For years the cannabis industry has argued about a single question that sounds academic until you are the one in pain: is the whole plant better than one molecule pulled out of it? Marketing has an answer. Science mostly has not. A Brazilian research team has now run the two ideas against each other in the same laboratory, on the same animals, in the same week — and the result is more interesting than either camp would like.
The study comes from the University of São Paulo’s Ribeirão Preto campus and was published in Psychopharmacology on 3 September 2026. You can read the full paper, which is open access, at https://doi.org/10.1007/s00213-026-07156-y. The researchers compared cannabigerol (CBG), a minor cannabinoid isolated and purified, against a standardized full-spectrum cannabis extract containing CBD, a trace of THC and a small terpene fraction. Both were tested in rats with two different kinds of inflammatory pain: the sudden kind, and the kind that grinds on for weeks.
Key Takeaways
Both formulations reduced pain, but on completely different schedules. In the acute test, the full-spectrum extract was the only one that kept working past the first fifteen minutes. In the chronic model, a single dose did nothing at all for either compound — the effects only appeared after ten to fifteen days of daily treatment, and only at the highest dose tested. The full-spectrum extract acted sooner, by day ten; CBG acted later but more completely, returning the animals to normal pain sensitivity from day fifteen onward. Both compounds lowered TNF-α, a key inflammatory signalling molecule, in the spinal cord and bloodstream, but neither touched it in the nerve clusters just outside the spine. Neither slowed the animals down. And every one of these findings comes from male rats, with roughly five to six animals per group, which is exactly as far as the evidence currently goes.
Two Very Different Bets on How Cannabis Should Work
CBG is often called the “mother cannabinoid” because the plant builds CBD and THC from it. It is not intoxicating, it appears in most cultivars in tiny amounts, and it has been studied far less than its famous descendants. The commercial pitch for CBG is purity: one known molecule, one known dose, no surprises.
The full-spectrum extract represents the opposite bet. The one used here was a pharmaceutical-grade whole-plant preparation made by supercritical CO₂ extraction from a CBD-dominant cultivar, standardized to 9.6% CBD, 0.2% THC and 3.8% essential oils, or terpenes. The logic behind it is the so-called entourage effect — the idea that cannabinoids and terpenes together do more than any of them alone. It remains a hypothesis rather than an established mechanism, and it is the kind of claim that gets far more confident treatment in product copy than in journals. If you want the wider context on what CBD does and does not have evidence for, our complete science-based guide to cannabidiol lays out the current state of play.
Forty-eight male Wistar Hannover rats were used across the two experiments. Doses of 1, 3 and 10 mg/kg were injected into the abdominal cavity — a delivery route that bypasses digestion entirely and does not correspond to how anyone takes a cannabis product at home.
The First Hour: A Sixty-Minute Window That Split the Two Compounds
The first experiment used the formalin test, a standard model in pain research. A small amount of irritant is injected into the paw, and the animal’s response unfolds in two distinct waves. Phase I, in the first fifteen minutes, is raw nerve activation — the pain fibres firing directly. Phase II, from fifteen to sixty minutes, is inflammation taking over, with chemical mediators recruiting the immune system into the process.
In Phase I, everything worked. CBG at 3 mg/kg cut pain behaviour by 87.1%. The full-spectrum extract cut it by 93.3% at 3 mg/kg and by 99.2% at 10 mg/kg. On the raw signal, all three treatments were close to complete blockade.
Phase II was where they parted ways. CBG’s effect vanished. Only the full-spectrum extract, and only at the lower 3 mg/kg dose, held on — reducing pain behaviour by 90.2% while inflammation was doing the driving. The higher 10 mg/kg dose did not. That inverted dose response is a recurring pattern with cannabinoids, where more is sometimes less, and it is one of the reasons dosing guidance in this field remains so unsettled.
Three Weeks of Daily Dosing, and Nothing for the First Ten Days
The second experiment is the one that matters more for anyone thinking about long-term pain. The researchers used Complete Freund’s Adjuvant, an injection that produces genuine, persistent inflammatory pain in a single paw lasting weeks.
The first result was a flat failure. Twenty-four hours after the injection, neither compound at any dose made any measurable difference. The animals were just as sensitive as untreated ones.
So the team kept dosing daily for twenty-one days, and the picture changed — slowly. The full-spectrum extract at 10 mg/kg produced a significant improvement in pain thresholds by day ten. CBG at 10 mg/kg took longer, but from day fifteen it restored the animals to their baseline sensitivity, meaning the paw responded to pressure the way it had before the injury. The lower doses, 1 and 3 mg/kg, did nothing at any time point.
Read that sequence again, because it undercuts a lot of consumer expectation. There was no immediate relief. The compounds behaved less like a painkiller and more like a treatment that had to accumulate, in a system that had to change before anything happened.
