Transnasal electrical stimulation (TnES) for brain modulation in an in vivo mouse model
DOI: Pending DOI
Synopsis
Transnasal electrical stimulation (TnES) is a recently proposed approach for minimally-invasive deep brain stimulation to overcome the limitations of both conventional deep brain stimulation (DBS) and non-invasive brain stimulation. Previous work showed through simulations and human cadaver studies that TnES could generate focal electric fields reaching some deep brain targets. In this study, we developed a novel platform to validate the brain activation elicited by TnES in an in-vivo mouse model. We placed electrodes inside the nasal pathway below the hypothalamus, and applied electrical stimulation while recording the resulting neural activity. We found that TnES could evoke neural responses in the thalamus and the hypothalamus.
Background
Traditional deep brain stimulation (DBS) is a commonly used treatment for Parkinson’s, essential tremor, and epilepsy, and is an emerging treatment for neuropsychiatric conditions1,2. However, the implantation requires costly surgery and carries infection risk. On the other hand, non-invasive brain stimulation methods, while not requiring any surgery, are largely unable to effectively target deep brain areas3.
An alternative approach, recently proposed by our team, utilizes nasal cavities that are air-filled spaces inside the head, accessible transnasally. Of these spaces, in humans, the sphenoid sinus is close to the pituitary gland, and the olfactory cleft is close to the orbitofrontal cortex. Simulations using a human head model showed that transnasal electrical stimulation (TnES) using electrodes in these spaces, can generate focal electric fields targeting some deep brain regions, notably those linked to mental health treatments (e.g. Brodmann area 25)4,5.
Prior studies placed electrodes in the nasal cavities in patients to restore the sense of smell6,7. Others8 placed electrodes in the nasal space of sheep to stimulate the brainstem, but required breaking the clival bone to activate the brainstem, as electrodes placed on the bone did not modulate the deep brain. In mice, electrical stimulation was applied through the tongue9 or the entrance of the nostril10 to target trigeminal nerves, as well as on the palate11 for wound healing.
Methods
A mouse (C57, 9 mo, Male) was anesthetized by ketamine/xylazine cocktail ((90 mg/10 mg)/kg) injected intraperitoneally, and placed on a stereotaxic apparatus. We inserted a commercial multielectrode stimulation probe (32 Pt/Ir electrodes, 50 μm diameter, 100 μm inter-electrode distance, on a 450 μm diameter shank) into its nose to ~18 mm depth from the nostril.
We drilled a hole in the right hemisphere of the skull (AP -1.8 mm, ML 0.5 mm from Bregma) and inserted a commercial recording probe (64 electrodes at 50 μm pitch) to DV 5.5 mm, such that the distal electrode targeted the hypothalamus. We measured neural responses to TnES at a sampling rate of 30 kHz.
To estimate the position of the stimulation electrodes with respect to the recording probe, sequentially applied biphasic current pulses (amplitude: 30 μA, width: 5 ms) using each nasal electrode as an anode and a common large electrode placed in the mouth as the cathode and measured the resulting electric field. For each nasal electrode, the electric field was estimated as the slope of the peak-to-peak stimulation artifact voltage over the deepest 1 mm of the recording probe. The stimulation probe position was adjusted to ensure that the recording probe was located in the middle of the stimulation electrodes span.
Subsequently, we sequentially applied anodic-first biphasic pulses (5-9 mA amplitude ; 1 ms pulse width) through pairs of nasal electrodes 500 μm apart, sweeping the entire range of the stimulation probe. For one pair of electrodes, we additionally applied currents with varying pulse widths.
Results
We successfully inserted the nasal stimulation probe and placed it below the ventral skull inside the nasal pathway without blocking breathing or causing fatal damage to the mouse. Figure (A) presents neural responses to nasal stimulation with a 2.5 mm inter-electrode distance (5 mA, 1ms/phase, anode-first biphasic square pulse). The brain areas where neural responses were observed were the dorsal thalamus (DT), the ventral thalamus (VT), and the hypothalamus (HT). The deepest part of the probe, where no neural activity was detected, was found to be inside the third ventricle.
To quantify the neural responses to TnES, stimulation pulses were blanked, with an additional 0.6-0.7 ms blanked before and after pulses to remove stimulation artifacts. The data were then bandpass filtered at [200 Hz; 2000 Hz] using a 4th order Butterworth filter. We compared RMS before ([-8 0] ms) and after stimulation ([0 10] ms):
RMS =(RMS_after - RMS_before)/RMS_before
The intensity of the hypothalamic response remains constant over 20 ms post-stimulation, while thalamic responses markedly decrease after ~15 ms. RMS values are 3.1, 4.4, and 1.6 at DT, VT, and HT respectively. RMS is slightly lower at DT than VT because of the larger pre-stimulation activity.
Notably, we required high current to observe neural responses (5-7 mA, 1 ms/phase). In addition, the RMS at HT is smaller than RMS at DT and VT. To confirm if these responses are neural responses rather than artifacts or noise, we compared signals before and after euthanasia. (Figure (B)). Immediately after the euthanasia, responses to stimulation disappeared, confirming the neural origin of the responses that we measured.
