Past exam of the biochemistry course of the University of Cambridge / 2025 / Paper 5

Partial proteolysis is one plausible explanation: intact SNAP25 and a slightly shortened form would migrate as two nearby bands in SDS-PAGE. Heterogeneous processing, a retained affinity tag, or an anomalously migrating species are other possibilities. The gel alone does not identify the chemical difference. Because the proteins were expressed in Escherichia coli, a mammalian post-translational modification should not simply be assumed. A Western blot establishing that both bands contain SNAP25, followed by mass spectrometry, could distinguish these explanations.

The liposomes could be separated from free proteins by density-gradient flotation: lipid-containing particles float to a different position from soluble protein. Size-exclusion chromatography is another possibility if it resolves the large vesicles from the unincorporated proteins. Simple sedimentation can also collect sufficiently large vesicles, but protein aggregates may sediment too. The important control is to show that the recovered SNARE proteins associate with the lipid bilayer, rather than merely copurifying with it.

The recovered vesicle fractions contain bands at the expected positions for the supplied SNARE proteins. The v-SNARE preparation contains its principal band, and the t-SNARE preparation contains both t-SNARE and the SNAP25 doublet. Thus protein-containing vesicle fractions have been recovered, consistent with successful reconstitution.

The SDS-PAGE experiment does not establish the orientation of the proteins, their proportion facing outward, the stoichiometry in an individual vesicle, or membrane fusion. Nor does a stained band prove that all of its protein is correctly folded and active. Those are separate functional and membrane topology questions.

After warming, more labelled v-SNARE remains full length: lane 2 is approximately 80–85% of the reference signal, compared with about 40–45% in lane 1. Formation of a protected SNARE complex provides a consistent explanation: the susceptible part of v-SNARE is less accessible after complex assembly.

Protection is not itself a unique readout of membrane fusion. Some protein may face the vesicle lumen, and SNARE complexes can protect protein without proving that the two aqueous compartments have merged. The inference depends on comparing these alternatives with the detergent and partner-free controls.

In lane 3, detergent removes the lipid bilayer barrier while both partner preparations are present. Their binding regions can then encounter one another even if they originally faced inward or were on poorly apposed lipid bilayers. The strong remaining full-length signal is consistent with extensive formation of protected SNARE complexes.

Thus detergent does not prevent partner association under the displayed conditions; it removes the topological restriction on that association. This lane is the normalization reference, not a direct measurement that 100% of all initially supplied protein is active. The weak signal in the detergent-treated partner-free control also argues against detergent simply preventing all cleavage.

Lane 4 combines detergent with the absence of the t-SNARE preparation. The lipid bilayer no longer protects inward-facing v-SNARE, and there is no t-SNARE partner to provide complex-dependent protection. Consequently little full-length labelled protein remains. This distinguishes membrane-dependent protection from partner-dependent protection in a protease protection assay. It also helps interpret lane 3: its strong signal requires the partner rather than lipid bilayer enclosure alone.

Use a membrane topology measurement on v-SNARE vesicles alone, so that partner binding cannot mask an outward-facing domain. A membrane-impermeant probe directed against the binding-domain region would label only the exposed population while vesicles remain intact. Compare that signal with total accessible protein after lipid bilayer disruption, using a trapped-lumen marker to check that the intact vesicles exclude the probe.

With comparable detection efficiencies and background subtraction,An analogous protease protection assay compares external accessibility with total accessibility. In either approach, lipid bilayer leakage and steric masking must be distinguished from genuinely inward-facing protein.

NBD and rhodamine are initially close together in the labelled vesicle lipid bilayer, allowing Förster resonance energy transfer and suppressing donor fluorescence. Mixing labelled lipid bilayer with unlabelled lipid bilayer dilutes the dyes, decreases energy transfer and increases the NBD signal.

The rise with the complementary SNARE proteins, but little rise with protein-free vesicles, therefore supports SNARE-dependent lipid mixing. It is consistent with membrane fusion, but does not by itself distinguish complete fusion from hemifusion or other causes of dye dilution. A separate content mixing assay with a leakage control would test merger of the aqueous compartments.

