15 h-1. Some cells expressed the ptsG reporter in conditions when no glucose was taken up via Glc-PTS. Also, low concentration of glucose in the medium feed (first column) led to the existence of a small subpopulation that does not engage in the glucose uptake via Glc-PTS. Transcriptional reporters for glucose transporters can only provide limited insights into the actual metabolic state of cells. Several selleck chemical recent papers have discussed discrepancies between transcriptional reporters and metabolic fluxes in specific parts of metabolic pathways [35, 36]. As a consequence, we need to be cautious when using data from transcriptional reporters to make inferences about the actual physiology of cells.

Additional experiments could provide complementary insights, for instance the analysis of sugar transporter synthesis or activity, together with analysis of sugar assimilation at the single-cell level [37]. Variation in the expression of glucose transporters across environments We next investigated how the variation in expression of reporters

for different glucose transporters changes across different environments. We first compared the results of this study with the results from a genome-wide study of promoter-mediated phenotypic variation [31]. Mean and variation of the expression of ptsG, mglB and rpsM reporters are shown in Figure  3 (plotted are mean values of replicates in different conditions). When power regression lines were fitted across different expression data from the same environment, all lines showed Cyclin-dependent kinase 3 the same trend, namely that the CV of log fluorescence values decreased INCB024360 nmr with mean log GFP expression (Figure  3). Our analysis suggests some general rules: variation in the expression from these three promoters was lowest in batch cultures supplemented with glucose, or glucose plus

acetate, and highest in batch or chemostats cultures with acetate as a sole carbon source. Figure 3 Phenotypic variation in gene expression in 13 different environments. The coefficient of variation (CV) of log expression of PptsG-gfp, PmglB-gfp and PrpsM-gfp was plotted against the mean log expression. Expression of the reporters in different environments was compared to data for 1522 E.coli promoters [31] (light blue diamonds) that were measured in the early exponential phase in batch cultures containing arabinose as a sole carbon source. Circles represent measurements in chemostat environments and triangles represent measurements in batch cultures. Different color of triangles and circles represents different reporters: ptsG (green), mglB (blue) and rpsM (red). Power regression (i.e. linear regression on log-transformed data) was fitted to each set of three promoters measured in the same environment. Colors of fitted lines mark different carbon sources in the feed; full lines mark chemostat environments and dashed lines mark batch cultures. Each data point is the average over 2–5 independent replicates (except for data from [31]).