The complex cell

Foundations

Something in your cells does not belong to you

Inside every one of your cells are several hundred small compartments that burn food with oxygen. They are called mitochondria, and they are peculiar in ways that took biology a century to explain.

They have their own DNA, separate from the DNA in your nucleus, arranged in a circle. Circular DNA is a bacterial arrangement. They have their own ribosomes, the machines that build proteins, and those ribosomes are bacterial-type rather than the type in the rest of your cell, which is why some antibiotics that target bacterial ribosomes also cause side effects in humans. They are wrapped in two membranes rather than one. And they divide on their own schedule, by splitting in half, rather than being manufactured by the cell.

Every one of those facts is odd for a cell component. Together they are not odd at all, if the component used to be a separate organism.

The claim nobody believed

In 1967 a young researcher named Lynn Sagan, later Lynn Margulis, published a paper arguing exactly that: mitochondria and chloroplasts were once free-living bacteria that took up residence inside another cell and never left.

The paper was reportedly rejected by around fifteen journals before it was published. The idea was considered fringe. Within twenty years, once it became possible to read genetic sequences, it was considered obvious. The closest living relatives of your mitochondria are a specific group of bacteria; the closest relatives of a plant’s chloroplasts are cyanobacteria, the same organisms from lesson 5.4.

The merger is no longer in question. Almost everything else about it still is.

Why this is a puzzle and not just a story

Evolution repeats itself constantly. Eyes have evolved independently dozens of times. Flight evolved at least four times, in insects, pterosaurs, birds and bats. Multicellularity evolved independently more than twenty times. Photosynthesis of one kind or another evolved several times.

The complex cell appears to have evolved once.

Bacteria and archaea had been running the planet for at least 1.7 billion years before it happened, in astronomical numbers, with generation times of minutes and a habit of swapping genes freely. They had every opportunity. And in all that time, and in the two billion years since, nothing else has independently produced a cell with a nucleus and internal compartments.

That singularity is the central fact of this lesson, and any explanation has to account for it. Something about this step was not merely difficult but nearly impossible.

Who the host was

For a long time the host cell was assumed to be some sort of primitive proto-eukaryote. Genetics said otherwise: the host belonged to the archaea, the third domain of life, which look superficially like bacteria under a microscope and are in fact more closely related to us than to bacteria.

Then, in 2015, a team analysing DNA from sediment near a hydrothermal field in the Arctic deep sea, at a site called Loki’s Castle, assembled the genome of an archaeon nobody had ever cultured. It carried genes that were supposed to be eukaryotic exclusives: genes for reshaping membranes, for an internal skeleton, for trafficking vesicles.

They named it Lokiarchaeum, and the group that followed took names from the same mythology: Thor, Odin, Heimdall. Collectively, the Asgard archaea. They are the closest known relatives of whatever cell hosted the merger.

In 2020 a Japanese team finally grew one in a laboratory. It took more than twelve years of patient culturing, and the organism divides roughly every fourteen to twenty-five days. It is small, it lives in partnership with other microbes, and it puts out long branching arms. Whether those arms are how the original engulfment happened is speculation, and it is very tempting speculation.

What the deal actually was

The intuitive story is that a cell swallowed a bacterium that was good at using oxygen, and got a power plant. That may be backwards.

Under the leading alternative, the hydrogen hypothesis, the partnership started as a metabolic trade in an oxygen-poor world. The archaeal host needed hydrogen. The bacterium produced hydrogen as waste. Living close together was useful to both, then living inside was better, and only later, as oxygen accumulated, did the relationship become about respiration.

If that is right, then the defining feature of complex life began as a waste-disposal arrangement between two microbes, which is a satisfying rhyme with the previous lesson.

Why it mattered so much

Nick Lane’s argument for why this one merger changed everything is about energy budgets, and it is worth following.

A bacterium generates energy across its outer membrane. As it gets bigger, its volume grows faster than its surface area, so its power supply falls behind its needs. That caps how large and how genetically complex a bacterium can be.

A eukaryote has hundreds or thousands of mitochondria, each with its own heavily folded internal membrane, all generating power inside the cell. Energy generation scales with the number of mitochondria rather than with the outside of the cell. On this accounting, a eukaryote has something like four to five orders of magnitude more energy available per gene than a prokaryote.

That is what buys a large genome. And a large genome is what buys the ability to be a nerve cell one day and a liver cell the next, to build a body, to specialise. Every organism you can see without a microscope is downstream of that energy budget.

The second merger

It happened again, at least once. Somewhere in an early eukaryote’s history, a cyanobacterium was captured and kept, and became the chloroplast. That is the origin of every plant and alga, and it is why the water-splitting machine from the last lesson is now inside a leaf.

Some algae then did it a third time, capturing an alga that had already captured a cyanobacterium, which leaves certain organisms with four membranes around their photosynthetic compartment, one set for each round. Evolution keeps the receipts.

