Brain Gene FTO Drives Obesity by Hijacking Appetite Neuron Cargo Delivery
FTO demethylase in AgRP neurons alters Kif1a splicing to boost neuropeptide release, promoting weight gain in mice.
Summary
Scientists have uncovered a precise molecular chain linking the obesity-associated gene FTO to weight gain in the brain. FTO, an enzyme that removes chemical tags from RNA, operates inside hypothalamic AgRP neurons — the brain's primary hunger-signaling cells. By stripping m6A tags from specific messenger RNAs, FTO changes how the motor protein KIF1A is assembled, producing a longer, more active form that ferries appetite-stimulating neuropeptides (NPY and AgRP) down axons for release. Mice engineered to overexpress FTO in AgRP neurons gained significantly more weight, while mice lacking FTO in those same neurons stayed lean. Crucially, silencing Kif1a reversed the obesity caused by FTO overexpression, pinpointing this axis as a druggable target for metabolic disease.
Detailed Summary
Obesity is one of the defining health crises of our era, yet the molecular mechanisms that translate environmental signals into lasting changes in appetite and energy balance remain incompletely understood. This study, published in The EMBO Journal, addresses that gap by tracing a precise epitranscriptomic pathway from the RNA-modifying enzyme FTO through alternative splicing of a motor protein gene to the secretion of appetite-stimulating neuropeptides in the hypothalamus. The findings reframe FTO not merely as an obesity-associated gene but as a master regulator of neuronal cargo trafficking in hunger circuits.
The researchers used conditional genetic mouse models to manipulate Fto expression specifically in agouti-related peptide (AgRP) neurons — hypothalamic cells that powerfully drive food intake and suppress energy expenditure. Mice with AgRP-neuron-specific Fto knockout were significantly leaner than controls, while mice overexpressing Fto in those same neurons gained substantially more weight, particularly under high-fat diet or fasting-refeeding conditions. These bidirectional phenotypes confirmed that FTO in AgRP neurons is both necessary and sufficient to modulate body weight, independent of FTO activity in adipose tissue or other brain regions.
To identify the molecular mechanism, the team performed m6A sequencing (MeRIP-seq) and RNA sequencing on AgRP neurons isolated from Fto-overexpressing mice. FTO demethylation was enriched on mRNAs encoding proteins involved in membrane trafficking and alternative splicing regulation. The most functionally compelling target was Kif1a, encoding the kinesin motor protein KIF1A. FTO demethylation of Kif1a mRNA promoted inclusion of exon 13, which encodes a segment of the protein's hinge domain. Structural and functional analyses showed that this exon-13-inclusive isoform has an expanded hinge region that facilitates KIF1A dimerization and enhances its processive motor activity along axonal microtubules.
The downstream consequence of increased KIF1A activity was striking: dense-core vesicles (DCVs) containing NPY and AgRP were transported more efficiently to axon terminals and secreted at higher rates. Live-cell imaging of vesicle dynamics in cultured AgRP neurons confirmed that FTO overexpression accelerated anterograde DCV trafficking, while Kif1a knockdown reversed this effect. In vivo, Kif1a knockdown in AgRP neurons of Fto-overexpressing mice significantly suppressed weight gain, validating the FTO→KIF1A→DCV secretion→obesity axis as causally connected rather than merely correlative. Plasma NPY and AgRP levels were elevated in Fto-overexpressing mice and normalized by Kif1a knockdown.
The study carries important implications for understanding how epitranscriptomic regulation intersects with neuronal physiology to control systemic metabolism. The FTO-KIF1A axis represents a previously unrecognized layer of appetite regulation operating at the level of RNA modification and axonal transport rather than transcription or receptor signaling. From a translational standpoint, KIF1A or the splicing machinery governing exon 13 inclusion could represent novel therapeutic targets for obesity. Caveats include the mouse-model focus, the complexity of AgRP neuron biology in humans, and the need to determine whether analogous FTO-KIF1A signaling operates in human hypothalamic neurons.
Key Findings
- AgRP-neuron-specific Fto knockout mice were significantly leaner than wild-type controls, while Fto-overexpressing mice gained substantially more weight under high-fat diet and fasting-refeeding paradigms.
- MeRIP-seq revealed FTO demethylation was enriched on mRNAs for membrane trafficking and alternative splicing factors in AgRP neurons.
- FTO demethylation of Kif1a mRNA promoted inclusion of exon 13, producing a KIF1A isoform with an expanded hinge domain that enhances dimerization and motor processivity.
- Live-cell imaging confirmed that FTO overexpression accelerated anterograde dense-core vesicle (DCV) trafficking in AgRP neuron axons, while Kif1a knockdown reversed this acceleration.
- Kif1a knockdown in AgRP neurons of Fto-overexpressing mice significantly suppressed weight gain, causally linking the FTO→KIF1A axis to obesity.
- Plasma NPY and AgRP neuropeptide levels were elevated in Fto-overexpressing mice and normalized following Kif1a knockdown, confirming the secretion mechanism.
- FTO overexpression increased food intake and reduced energy expenditure in mice, both of which were partially rescued by Kif1a silencing in AgRP neurons.
Methodology
The study used conditional Cre-lox mouse models (AgRP-Cre driver) to generate AgRP-neuron-specific Fto knockout and overexpression lines, with littermate controls. Epitranscriptomic profiling was performed via MeRIP-seq and RNA-seq on FACS-isolated AgRP neurons. Vesicle trafficking dynamics were quantified by live-cell fluorescence imaging of DCV markers in primary cultured AgRP neurons. Kif1a knockdown was achieved via AAV-delivered shRNA stereotaxically injected into the hypothalamus, and body weight, food intake, energy expenditure, and plasma neuropeptide levels were measured as primary outcomes.
Study Limitations
The study was conducted entirely in mice, and it remains unknown whether the FTO-KIF1A axis operates similarly in human hypothalamic AgRP neurons. The complexity of AgRP neuron circuitry and the potential off-target effects of AAV-mediated Kif1a knockdown in vivo were not fully characterized. The authors did not report specific conflicts of interest, and the work was supported by public research grants from JSPS and related Japanese funding bodies.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.