Spatiotemporal kinetics of the SRP pathway in live E. coli cells

Abstract

Mechanistic details of the signal recognition particle (SRP)-mediated insertion of membrane proteins have been described from decades of in vitro biochemical studies. However, the dynamics of the pathway inside the living cell remain obscure. By combining in vivo single-molecule tracking with numerical modeling and simulated microscopy, we have constructed a quantitative reaction–diffusion model of the SRP cycle. Our results suggest that the SRP–ribosome complex finds its target, the membrane-bound translocon, through a combination of three-dimensional (3D) and 2D diffusional search, together taking on average 750 ms. During this time, the nascent peptide is expected to be elongated only 12 or 13 amino acids, which explains why, in Escherichia coli , no translation arrest is needed to prevent incorrect folding of the polypeptide in the cytosol. We also found that a remarkably high proportion (75%) of SRP bindings to ribosomes occur in the cytosol, suggesting that the majority of target ribosomes bind SRP before reaching the membrane. In combination with the average SRP cycling time, 2.2 s, this result further shows that the SRP pathway is capable of targeting all substrate ribosomes to translocons.

Dynamic binding of the bacterial chaperone Trigger factor to translating ribosomes

“…For TF, a threshold of 0.5 µm 2 s -1 was used during coarse-graining to separate free and ribosome-bound TF. For SRP, 0.8 µm 2 s -1 was used to separate ribosome-bound from free SRP (4.5S RNA + Ffh) and free 4.5S RNA (in agreement with ref 31 ). For analysis of LacY-Halo, all trajectories were fitted to a 1-state model to obtain the spatial distribution of a membrane-bound protein.…”

“…However, at high TF-Halo level, SRPs bind RNCs anywhere in the cytosol, but dissociate from them mainly by the membrane (Fig. 7f), which reflects functional targeting of RNCs to translocons (see reference 31 ). Thus, in the absence of TF-Halo, SRP displays more unproductive RNC bindings, suggesting that the large number of TF molecules in a normal cell actually tune the selectivity of SRP.…”

“…Based on our expectations of TF function, we hypothesized that the slow diffusion corresponds to RNC binding and that the faster diffusion is free TF-Halo. To quantify these diffusion differences, we applied our Hidden Markov Modeling (HMM) approach in which all trajectories are fitted to a pre-defined number of discrete diffusion states based on the trajectory step lengths, with state diffusion coefficients and transition frequencies between these states as fitting parameters [29][30][31] . Each diffusion state, hence, is described by an average diffusion coefficient (D), the steady-state fraction of particles in said state (occupancy), and the average time spent in that state (dwell time).…”

“…To specifically investigate RNC binding events, we coarse-grained the larger HMM-fitted models into 2 states using a threshold at 0.5 µm 2 s -1 [29][30][31]33 , with state 1 corresponding to RNC-bound TF-Halo and state 2 to free TF-Halo. The occupancy and average dwell time, calculated as weighted averages from course-grained models 4-9, were 85% and 180 ms for the bound state, and 15% and 28 ms for the free state, respectively.…”

Real-time dynamics of individual chemoreceptor mRNA molecules reveals translation hotspots at the inner membrane ofEscherichia coli

“…Therefore, the lifetime of 3D and 2D diffusion of tsr mRNA until binding to the membrane and Sec will be very short. Indeed, recent work investigating the kinetics of SRP shows an average search time of approximately 750 ms until binding to Sec occurs (57). This short search time will consequently restrict exploration of space of tsr mRNA until initiation of translation, leading to the observed overlap in gene location and translation.…”