Most insulin used by diabetics today is made by genetically engineered bacteria or yeast that have been programmed with the human insulin gene. These microorganisms grow in large fermentation tanks, producing insulin that is chemically identical to what a healthy human pancreas would make. The process replaced animal-sourced insulin, which came from pig and cow pancreases and often caused immune reactions or varied in strength from batch to batch.
From Animal Pancreases to Engineered Bacteria
Before the 1980s, all commercial insulin came from slaughtered animals. Eli Lilly was among the first companies to extract insulin from animal pancreases, but the supply couldn’t keep up with demand, and potency varied by as much as 25% between batches. Improved purification methods eventually narrowed that variation to about 10%, but animal insulin still wasn’t a perfect match for the human version, and some patients developed allergic responses over time.
The breakthrough came in 1978, when researchers at Genentech figured out how to build a synthetic copy of the human insulin gene and insert it into the common gut bacterium E. coli. By 1982, the first recombinant human insulin products hit the market. For the first time, diabetics could inject insulin that was structurally identical to what their own bodies would produce.
How the Gene Gets Into the Bacteria
The process starts in a lab, where scientists construct a synthetic version of the human insulin gene. Bacteria carry small loops of DNA called plasmids that sit outside their main chromosome. Researchers cut open a plasmid, splice the insulin gene into it, and return the modified plasmid to the bacterial cell. Once inside, the bacterium reads the inserted gene and begins producing human insulin (or, more precisely, a precursor called proinsulin) as if it were one of its own proteins.
These engineered bacteria are then placed into large stainless-steel fermentation tanks, sometimes holding thousands of liters of nutrient broth. The bacteria multiply rapidly, and as each cell divides, it copies the insulin gene along with the rest of its DNA. Within hours, billions of cells are churning out proinsulin.
E. Coli vs. Yeast
Two organisms dominate insulin manufacturing: the bacterium E. coli and baker’s yeast (Saccharomyces cerevisiae). E. coli grows fast, needs simple nutrients, and produces high yields at low cost, which is why it became the first and most common host. But it has drawbacks. E. coli tends to bundle up the insulin protein into dense clumps called inclusion bodies inside the cell, which then need to be dissolved and refolded into the correct shape. The bacterium also can’t perform certain chemical finishing steps that human cells do naturally, like forming the specific bonds that hold the insulin molecule together properly.
Yeast handles some of these problems better. It can fold proteins more accurately and even secrete the finished product outside the cell, which simplifies collection. Several yeast species are used commercially, including Saccharomyces cerevisiae and Pichia pastoris. Between 2004 and 2013, about 24% of all FDA and EMA-approved biopharmaceuticals came from E. coli and 13% from yeast. Both remain central to insulin production today.
Turning Raw Protein Into Medical-Grade Insulin
What comes out of the fermentation tank isn’t ready for injection. The bacteria or yeast produce proinsulin, a longer precursor molecule that needs to be trimmed into its active form. Enzymes called trypsin and carboxypeptidase B do the cutting, snipping away a connecting segment to leave the two chains (A and B) that make up functional insulin. In some manufacturing routes, a chemical called cyanogen bromide is used first to remove an extra tag that helped the bacteria produce the protein.
After cleavage, the insulin goes through multiple rounds of chromatographic purification. Think of chromatography as a series of increasingly fine filters, each one exploiting a different physical property of the insulin molecule: its size, its electrical charge, or how it interacts with water. Reversed-phase chromatography and size-exclusion chromatography are commonly used in the final “polishing” steps. The goal is to strip away bacterial debris, leftover enzymes, DNA fragments, and any misfolded insulin, leaving a product that is extremely pure.
What Gets Added Before It Reaches Your Pen or Vial
Pure insulin protein on its own would clump together and break down quickly. To prevent that, manufacturers add a handful of carefully chosen ingredients to the final formulation. Zinc ions help insulin molecules organize into stable clusters of six (called hexamers), which resist degradation during storage. Antimicrobial preservatives like metacresol are included to prevent bacterial contamination in multi-use vials and pens. Metacresol does double duty: it also stabilizes the hexamer structure by fitting into a small pocket between insulin molecules and forming hydrogen bonds that hold the cluster together.
The choice of preservative matters for how the insulin behaves after injection. In newer ultra-fast formulations, manufacturers deliberately use different preservatives, like phenoxyethanol, that don’t stabilize the hexamer. Instead, they promote the smaller, faster-absorbing monomer form. By swapping just one additive, formulators can change how quickly insulin enters the bloodstream after a shot.
Engineering Insulin Analogs
The same genetic engineering that made human insulin possible also opened the door to designing improved versions called analogs. These are insulin molecules with small, intentional changes to their amino acid sequence that alter how fast or how long they work in the body.
Rapid-acting analogs are designed to break apart from hexamers quickly after injection so they reach the bloodstream faster. Insulin lispro, for example, simply swaps two amino acids near the end of the B chain (proline and lysine switch positions at spots 28 and 29). That tiny change weakens the bonds holding hexamers together, so the insulin disperses faster at the injection site. Insulin glulisine takes a different approach, substituting lysine and glutamic acid at different positions on the B chain to achieve a similar speed.
Long-acting analogs work the opposite way. Insulin glargine has two extra arginine molecules added to the tail of its B chain, which shifts its chemistry so it forms tiny crystals under the skin that dissolve slowly over 24 hours. Insulin detemir attaches a fatty acid to the B chain, which causes the molecule to bind to a blood protein called albumin and release gradually. Insulin degludec takes this further with a long fatty acid chain that causes the molecules to link into extended multi-hexamer chains at the injection site, creating an ultra-long depot that can last beyond 42 hours.
All of these analogs start the same way: scientists modify the synthetic insulin gene in the lab before inserting it into bacteria or yeast. The microorganism then faithfully produces the altered protein, and the same fermentation, cleavage, and purification pipeline turns it into a finished drug.
Biosimilars and Lower-Cost Alternatives
As patents on original insulin products expire, other manufacturers can produce biosimilar versions. A biosimilar is a biological product that is highly similar to an already-approved insulin, with no clinically meaningful differences in safety or effectiveness. The FDA requires extensive testing to confirm this before granting approval. For patients, a biosimilar works the same as the original product but typically costs less, since the manufacturer didn’t bear the full expense of developing the molecule from scratch. The first biosimilar rapid-acting insulin was approved by the FDA, expanding the number of affordable options available.
From a manufacturing standpoint, biosimilar insulin follows the same fundamental steps: gene insertion, microbial fermentation, enzymatic processing, chromatographic purification, and careful formulation. The challenge lies in matching the original product closely enough to meet regulatory standards, since even small differences in how a protein folds or clusters can affect how it works in the body.