Two Brakes, One Broken System: A Precision Medicine Framework for Bruxism and Psychogenic Jaw Clenching

New iPSC-based genomic evidence has identified two converging neurobiological failures at the root of bruxism: a serotonin 2A receptor variant that amplifies stress-driven motor activation, and a GABAergic inhibitory deficit that removes the brainstem’s ability to suppress the resulting rhythmic jaw activity. Matching pharmacotherapy to these two mechanisms, rather than managing downstream damage, is where the field needs to go.

Introduction

Most bruxism patients seen in an orofacial pain clinic have two overlapping problems: rhythmic grinding and clenching during sleep, and a psychogenic pattern of sustained jaw-muscle tension during the day, worsened by stress. These are not the same disorder, but they run on the same nervous system. Both involve excessive, involuntary jaw motor activity that is resistant to conventional management. Occlusal splints protect teeth; they do not stop the motor behavior. Botulinum toxin weakens the masseter for a few months but leaves the neural drive untouched. Benzodiazepines carry tolerance and dependence risk and are not a long-term answer. Nothing we currently have reaches into the brainstem and turns down the gain on the motor generators responsible for either disorder.

That gap may be addressable, at least in part, because recent genomic and neurophysiological work has identified the specific circuit-level failures underlying both problems. The work comes primarily from a research group at Showa University (Tokyo) led by prosthodontist Kazuyoshi Baba and stem cell neurobiologist Wado Akamatsu, using induced pluripotent stem cell (iPSC) technology to build the first cellular disease models of sleep bruxism. What they have found points to two distinct but interacting mechanisms: stress-driven serotonergic motor activation running through the 5-HT2A receptor, and a GABAergic disinhibition state that removes the neural brake on the brainstem masticatory motor pattern generator. Understanding both, and matching pharmacotherapy accordingly is a more tractable path forward than waiting for evidence that does not yet exist.

The Genetics: Where the Trail Starts

The mechanistic story starts with a genetic finding by Abe, Clark, Baba, and colleagues, published in 2012: a single-nucleotide polymorphism (SNP, rs6313) in the serotonin 2A receptor gene HTR2A is associated with a 4.25-fold increase in the risk of sleep bruxism (Abe et al., 2012).¹ That case-control study recruited a Japanese population, confirmed sleep bruxism with three-night masseter EMG recordings, and screened 13 polymorphisms across four serotonergic neurotransmission genes, SLC6A4, HTR1A, HTR2A, and HTR2C. The rs6313 C allele was the only variant to remain an independent predictor in multivariate regression. A fourfold effect size for a complex behavioral trait is not subtle, and it gave the Baba/Akamatsu group a genetically defined anchor for building patient-specific disease models.

The rs6313 polymorphism reduces 5-HT2A receptor expression. That matters more than it might initially appear. The 5-HT2A receptor is not simply a serotonin-binding site, it is a key regulatory node in the circuit connecting stress, serotonin, dopamine, and motor output. Reduced receptor expression from the variant allele alters the calibration of that entire circuit, with downstream consequences for both sleep-related motor control and daytime stress reactivity.

Mechanism One: Stress, Serotonin, and the 5-HT2A Motor Pathway

Under stress, serotonergic neurons in the dorsal raphe nucleus increase their firing and flood mesocortical circuits with 5-HT. What happens next depends critically on the 5-HT2A receptor in the medial prefrontal cortex: its activation by elevated serotonin drives dopamine release into mesocortical motor circuits. Pehek and colleagues (2006) demonstrated this using selective 5-HT2A antagonists in microdialysis studies, establishing that stress-induced dopamine efflux in the prefrontal cortex depends specifically on 5-HT2A receptor activation.¹⁹ More dopamine output through these pathways means more excitatory drive to masticatory motor neurons. The clinical consequence is the pattern seen in stress-driven jaw clenchers: sustained daytime muscle tension that worsens predictably with psychological load and does not need sleep to manifest.

