Skip to content
📢We are still a growing community and would love some help. Want to contribute? Join us by mailing an e-mail of interest to "r6394175@gmail.com"
Start helping our community →

Protein Structure, Folding & Function

Intermediate Prerequisites: Amino Acids Protein Chemistry IBO USABO biochemistry

Overview

A protein’s function is inseparable from its shape, and its shape is built in four hierarchical levels — primary, secondary, tertiary, and quaternary structure — each stabilised by a different balance of covalent and non-covalent forces. This page covers that hierarchy, the model system for allosteric regulation (hemoglobin), how cells prevent and correct misfolding, and how they destroy proteins deliberately. Notably, protein fold is more evolutionarily conserved than amino acid sequence — distantly related proteins with very different sequences often adopt near-identical folds, because the geometric and energetic constraints on a stable fold are stricter than the constraints on any single residue.

Key Concepts

The four levels of structure

Primary structure is simply the linear amino acid sequence, held together by peptide bonds (and, for cross-chain or cross-loop connections, disulfide bonds between cysteines).

Secondary structure is a local, repeating backbone conformation stabilised almost entirely by hydrogen bonds between the carbonyl oxygen and amide hydrogen of the polypeptide backbone — side chains are not directly involved. Two dominant motifs:

  • α-helix (one turn = 3.6 residues). High-propensity residues: Ala, Glu, Leu, Met, Lys (“MALEK”) — small enough to pack tightly, or able to form stabilising salt bridges (Glu, Lys), with no steric interference with the coil. Helix breakers: Proline (its side chain cyclises onto the backbone nitrogen, removing the amide hydrogen the helix needs and imposing a rigid kink) and Glycine (too flexible — entropically costly to lock into one rigid helical conformation).
  • β-sheet (≈180° turn per 4 residues). High-propensity residues: bulky aromatics (Trp, Tyr, Phe) and β-branched residues (Val, Ile, Thr) — their bulk disfavours the tight coiling of a helix and favours the more open, extended sheet geometry. Sheet breakers: Proline (same rigidity argument) and bulky charged residues (Glu, Lys) that clash sterically/electrostatically in the tightly packed strand.
  • β-turn: a third, shorter motif connecting two antiparallel β-strands.

α-helix (coiled ribbon) and pleated β-sheet secondary structures, with the four non-covalent/covalent interaction types that stabilise folded protein structure labelled: ionic bonds, hydrogen bonds, hydrophobic interactions, and van der Waals forces Source: unattributed pre-existing site asset

Amino acid propensities for secondary structure: residues branched at Cβ (Val, Ile) destabilise the α-helix but suit β-sheet; residues whose R group can hydrogen-bond (Ser, Asp, Asn) also destabilise the α-helix by competing with backbone H-bonds; Gly (too flexible) and Pro (too rigid, no amide H) are exceptions unsuited to either Source: unattributed pre-existing site asset

A useful rule of thumb: secondary-structure hydrogen bonding is local, roughly n to n+3/n+4 — only nearby residues interact directly.

Supersecondary structure (motifs) combine a small number of secondary elements in a recognisable pattern — α-α-α, α-β-α, β-β-β — without yet forming a full folded domain.

Eight example supersecondary motifs from solved PDB structures, combining helices (H) and extended strands (E) in patterns such as helix-loop-helix and sheet-loop-sheet, each labelled with its PDB source code Source: unattributed pre-existing site asset

Tertiary structure is the overall 3D fold of a single polypeptide, formed by packing secondary elements into compact domains. It is stabilised by, in rough order of contribution:

  1. Hydrophobic interactions — the dominant driving force; nonpolar side chains bury in the interior to avoid water, polar side chains stay solvent-exposed.
  2. Hydrogen bonds between polar side chains, or side chain–backbone — individually weak but numerous and directional.
  3. Ionic bonds (salt bridges) between oppositely charged side chains (e.g. Lys⁺···Asp⁻).
  4. Disulfide bonds — the only covalent tertiary-level bond, cross-linking two cysteine sulfurs.
  5. Van der Waals forces — weak, short-range, but numerous in a densely packed hydrophobic core.

A folded polypeptide backbone with the tertiary stabilising interactions labelled at their respective sites: an ionic bond between charged side chains, a hydrogen bond, hydrophobic interactions clustering nonpolar side chains together, and a covalent disulfide linkage between two cysteines Source: unattributed pre-existing site asset

Because most of these forces are non-covalent, tertiary structure is fragile: heat, extreme pH, or heavy metals disrupt them, causing denaturation — loss of function without necessarily breaking the primary sequence.

Quaternary structure arises when multiple independently folded (tertiary) subunits associate — often via the same non-covalent forces listed above — into one functional multimeric complex.

Quaternary protein structure examples: dimers (homodimer vs. heterodimer), a trimer (collagen’s triple helix), and a tetramer (hemoglobin’s α2β2 assembly) Source: unattributed pre-existing site asset

Hemoglobin: the model system for allostery and cooperativity

Allostery is regulation of a protein’s activity by a ligand binding somewhere other than the active/functional site, via an induced conformational change. Hemoglobin (Hb), a tetramer of two α- and two β-globin chains, is the canonical teaching example.

