The role of glycogen Metabolism in the Adaptation to environmental Changes of the cyanobacterium Synechocystis sp. PCC 6803

Doctoral Thesis Pablo Ortega Martínez Sevilla 2025 The role of glycogen metabolism in the adaptation to environmental changes in the cyanobacterium Synechocystis sp. PCC 6803 Glucose-1P ADP-Glucose Glycogen Glucose-6P CO 2 The role of glycogen metabolism in the adaptation to environmental changes in the cyanobacterium Synechocystis sp. PCC 6803 Doctoral Thesis Pablo Ortega Martínez Sevilla 2025 Supervisors Dr. Francisco Javier Florencio Bellido Dra. Sandra Díaz Troya INTRODUCTION ................................................................................................ 1 1 Photosynthetic organisms ................................................................................ 1 1.1 Cyanobacteria ............................................................................................................... 1 1.1.1 Synechocystis sp. PCC 6803 as a model organism ......................................................... 5 1.1.1.1 Substrains in Synechocystis ................................................................................................. 6 2 Oxygenic photosynthesis and respiration ....................................................... 6 2.1 Photosynthetic Electron Transport Chain ...................................................................... 7 2.1.1 Phycobilisomes ............................................................................................................... 11 2.2 Respiration and auxiliary electron transport ................................................................ 13 3 Carbon metabolism ......................................................................................... 15 3.1 Calvin-Benson-Bassham cycle .................................................................................... 17 3.1.1 Carbon Concentration Mechanism (CCM) ...................................................................... 18 3.2 Glycogen metabolism .................................................................................................. 20 3.2.1 Glycogen connection to central carbon metabolism ....................................................... 22 3.2.2 Consequences of impaired glycogen metabolism .......................................................... 23 3.2.3 Alternative glycogen/α-glucan pathways in prokaryotes ................................................. 24 3.3 Embden-Meyerhof-Parnas (EMP/glycolysis) .............................................................. 26 3.4 Oxidative pentose phosphate (OPP) pathway ............................................................ 27 3.5 Entner-Doudoroff (ED) and phosphoketolase pathways ............................................. 27 3.6 Tricarboxylic acid (TCA) pathway ................................................................................ 28 3.7 Salt acclimatation: Synthesis of compatible solutes .................................................... 29 4 Nitrogen metabolism ....................................................................................... 30 4.1 Nitrogen assimilation and amino acid metabolism ...................................................... 30 4.2 Chlorosis program in non diazotrophic cyanobacteria ................................................ 31 4.2.1 Phycobilisomes degradation ........................................................................................... 32 4.2.2 Recovery from chlorosis ................................................................................................. 33 5 Photomixotrophy ............................................................................................. 33 MATERIALS AND METHODS ......................................................................... 37 1 Organisms and culture conditions ................................................................. 37 1.1 Cyanobacteria ............................................................................................................. 37 1.1.1 Cyanobacterial strains used in this work ........................................................................ 37 1.1.2 Media and culture conditions .......................................................................................... 40 1.2 Escherichia coli ............................................................................................................ 41 1.2.1 Strains ............................................................................................................................ 41 1.2.2 Media and culture conditions .......................................................................................... 42 2 DNA analysis and manipulation ..................................................................... 43 2.1 Plasmids ...................................................................................................................... 43 2.1.1 Commercial and laboratory collection plasmids. ............................................................. 44 2.1.2 Plasmids generated in this work. .................................................................................... 45 2.2 Primers ........................................................................................................................ 47 2.3 DNA isolation ............................................................................................................... 51 2.3.1 Plasmid DNA isolation from E.coli .................................................................................. 51 2.3.2 Cyanobacterial genome DNA isolation ........................................................................... 51 2.4 DNA visualization/analysis, manipulation .................................................................... 52 2.4.1 DNA electrophoresis in agarose gels ............................................................................. 52 2.4.2 DNA quantification and sequencing ................................................................................ 52 2.4.3 Restriction assays and enzymatic manipulation of DNA ................................................. 52 2.5 DNA amplification by polymerase chain reaction (PCR) ............................................. 53 2.5.1 Overlapping PCR ............................................................................................................ 53 2.6 Organism transformation ............................................................................................. 55 2.6.1 Synechocystis transformation ......................................................................................... 55 2.6.2 E.coli transformation ....................................................................................................... 55 2.6.2.1 Heat shock.......................................................................................................................... 55 2.6.2.2 Cell preparation for heat shock ........................................................................................... 56 2.6.2.3 Electroporation.................................................................................................................... 56 2.6.2.4 Cell preparation for electroporation ..................................................................................... 57 3 Biochemical methods...................................................................................... 57 3.1 Protein expression in E. coli ........................................................................................ 