Supplemental Figure 2. Comparison of the Signal Suppression Effects Observed with The

Supplemental Figure 1. Capacity factors (k’) for several groups of compounds analysed with the PFPP-LC/MS and C18-UHPLC/MS systems.

The values are the means of at least 3 independent analyses of each tested compound.

Supplemental Figure 2. Comparison of the signal suppression effects observed with the PFPP-LC/MS

and C18-UHPLC/MS systems.

A) Capacity factor (k’) for 21 selected compounds.

B) Estimation of the signal suppression induced by the matrix effect for each compound.

Supplemental Figure 3. Influence of the PFPP-LC/MS and C18-UHPLC/MS systems on the detection of XCMS features.

Data were obtained in positive (A) and negative (B) ion modes with the PFPP-LC/MS (black bars) and C18-UHPLC/MS (white bars) systems. The variables were first selected for signal over noise ratio higher than 4. Then CV values (coefficient variation in %) were calculated for each sample (one extract from each growth condition). The numbers features with CV values below 100% that were retained for further consideration are indicated above the corresponding bar.

Supplemental Figure 4. Estimation of the intra-assay precision on the XCMS features detected with the PFPP-LC/MS system.

An extract from Synechocystis cells grown in standard light conditions in presence of 5 mM glucose was analysed over 36 hours in both positive and negative ion mode (18 successive chromatographic injection for each mode of ionisation. A) Time course evolution of LC/MS area for 4 selected metabolites. B) Global intra-assay precision measured on XCMS data sets CVs were calculated for each variable.

Supplemental Figure5. The impact of 3 freeze-thaw cycles of cell extracts on metabolite levels.

Metabolic fingerprints of SLlG-grown Synechocystis cell extracts were recorded with the PFPP-LC-MS system operated in negative ion mode. Data were mean centered and scaled to Pareto variance before multivariate statistical analysis. The PCA scores plot for individual extracts (represented by circles, squares and triangles) does not separate the data corresponding to each unfrozen samples (white symbols) from its frozen counterparts (black symbols).

Supplemental Figure 6. Schematic representation of the central carbon metabolism of Synechocystis

The carbon metabolites and pathways are abbreviated as follows: Glc: glucose; G1P: glucose-1P; G6P: glucose-6P; F6P: fructose-6P; F1,6BP: fructose-1,6-bisP; GA3P: glyceraldehyde-3P; 1,3-BPGA: 1,3-bisphosphoglycerate; 3PGA: 3-phosphoglycerate; 2PGA: 2-phosphoglycerate; PEP: phosphoenolpyruvate; OOA: oxaloacetate; E4P: erythrose-4P; X5P: xylulose-5P; S7P: sedoheptulose-7P; R5P: ribose-5P; Ru5P: ribulose-5P; RuBP: ribulose-1,5-BisP; OPP: oxydative pentose phosphate pathway; NOPP: non-oxydative pentose phosphate pathway; TCA: tricarboxylic acid cycle.

Supplemental Figure 7. Global visualization of the influence of growth conditions on Synechocystis metabolic fingerprint with multivariate statistical analyses.

Data were acquired with the PFPP-LC/MS system operating in negative ion mode, from three independant extracts of cells grown under each studied conditions: SL (), SLlG () and LLhG (). The PCA Pareto score plot analysis defined three groups (noted I, II and III) each corresponding to one growth condition tested.