There is no evidence for substantial lipid mixing during the cold preincubation. The trace begins near its low-fluorescence baseline when warming starts; a substantial pre-existing dilution of the dyes would already have raised that baseline. Nevertheless, vesicle docking or trans-SNARE complex formation can accumulate while the later membrane-rearrangement step remains slow. Absence of dequenching does not mean that the vesicles have had no interaction.

Cold preincubation can accumulate docked, fusion-ready intermediates. On warming, that population proceeds promptly to lipid mixing, whereas newly combined vesicles must first encounter, dock and assemble productive SNARE complexes. This explains the faster initial rise and larger signal at the displayed endpoint without requiring a different total amount of fluorescent lipid.

The original reconstitution experiments likewise distinguish docking from subsequent membrane fusion.^[ref]

Triton disperses the lipid bilayer into detergent-containing particles, separating the fluorescent lipids enough to remove most fluorescence quenching. Both traces approach the maximum-dilution reference even though their earlier lipid-mixing histories differed. The final fluorescence jump is detergent-induced dequenching, not an additional burst of physiological membrane fusion. It provides a useful normalization endpoint for the available dye signal.

The N-terminal 1–29 fragment and the 1–15 fragment inhibit invadopodium formation. Their values are around one fifth of the control, and gelatin degradation is strongly reduced too. Full-length t-SNARE, empty vector and the 15–29 fragment remain near control values. The images agree: the inhibitory fragments leave much more intact gelatin beneath the cells. The corresponding published syntaxin-4 study links this phenotype to impaired cargo trafficking.^[ref]

The 1–15 fragment has a slightly lower mean degradation signal and perhaps lower invadopodium formation than the longer N-terminal fragment. It may be the stronger inhibitor, but the displayed comparisons do not establish that difference. The asterisks compare groups with control, not necessarily the two active fragments with one another. Their error bars overlap, and overlap alone is neither a definitive test of equality nor a calibrated significance test.

A direct comparison needs replicate-level data and a suitable statistical hypothesis test, with an effect size and confidence interval. Comparable fragment abundance, cell viability and cell number are also needed before interpreting a phenotypic difference as greater molecular potency.

Three complementary readouts would separate local matrix degradation from invasive and metastatic behaviour:

Empty-vector and inactive-fragment controls are needed throughout. Neither reduced gelatin digestion alone nor fewer cells after treatment establishes a specific effect on metastasis.

The fragment might sequester another factor required for vesicle trafficking or actin organization, rather than acting through Munc18. For example, binding a different SNARE-associated protein could lower the concentration available for productive endogenous complexes. This is a possible dominant-negative effect, not evidence that the isolated N terminus contains the entire SNARE interaction surface.

A less specific alternative is cellular stress or reduced viability. Those effects could suppress both invadopodium formation and matrix degradation without selectively inhibiting the proposed trafficking pathway.

For the sequestration hypothesis, ask whether the fragment associates with the proposed alternative partner by co-immunoprecipitation, then compare endogenous complex assembly and cargo localization. An association should be evaluated against empty-vector and inactive-fragment controls, and an independent interaction readout would help distinguish direct binding from membership of a larger complex.

A useful functional test is whether restoring the limiting partner rescues trafficking and invadopodium formation without changing fragment expression or viability. Rescue would support the hypothesis, although overexpression can itself alter trafficking. Binding evidence, the endogenous phenotype and rescue should therefore agree; any one alone is insufficient.

The S61/E62 and S75/Q76 pairs are especially important for the extra stimulation attributable to Munc18. Their replacement leaves basal lipid mixing broadly present but makes the red and blue traces much closer together. In contrast, replacement of R47/V48 or D57/Q58 retains strong stimulation. The S75/E78/T79 replacement retains a smaller enhancement, suggesting a partial contribution under these conditions.

Compare the Munc18-dependent increment with each construct's own baseline, rather than comparing red endpoints alone. These results identify functionally important regions; they do not prove that every altered side chain directly contacts Munc18. The original reconstitution experiments also support cognate-SNARE recognition.^[ref]

V3 is a plausible responsive partner; V8 is less likely to show the same stimulation. V3 conserves the aligned S/E and S/Q pairs implicated by the V2 comparisons, whereas V8 differs at those positions. This supports a prediction of conserved recognition, not a conclusion that sequence identity guarantees the same kinetic response.

