Chloroplasts almost certainly originated from an ancient cyanobacterium that was engulfed by a larger cell roughly 2.1 billion years ago. This event, called primary endosymbiosis, is one of the most important moments in the history of life on Earth. Instead of being digested, the cyanobacterium survived inside its host, eventually becoming a permanent internal partner that provided the ability to convert sunlight into energy. Over deep time, that captured bacterium lost its independence and became the chloroplast found in every plant and alga today.
How Endosymbiosis Works
The basic story is surprisingly straightforward. An early eukaryotic cell, one that already had mitochondria and could survive with or without oxygen (a facultative anaerobe), swallowed a photosynthetic cyanobacterium. Rather than breaking it down for food, the host cell kept it alive. The cyanobacterium continued doing what it had always done: harvesting light and producing sugars. The host benefited from that energy, and the cyanobacterium benefited from the protected, nutrient-rich environment inside the host. Over hundreds of millions of years, the two became inseparable.
This wasn’t a common outcome. Cells engulf bacteria constantly through phagocytosis, and they almost always digest them. The establishment of a permanent photosynthetic endosymbiont appears to have succeeded only once. That single event gave rise to all chloroplasts in existence, a conclusion supported by the fact that all primary plastids share a core set of genes, structural features, and metabolic pathways traceable to one cyanobacterial ancestor.
Structural Evidence From Living Cells
When you look at a chloroplast and a cyanobacterium side by side, the family resemblance is hard to miss. Both contain internal flattened membranes called thylakoids, which are the surfaces where light-driven photosynthesis actually happens. Both are enclosed by a double membrane system: in cyanobacteria, these are the inner and outer cell membranes; in chloroplasts, they’re called envelope membranes. The architecture is essentially the same, just renamed.
One of the most striking pieces of physical evidence comes from a small group of freshwater algae called glaucophytes. Their chloroplasts still retain a layer of peptidoglycan, a rigid mesh-like material that sits between the two envelope membranes. Peptidoglycan is the signature structural component of bacterial cell walls, and in glaucophytes, its structure closely resembles the peptidoglycan found in living cyanobacteria. Exposing glaucophyte cells to beta-lactam antibiotics (which target peptidoglycan) blocks chloroplast division, just as it would block bacterial cell division. This is essentially a bacterial wall preserved inside a plant-like cell for over a billion years.
In most other plant lineages, the peptidoglycan layer was assumed to be completely lost. But researchers using high-resolution imaging detected peptidoglycan material between the envelope membranes of moss chloroplasts, and the genes responsible for building peptidoglycan turned out to be essential for chloroplast division in those plants. The wall material hasn’t fully disappeared; it’s just been reduced.
Genetic Evidence: Genes That Moved
If chloroplasts were once free-living bacteria, they should have their own DNA, and they do. Chloroplast genomes are small, circular molecules that look far more like bacterial genomes than like the nuclear DNA of the cells they inhabit. But those genomes are dramatically shrunken compared to the genomes of modern cyanobacteria. The reason: over evolutionary time, most of the original endosymbiont’s genes were transferred to the host cell’s nucleus.
This process, called endosymbiotic gene transfer, left clear fingerprints. Genes like rpl22 (found in legumes) represent textbook cases where a gene that once sat in the chloroplast genome now resides in the nucleus. The protein it encodes is still needed inside the chloroplast, so after being made in the cell’s cytoplasm, it gets shipped back into the chloroplast using a molecular address tag called a transit peptide. Other genes, like infA, appear to have jumped from the chloroplast to the nucleus independently in multiple plant lineages, meaning the transfer process isn’t a one-time event but something that recurs.
This gene migration created a dependency. The chloroplast can no longer survive on its own because it lacks the genetic instructions for many proteins it needs. Those instructions now live in the nucleus, which means the chloroplast is fully integrated into the host cell’s biology.