T (tense) vs. R (relaxed) state: deoxy-Hb sits in the T state, held by 8 inter-subunit salt bridges, with low O₂ affinity. When O₂ binds one heme iron, the iron (previously pulled slightly out of the porphyrin plane by the proximal histidine) moves into the plane, tugging the attached helix and breaking salt bridges — a conformational change that propagates cooperatively across all four subunits, flipping the whole tetramer to the R state with high O₂ affinity.

O2 saturation (Y) vs. pO2 for basal Hb compared with locked T-state and locked R-state polymerised hemoglobin (fitted curves plus experimental data points): the T-state curve sits far right (low affinity), the R-state curve sits far left (high affinity), and basal Hb’s sigmoidal curve falls between them Source: unattributed pre-existing site asset

Cooperativity means O₂ binding at one subunit raises affinity at the others, producing a sigmoidal binding curve (vs. the hyperbolic curve of monomeric myoglobin, which has no cooperativity):

$$ \theta = \frac{[O_2]^{n_H}}{K_d^{n_H} + [O_2]^{n_H}} $$

For Hb, the Hill coefficient n_H ≈ 2.8 (out of a theoretical maximum of 4 for perfect cooperativity); for myoglobin, n_H = 1. You do not need to derive this equation for most olympiad purposes — understand what n_H means (degree of cooperativity) and its bounds.

Generic fraction-bound vs. free-ligand-concentration curves for Hill coefficients of 4, 2, 1, 0.5, and 0.2: higher coefficients produce steeper, more sigmoidal curves (stronger positive cooperativity), a coefficient of 1 gives a hyperbolic curve (no cooperativity), and coefficients below 1 indicate negative cooperativity Source: unattributed pre-existing site asset

Heterotropic effectors — ligands other than O₂ that shift Hb’s affinity:

Effector Effect on O₂ affinity Mechanism
2,3-bisphosphoglycerate (2,3-BPG) Decreases Binds the central cavity of the T state, stabilising it
CO₂ Decreases Forms carbamino-Hb; also lowers pH
H⁺ (low pH, the Bohr effect) Decreases Protonates β-chain His HC3, stabilising T state
CO Increases (pathologically) Binds the same site as O₂ with >200× affinity; shifts remaining subunits toward R, reducing effective cooperativity and causing hypoxia despite normal Hb saturation

At the pO₂ found in respiring tissue (~40 mmHg), myoglobin — with its hyperbolic, high-affinity curve — retains far more bound O₂ than hemoglobin, which is exactly the point of the two proteins having different curve shapes: Hb is built to release O₂ readily in tissue, Mb to hold onto it as an intracellular O₂ store.

Oxygen saturation vs. pO2 for hemoglobin (blue, sigmoidal) and myoglobin (red, hyperbolic), with venous-tissue and lung pO2 ranges shaded and the ~40 mmHg tissue point marked: myoglobin sits at ~93% saturation there vs. hemoglobin’s ~76% Source: unattributed pre-existing site asset

Chaperones and protein folding

Anfinsen’s dogma: a protein’s final fold is fully encoded in its primary sequence — chaperones do not dictate the fold. What they do is prevent off-pathway aggregation, by transiently shielding exposed hydrophobic patches during and after translation, before folding completes.

GroEL/GroES (Hsp60 family) cycle: (1) the unfolded protein enters the GroEL barrel, whose inner surface is initially hydrophobic; (2) ATP binding flips the inner surface hydrophilic, creating a protected folding chamber; (3) the GroES cap closes over the chamber; (4) ATP hydrolysis (~10 s later) releases the protein, correctly or incorrectly folded; (5) incorrectly folded protein re-enters the cycle.

Hsp70 (DnaK in bacteria) binds extended peptide co-translationally, assisted by co-chaperones DnaJ (Hsp40) and the nucleotide-exchange factor GrpE.

Misfolding diseases

Disease Protein Misfolding event
Sickle cell anaemia β-globin (HbS, Glu6Val) The substituted valine creates a hydrophobic surface patch; deoxy-HbS polymerises into rigid fibres
Cystic fibrosis CFTR (most commonly ΔF508) Misfolded CFTR is retained in the ER and degraded by ERAD before ever reaching the plasma membrane
Prion disease (CJD, scrapie) PrP Native α-helix-rich PrPᶜ converts to β-sheet-rich PrPˢᶜ; PrPˢᶜ templates the same conversion in native PrPᶜ molecules it contacts — a self-propagating misfold requiring no nucleic acid
Alzheimer’s disease Amyloid-β / Tau Aβ aggregates into cross-β amyloid plaques; Tau forms intracellular neurofibrillary tangles
Parkinson’s disease α-synuclein Aggregates into Lewy bodies

Prion disease is a favourite exam topic precisely because it breaks the central dogma’s usual information flow: heritable, infectious biological information is propagated by protein conformation alone, with no nucleic acid template. This is also why prions resist UV/nuclease treatment (which target nucleic acids) and require harsh treatment (autoclaving with NaOH) for inactivation.

