Is There Endoplasmic Reticulum in Prokaryotic Cells?

Prokaryotic cells do not have an endoplasmic reticulum. The ER, with its sprawling network of membrane sheets and tubules dedicated to protein folding, lipid synthesis, and intracellular transport, is a hallmark of eukaryotic life. But the clean textbook distinction between “simple prokaryotes” and “complex eukaryotes” has eroded considerably in recent years, as researchers have discovered that many bacteria and archaea possess internal membrane systems, some of which look and behave in ways that are strikingly reminiscent of the ER.

Why Prokaryotes Were Long Considered Too Simple for Internal Membranes

For most of the twentieth century, biology textbooks drew a sharp line: eukaryotic cells have membrane-bound organelles, and prokaryotic cells do not. The prokaryotic cell was depicted as a bag of cytoplasm enclosed by a plasma membrane (and often a cell wall), with its DNA floating freely rather than tucked inside a nucleus. The endoplasmic reticulum, Golgi apparatus, lysosomes, and other membrane-bound compartments were treated as exclusively eukaryotic innovations.

This view was not baseless. Prokaryotic cells are, on average, much smaller than eukaryotic cells, and the majority of well-studied species in the mid-twentieth century really did appear to lack visible internal membranes. The problem was that researchers generalized from a handful of model organisms and assumed the rest of the prokaryotic world matched. As more diverse prokaryotes were examined with better tools, that assumption started to fall apart. Research over the past two decades has made clear that compartmentalization through both protein-bounded and lipid-bounded organelles is widespread in prokaryotes, challenging the old dogma that these cells are structurally simple and disorganized.¹

The Mesosome Debacle

The history of prokaryotic internal membranes includes one embarrassing chapter that deserves attention, because it shaped how cautiously biologists approached the topic for decades afterward. In the 1960s and 1970s, electron microscopy revealed what appeared to be elaborate membrane infoldings inside Gram-positive bacteria. These structures were named “mesosomes,” and for a time they were treated as a genuine prokaryotic organelle, potentially involved in DNA replication and cell division.

The excitement did not last. By the late 1970s, careful experiments demonstrated that mesosomes were artifacts of the chemical fixation process used to prepare samples for electron microscopy.² The harsh fixatives caused the plasma membrane to buckle inward, creating structures that did not exist in living cells. Further work confirmed that mesosomes appeared consistently across different bacterial species when the same fixation protocols were used, making the case for artifact even stronger.³ The mesosome episode left a lasting impression on the field: researchers became reluctant to claim that any prokaryote had genuine internal membranes, and the textbook narrative of “prokaryotes are simple” hardened further.

That reluctance turned out to be overcorrective. The mesosomes were indeed fake, but the lesson that some researchers drew from the episode, that prokaryotic internal membranes in general were suspect, was too broad. Genuine internal membrane systems in prokaryotes have since been documented using live-cell imaging and cryo-electron tomography, techniques that do not involve the chemical fixation that created mesosomes in the first place.

Internal Membranes That Prokaryotes Actually Have

Several groups of bacteria maintain elaborate internal membrane systems that are unambiguously real. These are not ER, but they demonstrate that prokaryotes can and do build lipid-bounded compartments inside their cells.

Among the best studied are the intracytoplasmic membranes (ICMs) found in photosynthetic bacteria. Purple nonsulfur bacteria, for example, develop extensive ICMs that house their photosynthetic machinery. Live-cell imaging of 14 species across nine genera revealed that while many species elaborate ICMs throughout the entire cell, species in at least two genera restrict these membranes to discrete, non-random sites near the cell poles, coordinated with cell growth and division.⁴ That level of spatial organization is exactly the kind of thing that was once considered a eukaryotic trait.

Cyanobacteria, the oxygen-producing photosynthetic prokaryotes, contain thylakoid membranes that are functionally analogous to the thylakoids inside chloroplasts. Methane-oxidizing bacteria (methanotrophs) develop stacked ICMs that house particulate methane monooxygenase, the enzyme responsible for their ability to consume methane. These ICMs are derived from the inner cell membrane, and cryo-electron tomography has shown that the enzyme assembles into ordered hexagonal arrays within them.⁵ In some methanotroph species, ICM formation is tunable: the extent of membrane development increases with copper concentration in the growth medium, plateauing once the membranes reach maximum capacity.⁶ Magnetotactic bacteria provide yet another example, forming magnetosomes: membrane-bounded compartments that begin as invaginations of the inner membrane and go on to house iron mineral crystals used for navigation along magnetic field lines.⁷

None of these are ER. They lack the characteristic ER functions of general protein folding, glycosylation, and lipid synthesis for distant membranes. But they show that the ability to form and maintain lipid-bounded internal compartments is not something prokaryotes are fundamentally incapable of. The raw cellular machinery for building internal membranes exists in the prokaryotic world; it is just deployed differently.