This is not just a story about anxiety causing clenching. It is a receptor-level cascade, stress → elevated raphe serotonin → 5-HT2A receptor activation → dopamine efflux in mesocortical circuits → jaw motor hyperactivation, that is pharmacologically accessible at multiple points. The rs6313 polymorphism alters 5-HT2A receptor expression, which plausibly explains why carriers show both elevated bruxism risk and the constitutive neuronal hyperexcitability phenotype documented by Sarkar and colleagues (2022) in iPSC-derived neurons.²

Buspirone: Addressing the Serotonergic Arm

Buspirone targets this pathway at the source. It is a partial agonist at presynaptic 5-HT1A somatodendritic autoreceptors on the cell bodies of raphe serotonergic neurons (Bostwick and Jaffee, 1999).²⁰ When those autoreceptors are activated, raphe neuron firing decreases and serotonergic output to mesocortical circuits drops. Less serotonin reaching 5-HT2A receptors means less stress-driven dopamine release and less motor activation of the jaw. Buspirone is not a sedative, does not act at GABA-A receptors, and carries no dependence risk. Its anxiolytic and motor-quieting effects come from this presynaptic serotonergic feedback mechanism.

Most of the literature on buspirone and bruxism focuses on SSRI-induced cases, where excess synaptic serotonin from the drug itself is the trigger (Bostwick and Jaffee, 1999).²⁰ What gets less attention is a 2021 case series by Kowacs and colleagues that evaluated buspirone specifically in patients with primary bruxism where anxiety was the predisposing condition and no serotonergic drug was involved. Four patients, two with sleep bruxism only, two with combined sleep and awake bruxism including one with a 20-year history, showed clinically meaningful improvement, with a mean self-reported bruxism reduction of approximately 65% (Kowacs et al., 2021).²¹ That is a small, uncontrolled series, but it is direct documentation of the clinical hypothesis: buspirone for stress/anxiety-driven bruxism in the absence of any pharmacological trigger.

One practical point that gets missed: buspirone is not fast. Its mechanism requires adaptive receptor-level changes to develop, not immediate pharmacological binding. Relief typically appears at two to four weeks, and some patients need the full therapeutic dose sustained for a month before significant improvement is apparent (Bostwick and Jaffee, 1999).²⁰ Evaluating buspirone on the same timeline as a benzodiazepine, a few days to two weeks, will reliably miss responders. Any trial needs a minimum of four weeks at therapeutic dose before conclusions are drawn. The effective dose range in reported cases is 5 to 30 mg daily in one to three divided doses; bedtime dosing concentrates the pharmacological effect at the clinically relevant window while adding essentially no daytime sedative burden.

Mechanism Two: GABAergic Disinhibition in the Brainstem

The second mechanism operates at a different level of the neuraxis and drives primarily the sleep component. Sato and colleagues (2024) performed the first comprehensive RNA-sequencing analysis of iPSC-derived bruxism neurons using both bulk and single-nucleus RNA-seq, and identified three major molecular lesions.³ Genes encoding receptor-operated calcium channels were pathologically upregulated. GRIN2B, an NMDA receptor subunit that increases synaptic excitability when overexpressed, was elevated specifically within GABAergic neuron populations. And CHRM3 (M3 muscarinic acetylcholine receptor) was broadly upregulated across neuronal subtypes. The GRIN2B finding in GABAergic neurons is the key: the inhibitory neurons whose job is to suppress excessive motor activity are themselves hyperexcitable and functionally impaired. The system that should be braking masticatory motor drive during sleep is broken.

The in vivo correlate was provided independently. Gutiérrez Herrera and colleagues (2024) cited MRS data showing directly measured GABA+ levels in the brainstems of sleep bruxism patients are reduced relative to controls.⁴ Transcriptomics and in vivo neurochemistry are pointing at the same deficit from different angles. Bruxism, in this framework, is as much a disinhibition disorder as an excitation disorder.