57 3.2 Preparation of cell extracts .......................................................................................... 58 3.2.1 Glass bead lysis ............................................................................................................. 58 3.2.2 Sonication lysis ............................................................................................................... 58 3.2.3 Whole cell preparation .................................................................................................... 58 3.3 Protein quantification ................................................................................................... 59 3.3.1 Bradford .......................................................................................................................... 59 3.3.2 Lowry .............................................................................................................................. 59 3.4 Protein electrophoresis in denaturing polyacrylamide gels (1D SDS-PAGE) ............. 59 3.4.1 Coomassie blue stainning............................................................................................... 59 3.4.2 Protein immunodetection (Western blotting) ................................................................... 60 3.5 Recombinant protein purification ................................................................................. 61 3.5.1 Affinity chromatography .................................................................................................. 61 3.6 Polyclonal antibodies production ................................................................................. 62 4 Biophysical methods....................................................................................... 62 4.1 Oxygen evolution ......................................................................................................... 62 4.1.1 Clark-type oxygen electrode ........................................................................................... 62 4.1.2 MIMS .............................................................................................................................. 62 4.2 Fluorescence measurements ...................................................................................... 63 4.2.1 Chlorophyll fluorescence analysis and Y(II) calculations ................................................ 63 4.2.2 NAD(P)H fluorescence measurements ........................................................................... 63 4.2.3 Determination of P700 redox kinetics. ............................................................................ 63 4.2.4 Reoxidation kinetics of QA .............................................................................................. 64 4.2.5 PC, P700 and Fd redox changes .................................................................................... 65 4.3 Reactive oxygen species measurements .................................................................... 65 4.4 Microscopy .................................................................................................................. 66 4.4.1 Fluorescence microscopy ............................................................................................... 66 4.4.2 Electron microscopy ....................................................................................................... 66 5 Enzymatic assays ............................................................................................ 66 5.1 Glycogen quantification ............................................................................................... 66 5.2 Glucose quantification in the media ............................................................................ 67 5.3 Determination of extracellular pyruvate and 2-oxoglutarate........................................ 67 5.4 In situ enzymatic activities ........................................................................................... 68 5.4.1 Glucose-6P crossroads .................................................................................................. 68 5.4.2 Glutamine synthetase activity ......................................................................................... 69 5.5 In vitro enzymatic activities .......................................................................................... 69 6 Analytical methods .......................................................................................... 70 6.1 Metabolite extraction and targeted metabolomics. ...................................................... 70 6.2 Amino acid and glutathione pools extraction ............................................................... 71 6.3 Quantification of intracellular osmolytes (GG and sucrose) ........................................ 71 6.4 Quantification of glucose and mannose in the media. ................................................ 71 6.5 Polysaccharide size characterization. ......................................................................... 72 7 Other methods ................................................................................................. 72 7.1 pH measurements ....................................................................................................... 72 7.2 Chlorophyll concentration determination ..................................................................... 73 7.3 Cell density spectrophotometric measurements ......................................................... 73 7.3.1 Whole cell spectra .......................................................................................................... 73 7.3.2 Graphic representation and statistical analysis. .............................................................. 73 OBJECTIVES ................................................................................................... 75 RESULTS......................................................................................................... 77 1 Analysis of the connection between glycogen and central carbon metabolism by phosphoglucomutases in cyanobacteria. ............................ 77 1.1 Homology analysis and abundance of the phosphoglucomutases in Synechocystis. 78 1.2 Generation of a ∆PGM mutant strain .......................................................................... 80 1.3 Effect of high light exposure ........................................................................................ 81 1.4 Impact of light-dark cycles ........................................................................................... 84 1.5 Response to nitrogen deprivation ................................................................................ 86 1.5.1 Metabolomic response to nitrogen depletion .................................................................. 88 1.5.2 Recovery from nitrogen deprivation ................................................................................ 93 1.6 Resistance to salt stress.............................................................................................. 95 1.7 The overexpression of PMM/PGM compensates for the lack of PGM. ....................... 97 1.8 Characterization of the essentiality of PMM/PGM and its contribution to glycogen synthesis in Synechocystis. ....................................................................................... 103 1.8.1 In search of the lethality: Metabolic response to PMM/PGM depletion in the ∆PMM_Pars:M strain. ................................................................................................... 