Deeper

The evidence for bacterial ancestry

Beyond structural features, the decisive evidence is phylogenetic. Mitochondrial genes for ribosomal RNA and for core respiratory proteins place mitochondria within the alphaproteobacteria, though the precise sister group remains contested and recent analyses favour a lineage outside the sampled alphaproteobacteria rather than within a known order. Plastid genes place chloroplasts within the cyanobacteria. Both organelles have lost the vast majority of their original genes: a human mitochondrion retains 37, against a few thousand in a free-living relative. Most of the rest were transferred to the nuclear genome rather than discarded, which is itself strong evidence of the original merger.

Endosymbiotic gene transfer

The migration of genes from organelle to nucleus is ongoing and observable. Fragments of mitochondrial DNA are found integrated into nuclear genomes across eukaryotes, including recent insertions in humans. This transfer explains why organelles cannot live independently: they are not merely reduced, they are hostages, dependent on proteins encoded in the nucleus and imported. It also creates a control problem, because the cell must coordinate two genomes with different inheritance patterns.

Why some genes stayed

A small core of genes remains in the mitochondrion in every eukaryote studied, and they are consistently genes for core respiratory chain components. The leading explanation is that these must be regulated locally and rapidly in response to membrane potential, which is difficult to do from the nucleus. This constraint may explain why complete gene transfer never occurs and why organelles remain semi-autonomous rather than fully absorbed.

Energy per gene, and its critics

Lane and Martin’s calculation compares the energy available per gene in a prokaryote and a eukaryote by normalising metabolic rate to genome size, arriving at a difference of roughly four to five orders of magnitude. The argument has been influential and is not universally accepted: critics point out that the comparison depends on choices about which cells to compare and how to treat scaling, and that some giant bacteria achieve large sizes with polyploid genomes distributed around the cell. What is not disputed is that no prokaryote has independently achieved eukaryote-like structural complexity.

The order of events

Whether the nucleus, the endomembrane system and phagocytosis preceded the mitochondrial merger or followed it is genuinely unresolved and is the main axis of disagreement in the field. Asgard genomes carry many of the relevant gene families, which supports substantial pre-existing complexity in the host and makes engulfment plausible. Alternative models propose that the association began between cells that could not phagocytose, mediated by membrane protrusions or syntrophy, in which case eukaryotic complexity was a consequence rather than a precondition.

Frontier

Was it really once

The claim rests on the observation that all known eukaryotes descend from a single common ancestor that already possessed mitochondria, including lineages that later lost aerobic respiration but retain reduced mitochondrion-derived organelles. No eukaryote lacking any mitochondrial derivative has been confirmed. That is strong but is an argument from what has been sampled, and deep-sea and subsurface environments continue to yield surprises.

Growing the ancestors

Culturing Asgard archaea is the current bottleneck. One species grown after twelve years, dividing on a timescale of weeks, is not enough to test hypotheses about cell biology directly. More cultures, and imaging of how these cells behave with their syntrophic partners, would settle questions that genomics alone cannot.

Transcript and shot list

Target runtime about 8 minutes. Built as a detective story: a set of suspicious clues inside the viewer’s own cells, a hypothesis that was rejected, and a culprit eventually found in deep-sea mud. Shot types: ANIM is rendered from code with Manim, AI is a generated shot from Higgsfield with the shared style reference, REAL is agency imagery with credit, CAM is the narrator.

[00:00] S01 · ANIM · Zoom into a human cell; several hundred mitochondria highlighted; four clues appear as labels: own DNA, own ribosomes, two membranes, divides itself.

There is something inside every one of your cells that does not belong to you. A few hundred small compartments that burn food with oxygen. And they are strange in four specific ways that took biology a century to explain.

[00:25] S02 · ANIM · Each clue examined: a circle of DNA beside the cell’s linear chromosomes; a bacterial-type ribosome beside a eukaryotic one; a double membrane; a mitochondrion pinching in two.

They have their own DNA, separate from yours, arranged in a circle. Circular DNA is a bacterial arrangement. They have their own protein-building machines, and those are bacterial-type, not the type in the rest of your cell, which is why some antibiotics aimed at bacteria cause side effects in people. They have two membranes instead of one. And they divide on their own schedule, by splitting, rather than being manufactured by the cell. Every one of those is odd for a cell part. Together, they are not odd at all, if the part used to be a separate organism.

[01:15] S03 · ANIM · Typographic: 1967, Lynn Sagan, “On the origin of mitosing cells”; a stack of rejection letters; then a portrait card.

In 1967 a young researcher called Lynn Sagan, later Lynn Margulis, published exactly that claim: mitochondria and chloroplasts were once free-living bacteria that moved in and never left. Her paper was reportedly turned down by around fifteen journals. The idea was considered fringe. Within twenty years, once anyone could read genetic sequences, it was considered obvious.

[01:50] S04 · ANIM · A phylogenetic tree: mitochondria nesting inside the alphaproteobacteria, chloroplasts inside the cyanobacteria.

The closest living relatives of your mitochondria are a specific group of bacteria. The closest relatives of a plant’s chloroplasts are cyanobacteria, the same organisms that poisoned the world in the last lesson. The merger is not in question any more. Almost everything else about it still is.

