Your cells make mRNA by copying specific genes from DNA in a process called transcription, then chemically modifying the raw copy into a finished molecule ready for protein production. Scientists can also manufacture mRNA synthetically in a lab, which is how mRNA vaccines are produced. Both routes follow a similar logic: start with a DNA template, build an RNA strand from it, then add protective features so the mRNA is stable enough to do its job.
How Cells Build mRNA From DNA
Inside your cells, mRNA production begins when a large molecular machine called RNA polymerase II lands on a stretch of DNA near a gene. This landing zone is called a promoter, a region spanning roughly 250 DNA letters upstream and 50 letters downstream from the exact spot where copying starts. The promoter acts like a signpost, telling the cell’s machinery precisely where to begin reading.
Before copying can start, a cluster of helper proteins assembles on the promoter to form what’s called the preinitiation complex. One of these helpers acts as a motor, grabbing the DNA downstream and reeling it toward the active site of the polymerase. This creates torsional strain that physically pries open the two strands of the DNA double helix, exposing about 12 DNA letters around the start site. Once the strands separate, RNA polymerase II locks down slightly and begins stitching together RNA building blocks (nucleotides) that match the exposed DNA template, one by one.
The polymerase then slides along the gene, reading DNA and assembling a growing RNA strand behind it. This elongation phase continues until the polymerase hits signals in the DNA that trigger it to release the new RNA transcript and detach. At this point, the cell has a raw, unfinished copy called pre-mRNA. It isn’t ready to make a protein yet.
Three Modifications That Finish the Job
Before leaving the nucleus, pre-mRNA undergoes three critical processing steps: capping, splicing, and tailing.
The 5′ Cap
Almost immediately after transcription begins, the leading end of the new RNA gets a protective cap. Three enzymes work in sequence: the first removes a phosphate group from the RNA’s tip, the second attaches a guanosine nucleotide in a reversed orientation, and the third adds a methyl group to that guanosine. The result is a structure known as a 7-methylguanosine cap. In many cases, additional methylation occurs on the first or second nucleotides of the transcript itself, creating slightly upgraded versions called Cap 1 or Cap 2. This cap shields the mRNA from being chewed up by enzymes, and it also serves as a recognition tag that your cell’s protein-building machinery (the ribosome) uses to latch on.
Splicing Out the Non-Coding Sections
Human genes are interrupted by long stretches of non-coding DNA called introns, mixed in with the coding segments called exons. The pre-mRNA copy includes all of them, so the cell needs to cut out the introns and stitch the exons together. This is handled by the spliceosome, a complex built from small RNA molecules and dozens of proteins.
The spliceosome identifies introns using short, conserved sequences at each intron’s boundaries and at an internal branch point. First, a small RNA component pairs with the sequence at the intron’s starting edge. Then a second RNA component pairs with the branch point, forcing a specific adenosine nucleotide to bulge outward. That bulging adenosine becomes the chemical attacker: its reactive group strikes the intron’s starting boundary, cutting it free and forming a looped structure called a lariat. A second chemical reaction then joins the two flanking exons together and releases the intron loop for recycling. This entire process happens with extraordinary precision, since even a single-nucleotide error would scramble the protein’s instructions.
The Poly-A Tail
At the trailing end of the mRNA, an enzyme adds a long string of adenine nucleotides, typically 150 to 250 of them. This poly-A tail protects the mRNA from degradation, helps it get exported from the nucleus, and influences how efficiently it gets translated into protein. Once the cap, splicing, and tail are all in place, the mature mRNA is transported out of the nucleus and into the cytoplasm, where ribosomes read it to build the corresponding protein.
How Synthetic mRNA Is Made in a Lab
Manufacturing mRNA for vaccines and therapies follows the same basic principle as what happens in cells: a polymerase enzyme reads a DNA template and produces a complementary RNA strand. But instead of happening inside a nucleus, the reaction takes place in a test tube or bioreactor in a process called in vitro transcription (IVT).
