Master 2 in Complex Systems

INTERNSHIP PROPOSALLaboratory name: Physico-Cimie CurieCNRS identification code: UMR 168Internship director’s surname: Castellanae-mail: michele.castellana@curie.frPhone number: +330156246789Internship location: 11 Rue Pierre et Marie Curie, 75237 ParisThesis possibility after internship: NOFunding: YES (Institut Curie)

Optimality principles in protein synthesis

Protein synthesis is a mechanism of fundamental importance for life. The molecular components involved in this process aremacromolecules called messenger RNAs (mRNAs) and ribosomes. Recent theoretical and experimental studies in the modelEscherichia coli bacterium show that these molecular components exhibit striking, marked spatial localization patterns, wheremRNAs and ribosomes are colocalized in the same intracellular region of space.

distribution common among the longest cells (Fig. 1B),but with enhanced spatial resolution. The endcap regionand the region at mid-cell between the two pairs of DNAlobes are essentially devoid of RNAP. The number densityof RNAP in the ribosome-rich region between nucleoidlobes is < 10% of that in the densest part of the nucleoidlobes.

Importantly, the y co-ordinate profiles of the RNAP mol-ecules, although noisy, clearly extend all the way to thecytoplasmic membrane (example in Fig. S4D). The pres-ence of RNAP very near the cytoplasmic membrane sug-gests the presence of some DNA there as well, but only inthe straight, cylindrical part of the cell. There is essentiallyno RNAP near the endcap portion of the cytoplasmicmembrane (Fig. 4D). This has implications for the previ-ously proposed transertion mechanism.

Copy numbers and number densities of ribosomesand RNAP

Because the S2-YFP gene replaces the normal S2 gene onthe chromosome, all copies of S2 should carry the YFP

label. We further assume that all S2-YFP copies are fluo-rescent and that essentially all S2 proteins are incorpo-rated into 30S subunits. In Supporting information, weshow that original and ‘revived’ YFP copies have the samefluorescence intensity under single-molecule imaging con-ditions (Fig. S16). We can then scale the total, pre-bleachYFP intensity to that of a single YFP molecule to estimatethe copy numbers of ribosomes and RNAP in each cell(Taniguchi et al., 2010). For ribosomes (meaning the sumof 30S subunits and complete 70S ribosomes), the result-ing estimated copy number ranges from ~ 30 000 to~ 70 000 ribosomes per cell (42 cells, Fig. S6). The meannumber of ribosomes per cell is ~ 55 000. The ribosomecopy number increases monotonically with cell volume.The estimated copy number of b′-yGFP ranges from~ 2000 per cell to ~ 10 000 per cell. The average number ofb′-yGFP per cell is 4600 (31 cells, Fig. S6). The number ofb′-yGFP is correlated with cell volume.

To accurately determine cell geometry, we have alsocarried out superresolution imaging of free YFP expressedfrom a plasmid (Fig. S3; Bakshi et al., 2011). The cell radiusR = 380 ! 20 nm is remarkably well conserved from cell to

−2 −1 0 1 20

0.20.40.60.8

1

Endc

ap Cylinder

x (µm)

P(x)

B

C

Endc

ap

A

D

1 µm

Fig. 3. Superresolution images of ribosomes (S2-YFP) within K-12 cells grown in EZRDM at 30°C. Each localization is plotted as a point atthe calculated centroid position.A. Nine representative cells.B. Image of two single molecules within a cell prior to image filtering. Cell outline based on phase contrast image.C. Expanded view of superresolution image of ribosomes in the same cell. A model spherocylinder is shown as a guide to the endcappositions.D. Relative number of ribosomes at each axial position, with data along y at each x summed into 100 nm bins. The grey background showsthe theoretical profile for a uniform distribution filling the model spherocylinder, taking account of measurement uncertainty and binning.Sectioning by the 1.49 NA objective does not affect the axial distribution significantly, as shown in Supporting information (Fig. S3).

26 S. Bakshi, A. Siryaporn, M. Goulian and J. C. Weisshaar !

© 2012 Blackwell Publishing Ltd, Molecular Microbiology, 85, 21–38

Ribosome localization (from S. Bakshi et al.. Mol. Microbiol., 85(1):21, 2012). Top: superresolution image of ribosomes (yellowdots) in a single Escherichia coli cell. Middle: spherocylindrical model of the cell. Bottom: relative number of ribosomes along theposition on the long cell axis, where the gray background shows the theoretical profile for a uniform distribution.

The basic, general expectation that protein synthesis in vivo should involve an efficient use and allocation of cellular of resourcesprovides a natural framework to understand these localization patterns in terms of an optimality principle. Namely, reaction-diffusion models allow for quantifying the protein synthesis rate associated with any given spatial distribution of mRNAs andribosomes, and thus for characterizing the optimal mRNA-ribosome distribution which provides the highest protein-synthesisrate. As a result, the strongly localized distributions of ribosomes and mRNAs observed in the experiments could be directlyunderstood in terms of a general principle of optimality for protein synthesis.Overall, this internship will involve both analytical and numerical, cross-disciplinary approaches, ranging from the mathematicalcharacterization of reaction-diffusion equations and constrained-optimization problems, to their numerical solution, as well asto the physical interpretation of the results. No previous knowledge of the biology of bacterial protein synthesis is required.Importantly, the project provides also natural grounds for a strong collaboration with molecular biologists and biotechnologists,in the effort to directly test the relevance of these optimality principles in vivo, and explore their potential applications to controland regulate protein synthesis and cell growth.