Noninvasive tFUS stimulation of the parabrachial nucleus suppresses lever pressing during extinction of cocaine self-administration through aversive circuit engagement

Gyeong Hee Pyeon, Jihong Lim, Jaeyong Lee, Jihyun Lee, Hyunjoo Jenny Lee, Yong Sang Jo

Brain Stimulation 2026, 103197 · 10.1016/j.brs.2026.103197

rodentsubstance use disorderhistology molecularbehaviour

Abstract

Introduction Drug addiction is a significant public health concern, and persistent drug-seeking behavior remains difficult to control in severe cases. While invasive interventions have been explored to suppress these behaviors, their use is limited by procedural complexity, highlighting the need for effective noninvasive neuromodulatory approaches. Transcranial focused ultrasound (tFUS) provides a noninvasive approach for deep-brain neuromodulation, but its ability to produce persistent behavioral effects by targeting the extralateral parabrachial nucleus (elPBN) remains unknown. Methods Mice were trained to self-administer cocaine using an operant lever-press paradigm. Following stable cocaine self-administration, active lever presses triggered response-contingent tFUS stimulation instead of cocaine delivery. Stimulation was delivered through the intact skull and targeted to the elPBN for four consecutive days. Results tFUS delivered to the elPBN elicited aversive-related behavioral responses consistent with those previously reported using optogenetic stimulation of this region. In the operant task, tFUS stimulation significantly reduced active lever responding compared with controls. Together, these results indicate that response-contingent engagement of aversive signaling within the elPBN suppressed operant responding previously reinforced by cocaine. This behavioral suppression persisted after a two-week abstinence period, indicating that the effect extended beyond the stimulation phase. Conclusion Noninvasive tFUS targeting the elPBN produced persistent suppression of operant responding previously reinforced by cocaine. Replacing cocaine delivery with response-contingent stimulation of an aversion-related brainstem region produced a lasting change in reward-associated behavior. These findings support further investigation of deep-brain tFUS as a noninvasive neuromodulation approach for modifying maladaptive drug-related behaviors.

Abstract via europepmc.

Speciesmouse

Subjects8, 10, 5swept animals

Sessions per subjectnot reported

Randomisednot reported

Blindingsingle

Sham / controlinactive transducer, active control site

Auditory controlnot reported

Readout timingboth

Anaesthesiaawake

Readoutshistology molecular, behaviour — Fos immunohistochemistry; freezing behavior scoring; real-time place preference (chamber avoidance); operant lever-pressing during cocaine self-administration, extinction, and retention testing; locomotor velocity

Direction of effectexcitatorytFUS of the elPBN increased Fos expression (neuronal activation), consistent with excitation of this aversion-signaling region; behaviorally this activation produced freezing, place avoidance, and a persistent suppression of cocaine-reinforced lever pressing.

Adverse eventsnone observedTemperature beneath the skull rose by less than 0.5 degrees C during a single 4.5-s or 9-s stimulation and by about 0.5 degrees C after 20 consecutive stimulations, levels the authors describe as unlikely to affect behavior; an off-target control receiving identical acoustic exposure showed no behavioral suppression, arguing against a nonspecific thermal effect.

Exposures

Exposure 1: elPBN tFUS

Target: parabrachial nucleus — “extralateral parabrachial nucleus (elPBN)”
Device: custom-built

Pulse timing
Waveform	pulsed
Fundamental frequency (kHz)	5,000	✓✓✓
Pulse duration (ms)	0.5	✓✓✓
Pulse repetition frequency (Hz)	1,000	✓✓✓
Duty cycle (%)	not reportedpulse duration × PRF gives 50%	⚑
Sonication duration (s)	9, 4.5swept	✓✓✓

Pressure and intensity, by domain
Free-field pressure (kPa)	not reported
Free-field Isppa (W/cm²)	2.44	✓✓✓⚑
Free-field Ispta (W/cm²)	not reported
In-situ estimate	not reported
In-situ pressure (kPa)	not reported
In-situ Isppa (W/cm²)	not reported
In-situ Ispta (W/cm²)	not reported

Protocol, in the paper’s words

Pulsed 5 MHz sine waves (PD 0.5 ms, PRF 1 kHz) were gated into 0.5-s bursts (BD) separated by 1-s inter-burst intervals (IBI). For the c-Fos assay and open-arena behavioral test, six or four such 9-s trains were delivered with 60-s inter-sonication intervals. For the RTPP test, bursts were delivered continuously for up to 1 min while the animal occupied the paired chamber, followed by a 30-s timeout. During cocaine self-administration extinction, each active lever press triggered a 4.5-s train (3 bursts) over a 4-day extinction phase.

Flags from extraction

n_subjects — Study comprises separate cohorts: n=8 tFUS mice for the Fos assay and n=10 tFUS mice reused across freezing/RTPP/cocaine behavioral experiments; a 5-mouse off-target subgroup (within the n=10 pooled control) also received real, mistargeted ultrasound and is not counted here; reported as the list of distinct tFUS group sizes.
n_sessions_per_subject — Session structure differs by sub-experiment (single session for Fos/freezing/RTPP vs. a 4-day extinction phase for the cocaine operant task), so no single per-subject session count is stated.
exposures[0].timing.duty_cycle_pct — Not stated as a percentage; PD (0.5 ms) and PRF (1 kHz) are given but duty cycle is not computed by the paper.
exposures[0].free_field.isppa_w_cm2 — Isppa (2.439 W/cm2) was measured in a degassed soybean oil tank during device characterization (pre-implantation), not in the skull/brain; treated as free-field. No in-brain/simulated intensity is reported.
blinding — Blinding to group identity is stated only for Fos-positive cell counting, not for the behavioral scoring or overall study design.
readout_timing — Some outcomes were assessed during stimulation (RTPP chamber time, lever pressing) and others well after (Fos, 60 min post-stimulation); coded as both.