A synaptic mitochondrion–ribbon–vesicle complex organizes photoreceptor neurotransmission

Photoreceptor synapses rely on mitochondria, synaptic ribbons, and vesicles, yet how these elements integrate into functional units remains unclear. Here we show, using focused ion beam-scanning electron microscopy (FIB-SEM) and electrophysiology, that they form synaptic mitochondrion–ribbon–vesicle complexes (sMRVcs) with fundamentally different architectures in rods and cones. Rod spherules exhibit a strict 1:1 pairing of a single megamitochondrion with one ribbon, sustaining continuous glutamate release in darkness, whereas cone pedicles contain multiple distributed sMRVcs. Disrupting the mitochondrial anchor protein syntaphilin shifts this ratio to 2:1 in rods, with additional mitochondria recruited near the ribbon, and enhances synaptic transmission. In a retinal degeneration model, the sMRVc progressively disassembles before functional decline, with fragmented mitochondria, detached ribbons, and dispersed vesicles. We also identify non-ribbon release sites selectively targeting horizontal cells. These findings establish the sMRVc as a conserved structural–metabolic module whose subtype-specific organization reflects distinct visual demands and whose integrity is essential for synaptic health. Researchers find photoreceptor synapses contain mitochondrion–ribbon–vesicle complexes: rods have a 1:1 megamitochondrion–ribbon pair; cones have distributed units. Disrupting mitochondrial anchoring shifts rods to 2:1 and boosts signaling, while degeneration breaks the complex apart.