The most direct circuit-level proof is coming from work in preparation by Onishi and colleagues in the Showa/Juntendo group.⁵ Using the same iPSC lines differentiated into ventral brainstem neurons (verified by TUBB3 and GAD1/2 expression) and a multielectrode array (MEA) platform, they moved from transcriptomics to live network electrophysiology, measuring spontaneous firing, burst frequency, synchrony, and inhibitory circuit dynamics across dozens of electrodes simultaneously. Bruxism patient-derived neurons show aberrant network activity consistent with genuine loss of inhibitory control, confirming that the gene expression signal translates into measurable circuit pathology, not just a list of differentially expressed genes.

Disclosure: Onishi M et al. is currently in preparation for submission and has not yet been assigned a DOI or journal citation. The author has reviewed the methodology and key findings directly. The citation will be updated upon publication.

Like what you’re learning?  Download a brochure for our Orofacial Pain and Oral Medicine certificate or master’s degree program.

Gabapentinoids: Targeting the Calcium Channel Arm

Because the Sato et al. (2024) data specifically implicate pathological upregulation of receptor-operated calcium channels, gabapentin and pregabalin have direct mechanistic relevance.³ Both bind the α2δ subunit of voltage-gated calcium channels and reduce excitatory neurotransmitter release, a mechanism distinct from GABA-A modulation (benzodiazepines) and GABA-B agonism (baclofen). Neither of these drugs acts on serotonin receptors. The calcium channel arm of bruxism pathophysiology is their specific target.

There is PSG-controlled clinical evidence for gabapentin. Madani, Abdollahian, Khiavi, and colleagues (2013) conducted a randomized study in 20 PSG-confirmed sleep bruxism patients comparing gabapentin to stabilization splint therapy over two months.¹¹ Both interventions reduced SB episodes per hour, bruxism time index, and episode duration. But only the gabapentin group showed additional significant improvements in total sleep time, slow wave sleep (stage III), and sleep efficiency. Brown and Hong (1999) documented complete resolution of antidepressant-induced bruxism with gabapentin 300 mg nightly,¹² and Soyata and Oflaz (2015) reported full remission of SSRI-induced bruxism with 150–300 mg daily, with concurrent resolution of restless legs syndrome.¹³

Pregabalin data for bruxism remain thin, one case report of awake bruxism reduction at 375 mg/day in a patient with generalized anxiety disorder, with recurrence on dose reduction (Nasr et al., 2020).¹⁴ No controlled sleep bruxism trial exists for pregabalin. Both gabapentinoids carry strong AASM-level recommendations for restless legs syndrome and have PSG-documented reductions in periodic limb movements (Garcia-Borreguero et al., 2010; Happe et al., 2001),^15,16 which is relevant given that PLMS and sleep bruxism share a nocturnal motor suppression deficit and frequently co-occur.

Oral Baclofen: Directly Addressing the GABA-B Deficit

If the core brainstem lesion is GABAergic disinhibition, the most direct pharmacological correction is GABA-B receptor agonism. Baclofen is selective for GABA-B receptors, and Janati, ALGhasab, and ALGhassab (2014) reported the clearest case of bruxism: a 16-year-old with severe, intractable bruxism following anoxic encephalopathy, whose bruxism resolved within days of oral baclofen at 10 mg three times daily, with sustained efficacy at 20 mg three times daily.⁶ The authors explicitly attributed this to GABAergic dysregulation as the causal mechanism. Diazepam, which works via GABA-A potentiation rather than GABA-B agonism, also reduces bruxism-like jaw activity in rodent stress models (Gutiérrez Herrera et al., 2024),⁴ confirming that enhancing GABAergic tone generally suppresses the behavior, but with the tolerance and sedation limitations that baclofen avoids at appropriate doses. Oral baclofen has real limitations: it does not reach brainstem GABA-B receptors at reliable concentrations by oral dosing, and abrupt discontinuation after sustained use carries withdrawal risk. At low-to-moderate doses titrated gradually, it is a reasonable pharmacological trial for patients with severe refractory sleep bruxism who have not responded to gabapentinoids and buspirone. The GABA-B deficit identified in the iPSC work is the mechanistic target, and oral baclofen is the most accessible tool we currently have against it.

