Probing the oligomeric re-assembling of bacterial fimbriae in vitro: a small-angle X-ray scattering and analytical ultracentrifugation study

Abstract

Capsular antigen fragment 1 (Caf1) is an oligomeric protein consisting of 15 kDa monomeric subunits that are non-covalently linked through exceptionally strong and kinetically inert interactions into a linear polymer chain. It has been shown that after its thermal depolymerisation into unfolded monomeric subunits, Caf1 is able to efficiently repolymerise in vitro to reform its polymeric structure. However, little is known about the nature of the repolymerisation process. An improved understanding of this process will lead to the development of methods to better control the lengths of the repolymerised species, and ultimately, to better design of the properties of Caf1-based materials. Here we utilize small-angle X-ray scattering to estimate the size of Caf1 polymers during the first 24 h of the re-polymerisation process. Analytical ultracentrifugation measurements were also used to investigate the process post-24 h, where the rate of repolymerisation becomes considerably slower. Results show that in vitro polymerisation proceeds in a linear manner with no evidence observed for the formation of a lateral polymer network or uncontrolled aggregates. The rate of Caf1 in vitro repolymerisation was found to be concentration-dependent. Importantly, the rate of polymer growth was found to be relatively fast over the first few hours, before continuing at a dramatically slower rate. This observation is not consistent with the previously proposed step-growth mechanism of in vitro polymerisation of Caf1, where a linear increase in polymer length would be expected with time. We speculate how our observations may support the idea that the polymerisation process may be occurring at the ends of the chains with monomers adding sequentially. Our findings will contribute towards the development of new biomaterials for 3D cell culture and bio-printing.

Keywords: Flexibility; Mass increment; Oligomeric growth; Repolymerisation.

Fig. 1

Structure and reversible thermal unfolding of Caf1 polymers. a Model of a segment of Caf1 polymer (generated from PDB entry 1P5U). The N-terminal donor strands (coloured orange) are inserted into the acceptor cleft of adjacent subunits. Protein structures were visualised using the CCP4MG molecular graphics package (McNicholas et al. 2011). b Cartoon depicting the reversible thermal unfolding of Caf1 polymers. When heated above the subunit melting temperature (T_m > 86 °C) the subunits (blue) thermally unfold with concomitant decomplexation of the N-terminal donor strands (orange), causing Caf1 to depolymerize into its unfolded monomeric form. When the subunits are incubated at room temperature they subsequently refold with donor strand recomplexation and concomitant repolymerisation, with a fraction of unfolded monomer still present (c)

Fig. 2

Experimental design of SAXS measurements for the Caf1 post-denaturation repolymerisation time course (a); a set of resulting scattering curves (the sample concentration was 5 mg/mL) in reciprocal space (b); and selected scattering curves from the same set represented as total scattering intensity versus scattering angle (c). These curves are shown together with the hypothetical total scattering intensity from a Gaussian coil (R_g = 41.3 Å, c = 5 mg/mL) (dashed line) which simulates the scattering of a disordered Caf1 monomer at zero time

Fig. 3

SAXS standard plots. a P(r) distributions show an increase in scattering particles’ maximal distance and similar positions for two main peaks on the distribution during the time course (the sample concentration was 5 mg/mL). b Guinier plot for the time-course sample at the concentration of 5 mg/mL. c Values of cross-sectional radius of gyration <R_c>_z do not substantially change with concentration and time for the repolymerazing sample. A smaller value of R_c was obtained for native Caf1^WT. d Shape of a dimensionless Kratky plot suggests multimeric scattering particles elongating with time (shown for the sample concentration of 5 mg/mL)

Fig. 4

Increase in z-average values of radius of gyration (a) and in weight-averaged number of chain subunits (b) of repolymerised Caf1 during the time course. The concentration of repolymerized Caf1 affects the rate of increase. <R_g>_z values were obtained from P(r) distribution and the average number of subunits (<N>_w) in the repolymerised chain was calculated on the basis of <$M$>_w values evaluated using the Guinier approximation. The samples at 1 mg/mL were examined for the fast stage only

Fig. 5

Linear model of Caf1 oligomer (cartoon in green-top panel) represented as various form-factor models: rigid cylinder (whole oligomer), flexible cylinder (every monomer or group of monomers in the chain is considered as fragments of flexible cylinder), and shape-independent mass fractal (spherical subunit monomers) and polymer with excluded volume—a basic model for scattering of various polymeric chains. The bottom panel shows an illustrative example of the experimental scattering curve fit to these models. The native Caf1^WT modelled as a flexible cylinder (black line) at 1 mg/mL, 2 mg/mL, 5 mg/mL and 10 mg/mL and a polymer with excluded volume (red line) at 10 mg mg/mL (a). Repolymerising Caf1 at 36 min post-denaturation, (sample concentration 5 mg/mL) modelled as a rigid cylinder (green line), flexible cylinder (black line) fractal (blue line) and a polymer with excluded volume (red line) (b)

Fig.6

Changes in the geometry of a repolymerized Caf1 chain in terms of fractal dimension. Fractal dimension drops very quickly during the first hour of repolymerisation suggesting the formation of polymers in the form of wiggly lines D ≅ 1.9, ultimately approaching the D value of 1.72 detected for the Caf1 sample repolymerised for more than 24 h in advance (the concentration of repolymerising sample was 13 mg/mL, whereas this sample concentration in SAXS experiment was 6.75 mg/mL). The changes in fractal dimension during first hour of repolymerisation are shown in the insert. The data obtained for the low concentration (2 mg/mL) repolymerized material, which polymerizes slower than at the higher concentrations, demonstrates slower linearization, probably on account of a relatively large amount of disordered monomeric material being present during the polymerisation

Fig. 7

Repolymerised Caf1 samples analysed by sedimentation velocity analytical ultracentrifugation experiments (the rotation speed was 20,000 rpm) modelled as 1D c(M) (a) and 2D c(M, f/f₀) (b) size distributions. Long storage time result in the appearance of heavier and longer species (marked with red arrows), while smaller species still exists (marked with yellow arrows) as seen from the 2D distribution c(M, f/f₀) (c)