How Science Works | How One Cell Became a Permanent Partner Inside Another

Cells can contain other organisms for many reasons. A captured cell may be digested. An intracellular parasite may reproduce at its host’s expense. A symbiont may exchange useful metabolites and persist across host generations. These outcomes must be distinguished rather than placed automatically on a ladder towards mitochondria.

An organelle-origin explanation needs to account for stable inheritance and integration. How is the internal population maintained? How do useful molecules move between partners? What prevents the arrangement from breaking down? How does the host’s reproduction reliably transmit the internal component to descendants?

There is no need to imagine that two cells recognised a future advantage and agreed to cooperate. Variants of an association can differ in survival and reproduction. Selection, genetic drift and changing ecological conditions can alter which associations persist. Dependence can increase when functions are lost or redistributed, even if the original relationship was very different from the later one.

“Permanent partner” is therefore shorthand for a stable evolutionary outcome, not a promise that every descendant retains exactly the same arrangement forever. Organelles can become highly modified, and particular lineages can lose functions or even lose an organelle under unusual circumstances.

For a simple conceptual diagram, show three distinct states: temporary internalisation, persistent association and integrated inheritance. Put a question mark over the arrows. The drawing identifies problems a historical explanation must solve; it does not claim that every modern intracellular association necessarily follows this route or that we know the exact order of all transitions.

Contents · Next chapter

Mitochondria and plastids retain genetic systems with a bacterial history. Comparing their genes and proteins with those of bacteria and eukaryotic hosts reveals patterns of relatedness. These comparisons are more informative than observing that both a bacterium and a mitochondrion look small and rounded.

Mitochondrial ancestry is closely associated with the alphaproteobacterial branch of bacterial evolution. Precisely where mitochondria attach relative to living and sampled bacterial lineages is harder to resolve. A 2018 phylogenomic study and an expanded 2022 analysis illustrate how additional sampling and models of sequence evolution can affect that placement.

This distinction is important: disagreement over the nearest branch does not erase the broad evidence for bacterial ancestry. A family relationship can be secure while the identity of the closest surviving cousin remains uncertain. The exact ancestral population may also be extinct or unsampled.

A phylogenetic tree is an inference from comparisons under a model of how sequences change. Very rapidly evolving sequences, biased amino-acid compositions, missing lineages and gene transfers can mislead an analysis. Researchers therefore examine multiple genes, alternative models and the effect of adding or removing samples.

The lesson is to ask what level of claim the evidence supports. “This organelle has bacterial ancestry” is broader than “this named modern species is its direct ancestor”. The latter is usually the wrong way to read the tree. Modern bacteria have continued evolving too; they are comparative relatives, not preserved copies of an ancient partner.

Contents · Previous chapter · Next chapter

Chloroplasts are photosynthetic members of the plastid family. Their ancestry connects them with cyanobacteria, bacteria capable of oxygen-producing photosynthesis. This is a different bacterial contribution from the lineage associated with mitochondria.

Genomic evidence helps reconstruct that relationship. The Cyanophora paradoxa genome study examined a strategically informative algal lineage and supported a shared primary-plastid origin in the major group containing plants and related algae. A comparative cyanobacterial study investigated where the plastid ancestor lay within cyanobacterial diversity.

Primary endosymbiosis refers here to the establishment of a bacterial-derived plastid inside a eukaryotic host. Later evolutionary events can move a plastid into another host through the incorporation of an already photosynthetic eukaryotic cell. Those secondary and more complex histories help explain why algal cell architecture is more varied than a single textbook chloroplast diagram suggests.

A plant cell consequently carries evidence of more than one major evolutionary integration. Its mitochondria and plastids do not represent two interchangeable versions of the same ancient acquisition. Each has its own ancestry and subsequent history, while both now cooperate with the nuclear genome and the rest of the cell.

Do not interpret photosynthesis itself as sufficient proof of a chloroplast. Cyanobacteria photosynthesise without possessing plant-style chloroplasts, and many plastids have functions other than photosynthesis. The historical question concerns the ancestry of a cellular compartment and its machinery, not merely whether a cell is green or can capture light energy.

Contents · Previous chapter · Next chapter

Mitochondria and plastids have features that make sense in light of bacterial ancestry: their own genetic material in many familiar examples, internal systems for gene expression, membranes and reproduction through growth and division of pre-existing organelles. These observations fit a history of integration from cellular ancestors.

The NCBI account of organelle genetic systems explains how organelle construction requires contributions from both organelle and nuclear genomes. This divided genetic responsibility is especially revealing. An organelle contains machinery with its own evolutionary history while depending heavily on products made under host nuclear control.

No single physical trait is a decisive test by itself. Many cellular structures have membranes. A roughly bacterial size can arise for different reasons. Genome shapes and sizes vary, and modern mitochondrial ribosomes have undergone extensive changes. “Circular DNA plus two membranes equals proof” is too simple to carry the full argument.

