Not only does it have a tongue-twisting name, 鈥渙rforglipron鈥� has a complex molecular structure that made its synthesis very challenging. A massive effort by chemists at the Eli Lilly pharmaceutical company finally resulted in a viable synthesis, but why this massive effort? Because the payoff also promises to be massive! Orforglipron is the first 鈥渟mall molecule鈥� to act as an oral 鈥淕LP-1 agonist". Its discovery takes us on a long and winding journey.
The term 鈥淕LP-1 agonist鈥� wriggled its way into our vocabulary with the discovery that weekly injections of semaglutide, the active ingredient in Ozempic and Wegovy, can cause significant weight loss. The weight loss was a fortuitous side effect because semaglutide was originally designed to be a treatment for diabetes.
鈥淕lucagon-like peptide-1鈥� is a hormone produced in the gut that travels through the bloodstream and attaches to a receptor found on the surface of cells mostly in the pancreas and the brain. A receptor is a protein with a very specific shape that can be thought of as a lock. When the right key is inserted, a series of activities is 鈥渦nlocked.鈥� In this case that key is GLP-1 and the unlocked activities are the release of insulin and the curbing of the production of glucagon. Insulin causes glucose to be absorbed into cells and glucagon causes glycogen in the liver to be broken down into glucose. The balance between insulin and glucagon keeps blood glucose levels stable which is why drugs that behave like GLP-1, the so-called 鈥淕LP-1 agonists,鈥� are used to treat diabetes.
However, it also turns out that interaction of GLP-1 agonists with receptors found on brain cells leads to feelings of satiety and reduced hunger. These agonists can also stimulate receptors on vagal nerve cells in the stomach with the result that the stomach empties more slowly leading to a feeling of fullness. The end result of all this activity is the curbing of appetite that then results in weight loss.
The most famous GLP-1 agonist, semaglutide, was developed by the Danish company Novo Nordisk. Like GLP-1 itself, semaglutide is a peptide, a molecule composed of a string of amino acids joined together. It is similar enough to GLP-1 to allow it to fit into the GLP-1 receptor and act as an agonist. But there is a problem. In order for semaglutide to travel to the pancreas and the brain it has to enter the bloodstream. However, peptides are readily broken down during digestion which is why Ozempic (approved for diabetes) and Wegovy (just a higher dose of semaglutide approved for weight loss) have to be injected.
Recently an oral version of Wegovy has been introduced but it needs a very high dose of semaglutide because only a small percentage is absorbed. Also, the drug needs to be taken on an empty stomach immediately on waking, with no more than 120 mL of water. Furthermore, no food, beverage or other medication can be consumed for another half hour. Because of these limitations, there was a desire for a GLP-1 agonist that could be taken orally at any time of the day without any food or water restrictions.
The first challenge to be met was ease absorption. That meant looking for small molecules that unlike peptides can be more readily absorbed into the bloodstream. But that is just the first requirement. The molecule has to be of just the right shape to fit into the GLP-1 receptor so that it can activate it. That means having to know the exact shape of the receptor. As a simple analogy, imagine you lost one piece of a jigsaw puzzle and had to make a duplicate. You would have to know the exact shape and dimensions of the hole into which it has to fit.
Once the exact structure of GLP-1, meaning the sequence of amino acids that make up the molecule, was identified in the 1980s, scientists were able to look for the receptor into which it would fit. Sort of like looking for the shape of a keyhole into which a given key would fit. That 鈥渒eyhole鈥� turned out to be a protein that is 463 amino acids long with a distinct shape that snaked back and forth through the cell membrane. Through a complicated technique known as Cryo-Electron Microscopy, the specific three-dimensional shape of this protein was determined and a 鈥減ocket鈥� into which the active part of the GLP-1 molecule fits was identified. Now the challenge was to find a small molecule that also fits that pocket. Not a small task.
Once the amino acid sequence of the GLP-1 receptor was determined, the gene that codes for this protein was identified and human cells were engineered to include the gene. These cells then produced the receptor protein. Thousands and thousands of known compounds were individually added to these cells using robotics and whether they had any influence on the receptors was detected. The molecules that had an effect were then chemically manipulated to see if the effect could be increased. After numerous manipulations on candidate molecules, one was arrived at that acted as a non-peptide GLP-1 agonist and was given the name 鈥渙rforglipron.鈥�
Now came the next challenge. This molecule had never existed before. Could it be synthesized economically on a large scale? ChemAIRS is an artificial intelligence program that optimizes 鈥渞etrosynthesis,鈥� the process of working backward from a target molecule to determine the best sequence of chemical steps to synthesize it. The software scans the 85 million known chemical reactions and their percent yields to come up with possible sequences of reactions that can synthesize the desired compound. Amazing! It was this process that Eli Lilly chemists used to come up with a commercially viable process for the manufacture of orforglipron.
Based on a 72 week-long clinical trial in which participants lost 12% of their body weight and showed no significant side effects, the U.S. Food and Drug Administration approved orforglipron which will be marketed as 鈥淔oundayo.鈥� While it may not exactly be like finding the Holy Grail, the drug promises to be hugely successful commercially. While it is not as effective as injectable GLP-1 agonists, orforglipron is more convenient to use and is cheaper to produce. An excellent example of a drug that is the product of human intelligence aided by artificial intelligence with a dose of serendipity thrown in.