[02:20] S05 · ANIM · A counter of independent origins: eyes dozens of times, flight four times, multicellularity twenty-plus times, photosynthesis several times; then the complex cell, counter reading ONE.

Here is the puzzle, and it is a real one. Evolution repeats itself constantly. Eyes have evolved independently dozens of times. Flight at least four. Multicellularity more than twenty. The complex cell appears to have evolved once. Bacteria and archaea had been running this planet for at least one point seven billion years before it happened, in astronomical numbers, with generation times of minutes, swapping genes freely. They had every opportunity. And in all that time, and in the two billion years since, nothing has done it again.

[03:05] S06 · ANIM · The three domains; the host cell moving from the bacterial branch to the archaeal branch.

So who was the host? For a long time people assumed some sort of primitive proto-eukaryote. Genetics said otherwise. The host was an archaeon: the third domain of life, which looks like a bacterium under a microscope and is actually more closely related to you than to one.

[03:35] S07 · REAL · The Loki’s Castle hydrothermal field on the Arctic seafloor (research image; confirm credit line). S08 · ANIM · Sediment cores; metagenomic assembly; a genome appearing with eukaryotic-signature genes highlighted.

Then, in 2015, a team sequencing DNA from sediment near a hydrothermal field in the Arctic deep sea, at a place called Loki’s Castle, assembled the genome of an archaeon nobody had ever grown. It carried genes that were supposed to be eukaryote-only: for reshaping membranes, for an internal skeleton, for moving cargo around inside a cell. They called it Lokiarchaeum. The relatives that followed got names from the same mythology. Thor. Odin. Heimdall. Together, the Asgard archaea, and they are the closest known relatives of whatever hosted the merger.

[04:20] S09 · ANIM · A culture vessel; a twelve-year counter; a small cell with long branching protrusions reaching around a partner microbe.

In 2020 a Japanese team finally grew one. It took more than twelve years of patient culturing, and the thing divides about once every two to three weeks. It is small, it lives in partnership with other microbes, and it puts out long branching arms. Whether those arms are how the original engulfment happened is speculation. It is very tempting speculation.

[04:55] S10 · ANIM · The intuitive story: a cell swallows an oxygen-using bacterium. Then it reverses: an archaeon consuming hydrogen produced as waste by a neighbouring bacterium; the two moving closer; then inside.

Now, what was the deal? The intuitive story is that a cell swallowed a bacterium that was good with oxygen and got itself a power plant. That may be backwards. Under the leading alternative, the partnership started as a metabolic trade in a world with very little oxygen. The archaeon needed hydrogen. The bacterium produced hydrogen as waste. Living close was useful to both. Living inside was better. And only later, as oxygen built up, did the relationship become about breathing. Which means the defining feature of complex life may have begun as a waste-disposal arrangement between two microbes.

[05:40] S11 · ANIM · A bacterium growing: volume rising as the cube, surface area as the square; power supply falling behind. Then a eukaryote with hundreds of internally folded mitochondria; a bar chart of energy available per gene, four to five orders of magnitude apart.

And why did it matter so much? Energy. A bacterium makes its energy across its outer membrane. As it gets bigger, volume grows faster than surface area, so the power supply falls behind the demand. That caps how big and how genetically complex a bacterium can get. A eukaryote has hundreds or thousands of mitochondria, each with its own heavily folded membrane, generating power inside the cell. On one influential calculation, that gives a eukaryote something like ten thousand to a hundred thousand times more energy available per gene. That is what buys a big genome. And a big genome is what buys being a nerve cell one day and a liver cell the next.

[06:35] S12 · ANIM · A second capture: a cyanobacterium becoming a chloroplast in an early alga; then a third capture, an alga inside an alga, leaving four membranes.

It happened again, at least once. Somewhere in an early eukaryote, a cyanobacterium was captured and kept, and became the chloroplast. That is the origin of every plant and every alga, and it is why the water-splitting machine from the last lesson now sits inside a leaf. And some algae went a third round, capturing an alga that had already captured a cyanobacterium, which leaves them with four membranes around the thing, one set per round. Evolution keeps the receipts.

[07:15] S13 · ANIM · The unresolved question: two competing orderings, nucleus-first and mitochondrion-first, shown side by side with an open verdict.

What is genuinely unresolved is the order. Did the nucleus and the ability to engulf things come first, making the merger possible? Or did the merger come first and drive the rest? The Asgard genomes carry many of the relevant genes, which supports a fairly complex host. But nobody has settled it, and the two orderings imply quite different stories about how difficult this step really was.

[07:45] S14 · AI into ANIM · A mitochondrion in a living cell, moving; pull back to the calendar.

So: one merger, once, about two billion years ago, between an archaeon and a bacterium that were probably just trading waste gases. Everything you have ever seen without a microscope is downstream of it. Next lesson: what happened when those cells finally started building bodies.

Shot list

One AI shot, twelve code-rendered shots, one research image. Estimated Higgsfield use: 2 to 4 generations including retries.