Shared Division Machinery
Chloroplasts don’t just look like bacteria. They divide like bacteria too. Both cyanobacteria and chloroplasts use a protein called FtsZ, a structural molecule that assembles into a ring at the middle of the cell or organelle and pinches it in two. In cyanobacteria, a single version of FtsZ handles this job. In plant chloroplasts, the system has been split into two cooperating versions: one that stabilizes the ring structure and one that promotes its dynamic behavior. This is exactly the kind of refinement you’d expect from billions of years of co-evolution, where a single ancestral protein’s functions were gradually divided between two specialized copies.
Protein Import: A System Built From Bacterial Parts
Because so many chloroplast genes moved to the nucleus, the cell needed a way to get the resulting proteins back into the chloroplast. The solution is an elaborate transport system embedded in the chloroplast’s two envelope membranes, with channel complexes in the outer membrane and inner membrane working in tandem. The remarkable finding is that the core components of this import system were repurposed from bacterial protein-export machinery that the cyanobacterial ancestor already had.
The main channel in the outer membrane is closely related to a type of bacterial membrane transporter still found in modern cyanobacteria. A key component spanning the inner membrane shares ancestry with another bacterial protein that helps shuttle molecules between a bacterium’s two membranes. Additional inner-membrane channel proteins also trace back to cyanobacterial genes, though they aren’t derived from any known bacterial transport system. In short, the chloroplast’s import system is a hybrid: bacterial transport components were adapted and combined with new host-derived elements to reverse the direction of traffic, pulling proteins in rather than pushing them out.
Primary, Secondary, and Tertiary Endosymbiosis
The original cyanobacterial capture, primary endosymbiosis, produced chloroplasts with two surrounding membranes. Only three lineages carry these primary plastids: glaucophytes, red algae, and green algae (including land plants). Each group’s chloroplasts have distinct features. Glaucophytes retained the peptidoglycan wall. Red algae contain a pigment called phycoerythrin that gives them their color. Green algae and plants use chlorophyll b alongside chlorophyll a and store starch inside the chloroplast.
But photosynthesis spread far beyond these three groups through a process called secondary endosymbiosis, where a non-photosynthetic eukaryote engulfed an entire red or green alga and kept its chloroplast. Most algal lineages got their chloroplasts this way. Euglenids and chlorarachniophytes acquired theirs from green algae. Haptophytes, cryptomonads, brown algae (heterokonts), dinoflagellates, and even the malaria-causing apicomplexans trace their plastids back to a red algal endosymbiont. Secondary plastids are recognizable because they’re surrounded by three or four membranes instead of two, the extra layers being remnants of the engulfed alga’s own membranes.
Some dinoflagellates have gone even further, replacing their original secondary plastid with a new one acquired from another organism that already had a secondary plastid. This tertiary endosymbiosis has produced dinoflagellates carrying plastids derived from cryptomonads, haptophytes, or diatoms. In one case, a dinoflagellate picked up an entirely new green algal endosymbiont, layering yet another chapter onto an already complex evolutionary history.
Open Questions About a Single Origin
The scientific consensus holds that primary endosymbiosis happened once, giving all chloroplasts a single common ancestor. But some researchers have raised questions about whether the picture is quite that simple. An analysis of chloroplast enzymes, particularly those involved in building chloroplast membranes, found that these enzymes trace back to surprisingly diverse sources and appear to have been acquired at different times rather than all at once. Membranes can’t replicate on their own the way DNA can, so the assembly of a functional chloroplast membrane system may have involved genetic contributions from multiple bacterial donors over an extended period.
These findings don’t overturn endosymbiotic theory, but they suggest the transition from free-living cyanobacterium to fully integrated organelle may have been messier than the clean “one cell swallowed another” narrative implies. Alternative hypotheses propose that the host cell may have played a more active role in constructing chloroplast membranes using genes gathered from various bacterial sources. The core event was still endosymbiosis, but the details of how chloroplasts were assembled and refined likely involved layers of genetic exchange that unfolded over a long stretch of evolutionary time.