Figure 1.
Schematic representation of the electron transport and reductant partitioning pathways in Cyanothece 51142.
Linear photosynthetic electron transfer: electrons from photosystem II (PS II) to photosystem I (PS I) are transferred through plastoquinone (Pq), cytochrome b6f complex (Cyt b6f), plastocyanin (Pc) and cytochrome c6 (Cyt c6). From PS I electrons can be transferred to ferredoxin (Fd) via ferredoxin:NADP+ reductase (FNR) and subsequently to generate reductant in the form of NADPH. Cyclic photosynthetic electron transport: electrons can flow from Fd to Pq (FdPq reaction). Respiratory electron transfer: includes two cytochrome oxidases (COX), two cytochrome-quinol oxidases (QOX), and two types of NADH dehydrogenases (NDH-1 and NDH-2). Alternative sinks for reductant beyond CO2 fixation: reduced Fd can be used by the nitrogenase (Nif) and by Mehler reactions to reduce O2. Bidirectional hydrogenase (Hox) can reversibly produce H2 using NAD(P)H as an electron donor, while the uptake hydrogenase (Hup) consumes H2 using Fd as an electron acceptor. Protons transferred across the thylakoid membrane are used by the ATPase to drive ATP synthesis.
Figure 2.
Impact of electron transport pathways on growth and metabolism of Cyanothece 51142.
(A) Effects of removing cyclic photosynthesis (via NDH-1, NDH-2, FdPq, G3PD_PQ, and SUCD_PQ) and alternative reductant sinks (H2 production, COX, QOX, and Mehler reactions). (B) Effect of removing alternative reductant sinks but including all routes for cyclic photosynthesis. Shaded regions indicate that multiple flux values can achieve maximal growth rate. (C) All photosynthetic and respiratory electron flow routes operate, except H2 production.
Figure 3.
Predicted effects of varying photon uptake rates on growth and electron transport pathways.
(A) 2-D phenotypic phase plane (PhPP) displaying maximum growth rates for varying photon uptake rates. The PhPP has 3 distinct regions – in regions 1 and 3, flux through a single photosystem limit growth rates, whereas in region 2 flux increases through either photosystem will increase growth rate. (B) Pathway maps of electron transfer reactions in different PhPP regions. PhPP flux variability analysis was performed to see which flux is always required (red arrows), optional (green arrows), and blocked (blue arrows) across each of the three PhPP regions.
Table 1.
Comparison of growth rates predicted by simulation model to those experimentally measured in batch cultures.
Figure 4.
Effect of nutrient limitation on biomass composition (normalized to ash-free dry weight).
Figure 5.
Predicted chemostat flux distributions in central metabolism including transcriptome and proteome data as constraints.
The flux values (mmol·g−1 AFDW·h−1) are those where the flux distribution best matches the transcriptome and proteome data (TPD) while also minimizing the magnitude of all fluxes in the network. The flux values in red and green represent ammonia-limited (AL) and light-limited (LL) conditions, respectively. Arrow colors indicate relative flux ratios between AL and LL conditions.
Table 2.
Flux variability analysis for model simulations in light-limited and ammonium-limited chemostat conditions.
Figure 6.
Effects of in silico reaction deletions on flux spans under light-limited conditions.
(A) Effects of deletions are compared to the cases where no reactions were deleted (red bar), or TPD were used as constraints (green bar). The values represent the average flux span across all reactions in central metabolism. Only deletions which lower the flux span by at least >1 mmol·g−1 AFDW·h−1 are presented. (B) Changes in flux spans for specific reactions catalyzed by ribulose bisphosphate carboxylase (RBC) and phosphoglucose isomerase (PGI) between simulations that (i) use TPD data as a constraint (green bars), (ii) delete single reactions (blue and purple bars), (iii) delete two reactions (yellow bar) or (iv) impose no additional constraints (red bars). Reaction abbreviations match those listed in Table S1.