**Background:** Atypical antipsychotics are associated with a high prevalence of metabolic disturbances, including weight gain, which contribute to increased mortality in patients with schizophrenia. The 5-HT2A/H1 hypothesis suggests that inhibition of these receptors leads to AMPK activation in the hypothalamus, promoting weight gain. However, this hypothesis does not fully explain the differential effects of various antipsychotics. L-β-aminoisobutyric acid (L-BAIBA), a GABA isomer and myokine, has been shown to activate AMPK and improve metabolism in peripheral tissues. This study investigates the role of L-BAIBA in the central nervous system, specifically its contribution to weight gain and metabolic complications induced by quetiapine (high-risk) and lurasidone (low-risk) antipsychotics.
**Methods:** Male Sprague-Dawley rats were chronically administered lurasidone (1 and 3 mg/kg/day) or quetiapine (10 and 30 mg/kg/day) for 14 days via osmotic pumps. Body weight was measured. Plasma and hypothalamic levels of L-BAIBA and D-BAIBA were measured using UHPLC. Intracellular levels of IP3, cAMP, AMP, ATP, and AMPK signaling were assessed in the hypothalamus. In vivo microdialysis was used to measure extracellular L-BAIBA and D-serine release in the hypothalamus following high potassium (HKMRS) stimulation, with or without the hemichannel inhibitor carbenoxolone. In vitro studies used primary cultured cortical astrocytes exposed to lurasidone (100 and 500 nM) or quetiapine (3 and 30 μM) for 14 days. Intracellular L-BAIBA, GABA, D-serine, and second messengers were measured. Ripple-burst stimulation was used to evoke astroglial transmitter release, and the effects of carbenoxolone and the ABAT inhibitor vigabatrin were examined. The interaction between quetiapine and the 5-HT7 inverse agonist SB269970 on AMPK signaling was also assessed.
**Key Results:** Chronic quetiapine administration (10 and 30 mg/kg/day) significantly increased rat body weight (F(2,15)=13.1, p<0.01), while lurasidone had no effect. Quetiapine increased plasma L-BAIBA levels (F(2,15)=6.6, p<0.01), whereas lurasidone did not. In the hypothalamus, quetiapine increased L-BAIBA levels (F(2,15)=6.7, p<0.01), while lurasidone decreased them (F(2,15)=11.0, p<0.01). D-BAIBA was not detected in the hypothalamus. Both antipsychotics decreased IP3 levels (lurasidone: F(2,15)=16.1, p<0.01; quetiapine: F(2,15)=29.6, p<0.01). Lurasidone decreased cAMP (F(2,15)=9.2, p<0.01) and increased AMP (F(2,15)=9.1, p<0.01). Quetiapine (30 mg/kg) decreased cAMP (F(2,15)=5.1, p<0.05) and increased AMP (F(2,15)=4.1, p<0.05). Neither affected ATP. Microdialysis showed that HKMRS-evoked L-BAIBA release was inhibited by carbenoxolone (F_carbenoxolone(1,10)=36.9, p<0.01). Lurasidone decreased (F_lurasidone(1,10)=5.7, p<0.01) and quetiapine increased (F_quetiapine(1,10)=9.3, p<0.01) HKMRS-evoked L-BAIBA release. Similar effects were observed for D-serine release. In astrocytes, lurasidone decreased intracellular L-BAIBA (F(2,15)=7.7, p<0.01), while quetiapine increased it (F(2,15)=5.6, p<0.05). Neither affected GABA or D-serine levels. Ripple-evoked L-BAIBA and D-serine release were inhibited by carbenoxolone. Lurasidone decreased and quetiapine increased ripple-evoked L-BAIBA and D-serine release. Vigabatrin decreased L-BAIBA levels and abolished the effects of both antipsychotics on L-BAIBA but not on D-serine. SB269970 (10 μM) attenuated the AMPK activation induced by 3 μM quetiapine but not by 30 μM quetiapine.
**Clinical Implications:** This study identifies L-BAIBA as a novel gliotransmitter that mediates the weight gain and metabolic complications associated with atypical antipsychotics. Quetiapine increases L-BAIBA synthesis and release, leading to hypothalamic AMPK activation and weight gain. In contrast, lurasidone decreases L-BAIBA levels, which may explain its lower risk for metabolic side effects. The biphasic dose-response of quetiapine on weight gain may be due to the opposing effects of 5-HT2A/H1 receptor inhibition (AMPK activation) and 5-HT7 receptor inverse agonism (AMPK suppression). These findings suggest that L-BAIBA could be a therapeutic target for preventing or managing antipsychotic-induced metabolic disturbances. Additionally, L-BAIBA release is stimulated by ripple-burst oscillations, which are involved in cognitive functions such as memory consolidation, indicating a potential dual role in both therapeutic and adverse effects.