107 2 Role of glycogen synthesis in the transition to mixotrophy. ..................... 110 2.1 Impact of photomixotrophy on ΔPGM and ΔAGP strains .......................................... 112 2.1.1 Physiological effects of glucose supplementation in WT, ΔPGM and ΔAGP strains .... 112 2.1.2 Photosynthesis is severely impaired in the glycogen mutant strains cultured with glucose. ........................................................................................................................ 125 2.1.3 Light-dependent NAD(P)H synthesis is altered in photomixotrophy. ............................ 134 2.1.4 Recovery of photoautotrophic growth following glucose exposition. ............................. 135 2.2 Impact of photomixotrophy on glycogen synthases and branching enzymes mutant strains in Synechocystis. ........................................................................................... 136 3 Integration of the metabolic and photosynthetic regulation of the bleaching process. ......................................................................................................... 142 3.1 Effect of photosynthesis inhibition and exogenous hexose supplementation during nitrogen deprivation. .................................................................................................. 144 3.1.1 Metabolic profile during nitrogen deprivation under photosynthesis inhibition and exogenous hexose supplementation. ........................................................................... 146 3.1.2 Exploring the role of specific metabolites in chlorosis: functional studies of 6PG and methionine. ................................................................................................................... 150 3.2 Early physiological responses of Synechocystis to glucose and mannose supplementation under nitrogen deprivation. ............................................................ 152 3.2.1 Growth, glycogen and sugar consumption.................................................................... 153 3.2.2 Early metabolic effect of glucose supplementation to the bleaching process. .............. 155 3.2.3 Early photosynthetic effect of glucose supplementation on the bleaching process. ..... 157 3.2.4 Effect of glucose addition at different times during the bleaching process. .................. 165 4 Alternative glycogen synthesis in a glycogen synthase mutant strain ..... 167 4.1 Identification of a glycogen-like polysaccharide in the mutant lacking glycogen synthases. ................................................................................................................. 167 4.2 Analysis of the physiological role of the glycogen-like polysaccharide in the ∆glgA mutant strain. ............................................................................................................. 170 4.3 Characterization of a ∆glgA strain that grows in salt free BG11C medium ............... 171 4.4 Role of the GgpS in an alternative glycogen synthesis pathway. ............................. 178 4.5 The essentiality of glucosylglycerol in alternative glycogen synthesis independent of glycogen synthases. .................................................................................................. 183 DISCUSSION ................................................................................................. 187 1 PGM is the main phosphoglucomutase in Synechocystis. ........................ 187 2 Glycogen synthesis is required for adequate photomixotrophic growth .. 193 3 Metabolic and photosynthetic regulation of the bleaching process .......... 198 4 Glycogen synthesis can be achieved using glucosylglycerol ................... 203 CONCLUSIONS ............................................................................................. 209 BIBLIOGRAPHY ............................................................................................ 211 1 INTRODUCTION 1 Photosynthetic organisms 1.1 Cyanobacteria Oxygenic photosynthesis is a metabolic process able to synthesize biomolecules and fuel biological processes using H2O, CO2 and harvested light energy releasing O2 in the process. The biosphere and trophic chains are sustained by the primary producers through this photoautotrophic metabolism. Oxygenic photosynthesis first evolved in cyanobacteria, an ancient lineage of gramnegative eubacteria that played a critical role in creating the environmental and evolutionary conditions that have shaped and influenced life on Earth (Schirrmeister et al., 2016). The release of O2 by photosynthesis led to the oxygenation of the environment around 2.4 billion years ago altering the composition of the atmosphere that transitioned from a primitive anaerobic and reducing composition to the aerobic and oxidizing environment in which we live today (Govindjee & Shevela, 2011; Lyons et al., 2014; Soo et al., 2017). This shift in the environmental conditions, referred as the Great Oxidation Event (Figure 1), derived in high O2 concentrations, forcing the species to adapt by developing Figure 1: Effect of oxygen concentrations in the atmosphere and the evolve of photosynthetic organisms. The figure is adapted from Agrisera Educational Poster 5 (D. Shevela, J. Kern, J. Whitmarsh, J. Messinger and G. Govindjee (2021). Photosystem II: Enzyme That Gives Us Molecular Oxygen). doi: 10.6084/m9.figshare.14802924.v1. 8 PsbZ, and PsbYcf12) and extrinsic (at least PsbO, U, and V) Psb proteins contributing to stability and integrity, optimizing PSII photochemistry (Müh & Zouni, 2020; Johnson & Pakrasi, 2022). Electrons from OEC are donated to the RC when the P680 gets excited and further transported through the redox active cofactors localized within them starting with YZ, consecutively followed by four similar energy levels chlorophylls, referred as P680 (ChlD1/PD1/PD2/ChlD2), then to a pheophytin. Finally, electrons are transferred sequentially to primary (QA) and secondary (QB) quinones (Figure 6). Although both QA and QB are plastoquinones, QA is permanently bound to PSII and accepts one electron, while QB is a lipophilic acceptor from the PQ pool loosely bound to PSII that requires two electrons and two protons to get fully reduced. Therefore, two consecutive steps of reduction are performed by the QA, being the partially reduced intermediate QBtightly bound until the second reduction occurs, incorporating two protons from the cytoplasmatic side. Once QB is fully reduced to PQH2 (plastohydroquinone), its binding affinity diminishes and leaves the QB binding site, being replaced by another oxidized plastoquinone from the PQ pool (Figure 6). Between the QA and QB, there is a non-heme iron that binds a bicarbonate ion (HCO3−), important in the electron transfer (and its electron potential) and protonation of QB (Brinkert et al., 2016; Shevela et al., 2020, 2023) (Figure 6). The PQ diffuses along the thylakoid membranes, connecting the PSII with the cytochrome b6f complex (Cyt b6f). Reduced PQ (PQH2) is oxidized in a process Figure 6: Representation of PSII principal domains with phycobilisome system and electron transfer from QA to QB. The figure is adapted from Agrisera Educational Poster 5 (D. Shevela, J. Kern, J. Whitmarsh, J. Messinger and G. Govindjee (2021). Photosystem II: Enzyme That Gives Us Molecular Oxygen). doi: 10.6084/m9.figshare.14802924.v1. 