A Bacterial Surface Layer Protein Exploits Multi-step Crystallization for Rapid Self-assembly

Abstract

24Surface layers (S-layers) are crystalline protein coats surrounding microbial cells. S-layer 25 proteins (SLPs) regulate their extracellular self-assembly by crystallizing when exposed to an 26 environmental trigger. However, molecular mechanisms governing rapid protein crystallization 27 in vivo or in vitro are largely unknown. Here, we demonstrate that the C. crescentus SLP readily 28 crystallizes into sheets in vitro via a calcium-triggered multi-step assembly pathway. This 29 pathway involves two domains serving distinct functions in assembly. The C-terminal 30 crystallization domain forms the physiological 2D crystal lattice, but full-length protein 31 crystallizes multiple orders of magnitude faster due to the N-terminal nucleation domain. 32 Observing crystallization using time-resolved electron cryo-microscopy (Cryo-EM) reveals a 33 crystalline intermediate wherein N-terminal nucleation domains exhibit motional dynamics with 34 respect to rigid lattice-forming crystallization domains. Dynamic flexibility between the two 35 domains rationalizes efficient S-layer crystal nucleation on the curved cellular surface. Rate 36 enhancement of protein crystallization by a discrete nucleation domain may enable engineering 37 of kinetically controllable self-assembling 2D macromolecular nanomaterials. 38 39 Keywords: protein self-assembly, time-resolved Cryo-EM, microbiology, biophysics 40 41 Significance Statement 42Many microbes assemble a crystalline protein layer on their outer surface as an additional 43 barrier and communication platform between the cell and its environment. Surface layer proteins 44 efficiently crystallize to continuously coat the cell and this trait has been utilized to design 45 functional macromolecular nanomaterials. Here, we report that rapid crystallization of a bacterial 46 surface layer protein occurs through a multi-step pathway involving a crystalline intermediate. 47Upon calcium-binding, sequential changes occur in the structure and arrangement of the protein, 48 which are captured by time-resolved small angle x-ray scattering and transmission electron cryo-49 microscopy. We demonstrate that a specific domain is responsible for enhancing the rate of self-50 assembly, unveiling possible evolutionary mechanisms to enhance the kinetics of 2D protein 51 crystallization in vivo. 52

High-resolution mapping of metal ions reveals principles of surface layer assembly in Caulobacter crescentus cells

“…Within this study and in previous attempts to crystallize RsaA (Bharat et al, 2017), we found that Mg 2+ could not mediate S-layer stabilization or oligomerization. Supplementing M2G medium with an excess of MgSO 4 to compensate for the exclusion of CaCl 2 resulted in non-fluorescent cells (Figure 1F), suggesting improper S-layer assembly or S-layer shedding, which is in agreement with previous biochemical reports (Herrmann et al, 2020).…”

“…When no additional CaCl 2 was added to the medium (Figure 1E), the resulting cell fluorescence readings were indistinguishable from unlabeled control cells (Figure 1G) and drastically lower than cell fluorescence intensities seen in cells in 500 or 250 mM CaCl 2 . These observations support the hypothesis that the formation of a viable S-layer is strongly dependent on the concentration of Ca 2+ in the growth medium and shows that even a small reduction in the available CaCl 2 can significantly affect the retention and display of the S-layer on cells, in agreement with other reports (Herrmann et al, 2020). To probe the Mg 2+ dependence of S-layer assembly, we included a sample of C. crescentus cells grown in M2G containing 0 mM CaCl 2 and 1,000 mM MgSO 4 (Figure 1F).…”

“…Previous studies have suggested that growth medium with CaCl 2 concentrations lower than 250 mM trigger shedding of the S-layer into the extracellular milieu and compromise C. crescentus fitness and viability (Herrmann et al, 2020). In addition to confirming those results, we further demonstrated that, under these conditions, newly secreted RsaA is inadequately incorporated into the S-layer in C. crescentus cells, but that the old S-layer remains at least partially attached to cells at CaCl 2 concentrations as low as 100 mM.…”

“…RsaA has a canonical bipartite arrangement (Figure 1A), with a lattice-forming C-terminal domain (RsaA CTD ), consisting of residues 251-1,026, and a cell-anchoring N-terminal domain (RsaA NTD ), consisting of residues 1-250. The assembly of the RsaA S-layer is divalent metal-ion dependent (Nomellini et al, 1997;Bharat et al, 2017;Herrmann et al, 2020), as, without divalent metal ions, S-layer formation is inhibited both in vitro and on cells (Bharat et al, 2017). Furthermore, cryoelectron microscopy (cryo-EM) and cryoelectron tomography (cryo-ET) investigations have demonstrated that binding of RsaA to lipopolysaccharide (LPS) molecules on the C. crescentus cell surface is also regulated by Ca 2+ ions (von K€ ugelgen et al, 2020).…”

“…Advances in structural biology and imaging techniques (Beck and Baumeister, 2016;Oikonomou and Jensen, 2017) have improved our understanding of S-layer biogenesis and assembly, revealing that S-layers have primarily a bipartite arrangement, often with separated lattice-forming and cell-anchoring domains in their constituent SLPs (Bharat et al, 2021;von K€ ugelgen et al, 2020;Gambelli et al, 2019;Phipps, 1990;Baumeister and Lembcke, 1992;Wildhaber and Baumeister, 1987;Pum and Sleytr, 2014;Veith et al, 2009). Another common feature of S-layers is lattice assembly mediated by the presence of divalent metal ions in the extracellular environment (Engelhardt, 2007a(Engelhardt, , 2007b, which has been observed in Archaea (Cohen et al, 1991;Kessel et al, 1988), Gram-positive (Baranova et al, 2012;Lupas et al, 1994), and Gram-negative bacteria (von K€ ugelgen et al, 2020;Bharat et al, 2017;Herrmann et al, 2020).…”