We additionally applied stimulation for all electrode pairs whose inter-electrode distance was 500 μm, to study the potential for focal stimulation and steering. We repeated each stimulation 10 times, and excluded outliers with large artifacts. We omitted from further analysis data points with less than four remaining recordings. Figure (C) shows RMS (mean ± std) at the three identified brain regions. x=0 corresponds to the electrode which is the closest to the recording probe (i.e. generating the largest electric fields). Cathodes are positioned 500 μm anterior from the anodes.
Figure (C) highlights stimulation locations where mean RMS exceeds half of the maximum value measured within each of the brain regions (1.06 for DT, 2.52 for VT and 0.54 for HT) at the highest injected current. Six stimulation locations exceeded this threshold in DT, three in VT and six in HT. VT exhibited the largest RMS but the fewest locations above its half-max threshold, potentially pointing to higher spatial selectivity. One location preferentially activated VT, and another location preferentially activated HT. Three locations activated all three brain regions.
Based on the results in Figure (C), we chose the electrode pair eliciting the largest response, whose anode was positioned at -200 μm from the reference electrode (and cathode at +300μm). We applied stimulation pulses with various pulse widths (amplitude: 7 mA) using both anodic-first and cathodic-first bipolar pulses. Figure (D) shows that different pulse widths led to significantly different RMS except for four cases (a two-sided, independent two-sample Student's t-test *: p < 0.05, **: p < 0.01, ***: p < 0.001).
Discussion
In this study, we developed the first in vivo TnES mouse model. We successfully applied electrical stimulation inside the nasal pathway and measured evoked neural responses in the brain. Neural responses were observed in multiple deep brain areas: the dorsal thalamus, the ventral thalamus, and the hypothalamus.
We believe that this is also the first report of hypothalamic neural responses to stimulation from electrodes positioned millimeters away. Unlike the thalamic responses where multiple firing populations (e.g. DT and VT) are observed, the HT responses are smaller and sparser.
TnES holds significant promise for minimally invasive deep brain stimulation. Over the 27 stimulation locations tested (anode spanning [-1000;1600] μm from the recording probe intercept), 3 (VT) and 6 (DT, HT) patterns could evoke RMS larger than half of the maximum RMS. We observed abrupt RMS changes between the adjacent patterns with small (100 μm) changes in electrode distance: e.g. RMS in VT, DT, and HT increased > 2✕ when moving the anode from -300 μm to -200 μm from the reference electrode. We also noted that thalamic responses sometimes occurred in the absence of substantial hypothalamic response (e.g. anode at -500 μm from the reference electrode), pointing to a likely indirect neuromodulation of shallower brain areas.
Lastly, in the cathodic-first waveform, increasing the pulse width from 500 μs to 1000 μs, increased RMS 5.5✕ in DT and 2✕ in VT without significantly changing the HT response. This indicates that pulse parameters can influence stimulation selectivity at the desired regions.
Disclosures
This work was supported in part by the Pennsylvania Infrastructure Technology Alliance and the Chuck Noll Foundation for Brain Injury Research. This work was supported by The Assistant Secretary of Defense for Health Affairs endorsed by the Department of Defense, in the amount of ($258 606), through the Peer Reviewed Medical Research Program under
Award No. HT9425-24-1-0018. Opinions, interpretations, conclusions, and recommendations are those of the author(s) and are not necessarily endorsed by The Assistant Secretary of Defense for Health Affairs endorsed by the Department of Defense.
References
1. Fang, J. Y. & Tolleson, C. The role of deep brain stimulation in Parkinson’s disease: an overview and update on new developments. Neuropsychiatr. Dis. Treat. 13, 723–732 (2017).
2. Lyons, M. K. Deep brain stimulation: current and future clinical applications. Mayo Clin. Proc. 86, 662–672 (2011).
3. Huang, Y. & Parra, L. C. Can transcranial electric stimulation with multiple electrodes reach deep targets? Brain Stimul. 12, 30–40 (2019).
4. Guo, Y. et al. DeepFocus: a transnasal approach for optimized deep brain stimulation of reward circuit nodes. J. Neural Eng. 22, 016048 (2025).
5. Forssell, M. et al. Transnasal implants in the nasal cavity and sphenoid sinus for minimally invasive deep brain stimulation: measurement of electric field in cadavers. IEEE Trans. Biomed. Eng. PP, 1–11 (2026).
6. Ishimaru, T. et al. Olfactory evoked potential produced by electrical stimulation of the human olfactory mucosa. Chem. Senses 22, 77–81 (1997).
7. Holbrook, E. H., Puram, S. V., See, R. B., Tripp, A. G. & Nair, D. G. Induction of smell through transethmoid electrical stimulation of the olfactory bulb. Int. Forum Allergy Rhinol. 9, 158–164 (2019).
8. Feng, J. et al. Transnasal ventral brainstem stimulation reverses anesthesia-induced unconsciousness in sheep. Brain Stimul. 19, 103068 (2026).
9. Majdi, A., Asamoah, B. & Mc Laughlin, M. Understanding novel neuromodulation pathways in tDCS: brain stem recordings in rats during trigeminal nerve direct current stimulation. Transl. Psychiatry 14, 456 (2024).
10. Xu, J. et al. Trigeminal nerve stimulation restores hippocampal dopamine deficiency to promote cognitive recovery in traumatic brain injury. Prog. Neurobiol. 227, 102477 (2023).
11. Ferreira, C. L. et al. Electrical stimulation enhances early palatal wound healing in mice. Arch. Oral Biol. 122, 105028 (2021).