9 called the Q-cycle upon reaching the Cyt b6f, releasing concomitantly the protons in the thylakoid lumen (Figure 7). The Q-cycle diverted the two electrons from the PQH2 in a high-potential pathway (through the cytochrome f to the soluble acceptor carriers plastocyanin; PC or Cytochrome c6; Cyt c6) and a low potential pathway (through heme groups). Since PC/Cyt c6 only accept one electron, the electron in the low potential pathway is recycled back to an oxidized PQ, staying in a semiquinone state until a second PQH2 donates its electron to the Cyt b6f, restoring a full PQH2 to the plastoquinone pool (Figure 7) (Proctor et al., 2022). PC/Cyt c6 are soluble proteins, single electron carriers ubicated in the thylakoid lumen that connect Cyt b6f to the PSI (Figure 5). The PC is a cuproprotein (encoded by petE) and the Cyt c6 is a ferroprotein (encoded by petJ) and they are alternatively expressed by a tightly regulated mechanism ultimately controlled by the availability of copper. When copper is not present, a transcription factor (PetR) induces the expression of petJ (Cyt c6) and represses petE (plastocyanin). When copper is available, there is a switch in the expression of petJ and petE mediated by a protease that degrades the transcription factor PetR (GarcíaCañas et al., 2021). Both PC and Cyt c6 present similar efficiency as electron donors to PSI. The electrons in PSI become excited similarly to PSII operation principles, acting in Figure 7: Schematic and sequential representation of the Q-cycle for plastoquinone oxidation at Cyt b6f level. See text for details. 10 this case as a light driven iron–sulfur PC:Ferredoxin oxidoreductase. The PSI core protein (P700) harbors a chlorophyll dimer with the peak for energy light absorption at 700 nm. The fate of the electrons is the reduction of the soluble electron acceptor Ferredoxin (Fd), an electron hub in the PETC. Therefore, Fd can be oxidized to maintain the FTR-Trx redox system (Fd thioredoxin reductasethioredoxin) during nitrogen assimilation, in alternative electron sinks, or mainly to donate electrons to the Federroxin:NADPH reductase (FNR) due to their high affinity (Shimakawa & Dietz, 2023). FRN is a soluble protein intimately associated with the thylakoid membrane in its cytoplasmatic side, as well as with the phycobilisomes, which generate the reduced form of NADPH (Figure 5) (Mullineaux, 2014). During this process, known as Linear Electron Transfer (LET), electrons are donated from H2O to NADPH and protons are accumulated in the thylakoid lumen through water oxidation and the proton pumping associated with the PQ oxidoreductive cycle. This generates an electrochemical gradient, or proton motive force (pmf), due to differences in electric field (Δψ) and proton concentration (ΔpH) across the thylakoid membrane. The pmf is harnessed by ATP synthase, releasing the protons from the lumen to the cytoplasm to convert ADP in ATP (Figure 5) (Nikkanen et al., 2021; Shimakawa & Dietz, 2023). As an alternative to the LET, the electrons from Fd can be diverted back to the PQ pool through the NAD(P)H dehydrogenase-like (NDH-1) complex and the rest of shared components of the PETC in a cyclic manner (Cyclic Electron Transfer; CET). This process facilitates proton pumping, thereby enabling additional ATP production without producing NADPH (Shimakawa & Dietz, 2023). Other alternative electron transfer processes such as the mediated by the aa3-type Cyt c oxidase (COX) respiratory terminal oxidase (RTO) (see section 2.2 Respiration and auxiliary electron transport) also contribute to the proton accumulation and thus to ATP production. Cells can adjust the rates of LET and CET based on their energetic requirements for the ATP/NADPH ratio under variable environmental conditions (Theune et al., 2021; Miller et al., 2022). 11 2.1.1 Phycobilisomes Cyanobacterial photosystem light-harvesting complexes (LHCs) consist of Chlorophyll a and phycobilisomes (PBSs). PBSs are massive hydrophilic protein complexes formed mostly by phycobiliproteins (Figure 8). The basal unit of the PBS consists of a heterodimer of α and β homologous apoproteins of 15-20 kDa covalently bound to key open-chain tetrapyrroles pigment molecules (Adir et al., 2020). The pigments are derived from oxidation of a heme group and are referred to as phycobilins, like phycocyanobilins and in some cyanobacteria phycoerythrobilins (Watanabe & Ikeuchi, 2013; Adir et al., 2020). The combination of the apoproteins with the phycobilins results in the different available phycobiliproteins: phycocyanin, allophycocyanin, phycoerythrin, and phycoerythrocyanin (Mullineaux, 2014; Singh et al., 2015). The type and ratio of phycobiliproteins in the phycobilisomes vary between organisms and environmental conditions (Mullineaux, 2014; Singh et al., 2015). The phycobiliproteins undergo oligomerization into trimeric (αβ)3 or hexameric (αβ)6 discs to form the phycobilisomes (Adir et al., 2020). Thanks to the chromophores, PBSs are able to absorb light energy between 490-650 nm, complementing the spectral range provided by chlorophylls. The absorbed energy is finally transferred to the chlorophylls in the RC of PSII and, to a lesser extent, PSI (Figure 8) (Watanabe & Ikeuchi, 2013; Strašková et al., 2019; Adir et al., 2020). In vegetative growing cells, phycobiliproteins are the most abundant proteins and Figure 8 Representation of cyanobacterial photosystem light-harvesting complexes. The figure is adapted from Agrisera Educational Poster 5 (D. Shevela, J. Kern, J. Whitmarsh, J. Messinger and G. Govindjee (2021). Photosystem II: Enzyme That Gives Us Molecular Oxygen). doi: 10.6084/m9.figshare.14802924.v1. 12 can account for up to 50% of the soluble protein content (Zavřel et al., 2019; Domínguez-Martín et al., 2022). The PBSs in Synechocystis are organized in a core of three allophycocyanin cylinders of four discs in a pyramidal shape. This allophycocyanin core is radiated by six rods composed of stacked phycocyanin discs (Singh et al., 2015; Calzadilla & Kirilovsky, 2020). The rods distribution is unequal, with one rod per cylinder in the base and four rods in the top cylinder, granting the PBS with a hemidiscoidal structure. Nevertheless, the phycobilisome is a dynamic structure, since some rods can be positioned in alternative conformations towards more compact structures (Domínguez-Martín et al., 2022). The assembly and scaffold for the core and rods arrangements is provided by attached hydrophobic linker proteins, like the Apc and CpcC proteins (Baier et al., 2014; Singh et al., 2015; Domínguez-Martín et al., 2022). In addition, some allophycocyanin subunits from the basal cylinders (at the membrane surface) are structurally replaced by some modified subunits such as ApcD, ApcE and ApcG that act as terminal emitters of the energy absorbed in the PBS to the RC of photosystems I and II, respectively (Calzadilla & Kirilovsky, 2020; DomínguezMartín et al., 2022; Espinoza-Corral et al., 2023). Therefore, the phycobilisomes structure efficiently increases the photon absorption area and delivers the excitation energy absorbed by any phycobilins to the terminal emitters within the PBSs (Calzadilla & Kirilovsky, 2020; Domínguez-Martín et al., 2022). Besides this role trapping incident sunlight and transferring it in a cascade towards the RC, they also provide energy efficiency optimization through the allocation of the excitation in the vicinity of the closed active RC until they are open again (Müh & Zouni, 2020). 