The ubiquitin-proteasome system

Beyond phosphorylation and glycosylation, the ubiquitin-proteasome system (UPS) is the cell’s primary route for regulated, targeted protein destruction. Ubiquitin (Ub), a 76-residue protein, is attached to a target lysine via an isopeptide bond in a three-enzyme cascade:

$$ \text{E1 (Ub-activating)} \rightarrow \text{E2 (Ub-conjugating)} \rightarrow \text{E3 (Ub-ligase — provides substrate specificity)} $$

The linkage type determines the outcome:

  • Lys48-linked polyubiquitination → targets the protein for proteasomal degradation
  • Lys63-linked polyubiquitination → non-degradative signal (DNA repair, endosomal sorting)
  • Monoubiquitination → histone regulation, membrane protein trafficking

The 26S proteasome is a 20S catalytic barrel (with chymotrypsin-like, trypsin-like, and caspase-like protease activities) capped by two ATP-dependent 19S regulatory particles, which recognise polyubiquitin chains, unfold the substrate, and thread it into the barrel. This system is the drug target of bortezomib, a proteasome inhibitor used in multiple myeloma.

Fibrous vs. globular proteins

Structural role correlates strongly with secondary-structure composition: α-keratin is helix-rich with disulfide cross-links, giving toughness; silk fibroin is β-sheet-rich, giving flexibility over tensile toughness; collagen forms a unique triple helix (distinct from the α-helix), giving very high tensile strength. Globular proteins (e.g. hemoglobin) fold compactly with a hydrophobic core, favouring solubility and functional flexibility (conformational switching, as in the T↔R transition above) over mechanical strength.

Comparative Structures

Feature α-Helix β-Sheet
H-bond pattern Intrachain, local (i to i+4) Interchain or distant intrachain, extended
Residues per turn/repeat 3.6 residues/turn ~2 residues per repeat unit, sheet is a zig-zag
High-propensity residues Ala, Glu, Leu, Met, Lys Trp, Tyr, Phe, Val, Ile, Thr
Breakers Pro, Gly Pro, bulky charged residues (Glu, Lys)
Typical role Membrane-spanning segments, structural coils (keratin) Structural sheets (fibroin), extended binding surfaces

Common Exam Questions

  • “Explain why proline and glycine disrupt α-helices” — always give the chemical reason (cyclic side chain / no amide H for Pro; excess backbone flexibility for Gly), not just “they’re different.”
  • “At tissue pO₂, which protein — hemoglobin or myoglobin — holds more oxygen, and why does this matter physiologically?” — tests whether you understand why the two curve shapes exist, not just that they differ.
  • Distinguishing whether a stabilising interaction is covalent (disulfide bond only) vs. non-covalent (everything else in tertiary structure) is a common trap in “which bond breaks first on denaturation” questions.
  • Prion disease questions often test the “no nucleic acid required” concept specifically — watch for answer choices that assume all heritable/infectious agents must involve DNA or RNA.

Visual Reference

Interactive

  • A clickable hemoglobin T-state/R-state toggle: view the tetramer, click to trigger the conformational shift, watch the salt bridges break and reform.

Static

(Static images are placed inline in Key Concepts above, next to the concept each one illustrates, rather than collected here. One item — a primary-structure diagram showing a peptide bond chain with a disulfide cross-link — has no sourced image yet: the original candidate, MCBBPICS/secondary.png, turned out to depict secondary structure instead and was placed there. Still needs sourcing.)

Practice Problems

1. Plot (conceptually) the O₂-binding curves of hemoglobin and myoglobin on the same axes. At pO₂ = 40 mmHg (venous tissue), which protein retains more O₂, and why is this the physiologically correct behaviour for each protein’s role?

Show answer

Myoglobin’s hyperbolic curve sits to the left of (higher affinity than) hemoglobin’s sigmoidal curve across the whole range, so at pO₂ = 40 mmHg myoglobin is still highly saturated while hemoglobin has released a substantial fraction of its bound O₂. This is functionally correct: myoglobin’s job is to store O₂ inside muscle cells and only release it under extreme local hypoxia (heavy exertion); hemoglobin’s job is to transport O₂ and unload a large, tunable fraction of it in respiring tissue — the sigmoidal curve, shifted further right by 2,3-BPG, CO₂, and low pH exactly where those signals of active metabolism are highest, makes this unloading responsive to local tissue demand.

2. A missense mutation converts a helix-interior alanine to proline in an otherwise stable α-helical domain. Predict the structural consequence, and explain the mechanism.

Show answer

Inserting proline mid-helix introduces a rigid kink and removes an amide hydrogen the helix’s hydrogen-bonding pattern requires at that position, locally disrupting or terminating the helix. The mutation is likely destabilising unless the position happens to sit at a natural turn/kink already present in the native structure.

3. Two proteins have identical Lys48-linked polyubiquitin chains but different fates: one is degraded within minutes, the other persists for hours despite similar chain length. Propose two non-ubiquitin-related explanations.

4. Explain, in mechanistic terms, why CO poisoning causes tissue hypoxia even when overall Hb-bound oxygen content may not be far below normal.