Other residues, protein folding, lipid bilayer orientation and partner compatibility matter too. In particular, comparing the aligned motifs is more informative here than comparing only the variable N-terminal tails. The prediction should be checked by equivalent functional comparisons, rather than inferred from the family names alone.

1 = AB, 2 = AC, 3 = BC. A and B are within the N-terminal portion of the chain and correspond to the darker, lower-error block. C lies in the later portion; the off-diagonal comparisons to it are paler and less confidently positioned.

The matrix reports predicted aligned error, not physical separation: a large matrix value does not mean that the two residues are far apart. It indicates uncertainty in their relative placement after alignment on one residue.^[ref]

The main diagonal contains relatively dark blocks separated by much paler off-diagonal regions. This supports a modular protein with internally more confidently predicted regions but uncertain relative packing. In particular, the broad early region ending near residue 650 and the later region are poorly positioned relative to one another. Smaller blocks within these regions suggest further domain subdivision.

Thus the compact-looking arrangement in the ribbon drawing should not be treated as a uniquely established structure of the whole enzyme. The matrix can suggest protein domain boundaries and flexible inter-domain relationships; it neither proves domain independence experimentally nor establishes that uncertainty is caused by a particular motion.

Using the supplied cluster counts,These are structural clusters represented across the three domains of life under the stated classification. They are not necessarily folds present in every individual species, nor must every cluster be a distinct topological fold. The distinction between a cluster and a fold depends on the structural-similarity criterion. The underlying large-scale analysis explicitly used structural clustering of predicted models.^[ref]

All three groups retain broadly shared structural solutions, while also containing more restricted clusters. The displayed domain-restricted counts are much larger for bacteria and eukaryotes, approximately $8.96\times 10^5$ and $8.32\times 10^5$, than for archaea, approximately $3.81\times 10^4$. Restricted clusters also occur at lower taxonomic levels, consistent with continuing diversification rather than a single early episode of structural innovation.

These counts are not direct rates of new-fold generation. The groups differ in sampling, numbers of represented sequences and species, evolutionary depth, and detectability of relationships. Restriction to a lineage can reflect divergence, differential loss or incomplete sampling as well as origin in that lineage. The figure therefore supports unequal observed structural diversity, but does not by itself identify how rapidly each lineage invents new folds.

There arepossible sequences of length 300 over the standard twenty-amino-acid alphabet. Sampling $10^{20}$ sequences would cover only about $4.9\times 10^{-371}$ of that space, even before allowing repeated sequences. We cannot conclude that all possible protein folds have been explored.

However, this sequence count is not a count of folds: many sequences can share a fold, and many arbitrary sequences may not form a stable folded protein. Evolution also samples sequences non-uniformly. The calculation establishes vastly incomplete sequence-space exploration; proving either exhaustive or non-exhaustive coverage of physically realizable folds needs an additional model relating sequences to structures. The observed structural-cluster count alone supplies no such proof.

Protein structure can remain similar after extensive protein sequence divergence has obscured a sequence-only relationship. A structural comparison may therefore reveal a conserved protein fold, plausible binding pocket or spatial arrangement of candidate catalytic residues that a direct sequence search misses. Such evidence can suggest a functional family and useful hypotheses.

A similar fold does not uniquely determine function. The same architecture can support different activities, and a predicted pocket does not demonstrate binding or catalysis. Confidence in the model, residue-level chemistry and independent biochemical evidence must be considered together. Structural classification complements rather than replaces sequence and experimental analysis.

Reusable protein domains make evolution combinatorial: a new domain combination can join an existing catalytic activity to a binding, localization or regulatory module without inventing every component anew. Gene duplication allows one copy to retain an established role while another diverges; domain shuffling and fusion can create new combinations. Connecting domains can also increase effective local concentration of interacting sites or enable allostery.

These are potential advantages, not guarantees that every fusion is beneficial: domains must still fold and work together. The domain-level analysis behind this example found extensive reuse and new combinations of recognized structural units.^[ref]

In surface plasmon resonance, target protein in the mobile phase passes over the immobilized antibody domain. Binding changes the optical response near the sensor surface; response increases during association and falls when unbound target is washed away. The successive injections give larger responses as the supplied target concentration increases.