13 2.2 Respiration and auxiliary electron transport In contrast to other photosynthetic organisms which have spatial separation of photosynthesis and respiration, cyanobacteria share the same thylakoid membrane for both processes (Mullineaux, 2014; Nikkanen et al., 2021). Respiration consists in the transport of electrons from metabolic intermediates through the electron transport chain, with predominantly oxygen as the ultimate electron acceptor and maintaining ATP production (Figure 9). The source of electrons in the respiratory pathways can have two origins. One source is the oxidation of intermediates by respiratory enzymes, such as succinate (by succinate dehydrogenase; SDH (Mullineaux, 2014)) or pyruvate (via pyruvateferredoxin oxidoreductase; PFOR (Wang et al., 2022)), which donate electrons to the PQ pool and Fd, respectively (Figure 9). The other source is the oxidation by the Ferredoxin:NADP-Oxidoreductase (FNR; Miller et al., 2022) of NADPH produced mainly in the OPP shunt by glucose-6P oxidation (Shimakawa & Dietz, 2023). Thus, within the thylakoid membrane, electrons are fueled into the PETC at the level of the PQ pool directly or indirectly (through the NDH-1 complex, also involved in CET, see section 2.1-Photosynthetic Electron Transport Chain) that accepts them from the Fd (Figure 9). From the PQ pool, electrons are transferred to the RTOs cytochrome bd quinol oxidase (Cyd) and COX, which O2 2H2O O2 2H2O ADP PSII H+ H+ QB QA O2 + 4H+ 2H2OOEC Cyt b6f PSI P680 P700 Fd FNR hv H+ SDH Suc Fum PBS H+ H+ NDH-1 Pyr AcCoA hv PFOR c6 PC NADP+ NADPH COX Cyd H+ H+ Flvs O2 + 4H+ 2H2O Lumen Cytosol Figure 9: Schematic representation of the cyanobacterial Photosynthetic Electron Transport Chain (PETC) in addition to respiration and auxiliary electron transport routes. Red arrows represent Linear Electron Transport (LET), orange arrows represent Cyclic Electron Transport (CET) and purple respiratory pathways (Cyd and COX). 14 ultimately reduces O2 to H2O (Pils & Schmetterer, 2001). Cyd accepts electrons directly from the PQ pool, while COX receives electrons from PC/Cyt c6, pumping protons to the luminal side in the process (Figure 9). Therefore, the PQ pool, Cyt b6f complex and PC/Cyt c6 are shared for both photosynthesis and respiration and are referred to as interchain components (Shimakawa & Dietz, 2023; Shimakawa et al., 2024). In the case of Synechocystis, RTOs can also act as alternative electron sink proteins to avoid excess energy supply on the PETC caused by several alterations in environmental conditions (Mullineaux, 2014; Lea-Smith et al., 2016; Ermakova et al., 2016; Shimakawa & Miyake, 2018b). Flavodiiron proteins (FLVs) also have a role as alternative electron sinks, dissipating excess electrons to O2 generating H2O, which is especially important under fluctuating light, low carbon conditions or other conditions where PETC is overreduced (Figure 9) (Allahverdiyeva et al., 2011; Mullineaux, 2014). The combination of these highly interconnected pathways helps dissipate the flux pressure and contributes to the balance between energy consuming and producing reactions. (Lea-Smith et al., 2016; Nikkanen et al., 2021; Shimakawa & Dietz, 2023). 15 3 Carbon metabolism As photosynthetic organisms, cyanobacteria rely on a photoautotrophic metabolism, using the ATP and NADPH produced in the PETC during illumination, to synthesize from CO2 their array of compounds, supporting growth and cell division. Synechocystis presents a highly interconnected metabolism, generating intermediates circuits of great complexity (Figure 10). Cyanobacteria present multitude of carbon pathways observed in nature: Calvin-BensonBassham (CBB) cycle, Embden–Meyerhof–Parnas (EMP) pathway, the oxidative pentose phosphate (OPP) pathway, the phosphoketolase pathway and a nonconventional tricarboxylic acid (TCA) pathway with several metabolic alternatives (Xiong et al., 2015, 2017; Makowka et al., 2020; Mills et al., 2020; Veaudor et al., 2020). The Entner–Doudoroff (ED) pathway was also proposed as an active metabolic pathway (Chen et al., 2016; Makowka et al., 2020), but its contribution is controversial (Schulze et al., 2022; Xie et al., 2024). These pathways partially share intermediates and even operate in opposite directions and are key as anabolic precursors for other biomolecules (nucleotides, amino acids or fatty acids). In addition, there are enzymes with bidirectional activities, multiple isoforms, differential expression patterns and allosteric regulations. Therefore, a tuned coordination of the metabolic flux is key to ensure a stable and equilibrated supply of energy and cellular constituents according to the requirements of the cell. Altogether, an in-depth comprehension of such metabolic fluxes and their versatility is crucial for redirecting metabolism towards desired compounds in biotechnological approaches. 16 Glucose-6P PGI 6P-GluconateG6PDH/zwf Gnd NADP+ NADPH NADP+ NADPH F6P F1,6BP DHAP 2K3DPG GAP Gluconate CO2 O2 2P-Glycolate Ribulose-1,5P FBPase PFK 1,3PG Ribulose-5P Xu5P R5P S7P E4P GAP Xu5P S1,7BP DHAP SBPase GAP F6P ATP 3PG PEP Pyr AKG Citrate Ac-CoA GOGAT GS NH4+ Gln Glu Glu Ala ATP ADP ATP ADP NO3NO2NR NiR Malate ME PEPC AcAcetyl-CoA PHB ATP ADP Edd Eda 2PG PGAM PK Eno PGK GAPDH Asp Asn Lys Thr Ac-P PK Val Leu Ile Ser GDP G-3P Glycogen ADP-glucose Glucose-1P ADP AGPase GlgP1/P2 GlgB PPi PGM GlgX Pi UDP-Glucose Sps Sucrose-6P Spp Sucrose CugP GG-3P GgpS GgpP GG ATP gghA Glucose Glycerol Isocitrate Succinate Aconitate Oxalacetate SSal Glucosein GK glcP Glucoseout GnK NADP+ NADPH Gdh Nitrogen assimilation GMPP Mannose-6P Mannose-1P GDP-Mannose PMM/PGM PMI PMM/PGM His TPI/TIM GABA CO2 Glyoxylate GDH Fdred Fdox Fdred Fdox SSADH GDC OGDC CO2 GABA-AT (ArgD) Fumarate MDH SDH IDH Fba N-Glycan Biosynthesis CO2 CO2 CO2 Rpe Rpi Tkt Tal Tkt NADPH Pi, NADP+ F6P ATP ADP RuBisCO Figure 10: Schematic representation of key metabolic pathways and their integration in cyanobacteria. The figure illustrates the core metabolic pathways in cyanobacteria, including glycolysis (grey), the pentose phosphate shunt (blue), Calvin-Benson-Bassham cycle (green, partially shared with pentose phosphate pathway, dark blue arrows), glycogen metabolism (purple), osmolytes synthesis (red), the tricarboxylic acid cycle (yellow) and nitrogen assimilation (light blue). Key metabolic intermediates and enzymes are indicated. 17 3.1 Calvin-Benson-Bassham cycle The Calvin-Benson-Bassham (CBB) cycle is the pathway that contributes most to carbon fixation on Earth (Berg, 2011; Veaudor et al., 2020). The CO2 molecule is incorporated into the intermediate metabolite ribulose 1,5 bisphosphate (RuBP) by the enzyme ribulose bisphosphate carboxylase/oxygenase (RuBisCO), releasing two molecules of 3P-Glycerate (3PG) (Figure 11: 1. CO2 fixation). A portion of these molecules are used for biosynthetic processes, while the majority are redirected back to the cycle to replenish the RuBP pools and thus, ensuring an efficient CO2 fixation. For this regeneration, 3PG is first phosphorylated by the 3-phosphlycerate kinase (PGK) and subsequently reduced by the glyceraldehyde-3P dehydrogenase (GADPH). These reactions consume ATP and NADPH produced in the PETC, respectively, rendering glyceraldehyde-3P (GAP) (Figure 11: 2. Reduction). Next, GAP molecules and its isomer dihydroxyacetone-P (DHAP) are combined and interconverted through several reactions to different sugar phosphates such as fructose-1,6BP, fructose-6P, erythrose-4P, sedoheptulose-1,7BP, sedoheptulose-7P, xyluose-5P, ribose-5P and finally ribulose-5P (Ru-5P) (Figure 11: 3. Regeneration). All these reactions are enzymatically the opposite to the oxidative pentose phosphate (OPP) pathway, the only one specific to the CBB cycle is the reaction catalyzed by sedoheptulose bisphosphatase (SBPase) (Figure 10 - green shadow). Ru-5P is Figure 11: Detailed schematic representation of the Calvin Benson Bassham (CBB) cycle. 