Primary and Secondary Structure

Two levels of structure, one molecule's entire personality

Forty-four amino acids. That's the length of the glucagon-like peptide-1 precursor fragment that sparked half the metabolic research conversation of the last decade. I keep that number in my head because it's a good reminder that even a relatively short chain can fold into something with enormous biological specificity. The sequence , the primary structure , is like a sentence. The shape that sentence folds into , the secondary structure; is like the tone of voice it's said in. Get either wrong and the meaning changes entirely.

This piece covers the biochemistry of both levels: what they are, how chemists describe them, and why the distinction matters if you're trying to understand how any peptide actually does what it does at a receptor.

Primary structure: the sequence is everything

Primary structure is the simplest of the four levels of protein/peptide organisation to define: it's the linear order of amino acids in a chain, read from the N-terminus (the free amino end) to the C-terminus (the free carboxyl end). That's it. No geometry. No folding. Just a sequence of residues held together by peptide bonds.

A peptide bond forms via a condensation reaction, one amino acid's carboxyl group reacts with the next residue's amino group, releasing water. The result is a planar, partially double-bond-character linkage that restricts rotation. This is important later when we get to secondary structure, because it means the backbone isn't a completely floppy chain; certain conformations are already constrained out of existence.

Each amino acid residue contributes a side chain, chemists call it the R group; and it's the identity and order of those R groups that defines primary structure in practical terms. Glycine has a hydrogen as its R group (the smallest possible, no chirality). Tryptophan has a bulky indole ring. Cysteine carries a thiol that can form disulfide bridges. Twenty standard amino acids, each with distinct size, charge, polarity, and reactivity at physiological pH.

The sequence is encoded genetically and synthesised either ribosomally (in biological systems) or chemically (in the lab, via solid-phase peptide synthesis, or SPPS). Changing a single residue, what's called a point substitution, can dramatically alter the peptide's stability, receptor binding affinity, or susceptibility to proteolytic degradation. Much of the research effort in peptide drug design involves deliberately modifying primary structure: swapping in D-amino acids (the mirror-image forms not recognised by most proteases), adding N-methylation to backbone nitrogens, or incorporating non-canonical residues entirely. All of that begins with understanding what the native sequence is doing first.

Why sequence determines destiny

There's a principle, attributed to the biochemist Christian Anfinsen's Nobel-winning work in the early 1970s, that a protein's three-dimensional structure is determined entirely by its amino acid sequence; that folding is thermodynamically encoded in the primary structure itself. Anfinsen showed this by denaturing and then refolding ribonuclease A; the enzyme recovered its activity spontaneously, without any external template.

The implication is striking: the sequence is not just a list of parts, it's already an instruction manual for assembly. Secondary and tertiary structure emerge from the primary structure given the right solvent conditions, temperature, and pH. That's why a single amino acid substitution in the wrong position can destabilise the whole fold, or, conversely, why researchers can engineer new sequences to adopt specific desired shapes.

For pharmacokinetics, primary structure also governs the very first challenge a therapeutic peptide faces: proteolytic stability. Peptidases, enzymes in blood, gut lumen, and target tissues; recognise specific sequence motifs and cleave at those sites. Understanding primary structure is therefore the starting point for designing molecules that survive long enough to reach their targets. This is directly relevant to how something like a GLP-1 receptor agonist is structurally modified from the native seven-amino-acid active core to a clinically studied molecule with a longer half-life.

Secondary structure: the chain finds its shape

Secondary structure describes the local, repeating three-dimensional arrangements that segments of a polypeptide chain adopt. It's not the overall fold of the whole molecule, that's tertiary structure. Secondary structure is the geometry of short stretches, driven primarily by hydrogen bonding between backbone atoms.

The two dominant secondary structural elements are the alpha helix and the beta sheet, and they were both predicted theoretically by Linus Pauling and Robert Corey in 1951 before X-ray crystallography fully confirmed them. Worth pausing on that: they worked it out from bond lengths, angles, and hydrogen bonding geometry alone. Honestly one of the more impressive pieces of structural reasoning in twentieth-century chemistry.

The alpha helix

In an alpha helix, the polypeptide backbone winds in a right-handed spiral. Each backbone carbonyl oxygen forms a hydrogen bond with the backbone NH of the residue four positions further along the chain (the i to i+4 hydrogen bond pattern). The result is a rod-like structure with the R groups pointing outward and roughly perpendicular to the helix axis.

There are 3.6 residues per turn, and the helix rises about 0.54 nanometres per turn. These numbers matter because they set geometric constraints: certain R groups are too bulky, or too charged in the wrong context, to sit in helical conformations without destabilising them. Proline, for instance, lacks the NH needed for the hydrogen bond and introduces a fixed kink, it's sometimes called a "helix breaker" for that reason. Glycine, with its tiny R group, allows so much backbone freedom that it tends to disrupt regular secondary structure too.

Beta sheets

Beta sheets arise when two or more segments of the backbone align side by side and form hydrogen bonds between adjacent strands rather than within a single strand. Strands can run parallel (both N-to-C in the same direction) or antiparallel (running in opposite directions), and the hydrogen bond geometry differs between the two. Antiparallel beta sheets are somewhat more stable energetically.

The R groups alternate above and below the sheet plane, and the overall structure has a pleated, almost corrugated appearance; hence the older term "beta pleated sheet." Beta sheets are common in structural proteins and in the core domains of many globular proteins. They're also implicated in amyloid formation, where normally soluble proteins misfold into extended, stable beta-sheet fibrils. That's a whole separate rabbit hole.

Loops, turns, and everything in between

Not every residue sits in a helix or sheet. Loop regions, also called random coil, though that's a bit misleading since they're often structurally well-defined, connect secondary structural elements. Beta turns reverse the chain's direction over four residues and are stabilised by a single i to i+3 hydrogen bond. These regions are frequently where receptor-binding happens: the specific geometry of a loop or turn presents side chains in the precise orientation needed for molecular recognition.

I find the loops the most interesting bit, actually. They're often less conserved in sequence but critically important in function; a reminder that primary structure and secondary structure don't map onto each other in a simple one-to-one way. You can swap residues in a helix and maintain the fold; swap the wrong residue in a binding loop and you've lost the interaction entirely.

How the two levels interact in peptide research

In practice, peptide researchers think about primary and secondary structure together, not in isolation. Algorithms like PSIPRED (developed at University College London) predict secondary structure from sequence with reasonable accuracy for well-characterised fold families. Circular dichroism (CD) spectroscopy in the lab gives an empirical read on secondary structural content in solution, useful because crystals don't always represent the solution conformation. NMR spectroscopy at facilities like the National Deuteration Facility at ANSTO in Lucas Heights, NSW, goes further and can resolve backbone geometry residue by residue in solution state.

When a peptide is being assessed for potential therapeutic development, understanding both levels of structure feeds directly into stability studies, formulation decisions, and the mechanistic rationale behind any observed receptor interaction. That mechanistic picture is what distinguishes serious structural science from the handwavy claims that sometimes accompany less well-characterised molecules.

And that, in my view, is where the field earns its credibility, or loses it. Structural data is verifiable. Sequence is verifiable. A lot of the noise around peptides as a category comes from marketing that bypasses this entirely. The chemistry doesn't care about the marketing.

Sources

• Protein Structure and Function; NCBI Bookshelf (Molecular Biology of the Cell)
• Peptide therapeutics: current status and future directions, NCBI / PubMed Central
• Regulatory framework for biologicals in Australia, TGA
• Biochemistry, Amino Acids; NCBI Bookshelf (StatPearls)

, Priya Nandakumar, Pharmacology writer, peptides & PK