A Note on Invertebrate Motor Circuit Evidence

The principle that GABAergic inhibitory failure drives persistent motor neuron hyperexcitability has been supported even outside mammalian systems. In the feeding motor network of Aplysia californica, Jing and Weiss (2003) showed that pharmacological blockade of GABAergic inhibition with picrotoxin increases motor neuron firing and disrupts motor program timing.¹⁷ Separately, Brezina and colleagues (2000) found that repeated stimulation of the GABAergic interneuron B40 produces a persistent increase in motor neuron B8 excitability that is specifically occluded by baclofen.¹⁸ The parallel is mechanistic, not anatomical, Aplysia is not a bruxism model, but the principle that GABA-B agonism reverses GABAergic disinhibition-driven motor hyperexcitability holds across preparations separated by 600 million years of evolution.

A Layered Treatment Framework

What the evidence produces is not a single drug target but a framework for phenotype-matched treatment. Most patients presenting with both sleep bruxism and stress-driven daytime clenching have both mechanisms running simultaneously: the serotonergic motor activation pathway driving the daytime component, and the GABAergic disinhibition state permitting the nocturnal component. They are not independent, the HTR2A variant plausibly contributes to both through the same downstream cascade, but they respond to different drugs.

The logical stepped-care approach, matched to pathophysiology:

• Buspirone (5–30 mg at bedtime, minimum four-week evaluation period), targets the serotonergic motor activation pathway via 5-HT1A autoreceptor-mediated suppression of raphe output; the anxiolytic effect also directly addresses the psychogenic component of daytime clenching. Primary target: stress-driven awake bruxism and anxiety-amplified sleep bruxism.
• Gabapentin (300–1800 mg at bedtime), targets the calcium channel arm of neuronal hyperexcitability identified by Sato et al. (2024); also improves slow wave sleep architecture. Primary target: sleep bruxism and nocturnal hyperexcitability.
• Oral baclofen (5–20 mg at bedtime, titrated cautiously), targets the GABA-B receptor deficit directly; appropriate as a third step for patients who have not achieved adequate control with buspirone and gabapentin alone, and in whom the GABAergic disinhibition mechanism appears dominant.

These three mechanisms do not overlap. Buspirone does not act on GABA receptors. Gabapentin does not act on GABA receptors or serotonin receptors. Baclofen does not act on calcium channels or 5-HT1A receptors. Used together, they provide orthogonal coverage of three distinct pathophysiological targets rather than additive sedation. The practical CNS depression concern is real with gabapentin and baclofen and warrants starting at the low end of each range and titrating slowly. Buspirone adds essentially no sedative burden and is a reasonable first step.

When Everything Else Has Failed: The Question of Intrathecal Delivery

For the small number of patients who exhaust all of the above, PSG-confirmed severe bruxism with documented irreversible tissue damage, failed splint therapy, failed gabapentinoid and buspirone trials, failed botulinum toxin, the question of more direct central delivery of GABAergic agents is legitimate, even if the evidence base for it is thin.

Intrathecal baclofen (ITB) pump therapy delivers drug directly into the CSF, achieving roughly 100-fold higher receptor-level activity at the spinal cord compared to oral administration, with programmable nocturnal dosing and avoidance of systemic cognitive side effects. PSG evidence supports ITB for nocturnal motor activity reduction in spasticity and PLMS populations: Kravitz and colleagues (1992) documented ITB-driven reduction in tibialis anterior EMG activity per sleep hour,⁷ Bensmail and colleagues (2006) showed ITB improved sleep continuity compared to oral baclofen,⁸ and a subsequent pilot study confirmed PSG-documented PLMS reduction after pump implantation (Bensmail et al., 2012).⁹

The anatomical problem for bruxism is that standard lumbar catheter placement delivers drug to spinal level, while the GABA-B receptors relevant to bruxism sit in brainstem populations, the LDT/PPT, trigeminal motor nucleus, and reticular formation circuits regulating sleep-state motor atonia. Cervical catheter tip placement at C1–T4 has been used for supraspinal movement disorders (Jacobs et al., 2021)¹⁰ and could in principle reach the relevant targets, but that approach adds technical complexity and a respiratory caveat: Bensmail and colleagues (2012) found that ITB increased central apneas and respiratory disturbance index in bolus delivery mode, with continuous infusion appearing safer from a respiratory standpoint.⁹ Any bruxism application would need continuous infusion mode and careful baseline respiratory assessment.