The starting point is a DNA template containing the gene of interest. This template is typically carried in a circular piece of DNA called a plasmid, which must first be cut open (linearized) using restriction enzymes or copied via PCR. Linearizing the template ensures the polymerase has a defined stopping point and doesn’t keep reading into irrelevant sequences.
The DNA template includes a special promoter sequence recognized by a viral RNA polymerase, most commonly T7 polymerase (originally derived from a bacteriophage). The reaction mix is straightforward: the linearized DNA template, T7 polymerase, the four RNA nucleotide building blocks (ATP, UTP, GTP, and CTP), and a buffer solution containing magnesium ions and a few stabilizers. The polymerase binds the promoter and begins synthesizing RNA, producing large quantities of identical mRNA copies. Modern IVT reactions can yield between 2 and 5 grams of mRNA per liter under standard conditions, and optimized reactions using engineered promoter variants have pushed yields as high as 12 to 14 grams per liter in under two hours.
Swapping in Modified Building Blocks
One of the key innovations in synthetic mRNA is replacing the standard uridine nucleotide with a modified version called N1-methylpseudouridine. The reason is immunological: your immune system has sensors that detect foreign single-stranded RNA, and these sensors are particularly good at recognizing stretches rich in uridine. Swapping in the modified nucleotide changes the RNA’s shape just enough that these sensors no longer bind it effectively. The result is mRNA that produces less inflammation, is better tolerated, and gets translated into protein more efficiently. This modification was central to the COVID-19 mRNA vaccines.
Capping and Tailing Synthetic mRNA
Just like cellular mRNA, synthetic mRNA needs a 5′ cap and a poly-A tail to function. Early methods required three separate enzymatic reactions to build the cap, but current approaches use a single enzyme (Vaccinia Capping Enzyme) that combines all three activities to produce a Cap 0 structure in one step. Adding an additional methylation enzyme converts this to Cap 1, which is more effective at driving protein production. The capping efficiency of this process can reach 100%. The poly-A tail is typically encoded directly into the DNA template, so it gets transcribed automatically during the IVT reaction.
Purifying the Final Product
The raw output of an IVT reaction is a messy mixture. It contains the desired mRNA but also leftover DNA template, unused nucleotides, enzymes, truncated RNA fragments, uncapped molecules, and double-stranded RNA byproducts. Double-stranded RNA is a particularly problematic contaminant because it triggers strong immune responses that can reduce the mRNA’s effectiveness and cause side effects.
Purification relies heavily on high-performance liquid chromatography (HPLC). Reversed-phase HPLC is especially effective at separating double-stranded RNA from the single-stranded mRNA product. Additional steps remove the DNA template and other impurities. The goal is a highly pure, single-stranded mRNA that will produce protein cleanly once inside a cell.
Packaging mRNA for Delivery
Naked mRNA breaks down within minutes in the body, so it needs to be wrapped in a protective shell. The standard approach uses lipid nanoparticles (LNPs), tiny fat-based spheres roughly 80 to 100 nanometers across.
Manufacturing LNPs involves microfluidic mixing. One stream carries the lipid components dissolved in ethanol: a positively charged (cationic) or pH-sensitive lipid, a structural phospholipid, cholesterol for stability, and a PEG-coated lipid that prevents the particles from clumping together. A second stream carries the mRNA dissolved in an acidic buffer. When the two streams meet inside a microfluidic chip, the positively charged lipids and negatively charged mRNA attract each other through electrostatic forces, forming complexes. These complexes then self-assemble with the remaining lipids as the ethanol is diluted, producing uniform nanoparticles with the mRNA safely encapsulated inside. The acidic buffer keeps the pH-sensitive lipids in their charged state during assembly, ensuring tight packaging. Once injected into the body, the nanoparticles are taken up by cells, the lipid shell dissolves in the cell’s acidic interior compartments, and the mRNA is released to begin producing protein.