The maths of many steps
Suppose each step in making a peptide works 99% of the time. That sounds excellent. For a 10 amino acid peptide, about 90% of your chains come out complete. For a 30 amino acid peptide, about 74%. For a 70 amino acid peptide, about 50%.
The rest are faulty chains: missing one amino acid here, stopped short there. And many of them look chemically almost identical to the product you wanted.
That simple arithmetic explains one of the most important lines in Peptide Synthesis and Applications: peptides up to about 15 to 20 amino acids are routinely available from specialist companies, but longer peptides pose challenges. Another chapter, on protein ligation, puts the practical ceiling of standard solid-phase synthesis at about 50 amino acids in reasonable yield and purity.
The two main culprits
The introductory chapter by Knud Jensen names the main reasons long syntheses fail: steric hindrance and packing between chains.
Steric hindrance means physical crowding. Some amino acids are bulky near the reacting group. The chapter singles out the "beta-branched" amino acids valine and isoleucine as sometimes difficult to add completely. When a bulky amino acid has to attach next to another bulky one, the reaction slows or stalls.
Aggregation is the bigger problem. As chains grow on the resin, they can start folding and sticking to each other, especially by forming beta-sheet structures, the same kind of flat, stacked arrangement seen in amyloid plaques. The chapter explains that amino acids prone to beta-sheets often cause aggregation during chain growth through hydrogen bonding and hydrophobic effects.
Once chains clump together, the reactive ends are buried. New amino acids cannot reach them. The result is premature stops (truncations) and missing amino acids (deletions), which the chapter notes can be tedious to purify out afterwards.
Deletion sequences: the silent contaminant
This is worth dwelling on. A "deletion sequence" is a chain that is correct except for one missing amino acid. It has almost the same size, almost the same chemistry and almost the same behaviour as the correct peptide.
That makes it very hard to separate. And because it is so similar, it might even have some biological activity, or a different one. When you see a purity figure on a peptide product, the impurities are often exactly these near-identical chains.
The tricks chemists use
The manual describes several clever countermeasures.
Low-loading resins. Spacing chains further apart on the resin reduces their chance of sticking together.
PEG-based resins. Resins made with polyethylene glycol keep chains better solvated. The chapter notes these often give higher purity, especially for longer or aggregation-prone sequences.
Pseudoprolines. Proline is a natural "helix breaker" that kinks the chain. Chemists can temporarily turn a serine or threonine into a proline-like ring structure during synthesis, disrupting aggregation. The ring opens again in the final acid step, leaving the normal amino acid. It is like installing a temporary kink to stop the chain folding up, then removing it when the job is done.
Backbone protection. Groups called Dmb and Hmb can be temporarily attached to the backbone nitrogen, removing a hydrogen bond donor and preventing sheet formation. The chapter notes these are especially useful at glycine-glycine and aspartate-glycine sites. The trade-off is cost.
Heat. Microwave heating during synthesis (Part 7 of the follow-up series looks at this critically) helps break up aggregates, particularly for sequences that form beta-sheets and for bulky couplings.
A 1960s feat
The instruments chapter includes a detail that puts all this in perspective. Using one of the first automated synthesisers, Merrifield's group made ribonuclease A, a 124 amino acid enzyme. The chapter records it took 369 chemical reactions and 11,391 steps.
With the chemistry and analytical tools of the 1960s, that was astonishing. It also showed what a heroic effort making a whole protein by stepwise synthesis is.
Stitching instead of building
So how do chemists make proteins longer than about 50 amino acids? They mostly stop trying to build them in one go.
Instead, they make shorter peptide segments separately, purify each one, and then join them. The most important method for this, native chemical ligation, is covered in detail in our follow-up series. The introductory chapter describes it briefly: one segment carries a special reactive end (a thioester), the other starts with a cysteine, and they join to form a normal peptide bond.
Why this matters to you
The length of a peptide tells you something about how hard it was to make well.
A short peptide, like the three amino acid GHK or the five amino acid ipamorelin, is straightforward chemistry. A 30 or 40 amino acid peptide, like many of the GLP-1 family drugs, is a serious synthetic challenge. Making it reliably at high purity requires optimised chemistry, good equipment and careful purification.
That is part of why pharmaceutical-grade versions of long peptides cost what they do. And it is why, for long peptides especially, the question "who made this, and how well?" is so important. The difference between a well-made and a poorly made batch may be invisible to the eye and even hard to spot without proper analysis. Part 6 explains how chemists check.
This series explains laboratory chemistry for general readers. It is not a guide to making or using any compound. Nothing here is medical advice.
