For most of the history of acupuncture research, scientists have had to infer what's happening inside the body indirectly — measuring a cytokine here, a hormone level there, days after treatment. A study published in Nature Biotechnology on 15 July 2026 does something different: it watches immune cells behave, live, at subcellular resolution, and records what happens to them when a classic acupuncture point is stimulated.
The finding that's generating the most interest among practitioners: electroacupuncture at ST36 (Zusanli) measurably calmed a runaway inflammatory response — and the researchers could see it happening, cell by cell, in real time.
A new way to watch the immune system at work
The team built a new imaging system called RUSH3D-HR — a microscope capable of tracking thousands of individual immune cells simultaneously across a wide field of view, at a resolution fine enough to see subcellular structures, without the tissue damage that usually limits how long this kind of imaging can run. In practice, that meant they could follow the same population of cells continuously for hours inside a living animal, rather than taking a single snapshot.
Using this tool, the researchers induced acute systemic inflammation in mice and watched neutrophils — a type of white blood cell that's usually first on the scene during an immune response — behave in the spleen. What they saw was a pattern nobody had directly observed in a living spleen before: neutrophils would spontaneously gather into dense, temporary clusters, or "swarms," lasting anywhere from 20 minutes to 3 hours, before dispersing again. It's a known behaviour in cell cultures; this is thought to be among the first times it's been captured happening inside a living organ.
What happened at ST36
The researchers then applied electroacupuncture bilaterally at ST36 — a point long used in Chinese medicine for its systemic and digestive effects — using standard experimental parameters (0.5 mA, 10 Hz, 20 minutes). Compared with untreated inflamed mice, the number of swarming events fell by more than fivefold. Needling the same point without electrical stimulation did not produce the same effect.
Because the imaging is direct and continuous, this isn't a study measuring a downstream marker and inferring a mechanism — it's a study watching the immune cells themselves change behaviour in response to stimulation at a specific point. That's a genuinely novel piece of evidence, and worth taking seriously.
Where this sits alongside existing research — and what's still speculative
It's tempting to read a striking new finding like this as confirmation of everything acupuncture is thought to do for the immune system. It's worth being precise about what is and isn't established here, because the surrounding evidence base is a mix of very different levels of certainty.
Reasonably well-supported, from animal and mechanistic research:
Electrical stimulation at ST36 has, across multiple independent studies, been shown to influence measurable immune parameters in animal models — from neutrophil behaviour (this study) to natural-killer-cell activity and antibody responses in earlier work.
There's a plausible, partially mapped neural pathway for at least some of these effects: a 2004 study found that the natural-killer-cell effect of moxibustion at ST36 disappeared when the sympathetic nervous system was blocked, pointing to a nerve-mediated mechanism rather than a purely local one — consistent with more recent neuroanatomical work on ST36 and the vagal–adrenal axis.
Genuinely interesting but still early-stage:
Moxibustion studies linking ST36 stimulation to specific molecular pathways (e.g. JAK2/STAT3 signalling) or to markers like PD-1/PD-L1 in tumour models are promising mechanistic leads, not established clinical effects. These are rodent studies with specific disease models, not evidence about how moxibustion performs in human patients with those conditions.
Older literature (1980s) on moxibustion enhancing antibody responses to T-cell-dependent antigens is consistent with the newer findings but is decades old and, again, animal-based.
What none of this shows:
These are all animal studies. None of the research discussed here is a human clinical trial, and none of it demonstrates that acupuncture or moxibustion treats, cures, or prevents any specific disease in people.
Why this still matters
Mechanistic evidence like this closes a real gap. It's one thing to observe that patients feel different after treatment; it's another to see, directly, that a specific and reproducible cellular process changes at a specific acupuncture point under controlled conditions. This kind of imaging is what allows researchers to move from "we don't know why this might work" toward testable, falsifiable mechanisms — which is exactly the kind of evidence that strengthens acupuncture's standing as a subject of serious scientific inquiry, rather than settling the clinical question on its own.
For patients, the practical takeaway isn't "this proves acupuncture treats condition X." It's that there is now a growing, methodologically serious body of laboratory evidence for how certain acupuncture points may influence the immune system at a cellular level — evidence that continues to build alongside, not instead of, clinical research and clinical experience.
Source: Lu, Z., Wang, M., Chen, W. et al. High numerical aperture confocal volumetric mesoscope reveals mesoscale subcellular dynamics in vivo. Nature Biotechnology (2026). https://doi.org/10.1038/s41587-026-03231-z
This article summarises findings from animal research for general interest and educational purposes. It is not medical advice and should not be taken as a claim that acupuncture or moxibustion treats any specific condition. If you have questions about your own care, please discuss them directly with your practitioner.