Translation, the process of building proteins from messenger RNA instructions, occurs primarily on ribosomes in the cytoplasm of cells. But “the cytoplasm” is an oversimplification. In eukaryotic cells (plants, animals, fungi), translation happens in several distinct locations: on free ribosomes floating in the cytoplasm, on ribosomes attached to the endoplasmic reticulum, and inside mitochondria and chloroplasts. In prokaryotic cells (bacteria and archaea), translation happens in the cytoplasm and can begin while the mRNA is still being made.
Translation in Prokaryotic Cells
Bacteria and archaea lack a nucleus, so their DNA sits directly in the cytoplasm in a dense region called the nucleoid. This means transcription (copying DNA into mRNA) and translation (reading that mRNA to build a protein) can happen almost simultaneously. As soon as one end of an mRNA strand emerges from the enzyme making it, a ribosome can latch on and start translating. This is called coupled transcription-translation, and it’s a defining feature of prokaryotic gene expression.
The physical details are more nuanced than textbooks typically show. In bacteria like E. coli, ribosomes and the nucleoid DNA are actually segregated from each other rather than mixed together. Ribosomes concentrate in ribosome-rich zones outside the nucleoid. Current evidence suggests that coupled transcription-translation likely occurs at the surface of the nucleoid, where the mRNA-making machinery and ribosomes can meet. Free ribosome subunits can penetrate into the nucleoid to initiate translation, but once a full ribosome assembles on the mRNA, physical forces push the whole complex outward to the nucleoid’s surface, where additional ribosomes can pile on.
For genes that encode membrane proteins, something even more interesting happens. The DNA region containing those genes migrates toward the inner membrane of the cell. There, transcription, translation, and insertion of the new protein into the membrane all occur together in one coordinated process called transertion.
Prokaryotic ribosomes are designated 70S, with a small 30S subunit and a large 50S subunit. They have a molecular mass of roughly 2.3 million daltons, making them considerably smaller than their eukaryotic counterparts.
Free Ribosomes in the Cytoplasm
In eukaryotic cells, the majority of protein synthesis takes place on ribosomes floating freely in the cytoplasm. The roughly 30,000 types of polypeptides in human cells are mostly synthesized here. These free ribosomes produce proteins that will stay and work in the cytoplasm, or proteins destined for the nucleus, mitochondria, chloroplasts, or peroxisomes. Once translation is complete, the finished protein is released directly into the cytoplasm, where it either folds into its functional shape on the spot or gets picked up by transport machinery that shuttles it to the correct organelle.
Eukaryotic ribosomes are 80S, composed of a 40S small subunit and a 60S large subunit, with a combined molecular mass of about 4.3 million daltons. That’s nearly twice the size of bacterial ribosomes.
Polyribosomes: Multiple Ribosomes on One mRNA
Whether free or membrane-bound, ribosomes rarely work alone. Several ribosomes typically bind to a single mRNA strand at once, forming structures called polyribosomes (or polysomes). This arrangement lets the cell produce multiple copies of the same protein simultaneously from one mRNA molecule. In eukaryotic cells, polyribosomes form a left-handed helical structure where the mRNA strand bridges the exit site of one ribosome to the entry site of the next, creating a continuous channel that prevents the mRNA from looping loosely between ribosomes.
The Rough Endoplasmic Reticulum
Not all eukaryotic translation stays in the open cytoplasm. Proteins destined for secretion outside the cell, or for insertion into the cell’s outer membrane, the endoplasmic reticulum, the Golgi apparatus, or lysosomes, are translated on ribosomes that dock onto the rough endoplasmic reticulum (rough ER). The “rough” appearance comes from these ribosomes studding its surface.
The targeting process works through an elegant molecular signal system. As a ribosome begins translating certain mRNAs, the first stretch of amino acids to emerge is a short signal sequence. A molecule called the signal recognition particle grabs this sequence and temporarily pauses translation. It then ferries the entire complex (ribosome, mRNA, and partial protein) to a receptor on the ER membrane. Once docked, translation resumes and the growing protein is threaded through a channel directly into the interior of the ER. From there, the protein travels through the Golgi apparatus and onward to its final destination.
Roughly 10,000 types of soluble and membrane proteins in human cells enter the ER during their production. Interestingly, some ribosomes attached to the ER also translate proteins that end up back in the cytoplasm. In those cases, the protein fails to engage the ER’s import channel and is released back into the surrounding cytoplasm.
Translation Inside Mitochondria and Chloroplasts
Mitochondria (in nearly all eukaryotic cells) and chloroplasts (in plants and algae) carry their own DNA and their own ribosomes, and they perform their own protein synthesis. These organelles are thought to descend from ancient bacteria that were engulfed by early eukaryotic cells, which is why their ribosomes resemble bacterial 70S ribosomes rather than the 80S ribosomes found in the cytoplasm.
Only a fraction of the proteins these organelles need are actually made inside them. In the green alga Chlamydomonas, for example, about 5 to 6 of the chloroplast ribosome’s large subunit proteins are made inside the chloroplast, while 26 to 27 are made on cytoplasmic ribosomes and imported. For the small subunit, 14 of 31 proteins are chloroplast-made, with the rest coming from the cytoplasm. Mitochondria follow a similar pattern: they translate a small set of essential proteins (mostly components of the energy-production machinery), while the cell’s nucleus encodes the rest and ships them in.
Localized Translation in Neurons
One of the more striking examples of translation happening in unexpected places is in nerve cells. Neurons can be extraordinarily long, with axons stretching up to a meter in humans. If every protein had to be made in the cell body and shipped to distant nerve endings, responses that require new proteins would be far too slow. Instead, neurons transport specific mRNAs out to their dendrites and axon terminals, where those mRNAs sit waiting until they’re needed.
When a synapse needs to become stronger or weaker (a process called synaptic plasticity), local ribosomes translate these stored mRNAs on the spot. The pool of mRNAs being locally translated changes dramatically during learning tasks, suggesting that local translation acts as a checkpoint for how neurons adapt. Scaffolding proteins that organize the receiving side of a synapse are preferentially made locally, while the receptor proteins that detect neurotransmitter signals are primarily synthesized back in the cell body and transported out. This division of labor lets neurons fine-tune individual synapses independently, which is fundamental to how memories form and circuits rewire.
Translation in the Nucleus
Textbooks have long taught that translation is strictly a cytoplasmic event in eukaryotes, with mRNA being exported through nuclear pores before ribosomes ever touch it. That picture is now being challenged. Evidence stretching back decades has hinted that some translation occurs inside the nucleus, particularly in the nucleolus (the structure within the nucleus where ribosomal RNA is made). Early experiments showed that purified nuclei could incorporate amino acids into new proteins, and more recent imaging techniques have confirmed the presence of translation components, including mature transfer RNAs, translation initiation factors, and assembled 80S ribosomes, inside the nucleus.
A technique called ribopuromycylation has visualized active protein synthesis occurring in the nucleolar compartment of human cells, and researchers have confirmed this isn’t simply contamination from cytoplasmic proteins leaking in. Most newly synthesized nuclear peptides appear to break down within minutes, suggesting this nuclear translation may serve a quality-control function, helping the cell detect and destroy faulty mRNAs before they ever reach the cytoplasm. One quality-control system called nonsense-mediated mRNA decay, which identifies mRNAs with premature stop signals, has been observed operating in the nucleus and requires active translation to work.
Nuclear translation appears to be minimal in normal cells but is substantially elevated in cancer cells, where it correlates with rapid growth. Whether enhanced nuclear translation drives cancer progression or is simply a byproduct of it remains an open question.