For a simple 1:1 model,The reported fit gives $K_D=78\mathrm{nM}$. This is a binding estimate for the tested preparation and model. Surface heterogeneity, transport limitations or rebinding could cause deviations; the visible mismatch between parts of the trace and fit discourages treating the binary model as exact.

Antibody specificity requires comparison with alternatives. Measure binding to unrelated proteins and to appropriate related antigens under comparable conditions, with a reference surface controlling nonspecific sensor binding. Preferential binding to the intended target would support specificity within that tested panel.

Competition with a soluble target, or an independent interaction assay, can help establish that the signal reflects the intended interaction. Specificity for a particular epitope is a stronger claim than specificity for the target protein and needs evidence localizing the contact. A single good-looking binding curve establishes neither universal absence of cross-reactivity nor epitope identity.

A complete bivalent antibody could show greater apparent avidity, provided both protein binding sites can engage suitably spaced antigens on the particle. Dissociation of one arm then need not release the antibody because the other arm retains it nearby. The observed release can consequently be slower than for an isolated monovalent domain.

This does not imply that the intrinsic single-site antibody affinity has improved. Geometry, accessibility and the antibody context could prevent bivalent engagement or even impair the isolated domain's interaction. Thus an altered apparent binding strength is plausible, but its direction and size cannot be inferred solely from conversion to a full antibody.

The unliganded and peptide-bound preparations elute at almost the same volume, near 14 ml. They therefore have similar effective hydrodynamic sizes in this separation. A conformational change can occur without a large change in that measurement. The GFP–peptide-containing complex elutes earlier, near 10.5–11 ml, consistent with a larger particle; the later peak is consistent with excess unbound GFP–peptide.

The experiment supports ligand association but does not alone determine subunit number. Size-exclusion chromatography separates by access to pores, which depends on size and shape rather than mass alone. The similar peptide-free and peptide-bound elution volumes therefore do not establish identical conformations or exact molecular masses.

The main particles are approximately 120 kDa in both conditions. With 30 kDa per protein subunit, this supports a tetramer:The small peptide contribution is not clearly resolved as a separate mass increase at this scale. Together with the microscopy, the results support a conformational change of an existing tetramer, rather than a large change in oligomer number. Mass photometry reports the scattering-derived mass distribution of detected particles; it does not by itself identify which conformation a particle adopts or count small bound ligands precisely.

Yes: the binding data support strong positive cooperativity. On the linear concentration axis, occupancy rises sigmoidally, with reported Hill coefficient $n=3.97\pm 0.13$, far above the noncooperative value one. The Hill equation gives the useful empirical descriptionThe steep response is consistent with the independent evidence for a four-subunit assembly. The fitted half-occupancy concentration is not automatically the microscopic dissociation constant of one isolated site.

Binding alone does not demonstrate which conformational change is coupled to occupancy. Here the microscopy supplies the complementary structural evidence, so together the observations support a cooperative ligand-dependent switch. A Hill coefficient near four is consistent with strongly coupled binding in a tetramer, but is not an independent exact count of ligands.

A: downward when ligand is present; B: downward; C: upward, over the intended protein and ligand concentration range. Ligand should favour its bound Y state, and the ligand-bound subunits should favour tetramer assembly. Without ligand, the unassembled state should predominate, rather than forming a stable ligand-free tetramer.

These preferences are thermodynamically coupled. The ligand must bind preferentially to states participating in the assembled structure strongly enough to compensate for the cost of shifting the unliganded conformational and assembly equilibria. At extremely low protein concentration, association entropy can still favour monomers; “ligand-dependent” does not mean concentration-independent assembly. This is a thermodynamic linkage argument, rather than three independently adjustable arrows.

A single predicted monomer structure is insufficient. It may suggest a possible binding pocket, domain hinge or subunit interface, and a confident multimer model may support a particular assembly hypothesis. Those features do not establish that alternative conformations are populated, that ligand binding shifts their populations, or that occupancy at one site changes another site's binding strength.

Cooperative allostery concerns free-energy relationships among states, not only one static geometry. Structural comparisons across conditions, independent oligomeric-mass measurements and ligand-dependent binding or activity curves are needed to test the proposal. A low-confidence inter-domain orientation is likewise uncertainty about a model, not a prediction that the protein is necessarily an allosteric oligomer.