More than a billion years ago, a eukaryotic cell swallowed a cyanobacterium and kept it alive. The captured cell became the chloroplast. Over evolutionary time, most of the captive's genes migrated to the host nucleus — a transfer so thorough that modern chloroplasts encode only about 80 proteins from their own genome, while thousands more are imported from nuclear genes. The migration makes energetic sense: a single nuclear copy is cheaper to maintain than organellar copies in every chloroplast. So why did those 80 genes stay?
Allen (arXiv:2512.10588, December 2025) proposes that the answer is physical proximity. The genes that remain in the chloroplast genome encode components of the photosynthetic electron transport chain. These proteins must be produced at rates that match the redox state of the thylakoid membrane — the internal surface where photosynthesis occurs. If photosystem II is producing electrons faster than photosystem I can consume them, the intervening electron carriers become over-reduced, and the chloroplast needs to adjust protein stoichiometry to restore balance.
The adjustment must be fast. Redox imbalances damage the photosynthetic machinery within minutes. A signal sent to the nucleus — translated in the cytoplasm, the protein imported back through the chloroplast envelope — takes too long. The genes stay in the chloroplast because they must be next to the membrane whose redox state they respond to. The Co-Location for Redox Regulation (CoRR) hypothesis: the genome is tethered to the thylakoid not for archival purposes but for regulatory speed.
The tethering is literal. The chloroplast nucleoid — the structure that holds the organellar DNA — is physically anchored to the thylakoid membrane. The electron transport chain senses its own redox state and modulates transcription of the adjacent genes. The feedback loop is spatial: the sensor and the responder share a membrane. Moving the genes to the nucleus would not lose the genetic information — it would lose the geometry of the control circuit.
The same logic applies to mitochondria, which retain genes encoding respiratory chain components for the same reason: redox regulation requires co-location. The two organelles arrived independently (chloroplasts from cyanobacteria, mitochondria from alphaproteobacteria), diverged for a billion years, and converged on the same solution. Both keep the genes that must respond fastest to the membrane they serve.
The evolutionary puzzle — why haven't these genes migrated like the rest? — dissolves once you see the constraint. The question is not why the genes stayed. The question is why any genes left. The ones that could afford the transit time moved to the nucleus. The ones that couldn't are still at the membrane, tethered to the signal they serve, because the speed of response cannot survive the distance.
Allen, "Co-Location for Redox Regulation of gene expression (CoRR): The chloroplast mesosome or nucleoid," arXiv:2512.10588 (December 2025).