Deleting Dopamine D2 Receptors from Cortical Interneurons Blunts Drug-Induced Hyperactivity in Mice

When mice lose dopamine D2 receptors from a specific set of forebrain interneurons, the frenzied locomotor response triggered by two widely used NMDA receptor blockers is dramatically toned down, according to a new study published in Discover Neuroscience.

The study focused on a conditional knockout mouse line in which the Drd2 gene, which encodes the dopamine D2 receptor, was deleted selectively from GABAergic interneurons derived from the medial ganglionic eminence, a transient embryonic structure that gives rise to the majority of the cerebral cortex’s parvalbumin-expressing and somatostatin-expressing inhibitory neurons. The researchers crossed mice carrying a floxed Drd2 allele with an Nkx2.1-Cre driver line, producing animals in which D2 receptors were removed from developing Nkx2.1-lineage interneurons while littermate controls retained the receptor. Because the two genotypes were generated as littermates, genetic background and rearing environment were controlled, and statistical analyses of the two control genotypes revealed no significant differences between them.

Before any drug challenge, the team established that the knockout animals were fundamentally healthy in terms of motor function. Body weights were equivalent across genotypes, with a large analysis of 18 to 24 animals per sex and genotype group confirming that only sex, not genotype, predicted body mass. Across two consecutive saline habituation sessions in the open field, ambulatory distance did not differ by genotype or sex, and within-session habituation was significant in both groups, indicating that the knockout mice learned to settle down in the chamber just as their controls did. The only baseline deviation was a modest reduction in stereotypic counts among knockout mice on the second habituation day, a subtle effect the authors noted but did not overinterpret.

The core experiment used a three-day open-field protocol designed to maximize signal and minimize noise. On days one and two, mice received saline injections after a 30-minute habituation period and were then monitored for 120 minutes. On day three, the same procedure was repeated with the drug instead of saline. For the dose-response study, the non-competitive NMDA receptor antagonist MK-801, also known as dizocilpine, was administered at 0.1, 0.3 and 0.6 milligrams per kilogram. Ninety animals contributed data on the test day, with six to eight mice per dose, sex and genotype group. Infrared beam arrays recorded ambulatory distance, stereotypic counts and vertical activity in five-minute blocks throughout each 150-minute session.

The results were striking. At the two behaviorally active doses, 0.3 and 0.6 milligrams per kilogram, the knockout mice traveled significantly less than controls after MK-801 injection, with the genotype effect reaching statistical significance in dose-specific mixed models. At the lowest dose, 0.1 milligrams per kilogram, which produced minimal locomotor activation in either genotype, no genotype difference emerged. A mixed-model analysis across all doses confirmed significant main effects of both dose and genotype, with no significant contribution of sex to locomotor activity. To rule out baseline differences as a confound, the researchers normalized post-injection activity to each animal’s own pre-injection baseline, and the genotype effect on percent change scores remained significant, confirming that the blunted response reflected a genuine difference in drug-induced activation rather than a pre-existing difference in how much the mice moved before injection.

Vertical activity, or rearing, told a similar story. Across the three doses, three-way analyses of variance on both vertical time and vertical counts revealed significant main effects of genotype, dose and sex, along with significant dose-by-genotype interactions. Within-dose analyses showed that at 0.3 and 0.6 milligrams per kilogram, control mice and male mice reared more than their counterparts, mirroring the pattern seen for horizontal locomotion. The authors noted that these vertical activity results closely resembled the horizontal locomotor findings rather than the stereotypy pattern, reinforcing the idea that the knockout selectively altered one component of drug-induced behavioral activation.

