The bacterial chaperone CsgC inhibits functional amyloid CsgA formation by promoting the intrinsically disordered pre-nuclear state

Abstract

Escherichia coli assembles a functional amyloid called curli during biofilm formation. The major curlin subunit is the CsgA protein, which adopts a beta-sheet-rich fold upon fibrillization. The chaperone protein CsgC inhibits CsgA amyloid formation. CsgA undergoes a 3-stage aggregation process: an initial lag phase where beta-rich nuclei form, an exponential elongation phase, and a plateau phase. It is currently not known whether CsgC inhibits amyloid formation by inhibiting the formation of a pre-fibril nucleus or whether CsgC inhibits a later stage of amyloid formation by blocking monomer addition. Here, CsgC homologs from Citrobacter youngae, Cedecea davisae, and Hafnia alvei were purified and characterized for their ability to interrogate CsgA amyloid formation. Each of the CsgC homologs prolonged the lag phase of E. coli CsgA amyloid formation, similar to E. coli CsgC. Additionally, we found E. coli CsgC interacted transiently and weakly with a monomeric, pre-nucleus species of CsgA, which delayed amyloid formation. A transient CsgC-CsgA heterodimer was observed using ion mobility-mass spectrometry. When CsgC was added to actively polymerizing CsgA, exponential growth commonly associated with nucleation-dependent amyloid formation was lost. Adding preformed CsgA seeds did not rescue exponential growth, indicating that CsgC also has inhibitory activity during fibril elongation. Indeed, CsgC interacted strongly with CsgA fibers, suggesting that the interaction between CsgC and CsgA fibers can slow new fiber growth. CsgC displays unique inhibitory activity at multiple stages of amyloid formation and acts as an energy-independent chaperone that transiently interacts with prefibrillar CsgA and an amyloid fiber.

Keywords: amyloid; chaperone; inhibition mechanism; mass spectrometry; surface plasmon resonance.

Figure 1

Comparing inhibition of E. coli CsgA amyloid formation by EC CsgC and its homologs.A–D, AlphaFold 2.0 predicted structures for CsgC EC, CY, CD, and HA. E–H, four different CsgC homologs were purified and tested for their ability to inhibit CsgA in vitro. In all cases, 20 μM CsgA was freshly purified and mixed in the stated stoichiometric ratio with CsgC homologs. The data points shown represent the average of triplicate experiments, the error bars show the SEM, and the curves were fit using the sigmoidal logistic function described in (24). I, the calculated lag phase for each homolog according to the fit curves. The error bars represent the SEM for each value. “nd” was used to denote values that were not defined by the equation used. Significance was attributed using a one-way ANOVA analysis comparing all values to the uninhibited CsgA condition; ∗p < 0.05, ∗∗∗∗p < 0.00005.

Figure 2

Addition of CsgC to actively polymerizing CsgA.A, 200 nM (1:100) CsgC was added to actively polymerizing CsgA at the times indicated and amyloid aggregation was monitored in a ThT binding assay. B, in a similar assay, CsgC was added to a 2% (400 nM) seeded reaction of CsgA at various ratios as indicated in the key above (All olive conditions also contain 2% seeds).

Figure 3

Sensograms of CsgA fibrils and monomers probed by CsgC.A, sensogram data of CsgA fibrils and CsgC, showing an equilibrium dissociation constant (K_D) of 360 ± 10 nM. The concentrations labeled “A” represent the first injection, while “B” denotes the secondary injection. The fitted curve represents the curve fitting performed using Scrubber2. B, sensogram data of CsgA monomers and CsgC across various tested concentrations. Residual standard deviation = 3194 RU.

Figure 4

Passing monomeric CsgA by immobilized CsgC delayed the formation of ThT-positive aggregates.A, affinity-purified CsgA was passed through a column of NHS agarose resin that was either amino-coupled to CsgC or Tris base to quench the free NHS binding spots. B, eluent CsgA was then diluted to 10 μM in phosphate buffer for a ThT binding assay to observe amyloid aggregation kinetics. C, quenched and CsgC resin was incubated in phosphate buffer for 24 h. The incubated phosphate buffers were used to dilute freshly purified CsgA to 10 μM and added to a ThT binding assay. D, a measurement of the lag phase (T_lag) and time to half maximum fluorescence (T_1/2) was calculated using a sigmoidal curve function. Error bars represent the standard error of the mean across three technical replicates. Significance was attributed using a student’s t test analysis comparing the two conditions; ∗p < 0.05.

Figure 5

IM-MS to detect CsgC-CsgA interaction.A, proteins were passed through an ion mobility drift cell that separates based on size. Next, proteins pass through a time-of-flight mass analyzer. In combination, IM-MS can provide three modes of information for a given ion: mass to charge ratio, drift time, and the intensity of species with those properties. B, mass spectra for CsgA incubated with CsgC in a 1:1 M ratio at 37 °C. Monomeric CsgA (blue circles), dimeric CsgA (blue double circles), monomeric CsgC (purple triangles), dimeric CsgC (double purple triangles) and 1:1 CsgA:CsgC complexes (blue circle and purple triangle) are annotated. A magnified MS spectrum (C) showed that the complex is seen flanked on either side by dimeric CsgA and dimeric CsgC. D, arrival time distribution of the 11+ monomer of CsgA in three experimental conditions: Apo CsgA in solution at 0 h (olive trace), apo CsgA in solution after 3 h (light purple trace), and CsgA in solution with CsgC after 3 h (dark grey trace). Traces are overlapped to show the changes in ATD after 3 h of incubation.

Figure 6

Schematic of CsgC inhibitory mechanism. CsgC inhibits CsgA nuclei formation leading to a prolonged lag phase. This interaction between prenuclear CsgA and CsgC is transient (left). CsgC can also bind stably to CsgA fibrils when added to a mixed CsgA population containing monomers, oligomers, and growing fibrils. This stable interaction prevents fibril growth by either preventing fibril elongation or secondary nucleation (right).