The preprint demonstrates that E. coli H-NS and its paralog StpA (58% sequence identity) display diametrically opposed phase behaviors: purified H-NS forms fluid condensates with csat ≈25 µM (reduced to ≈5 µM with 150 bp DNA), whereas StpA precipitates as insoluble fibrils even at 1.25 µM. When mixed, H-NS dissolves pre-formed StpA fibrils into liquid co-condensates, reducing the effective csat by ~20-fold, and the StpA:H-NS ratio tunes fluidity, salt resistance (from dissolution at 100 mM NaCl up to fibril-like stability at 500 mM), and thermal stability. In vivo, StpA:H-NS increases ~4-fold from 30°C to 42°C, and ΔstpA strains lose foci formation, gene silencing (dps, gadA, hdeB), and competitive fitness specifically under heat/osmotic stress .
All-atom MD (200 ns, 0.5 M KCl, GROMACS/CHARMM36m) shows StpA homotetramers adopt pronounced curvature with a stable D2 interface (Lys57, Arg62, Asp68) and DBD–D1 inter-dimer contacts absent in linear, stress-labile H-NS oligomers. Domain-swapping confirms the oligomerization domain dictates phase behavior: the H-S-S chimera phase-separates at csat ≈5 µM while S-H-H is insoluble. The H-NS L-A3R2 linker mutant provides a key control: it forms nucleoprotein filaments but depends on StpA to form condensates, linking condensation loss to dps derepression and growth defects—direct physiological evidence that condensates matter, not just filaments .
The authors reframe StpA from 'molecular backup' to stoichiometric support system—an idea consistent with broader paralog-evolution literature showing that paralogs frequently retain complex membership and diverge asymmetrically rather than acting redundantly . The HP1 paralog analogy (HP1β dissolving HP1α/γ condensates) is a credible eukaryotic parallel, though it remains speculative. Key limitations the authors acknowledge: no high-resolution structure of mixed fibrils, no exploration of Hha or PTMs, single strain (MG1655). Additional blind spots BGPT identifies: fluorescent tags could perturb condensate dynamics (partially addressed with 3xFLAG controls); the 4-fold StpA:H-NS shift conflates transcriptional, translational, and Lon-proteolysis contributions; and competitive fitness assays at 42°C cannot isolate condensate-stabilization from StpA's other nucleic-acid chaperone roles. Falsification would require showing ratio-independence of condensate stability or stress-independent gene silencing in ΔstpA—neither observed here.
Know what changed, what holds up, and what remains uncertain. Every Friday. No ads.