ITB for bruxism has not been tested in any formal study. This section belongs in the category of scientific hypothesis, not clinical proposal. We flag it because the molecular biology supports the mechanistic rationale, not because a pathway to clinical use currently exists. It belongs at the end of an exhaustive stepwise trial, with multidisciplinary neurosurgical evaluation and polysomnographic documentation at each stage.

The Precision Medicine Case

What the Baba/Akamatsu iPSC work makes possible, beyond any individual drug target, is the beginning of mechanism-matched clinical reasoning in bruxism. For the first time it is possible to ask not just whether a patient has bruxism, but which mechanisms are active in that patient. A patient presenting primarily with daytime stress-driven clenching, carrying the HTR2A rs6313 variant, and with anxiety as a major comorbidity represents a different neurobiological problem from a patient with purely nocturnal grinding and no psychogenic component. They likely need a different pharmacological approach.

The rs6313 variant is not yet part of routine clinical workup, and the Sato et al. (2024) paper explicitly flags its small sample (n=3 patients) and the inherent limitations of in vitro modeling. This is still preliminary science.³ But the direction it points is toward computable phenotyping of bruxism subtypes, using genetic, neurochemical, and polysomnographic data together to characterize the dominant pathophysiological mechanism in each patient and match treatment accordingly.

The evidence described here is not yet sufficient to change clinical guidelines. What it does justify is a change in how treatment sequencing is approached. Rather than starting with a mouth guard and cycling through sedatives, the question should be: which mechanism is dominant in this patient, and which pharmacological intervention targets it most directly? Buspirone for the serotonergic component. Gabapentinoids for the calcium channel and sleep architecture component. Oral baclofen for the GABAergic disinhibition component. The patients who break through every splint we make, who grind through crowns in months, who wake their partners with audible clenching night after night, they deserve that level of mechanistic thinking applied to their care. The molecular biology is now pointing clearly enough to warrant it.

Are you interested in a variety of issues focused on orofacial pain, medicine and sleep disorders? Consider enrolling in the Herman Ostrow School of Dentistry of USC’s online, competency-based certificate or master’s program in Orofacial Pain and Oral Medicine. 