Membrane count is particularly easy to overread. Membranes can be lost, modified or added during complex endosymbiotic histories. One should not assign every present membrane to an exact ancient source purely by counting rings in a cartoon. Biochemical composition, molecular machinery and comparative history provide additional constraints.

Imagine finding a previously unstudied intracellular body with a surrounding membrane. That observation earns the description “membrane-bound structure”. Finding an internal genome earns another claim. Demonstrating that multiple genes consistently relate it to a bacterial lineage earns a stronger historical inference. The strength comes from the combined pattern, not from upgrading the first striking resemblance into a complete explanation.

Contents · Previous chapter · Next chapter

Over evolutionary time, many genes from endosymbiotic ancestors have been lost or transferred to the host’s nuclear genome. Loss and transfer are different outcomes. A missing organelle gene does not automatically mean its entire function disappeared, and a gene in the nucleus does not automatically mean its product works only in the nucleus.

DNA movement from organelles to the nucleus can still occur. Direct measurements of chloroplast DNA transfer in tobacco demonstrated that this kind of transfer is an observable biological process. Such experiments test a component of the evolutionary account; they do not recreate the whole origin of a plastid.

A DNA fragment arriving in the nucleus is not yet a complete functional relocation. For a transferred gene to replace an organelle copy, it must be expressed appropriately, and its product must reach wherever it is needed. Regulatory compatibility and protein targeting therefore become part of the evolutionary problem.

Some transferred sequences may become inactive fragments. Some genes may acquire functions elsewhere in the cell. Other functions may be replaced by proteins with different histories. Modern organelles are consequently evolutionary mixtures, rather than intact ancient bacteria whose entire instruction manuals were merely moved to a new shelf.

A useful diagram places “gene location” and “protein destination” in separate columns. Draw an arrow from a nuclear gene to a protein made in the cytoplasm, then to its working location inside an organelle. That arrow makes the central distinction visible: the address of the DNA and the address of the protein’s activity can differ. The cell must maintain the connection between them.

Contents · Previous chapter · Next chapter

Most mitochondrial proteins in familiar model systems are encoded by nuclear genes and made outside the mitochondrion. They must be recognised and delivered to appropriate mitochondrial locations. Import complexes and targeting information connect nuclear control with organelle construction and function.

The TOM complex provides a major entry route across the outer mitochondrial membrane. Structural research on the mitochondrial import gate reveals channels and associated components that help explain how precursor proteins enter. Additional machinery directs proteins towards different internal destinations. One gate is not the whole import pathway.

Plastids have their own import systems, and the machinery is not simply identical to that of mitochondria. The common evolutionary problem is how proteins produced under host genetic control reach an internal compartment whose functions depend on them. The particular solution must be studied in its own lineage.

This creates a strong kind of dependence. An organelle that has lost the genes for essential components cannot necessarily live independently, even if it retains DNA and bacterial-derived machinery. Having a genome is not the same thing as possessing all the capacities of a free-living cell.

Protein import also sharpens the evidence for integration in more recently evolved systems. Nowack and Grossman’s study of Paulinella examined trafficking of nucleus-encoded proteins into its photosynthetic compartment. Paulinella provides a valuable comparison with the older plastids of plants and algae. It is not a surviving intermediate on the direct ancestral route to a modern plant chloroplast.

Contents · Previous chapter · Next chapter

Original hypothetical model, not an ancestral genome reconstruction. Begin with 100 labelled gene families in an imagined bacterial partner. In a later descendant, suppose 15 remain encoded in the organelle, 35 have functional nuclear copies whose products return to the organelle, and 50 have been lost without being included in our later functional set.

The ancestral bookkeeping closes: 15 + 35 + 50 = 100. Now suppose the host contributes 25 additional nuclear gene families to the organelle’s functioning. Our deliberately defined present-day set then contains 15 + 35 + 25 = 75 gene families.

Only 15 of those 75 are encoded locally in the organelle: 20%. The other 60 are nuclear-encoded: 80%. Yet 50 of the 75 families in the model have ancestry in the bacterial partner, because we counted both the 15 retained and the 35 relocated families. Gene location and evolutionary origin therefore give different classifications.

These percentages describe this invented account only. Real genomes contain duplications, replacements, transfers, losses and uncertain histories. Gene families also do not translate one-to-one into equal contributions to a cell’s function. We cannot infer an actual organelle’s evolutionary composition from these chosen numbers.

The model does identify an error to avoid. Looking only at genes still inside an organelle can underestimate the genetic contribution of its ancestor. Conversely, classifying every nuclear gene used by an organelle as transferred from its ancestor can overestimate that contribution, because host-derived proteins can be recruited too.

For a paper activity, use 100 numbered cards for the initial families and sort them into the three outcomes. Add 25 differently coloured host cards. Ask two questions separately: “Where is the gene now?” and “Which ancestral group did this model assign it to?” No biological samples or experimental manipulation are needed.