24 able to develop an adequate response to nitrogen deprivation. glgC mutant cells do not degrade their phycobiliproteins -maintaining their blue-green aspectand stop dividing immediately after nitrogen removal, in contrast to wild type cells, that divide once. These mutants also present a metabolite overflow phenotype, excreting organic acids, mainly pyruvate and 2-oxoglutarate, in a process suggested to act as a compensation for the lack of the carbon buffer function of glycogen (Carrieri et al., 2012; Gründel et al., 2012; Cano et al., 2018). Also, metabolic overflow of glutamate has been described in the presence of nitrogen in a ΔglgC mutant (Kato et al., 2024). In contrast to the work of Gründel et al., 2012, a double ΔglgA1ΔglgA2 knock-out strain could not be obtained in our laboratory or in (Yoo et al., 2014). The absence of glycogen synthases caused a toxic accumulation of ADP-glucose, compromising viability. However, if the ADP-glucose was diverted towards the synthesis of the osmoprotectant GG in the presence of saline media, viability was restored and thus allowed for complete segregation (Díaz-Troya et al., 2020). 3.2.3 Alternative glycogen/α-glucan pathways in prokaryotes In addition to the classical bacterial glycogen metabolism, alternative pathways for glycogen biosynthesis have been described in the prokaryotic kingdom. Those pathways do not use glucose nucleoside diphosphates, relying instead on disaccharides either phosphorylated such as maltose-1P or non-activated such as sucrose (Figure 14, Cifuente et al., 2024). In the case of maltose-1P pathway, the GlgE enzyme transfers a maltose to a linear glucan, where the GlgB enzyme introduces ramifications (Figure 14a, Rashid et al., 2016). In some organisms. maltose-1P can be directly synthesized from glucose-1P and ADP-glucose, while in other organisms, maltose-1P synthesis is closely related to trehalose metabolism, a maltose isomer (Qiao et al., 2020; Cifuente et al., 2024). The sucrose-dependent glycogen synthesis is performed through the cooperation of an amylosucrase (AMS), that transfers a glucose to an α-glucan chain, releasing fructose, and the GlgB enzyme that introduces the ramifications (Figure 14b and Cifuente et al., 2024). 25 Due to this diversity in polysaccharide synthesis, it is not surprising to find, in some cyanobacteria, α-glucans/polyhexoses independent of glycogen synthases (Xu et al., 2013) and even starch-like/semi-amylopectin polysaccharides (Nakamura et al., 2005; Suzuki et al., 2013; Kadouche et al., 2016). However, Synechocystis lacks trehalose metabolism and neither AMS nor GlgE enzymes are annotated in its genome (Chandra et al., 2011). In addition, most cyanobacteria have undetectable glycogen when either GlgC or GlgA proteins are absent (Suzuki et al., 2010; Gründel et al., 2012). Nevertheless, variants in polysaccharide or other kind of α-glucans or polyhexoses synthesis cannot be discarded. Figure 14: Detailed schematic representation of glycogen synthetic pathways independent of glycogen synthases. Glycogen production via (a) GlgE and (b) amylosucrase enzymes. The elongation reaction is depicted by a pink arrow. Abbreviations: OtsA (trehalose-6P synthase), OtsB (trehalose-6P phosphatase), TreS (trehalose synthase), Pep2/Mak (maltose kinase), GlgM (maltose 1-phosphate-producing glucosyltransferase), AMS (amylosucrase) Figure based on Cifuente JO, Colleoni C, Kalscheuer R, Guerin ME. Architecture, Function, Regulation, and Evolution of α-Glucans Metabolic Enzymes in Prokaryotes. Chem Rev. 2024 Apr 24;124(8):4863-4934. 26 3.3 Embden-Meyerhof-Parnas (EMP/glycolysis) The EMP is a metabolic pathway that allows for the catabolism of G6P to smaller carbon molecules with energy production (Figure 10, grey shadow). The EMP is a reversible (gluconeogenesis) and mostly bidirectional pathway that can synthesize G6P. G6P is an important crossroad substrate for both glycogen synthesis and the OPP pathway (and thus ED and PK pathways). The EMP pathway is enzymatically highly conserved across bacteria although some cyanobacterial strains like Prochlorococcus lack the phosphofructokinase (PFK) enzyme, relying on other glycolytic routes such as the OPP pathway or the ED pathways for glucose metabolism (Chen et al., 2016; Moreno-Cabezuelo et al., 2023). From glucose-6P, acetyl-CoA is obtained in 9 steps with ATP and NADPH production. First, glucose-6P is isomerized to fructose-6P by phosphoglucose isomerase (PGI). Then, phosphofructokinase (PFK) phosphorylates fructose-6P to fructose-1,6BP with ATP consumption. Fructose-1,6BP is split in two phosphotrioses; one glyceraldeyde-3P (GAP) and one dihydroxyacetone-P (DHAP) by fructose-bisphosphate aldolase (FBA). These trioses are isomers interconverted by triosephosphate isomerase (TPI). The GAP undergoes dehydrogenation coupled to an oxidative phosphorylation by glyceraldehyde-3phosphate dehydrogenase (GAPDH) and is then converted into 3PG by the phosphoglycerate kinase (PGK), with ATP production. Finally, 3PG is transformed into pyruvate through sequential conversions by the phosphoglycerate mutase (PGAM), enolase and pyruvate kinase (PK) with 2Pglycerate and phosphoenolpyruvate as intermediates. Pyruvate is then dehydrogenated to acetyl-coenzyme A (Ac-CoA) by the pyruvate dehydrogenase (PDH), with NADH production and CO2 release. In this pathway, most of the enzymes operate also in the anabolic direction, with the exception of PFK and PK. The reverse reaction for PFK is performed by FBPase, the same enzyme from the CBB cycle, since reactions from 3PG to F6P are shared with part of steps involved in the regeneration of Ru15BP. 27 3.4 Oxidative pentose phosphate (OPP) pathway The OPP pathway involves the catalytic dehydrogenation of glucose-6P by glucose-6P dehydrogenase (G6PDH) and 6P-gluconate dehydrogenase (Gnd), generating 2 NADPH molecules and ribulose-5P (Figure 10). These reactions are referred as the OPP shunt. Ribulose-5P is then converted, in an inversed way to the CCB cycle, into other pentoses and intermediates shared by these pathways through the action of the enzymes epimerase, isomerase, transketolase and transaldolase, until the obtention of GAP and fructose-6P. These metabolites are employed to regenerate the glucose-6P pools by the gluconeogenic activity of the EMP pathway involving the enzymes FBA, PKF and PGI (Xiong et al., 2017; Veaudor et al., 2020). Thus, the only enzymes exclusive in the CBB cycle are the ones involved in the ribulose-5P phosphorylation and its subsequent carboxylation (PRK and RuBisCO) and the dephosphorylation of the S1,7BP (SBPase). A bypass to G6PDH has been described by the concerted action of glucose dehydrogenation by a glucose dehydrogenase (Gdh, sll1709) to gluconate and further phosphorylation to 6P-gluconate (6PG) by a gluconate kinase (GnK) (Figure 10) (Chen et al., 2016). However, the contribution of this bypass under photoheterotrophic conditions (supplemented with glucose and the herbicide 3- (3,4-dichlorofenil)-1,1-dimethylurea; DCMU) must be negligible since a G6PDH (zwf) mutant strain is not viable (Chen et al., 2016). 