References

1. Abe Y, Suganuma T, Ishii M, Yamamoto G, Gunji T, Clark GT, Tachikawa T, Kiuchi Y, Igarashi Y, Baba K. Association of genetic, psychological and behavioral factors with sleep bruxism in a Japanese population. J Sleep Res. 2012;21(3):289–296. doi:10.1111/j.1365-2869.2011.00961.x
2. Sarkar AK, Nakamura S, Nakai K, Sato T, Shiga T, Abe Y, Hoashi Y, Inoue T, Akamatsu W, Baba K. Increased excitability of human iPSC-derived neurons in HTR2A variant-related sleep bruxism. Stem Cell Reports. 2022;17(4):893–908. doi:10.1016/j.stemcr.2022.01.009
3. Sato T, Yamaguchi A, Onishi M, Abe Y, Shiga T, Ishikawa K, Baba K, Akamatsu W. Comprehensive gene expression analysis using human induced pluripotent stem cells derived from patients with sleep bruxism: a preliminary in vitro study. Int J Mol Sci. 2024;25(23):13141. doi:10.3390/ijms252313141
4. Gutiérrez Herrera MA, Villanueva I, Peña-Ortega F, et al. The neural substrates of bruxism: current knowledge and clinical implications. Front Neurol. 2024;15:1451183. doi:10.3389/fneur.2024.1451183
5. Onishi M, [co-authors], Baba K, Akamatsu W. GABAergic inhibitory circuit impairment in sleep bruxism: multielectrode array analysis of iPSC-derived brainstem neurons. In preparation. 2026.
6. Janati AB, ALGhasab NS, ALGhassab FS. Bruxism associated with anoxic encephalopathy: successful treatment with baclofen. Case Rep Dent. 2014;2014:197461. doi:10.1155/2014/197461
7. Kravitz HM, Corcos DM, Hansen G, Penn RD, Cartwright RD, Gianino J. Intrathecal baclofen: effects on nocturnal leg muscle spasticity. Am J Phys Med Rehabil. 1992;71(1):48–52.
8. Bensmail D, Quera Salva MA, Roche N, et al. Effect of intrathecal baclofen on sleep and respiratory function in patients with spasticity. Neurology. 2006;67(8):1432–6. doi:10.1212/01.wnl.0000239827.38036.23
9. Bensmail D, Marquer A, Roche N, et al. Pilot study assessing the impact of intrathecal baclofen administration mode on sleep-related respiratory parameters. J Rehabil Med. 2012;44(4):344–7. doi:10.2340/16501977-0939
10. Jacobs NW, Maas EM, Brusse-Keizer M, Rietman HJS. Effectiveness and safety of cervical catheter tip placement in intrathecal baclofen treatment of spasticity: a systematic review. J Rehabil Med. 2021;53(7):jrm00215. doi:10.2340/16501977-2857
11. Madani AS, Abdollahian E, Khiavi HA, Radvar M, Foroughipour M, Asadpour H, Hasanzadeh N. The efficacy of gabapentin versus stabilization splint in management of sleep bruxism. J Prosthodont. 2013;22(2):126–131. doi:10.1111/j.1532-849X.2012.00914.x
12. Brown ES, Hong SC. Antidepressant-induced bruxism successfully treated with gabapentin. J Am Dent Assoc. 1999;130(10):1467–1469. doi:10.14219/jada.archive.1999.0057
13. Soyata AZ, Oflaz S. Gabapentin treatment in bruxism associated with fluoxetine. J Clin Psychopharmacol. 2015;35(4):481–483. doi:10.1097/JCP.0000000000000337
14. Nasr S, et al. Awake bruxism treated with pregabalin in a patient with generalized anxiety disorder. Psychiatr Danub. 2020;32(3–4):435–437.
15. Garcia-Borreguero D, Larrosa O, Williams AM, et al. Treatment of restless legs syndrome with pregabalin: a double-blind, placebo-controlled study. Neurology. 2010;74(23):1897–1904. doi:10.1212/WNL.0b013e3181e1ce96
16. Happe S, Klösch G, Saletu B, Zeitlhofer J. Treatment of idiopathic restless legs syndrome with gabapentin. Neurology. 2001;57(9):1717–1719. doi:10.1212/WNL.57.9.1717
17. Jing J, Weiss KR. Concerted GABAergic actions of Aplysia feeding interneurons in motor program specification. J Neurosci. 2003;23(12):5283–5294. doi:10.1523/JNEUROSCI.23-12-05283.2003
18. Brezina V, Orekhova IV, Weiss KR. The neuromuscular transform: the dynamic, nonlinear link between motor neuron firing patterns and muscle contraction in rhythmic jaw movements of Aplysia. J Neurophysiol. 2000;83(1):207–240. [See also: PMC3652222 re GABA-B occlusion of B8 persistent excitability increase]
19. Pehek EA, Nocjar C, Roth BL, Byrd TA, Mabrouk OS. Evidence for the preferential involvement of 5-HT2A serotonin receptors in stress- and drug-induced dopamine release in the rat medial prefrontal cortex. Neuropsychopharmacology. 2006;31(2):265–277. doi:10.1038/sj.npp.1300807
20. Bostwick JM, Jaffee MS. Buspirone as an antidote to SSRI-induced bruxism in 4 cases. J Clin Psychiatry. 1999;60(12):857–860. doi:10.4088/JCP.v60n1209
21. Kowacs DP, Folchini CM, de Moura Vieira KR, Giraldes JA, Kowacs PA. Use of buspirone in the treatment of nonpharmacological bruxism: about 4 cases. Clin Neuropharmacol. 2021;44(6):247–249. doi:10.1097/WNF.0000000000000486