The Closest Thing to ER Found in a Bacterium

If any prokaryote comes close to possessing something like an ER, it is the planctomycete Gemmata obscuriglobus. This unusual bacterium, a member of the PVC superphylum, has an internal organization that made headlines precisely because it mirrors features of eukaryotic cells.

Three-dimensional reconstructions of Gemmata using electron tomography revealed a membrane-enclosed nuclear body containing the cell’s DNA, along with additional compartments defined by internal membranes.⁸ Even more provocative, researchers performing cryo-electron tomography on related planctomycete species observed ribosome-studded extended membrane sheets inside the cell, directly resembling eukaryotic rough ER. The lumen of these sheets averaged about 20 nanometers in width, slightly narrower than the roughly 30 nanometers measured in yeast ER sheets and 50 nanometers in mammalian ER sheets.⁹

Ribosome-covered membrane sheets are one of the defining structural features of rough ER in eukaryotes, where they serve as the site where secreted and membrane proteins are translated and inserted into the membrane. Finding a structurally similar arrangement in a bacterium raises obvious questions. Is this an independent evolutionary invention? A relic of an ancient shared ancestor? Or convergent evolution driven by similar functional pressures? These questions remain open, and the planctomycete endomembrane system is still being characterized. But it is the strongest example yet that the structural blueprint for ER-like membranes is not inherently restricted to eukaryotic cells.

Shared Molecular Machinery at the Membrane

Even in ordinary bacteria that lack visible internal membrane compartments, the plasma membrane performs several tasks that the ER handles in eukaryotic cells. Protein translocation across membranes is one of the clearest examples, and the molecular machinery responsible turns out to be deeply conserved across all domains of life.

In eukaryotes, newly synthesized proteins destined for secretion or membrane insertion are threaded through the Sec61 complex embedded in the ER membrane. In bacteria, the equivalent job is done by the SecYEG complex in the inner membrane. These are not merely analogous systems that evolved separately to solve similar problems; they are homologous, descended from a shared ancestor. The SecYEG translocon in bacterial membranes and the Sec61 translocon in ER membranes are both three-subunit protein-conducting channels that represent an evolutionarily conserved mechanism for delivering proteins to compartments outside the cytoplasm.¹⁰

The targeting process is also conserved. In both bacteria and eukaryotes, ribosomes translating membrane proteins are recognized by a signal recognition particle (SRP) and directed to the translocon at the membrane. In bacteria, the SRP receptor FtsY guides the ribosome-nascent chain complex to the SecYEG translocon at the inner membrane, where the growing protein is inserted into the lipid bilayer.¹¹ In eukaryotes, the SRP receptor sits on the ER membrane and routes ribosomes to Sec61. The bacterial inner membrane, in this functional sense, is doing ER work. It just does it without the extensive sheet-and-tubule architecture.

Visualization studies have shown that SecYEG complexes in bacteria are not randomly scattered across the membrane; they cluster into defined zones.¹² After proteins pass through the translocon, quality control in the bacterial periplasm depends on chaperones, folding catalysts, and proteases, a streamlined parallel to the ER’s own elaborate protein-folding quality control system.¹³

Protein Shells as an Alternative to Membranes

Not all prokaryotic compartments are bounded by lipid membranes. Many bacteria have evolved protein-based shells that enclose metabolic enzymes, creating microenvironments that improve reaction efficiency. These are called bacterial microcompartments (BMCs), and they represent a fundamentally different solution to the problem of intracellular organization.

The best-known BMCs are carboxysomes, found in cyanobacteria and some chemoautotrophs, which concentrate carbon-fixing enzymes inside a selectively permeable protein shell. The shell lets in substrates while keeping intermediates from leaking out, boosting the efficiency of carbon fixation in a way that would be impossible without compartmentalization.¹⁴ Other BMCs handle reactions involved in breaking down compounds like propanediol and ethanolamine, sequestering toxic aldehyde intermediates so they do not damage the rest of the cell.

Anammox bacteria take a different approach: they build a membrane-bounded compartment called the anammoxosome using unusual lipids with concatenated cyclobutane rings, known as ladderane lipids, which create an unusually dense and impermeable membrane.¹⁵ The anammoxosome is where these bacteria carry out anaerobic ammonium oxidation, a reaction central to the global nitrogen cycle. The special membrane prevents the loss of reactive intermediates and maintains the steep proton gradient needed for energy generation.

Acidocalcisomes round out the picture. These membrane-bounded compartments store calcium, polyphosphate, and other ions, and they participate in pH regulation and osmotic balance.¹⁶ What makes acidocalcisomes especially interesting is that they are found across all three domains of life: bacteria, archaea, and eukaryotes. Their widespread distribution suggests they are an ancient feature, possibly predating the split between prokaryotes and eukaryotes.

How the ER May Have Originated from Prokaryotic Ancestors

The fact that prokaryotes lack a true ER raises an obvious evolutionary question: where did the eukaryotic ER come from? The answer, or at least the leading hypotheses, involves prokaryotes directly.