3.5 Entner-Doudoroff (ED) and phosphoketolase pathways As in other bacteria, the ED pathway was previously identified in cyanobacteria (Chen et al., 2016). This pathway would employ the intermediate of the OPP shunt, 6PG, to generate by a dehydratase (Edd) the metabolite 2-keto-3-deoxy6-phosphogluconate (2K3DPG), exclusive of this metabolic route. Then, the 2K3DPG is split into GAP and pyruvate by an aldolase (Eda) (Figure 10). However, recent allegations from the authors have ruled out the existence of an active Edd enzyme (Gutekunst, 2024), claiming there is no flux towards 2K3DPG from 6PG, in agreement with fluxomic studies (Schulze et al., 2022). In addition, 28 in Synechococcus elongatus PCC 7942 the ED pathway is incomplete, and it has been proposed that Eda enzyme acts as an oxaloacetate decarboxylase (Xie et al., 2024). In addition to the described carbon metabolic pathways, there is another pathway that employs the pentose xylulose-5P, the phosphoketolase pathway. The phosphoketolase enzyme cleaves the pentose into GAP and the two-carbon molecule acetyl-P (Ac-P) (Figure 10). Ac-P can be easily converted into Ac-CoA. Thus, this pathway provides cells with an alternative pathway for the production of Ac-CoA, bypassing the carbon loss concomitant with the dehydrogenation of pyruvate by PDH at the cost of the NADPH generation of this reaction (Xiong et al., 2015). However, exploring the PK from Synechococcus elongatus PCC 7942, it was identified as a phosphoketolase able to use sedoheptulose-7P, fructose6P and xylulose-5P when ATP levels drop, promoting cessation of the CBB to avoid futile CO2 fixation (Lu et al., 2023). 3.6 Tricarboxylic acid (TCA) pathway In most organisms with TCA cycle (Figure 10 - yellow), this pathway serves as a catabolic valve to produce reducing power employed in respiration. The 4-carbon molecule oxalacetate is combined with two carbons from Acetyl-CoA by citrate synthase (CS) to generate citrate. Citrate would undergo metabolic transformations with two decarboxylation steps by isocitrate dehydrogenase (IDH) and 2-oxoglutarate (2-OG) dehydrogenase generating NADH and succinate. Succinate, which is a 4 carbon atoms molecule, is further transformed into fumarate by the succinate dehydrogenase (SDH). Fumarate is converted to malate and this to oxalacetate, closing the cycle. However, cyanobacteria present an unusual TCA cycle due to the lack of the 2-OG dehydrogenase complex. Therefore, the cycle had been considered to be incomplete, and their main role proposed as an anaplerotic pathway for both 2-OG and succinate synthesis, key in nitrogen fixation (GS/GOGAT cycle) and respiration, respectively (See sections 4.1 Nitrogen assimilation and amino acid metabolism and 2.2 Respiration and auxiliary electron transport) (Esteves-Ferreira et al., 2018). However, alternatively to the 2-OG dehydrogenation, in cyanobacteria has been described 29 up to 3 pathways that bypass this reaction: the 2-oxoglutarate decarboxylase (OGDC) shunt, that involves the decarboxylation of 2-OG to succinate via succinic semialdehyde (SSal) as intermediate; the gamma-aminobutyric acid (GABA) shunt, which implies the decarboxylation of glutamate to succinate via gamma-aminobutyric acid (GABA) and SSal as intermediates; and finally, succinate and malate production through reactions involving isocitrate and acCoA as substrates and with glyoxylate as an intermediate (Xiong et al., 2017). In Synechocystis, only the OGDC shunt and the GABA shunt have been described (Zhang & Bryant, 2011; Xiong et al., 2014). Although the GABA shunt has been proposed to have the greatest contribution in Synechocystis, its implementation into Synechococcus sp. PCC 7002 showed less efficiency than the OGDC shunt (Zhang et al., 2016). In addition, the cyanobacterial TCA cycle receives another carbon supply/influx besides the incorporation of Ac-CoA. Phosphoenolpyruvate carboxylase (PEPC) synthesizes oxalacetate from PEP by an irreversible carboxylation reaction (Scholl et al., 2020). In a counterpart, the malic enzyme catalyzes the decarboxylation of malate to pyruvate. From Ac-CoA, Synechocystis and some other cyanobacterial strains can also accumulate polyhydroxybutyrate (PHB) (Figure 10), but mainly derived from previously synthesized glycogen reserves and under long acclimatation to nitrogen deprivation (Koch et al., 2019). 3.7 Salt acclimatation: Synthesis of compatible solutes Both aquatic and land habitats present several environmental threats, of which salinity is a mayor one. To maintain turgor pressure and ion homeostasis, cyanobacteria accumulate several soluble low-molecular mass compounds with osmoprotective function. Thus, sugars like sucrose and trehalose, polyols such as glycerol or sorbitol, heterosides like glucosylglycerol (GG) or glucosylglycerate, or amino acids and their derivatives such as glycine betaine can be found as osmoprotectants in cyanobacteria depending of the saline concentrations (Kirsch et al., 2019). In response to salt stress, Synechocystis, 30 due to its freshwater nature, presents a low-mid salt tolerance thanks to the synthesis of both sucrose and GG (Figure 10 red shadows and Figure 12). These compounds are both synthesized from a sugar-nucleotide in combination with a phosphorylated sugar to generate a phosphorylated intermediate releasing the nucleotide. The final step involves the generation of the unphosphorylated product. In the synthesis of sucrose, a disaccharide of glucose and fructose, the sucrose-P synthase (Sps) and the sucrose-P phosphatase (Spp) produce sucrose-P and sucrose respectively starting from UDP-glucose and fructose-6P (Kirsch et al., 2019). In the synthesis of GG, glucosylglycerol-phosphate synthase (GgpS) catalyzes the formation of the intermediate glucosylglycerol-phosphate (GG-3P) from ADP-Glc and glycerol-3-phosphate, and glucosylglycerolphosphate phosphatase (GgpP) dephosphorylates this intermediate to produce GG (Klähn & Hagemann, 2011) It is interesting to highlight the interconnection between osmolyte and glycogen biosynthetic pathways. Both GG and sucrose synthesis, as well as glycogen metabolism, share glucose-1P as a precursor. Furthermore, the synthesis of both glycogen and GG are carried out from the intermediate ADP-Glc. GgpS becomes highly expressed in response to osmotic stress. In addition, GgpS presents a biochemical regulation mediated by nucleic acids interactions to prevent its activity in free NaCl media (Kirsch et al., 2019). Due to the stronger effect of GG as compatible solute, Synechocystis mutants defective in its synthesis show a phenotype of salt-sensitivity (Marin et al., 1998). 4 Nitrogen metabolism 4.1 Nitrogen assimilation and amino acid metabolism Cyanobacteria can employ multiple nitrogen sources such as ammonium, nitrate, nitrite, urea, cyanate, some amino acids and, in the case of nitrogen fixing strains, even atmospheric nitrogen (Esteves-Ferreira et al., 2018). Combined nitrogen is acquired mainly through transport systems (such as the NrtABCD or NrtP). In the case of nitrate, it needs to be reduced to ammonium in a sequential reaction by nitrate reductase (NarB) and nitrite reductase (NirA). These two enzymes are Fd 31 dependent and consume two and six electrons, respectively. (Esteves-Ferreira et al., 2018). Ammonium is then incorporated to the metabolism through the Glutamine Synthetase/Glutamate Oxoglutarate Aminotransferase (GS-GOGAT) cycle (Figure 10 – light blue). First, ammonium is transferred to glutamate by GS, releasing glutamine and consuming ATP in the process. Then, GOGAT transfers one amide group from glutamine to 2-OG to form two molecules of glutamate at expenses of consuming electrons from either reduced Fd or NADPH (EstevesFerreira et al., 2018). Glutamate is further employed as an amino donor to other metabolites/carbon skeletons during the synthesis of other amino acids. 