Stereotypy, by contrast, behaved differently. Repetitive, non-ambulatory movements such as sniffing in place, head bobbing and grooming were scored separately from locomotion, and the genotype effect on this measure was far weaker. A significant dose-by-genotype interaction emerged for MK-801-induced stereotypy, driven by a directional crossover: knockout mice showed significantly lower stereotypic counts than controls at the intermediate dose, but at the highest dose the direction reversed, with knockout mice numerically exceeding controls without reaching within-dose significance. Across the total post-injection period, sex rather than genotype was the dominant predictor of stereotypic behavior, with males exhibiting significantly higher counts than females at both active doses. This dissociation between locomotion and stereotypy is consistent with a well-established neuroanatomical framework in which mesolimbic dopamine projections to the nucleus accumbens preferentially gate locomotor hyperactivity, while nigrostriatal dopamine in the caudate-putamen underlies stereotyped behavior.

To confirm that the effect was not specific to MK-801’s particular pharmacological profile, the team replicated the experiment in an independent cohort of 48 mice using memantine, a pharmacologically distinct NMDA receptor antagonist that shares the channel-pore binding site with MK-801 but has substantially lower binding affinity and faster off-rate kinetics. At a single dose of 30 milligrams per kilogram, knockout mice again traveled significantly less than controls, with the genotype effect sustained across the post-injection session rather than concentrated at any particular phase. Once again, stereotypic counts showed no significant genotype or sex effects. The authors argued that this replication across two compounds with different kinetic and clinical profiles confirms that interneuron D2 receptor signaling contributes to NMDA antagonist-induced motor activation regardless of the specific drug used.

The mechanistic implications reach into one of the most debated areas of schizophrenia research. The hypothesis that cortical GABAergic interneuron maturation is disrupted in the disorder has deep roots, and parvalbumin-expressing interneurons are known to depend heavily on NMDA receptor signaling to sustain their high firing rates. When these fast-spiking cells are acutely silenced by channel blockade, cortical inhibitory tone drops, pyramidal neurons become disinhibited, and downstream dopamine release in the striatum and limbic forebrain drives the locomotor hyperactivity characteristic of NMDA antagonist exposure. The new results position D2 receptors on these MGE-derived interneurons as modulators of that cascade rather than primary drivers, and the selectivity for locomotion over stereotypy argues against a global reduction in dopamine system responsiveness.

The authors were careful to outline the limitations of their approach. The Nkx2.1-Cre driver targets a broad population of MGE-derived neurons, including both parvalbumin-expressing and somatostatin-expressing subtypes as well as additional subcortical structures, so future work with more restricted Cre drivers will be needed to identify which populations are primarily responsible. The model involves a lifelong genetic alteration, meaning the observed effects could reflect developmental loss, ongoing adult loss of the receptor, or a combination of both. The study was also limited to acute drug challenges, and whether the same deletion alters locomotor sensitization or chronic NMDA hypofunction models remains an open question. Direct physiological measurements of interneuron firing, cortical excitatory-inhibitory balance and dopamine dynamics will be essential to establish the causal circuit-level mechanisms. Even so, the consistency of the locomotor attenuation across two drugs, two cohorts and a systematic dose-response range makes a compelling case that dopamine D2 signaling within cortical interneurons shapes how the brain translates NMDA receptor hypofunction into movement, a finding that may ultimately illuminate the early developmental processes that shape vulnerability to neuropsychiatric disease.

Subject of Research: Conditional deletion of dopamine D2 receptors from cortical GABAergic interneurons attenuates NMDA antagonist-induced locomotor hyperactivity in mice

Article Title: MK-801- and memantine-induced locomotor hyperactivity is attenuated by conditional loss of dopamine D2 receptors in select forebrain neurons

Article References: Thabet, Y., Neville, S. E., Shafau, A. O., Graham, D. L., & Stanwood, G. D. (2026). MK-801- and memantine-induced locomotor hyperactivity is attenuated by conditional loss of dopamine D2 receptors in select forebrain neurons. Discover Neuroscience, 21(1), Article 108. https://doi.org/10.1186/s13064-026-00338-6

Image Credits: AI Generated

DOI: 10.1186/s13064-026-00338-6

Keywords: dopamine D2 receptor, NMDA receptor antagonists, MK-801, memantine, cortical interneurons, medial ganglionic eminence, schizophrenia, locomotor hyperactivity, stereotypy, open field test, conditional knockout, GABAergic neurons