One influential model proposes that the eukaryotic endomembrane system originated from outer membrane vesicles (OMVs) released by the bacterial ancestor of mitochondria. According to this idea, the ancestral mitochondrial endosymbiont, living inside an archaeal host cell, shed vesicles from its outer membrane into the host’s cytoplasm. Trapped within the host, these vesicles accumulated and fused with each other or with the host’s plasma membrane, eventually giving rise to a primitive secretory ER.¹⁷ In this scenario, the ER is literally built from bacterial material, making it a prokaryotic contribution to eukaryotic complexity.

Supporting this idea, comparative genomics has found that many of the proteins responsible for shaping and maintaining the ER in modern eukaryotes, including reticulons, atlastins, and several SNARE proteins, are highly conserved and appear to trace back at least to the last eukaryotic common ancestor. This suggests the machinery for building and maintaining an ER was assembled very early in eukaryotic evolution.¹⁸ Multiple mechanisms, both endosymbiotic and autogenous, contributed to the genesis and increasing complexity of eukaryotic organelles and molecular machines.¹⁹

The Asgard archaea, a group of archaea considered the closest living relatives of the eukaryotic ancestor, add another piece to the puzzle. Genomic analyses of these organisms have identified genes encoding proteins with homology to eukaryotic intracellular trafficking components, including TRAPP-domain proteins, Sec23/24-like proteins, and roadblock domain proteins involved in vesicle transport.²⁰ These are not full endomembrane systems, but they hint that the molecular toolkit for building one may have been partially present in prokaryotic ancestors before the eukaryotic cell ever assembled.

Why Internal Membranes in Prokaryotes Remain Different from ER

Given all these examples of internal membranes, shared molecular machinery, and evolutionary connections, it is worth being precise about what makes the eukaryotic ER distinct and why nothing in the prokaryotic world fully replicates it.

The ER is not just a membrane inside a cell. It is an interconnected network that spans much of the cell’s volume, continuously connected to the nuclear envelope, and functionally divided into rough regions (studded with ribosomes for protein synthesis) and smooth regions (focused on lipid synthesis and detoxification). It serves as the entry point for the entire secretory pathway: proteins made on the rough ER are handed off to the Golgi apparatus, sorted, and shipped to their final destinations. The ER also acts as the cell’s primary calcium store and plays a central role in detecting and responding to misfolded proteins through a sophisticated stress-response system.

Prokaryotic internal membranes, by contrast, tend to be specialized for a single function. ICMs in phototrophs house photosynthetic pigments. Magnetosome membranes mineralize iron. Anammoxosome membranes run a particular metabolic reaction. Even the ER-like ribosome-studded sheets in planctomycetes have not been shown to support the full suite of ER functions. No prokaryote characterized so far has an interconnected membrane network that simultaneously handles protein folding, lipid biosynthesis, calcium storage, and secretory trafficking the way a eukaryotic ER does. The individual parts exist in scattered form across various prokaryotic lineages, but they have not been integrated into a single multifunctional system in any known prokaryote.

The bacterial plasma membrane compensates for much of this. It handles lipid synthesis, protein translocation, and energy generation all at once, packed into a single membrane layer. In a sense, the evolution of the ER involved pulling some of these functions off the plasma membrane and housing them on a dedicated internal membrane system, gaining the advantages of specialization, increased surface area, and spatial separation of incompatible processes. Why that step was taken in the eukaryotic lineage but not (or not fully) in prokaryotes is one of the enduring questions in cell biology.

Methanotroph ICMs and the Copper Connection

Methanotroph ICMs are among the most dynamic internal membranes in the prokaryotic world, and they illustrate something important about how prokaryotic compartmentalization differs from the relatively stable architecture of a eukaryotic ER. In methanotrophs, ICM production is regulated by environmental copper availability. When copper is scarce, ICM formation is minimal. As copper concentrations rise, the membranes expand, because the pMMO enzyme that fills them requires copper to function. In one well-studied species, Methylotuvimicrobium alcaliphilum, ICM formation increased with copper until plateauing at about 4.5 micromolar, even though the cells could tolerate higher copper levels without growth inhibition.⁶

This kind of environmentally tuned membrane production has no real equivalent in the eukaryotic ER, which is a permanent fixture of the cell that adjusts its volume modestly in response to secretory demand but never vanishes or appears de novo based on a single nutrient. Methanotroph ICMs are better understood as conditional extensions of the inner membrane, produced when the cell needs extra surface area for a specific enzyme and scaled back when it does not. The fact that these membranes are derived from the inner cell membrane, rather than forming independently, further distinguishes them from the ER, which has its own continuous identity separate from the plasma membrane.⁵ The dynamic, function-first nature of prokaryotic internal membranes is a recurring theme: they arise to solve a specific metabolic problem and are tightly regulated by the conditions that create that problem.