Stimulated by Dong et al 2025.[1]
EA – electroacupuncture
GI – gastrointestinal function (but mainly gastric here)
IF – impact factor
DMV – dorsal motor nucleus of the vagus
ChAT – choline O-acetyltransferase
GECI – genetically encoded calcium indicator
GCaMP6s – a fusion protein combining a fluorescent reporter, a calcium binding protein, and a CaM-binding domain
CaM – calmodulin, a calcium binding protein
TRPV1 – transient receptor potential vanilloid 1
Adra2a mRNA – a marker for CGRPγ neurons that exclusively innervate deep tissue
Oxtr – oxytocin receptor
EGG – electrogastrogram
FD – functional dyspepsia– key to acronyms
This is an experimental paper in mice and humans from Shanghai. The last author on this paper was the first author on that famous paper in Nature from 2021,[2] and this paper has a lot of similarities in terms of the experimental approaches used.
The paper is published in the journal Neuron (IF 15), which we have seen before on here, most notably ST36 vs ST25 EA in sepsis from September 2020.[3]
This paper demonstrates location and intensity specific effects of EA on gastric motility in mice and then goes on to confirm the effects of the same stimulus parameters in human subjects. The paper includes numerous individual experiments performed to determine and confirm each step of the mechanistic pathway – it is a fabulous paper, and not just because it uses my favourite point, or pairs of needles placed in the same tissue to deliver EA.
I count 10 different experiments in just Figure 1 of the paper. The paper has 6 more large composite figures in the main report and a further 7 in the supplement.
The experiments start with the application of different intensities of EA via pairs of needles placed deep at ST36 in mice. It sounds straightforward enough, but these mice were modified so that a single variety of neurons from the DMV to the stomach wall (specifically, those that express ChAT) fluoresced when activated. The fluorescence came from a genetically encoded calcium indicator (GECI) called GCaMP6s. So, just to get to this step, the team needed to cross two different genetically modified mouse lines, one that essentially labelled the genome of cells expressing ChAT and the other that used the labelling device to add a genetic sequence encoding the GCaMP6s.
The fluorescence was monitored in vivo from the gastric wall via something called two-photon calcium imaging, which allows detection of fluorescence in deeper layers than standard confocal microscopy.
Stimulation at 2mA activated just over 40% of gastric ChAT enteric neurons, but 0.5mA (that had previously worked in sepsis) activated none of them, as did 0mA (ie needling alone).
I have now summarised just Figure 1A, and the whole of Figure 1 runs from A to N. We could be here all summer if I don’t take some shortcuts!
The team went on to confirm that the stomach wall contraction increased with EA and that stomach emptying was enhanced. They determined that EA did not work in superficial tissues over ST36 or at the sural nerve, or in certain superficial muscles of the lower leg such as gastrocnemius and semitendinosus, but it did work when the needles were placed deep into the soleus muscle as well as when deep in tibialis anterior.
The effect required intact common peroneal and vagus nerves, but not all innervation areas of the common peroneal nerve worked, as illustrated above (superficial tissues were ineffective for EA stimulus). So, the team went on to determine a subset of nociceptive fibres (both Aδ and C fibres) that only derived from deep tissues. These fibres were characterised by co-expression of TRPV1 and Adra2a mRNA (a marker for CGRPγ neurons).
Next the focus was on the DMV and finding a subset of the ChAT+ neurons activated by the EA stimulus (2mA deep at ST36) that mediated gastric reflex activity. The complexity of the methods here is staggering! Anyway, to jump past that, they found a subtype of ChAT+ DMV neurons that also expressed an oxytocin receptor (Oxtr) and these made up over 80% of the ChAT+ DMV neurons stimulated by the TRPV1 neurons deep at ST36.
I have mentioned here before that you can genetically manipulate specific neurons so they become sensitive to photons of a certain wavelength. So, the team did this with the subset of neurons in the DMV expressing both ChAT and Oxtr, and went on to demonstrate that gastric activity could be stimulated by 470nm photons via an optic fibre in the DMV. By adding a diphtheria toxin receptor to this subset of neurons (through more genetic manipulation), they demonstrated that these neurons were essential for ST36 EA-induced gastric contractions.
There is lots I have not yet mentioned, but it is time to jump to the human element of this research to close the loop. In 48 healthy subjects, EGG monitoring demonstrated that only high intensity ST36 EA (2Hz at 3mA) was capable of increasing the amplitude (but not the frequency) of gastric activity. In 40 FD patients, high intensity ST36 EA significantly increased the amplitude of gastric activity compared with a superficial non-penetrating sham EA as well as improving dyspepsia symptoms on 2 different scoring outcomes.
I will show you all the beautiful and complex diagrams and try to give a simple overview of the methods at the webinar on Wednesday evening.
References
1 Dong S, Zhao L, Liu J, et al. Neuroanatomical organization of electroacupuncture in modulating gastric function in mice and humans. Neuron. 2025;S0896-6273(25)00504-5. doi: 10.1016/j.neuron.2025.06.023
2 Liu S, Wang Z, Su Y, et al. A neuroanatomical basis for electroacupuncture to drive the vagal-adrenal axis. Nature. 2021;598:641–5. doi: 10.1038/s41586-021-04001-4
3 Liu S, Wang Z-F, Su Y-S, et al. Somatotopic Organization and Intensity Dependence in Driving Distinct NPY-Expressing Sympathetic Pathways by Electroacupuncture. Neuron. 2020;108:436-450.e7. doi: 10.1016/j.neuron.2020.07.015
The BMAS Blog 
You must be logged in to post a comment.