What the Blood Said, and What the Nerves Refused to Say
To understand why, the researchers measured two immune signalling molecules — TNF-α, which drives inflammation and sensitizes pain neurons, and IL-10, which calms inflammation down — in three places: the bloodstream, the spinal cord, and the dorsal root ganglia, the clusters of sensory nerve cell bodies sitting just outside the spinal column.
Inflammation raised TNF-α in all three. Both treatments brought it back down in the blood and in the spinal cord. In the dorsal root ganglia, neither did. Not at any dose, not after three weeks.
The authors’ explanation is that by week three the inflammation in those nerve clusters has become entrenched, driven by stable changes in how the cells read their own DNA — a more consolidated state that no longer responds to the same signals. In the spinal cord, by contrast, TNF-α comes mostly from support cells that cannabinoids are known to influence. Whatever the reason, it is an honest complication: the animals felt better in a region where the inflammation had not budged.
IL-10 did not change anywhere, which suggests the pain relief did not come from boosting the body’s anti-inflammatory side.
No Sedation, No Slowdown
One practical worry with cannabinoid analgesia is that the animal simply stops moving and looks less pained. The team checked, measuring spontaneous walking and rearing after three weeks. Inflammation had reduced both. Both compounds prevented that decline, with the sole exception of the lowest CBG dose. Movement went up, not down, which points to functional recovery rather than sedation — and importantly, activity never rose above the level of healthy control animals, so neither compound was acting as a stimulant.
What This Study Cannot Tell You
Everything here happened in male rats. The authors say plainly that they do not know whether the findings extend to females — a genuine gap, since previous work from this same field has repeatedly found sex differences in how cannabinoids affect pain. Group sizes were five to six animals. Dosing was by injection into the abdominal cavity, not by mouth. And rat milligrams per kilogram do not translate into human doses by simple arithmetic.
There is also a supply-chain question worth naming. The CBG came from PurMed Global and the extract from Clever Leaves Green Care, and one of the senior authors discloses consulting relationships with both companies, alongside a broader set of pharmaceutical ties and cannabinoid patents. This is disclosed in the paper, as it should be, and it does not invalidate the data. It is simply part of reading it accurately.
Finally, the composition of the extract is doing real work in these results — which means the findings apply to this preparation, at this ratio, and not to full-spectrum products generally. Standardization is precisely what most consumer products lack: when researchers tested 186 CBD oils sold across Europe, roughly half did not match their own labels. And the assumption that adding CBD to THC reliably improves outcomes has itself taken a hit recently, with the first meta-analysis of the CBD–THC interaction finding effects that ran opposite to the popular story.
The Useful Conclusion
What this study actually delivers is not a winner but a shape. Two cannabinoid formulations, tested side by side, produced complementary rather than competing effects: one faster onset, one deeper recovery, both requiring sustained daily treatment before anything happened at all. That pattern is more clinically interesting than a simple victory would have been, because it suggests the choice between isolate and whole extract may be a question of timing rather than superiority.
It is also, still, a rodent study. The authors call for clinical investigation, which is the correct next step and the one that has been in short supply for a very long time.
FAQ
Not in any simple sense. In this study CBG worked more slowly than the full-spectrum extract in chronic pain but produced a more complete return to normal sensitivity once it took effect. In the acute test it was the weaker of the two, losing its effect after the first fifteen minutes. “Stronger” depends entirely on which phase of pain you are asking about.
No. This was a preclinical study in rats, using injected doses and a laboratory pain model. It establishes a rationale for human trials, not evidence that any product works in patients. No cannabis product should be treated as a substitute for medical care or prescribed pain treatment.
That is one of the study’s more striking findings. A single dose had no effect on chronic pain at all, and the researchers had to extend daily treatment to twenty-one days before benefits appeared. It suggests these compounds act by gradually shifting an inflammatory state rather than by blocking pain signals on demand.
Legal status varies enormously between countries and often between regions within a country, and it can hinge on the THC content of the finished product, on whether it is sold as a food, a cosmetic or a medicine, and on how the extract was produced. Anyone with a practical question about a specific product should check current national regulations and take qualified advice locally rather than relying on general summaries.
Disclaimer
This article is journalism about published scientific research and is provided for informational purposes only. It is not medical advice, a diagnosis, a treatment recommendation, or legal advice. The research described was conducted in laboratory animals and its findings cannot be applied directly to humans. Cannabis and cannabinoid products are regulated differently across jurisdictions, and their legal status may change. Readers with medical conditions, including chronic pain, should consult a qualified healthcare professional before making any decision about treatment, and should never discontinue or alter prescribed medication on the basis of a news report.
