinorin A dose of 0 mg/kg) at p<0.05. Adapted from McCurdy et al (2006).


Although these three tests don’t really measure the whole spectrum of different types of pain, they demonstrate that Salvinorin A does indeed reduce responses to pain. However in all three tests, the effects are short-lived, so clearly Salvinorin A would not be an ideal long-term painkiller. Additionally, the pain-killing effects of Salvinorin A seem to be very dose dependent. In both the heat-related pain-killing tests, higher doses had a greater effect. But in the abdominal constriction test, the lower dose of 0.5mg/kg had the greatest pain-killing effect. In the same way that we see dose-dependent effects of Salvinorin A on dopamine levels, the pain-killing properties of Salvinorin A are probably dose-dependent too.

An issue that is not mentioned in McCurdy’s study is that of the subjective effects of Salvinorin A on mouse consciousness. It is possible, knowing that humans experience hallucinations at much lower doses, that the mice are experiencing intense psychedelic effects that could possibly interfere with nociception; for example, if the mice are extremely terrified by the experience, they may resist painful stimuli more than usual. We would hope that the psychedelic effect of Salvinorin A is reduced by injecting it ten minutes before starting the pain tests; however we have no idea how long-lasting the psychedelic effects of Salvinorin A are in rodents. Unfortunately, since the subjective psychedelic effects of Salvinorin A are mostly unknown in rodents, it is difficult to estimate how much of an effect they are having on nociception. This is why a human study of Salvinorin A’s effects on nociception would be an interesting step forward.

Challenges of developing a painkiller from Salvinorin A

Clearly there are still issues surrounding the use of Salvinorin A as a painkiller. The initial problem that becomes apparent from the McCurdy study is that Salvinorin A only has a short duration of action. After fifteen minutes or so, nociception returns to normal. This would not be ideal for a painkiller, especially if we would like it to have applications for the treatment of chronic pain. Another apparent issue with Salvinorin A is the fact that it produces dysphoria in humans; in McCurdy’s study, relatively high doses of Salvinorin A were required to reduce heat nociception, and as we know from previous studies on mice, these high doses often lead to aversion. If high doses are required for pain-killing, it’s likely that Salvinorin A will also cause dysphoria, along with the typical psychedelic and derealisation effects associated with human consumption of Salvia.

The way of getting around these problems almost certainly involves changing the structure of Salvinorin A. Kivell & Prisinzano (2010) suggest producing a form of Salvinorin A that only works in the peripheral nervous system, meaning it induces no dysphoria; however, experiments with peripherally restricted KOR activators have given mixed results. Hooker et al (2009) suggest creating versions of Salvinorin A that are more resistant to metabolism, increasing their duration of action. Teksin et al (2009) present the idea of creating a form of Salvinorin A that can defend itself against our natural drug-resistance, again letting it last longer in the body.

As well as combatting these challenges associated with Salvinorin A, researchers are attempting to use the structure of Salvinorin A to create a drug that activates the MOR, but has no addictive properties. Harding et al (2005) have managed to create a scaffold from Salvinorin A that can activate the MOR, potentially leading to analgesia. Groer et al (2007) have developed an MOR agonist derived from Salvinorin A, called ‘Herkinorin’, that has very promising anti-addictive properties.  Additionally, it has been suggested that KOR activators, like Salvinorin A, could be administered in conjunction with typical MOR activator painkillers in a joint therapy (Wang et al, 2010).

All of these are just a few examples; all over the world, scientists are working with Salvinorin A  as a key to produce a powerful non-addictive painkiller.


Groer CE, Tidgewell K, Moyer RA, Harding WW, Rothman RB, Prisinzano TE & Bohn LM (2007) An opioid agonist that does not induce mu-opioid receptor-arrestin interactions or receptor internalization. Molecular Pharmacology 71(2):549-557

Harding WW, Tidgewell K, Byrd N, Cobb H, Dersch CM, Butelman ER, Rothman RB & Prisinzano TE (2005) Neoclerodane diterpenes as a novel scaffold for mu opioid receptor ligands. J Med Chem 48:4765-4771

Hooker JM, Munro TA, Beguin C, Alexoff D, Shea C, Xu Y & Cohen BM (2009) Salvinorin A and derivatives: protection from metabolism does not prolong short-term, whole-brain residence. Neuropharmacology 57:386-391

Kivell B & Prisinzano TE (2010) Kappa opioids and the modulation of pain. Psychopharmacology 210:109-119.

McCurdy CR, Sufka KJ, Smith GH, Warnick JE & Nieto MJ (2006) Antinociceptive profile of salvinorin A, a structurally unique kappa opioid receptor agonist. Pharm Bio Chem Beh 83:109-113.

Teksin ZS, Lee IJ, Nemieboka NN, Othman AA, Upreti VV, Hassan HE, Syed SS, Prisinzano TE & Eddington ND (2009) Evaluation of the transport, in vitro metabolism and pharmacokinetics of Salvinorin A, a potent hallucinogen. European Journal of Pharmaceutics and Biopharmaceutics 72:471-477

Wang Y, Sun J, Tao Y, Chi Z & Liu J (2010) The role of kappa-opioid receptor activation in mediating antinociception and addiction. Acta Pharmacologica Sinica 31:1065-1070