Contents · Previous chapter · Next chapter

Consider two explanations for a particular organelle. One says it arose through integration of a bacterial lineage. Another says it developed entirely within the host without such an ancestral cellular contribution. The useful test asks which observations should be more expected under each account.

Bacterial-related organelle genes, characteristic molecular machinery and functional dependence through nuclear contributions form a connected pattern. Endosymbiotic ancestry provides a common historical explanation for that pattern. A purely internal-origin account would need to explain those bacterial relationships through additional events and mechanisms.

However, not every bacterial-looking gene in a eukaryotic nucleus establishes an endosymbiotic event. Individual genes can move horizontally, contaminants can enter sequence datasets, and phylogenetic inference can be wrong. The argument becomes stronger when multiple lines of evidence converge on an integrated compartment and a coherent evolutionary relationship.

For an original evidence exercise, imagine three findings: a membrane-bound body, one gene resembling a bacterial sequence, and demonstrated import of host-produced proteins into that body. Each supports a different part of the story. None alone identifies the ancient donor, dates the integration or proves every proposed stage.

Now add many well-supported gene relationships and consistent inheritance of the body through host reproduction. The historical case improves because more of the proposed system has an observable counterpart. The investigator should still ask whether those observations are independent. Ten genes copied together in one transfer do not necessarily supply ten independent historical events.

This is how historical biology tests explanations without rerunning ancient Earth. Present-day descendants retain records, and those records can make competing histories more or less plausible. The Evolutionary Biology guide develops that logic through common descent, phylogenetic models and the limits of molecular clocks.

Contents · Previous chapter · Next chapter

Living associations let researchers examine problems that ancient organelle origins also had to solve: exchange, inheritance, protein targeting and coordination. They supply comparisons and mechanisms rather than a literal recording of the first mitochondrion.

A recent example is the 2024 study of a nitrogen-fixing organelle in a marine alga. The researchers analysed the integration of a cyanobacterial-derived compartment, including host-related control and protein contributions. Its importance lies in evidence about organelle-level integration. It does not mean that a modern alga is replaying the exact historical origin of a chloroplast or that every nitrogen-fixing symbiont is an organelle.

Loss provides another informative comparison. Research on Monocercomonoides identified a eukaryote lacking a mitochondrial organelle, with an alternative system handling an important function. Such a lineage is understood in terms of secondary evolutionary loss, rather than simply being a preserved representative of a stage before mitochondrial ancestry.

These examples warn against frozen definitions based on one familiar organism. An organelle can change dramatically after its origin. A reduced structure may preserve a historical relationship even after losing a feature we associate with a textbook mitochondrion. Conversely, retaining a small genome does not automatically settle where to draw every boundary between a symbiont and an organelle.

The most useful questions are functional and historical together: what is inherited, which genes contribute, where proteins are made, how material crosses boundaries, and how independently each component can persist? The label becomes more informative when attached to that evidence.

Contents · Previous chapter · Next chapter

The endosymbiotic bacterial ancestry of mitochondria and plastids is a central conclusion supported by converging molecular and cellular evidence. That conclusion is broader and firmer than any single detailed scenario for their initial establishment.

Researchers continue to investigate the mitochondrial host’s nature, the ecological relationship between partners, the sequence of cellular innovations and the placement of the ancestral bacterium relative to sampled lineages. An image of a fully modern-looking cell swallowing a bacterium can be a teaching cartoon, but it should not silently settle those questions. In particular, the timing of mitochondria relative to other features of eukaryotic complexity needs evidence rather than an assumed storyboard.

Draw a final diagram with two separate histories: one bacterial contribution leading to mitochondria, and a cyanobacterial contribution leading to primary plastids in their host lineage. Add arrows for gene transfer and protein import, pointing in their appropriate directions. Include a note that the diagram omits intermediate states and later secondary endosymbioses.

Does every cell inside another become an organelle? No. Persistence and integration require additional evolutionary changes. Does an organelle genome prove independence? No. Many essential contributions can come from the nucleus. Does one matching gene prove the entire history? No. Alternatives and wider evidence must be tested. Do unresolved early details undermine bacterial ancestry? No. Different claims can have different levels of support.

The core story is neither an act of foresight nor a single moment of engulfment. It is the evolution of a relationship in which inheritance, metabolism and genetic control became increasingly connected. Return to Evolutionary Biology for the historical framework or How Science Works for the wider evidence-building process.

Sources and example boundaries

The linked primary studies address phylogenetic placement, plastid ancestry, DNA transfer, protein import and modern examples of integration or loss. Textbook material supplies basic organelle terminology. The 100-family account, card exercise and comparison scenarios are original teaching constructions. They are not measured gene counts, reconstructed ancestral genomes, dates or experimental protocols.

Contents · Previous chapter · Continue to Evolutionary Biology