4.2 Chlorosis program in non diazotrophic cyanobacteria As stated in Section 1.1.1 Synechocystis sp. PCC 6803 as a model organism, Synechocystis is a non-diazotrophic cyanobacterium and thus cannot fixate atmospheric nitrogen (N2). Upon combined nitrogen limitation, Synechocystis undergoes an adaptative program known as bleaching or chlorosis to prepare them to face these unfavorable conditions (Forchhammer & Schwarz, 2019). The disbalance in C/N metabolism is reflected by the limitation in the GS-GOGAT cycle and subsequent increase of the substrate 2-oxoglutarate. This metabolic intermediate is sensed by regulatory proteins such as the signal transduction protein PII, that undergoes conformational changes affecting its interaction with other proteins, and the global nitrogen regulator NtcA (Alfonso et al., 2001; Esteves-Ferreira et al., 2018; Forchhammer & Schwarz, 2019). These proteins coordinate a transcriptomic and metabolic response for the acclimation to the lack of nitrogen (Esteves-Ferreira et al., 2018). Briefly, the immediate transcriptomic response involves the increase of uptake and transport of nitrogen systems, the expression of NblA, and then a decline in the transcripts of carbon fixation, and phycobiliprotein genes (Krasikov et al., 2012). During bleaching, since energy demanding reactions such as nitrogen assimilation and synthetic/anabolic pathways (nucleic acids and amino acids) that required nitrogen become restricted, cells respond modifying their photosynthetic machinery and metabolism. One of the major consequences is the degradation of the phycobilisomes, shifting the cell blue-green appearance to 32 an orange-yellowish pigmentation. This is a strategy with a dual function. First, it allows the recycling of the nitrogen present in these large protein complexes as a supply for de novo synthesis of proteins to adapt to this new nutrient situation. Second, the downsize in the light harvesting complexes diminish the excitation pressure in the PETC being adjusted to the current cell status and thus avoiding potential overreduction and photodamage (Salomon et al., 2013; Baier et al., 2014; Levi et al., 2018; Forchhammer & Schwarz, 2019). As an outcome, glycogen reserves are highly increased with the carbon from the phycobiliprotein degradation and the newly fixated CO2 (Forchhammer & Schwarz, 2019). This is promoted by the redirection of the carbon flux towards gluconeogenic reactions by the inactivation of the phosphoglycerate mutase (PGAM). This regulation is caused by the association of the PGAM with the small protein CfrA/PirC, an interactor of PII that becomes released and highly expressed under nitrogen depleted conditions (Muro-Pastor et al., 2020; Orthwein T et al., 2021). A prolonged nitrogen deficiency culminates with the halt of cell division and the entry into a dormant-like state, in which they retain viability for extended periods of time under these starvation conditions (Forchhammer & Schwarz, 2019). In that state, metabolism is reduced to a minimum and residual photosynthetic activity is focused mainly in CET around PSI, providing the cells with a basal stable amounts of ATP (Krasikov et al., 2012; Doello et al., 2018). 4.2.1 Phycobilisomes degradation Phycobilisomes degradation is an orchestrated process and requires the action of several components. Eventually, the regulation response lead by NtcA, among others, conclude with the expression of the proteins NblA, NblB and NblD (Non bleaching proteins) (Baier et al., 2014; Forchhammer & Schwarz, 2019; Krauspe et al., 2021). In Synechocystis, NblA is a heterodimeric dimmer (NblA1/NblA2) that interacts with both the N-terminus of β-phycocyanin and allophycocyanin within the PBS destabilizing the protein assembly (Sendersky et al., 2015; Nguyen et al., 2017). This protein binding is required for the pigment 33 disassociation by the bilin lyase activity of NblB (Levi et al., 2018). In addition, NblA also acts as an adaptor for the recruitment of an ATP-dependent Clp protease complex (ClpC-ClpP1ClpR) to further phycobiliprotein disassembly and degradation (Karradt et al., 2008; Baier et al., 2014). The process begins from the outer rods progressing to the core complex in a sequential process until nearly complete PBS degradation (Sendersky et al., 2015). 4.2.2 Recovery from chlorosis Upon nitrogen is added to chlorotic cultures, cells revert to vegetative growth in a stepwise process starting in heterotrophic metabolic mode (Neumann et al., 2021). This is characterized by fast ATP synthesis, nitrogen assimilation, since GS is present in an active state, and glycogen consumption through respiration (Doello et al., 2018). Glycogen catabolism provides the energy and carbon needed for the synthesis of the translational machinery and the PETC, which become fully functional within 24 hours. 5 Photomixotrophy Although many strains are obligated photoautotrophs, some cyanobacteria including Synechocystis can supplement photosynthesis by using a wide variety of organic carbon compounds from the environment, such as glucose, sucrose, glycerol or acetate in a photomixotrophic growth mode (Rippka et al., 1979; Matson & Atsumi, 2018). The dual carbon source enhances growth, offering attractive possibilities for biotechnology (Wan et al., 2015; Kanno et al., 2017; Luan et al., 2019). Strains growing in mixoor photoheterotrophy conditions undergo metabolic rewiring to optimize resource utilization, integrating their multiple metabolic pathways (Figure 15) (You et al., 2014, 2015; Nakajima et al., 2014; Schulze et al., 2022) although this exogenous organic compound supply induced intricate physiological modulations at various cellular levels that remain unclear (Hihara and Ikeuchi, 1997; Kahlon et al., 2006; Trautmann et al., 2012; Burnap et al., 2015; Zavřel et al., 2017, Muth-Pawlak et al. 2022). However, transcriptomic analyses have shown only minor differences in the expression of carbon 40 1.1.2 Media and culture conditions Synechocystis strains were cultivated under photoautotrophic, photomixotrophic or photoheterotrophic conditions in BG11 medium detailed in Table 2 (Rippka et al., 1979). To obtain photomixotrophic conditions 2 mM glucose was supplemented. Table 2: BG11 Medium composition When indicated this medium was supplemented with NaHCO3 at 12 mM (of otherwise explained in the figure) referred as BG11C. When grown in BG11, media was buffered with 20mM HEPES pH 7.5. Nitrogen-free medium (BG110C) has the same composition as BG11C without the nitrogen source NaNO3. When indicated, BG11C was supplemented with NaCl at the concentration needed from a 5M stock solution. Cultures were grown in growth chambers at 30 ºC with illumination provided by 4500k LED lights. Pre-cultures inoculated in liquid media from Petri dishes were grown in Erlenmeyer flasks with 20% of maximum volume under constant light at 20-40 μmol photons m−2 s−1. Flasks were continuously shaken at 100 rpm in an orbital incubator (IKA Laborthechnik KS501) under atmospheric CO2 until they reached mid-logarithmic phase. These precultures were adjusted to 80% of the desired experimental OD750 nm 24 hours prior to the experiments to ensure that the cultures were in a similar and near growth phase for the onset. BG11 composition NaNO3 17.6 mM Ammonium iron (III) citrate 6 mg/l MgSO4 0.30 mM MnCl2 9,1 μM CaCl2 0.24 mM Na2-EDTA 2,4 μM Na2CO3 0.20 mM Na2MoO4 1,6 μM K2HPO4 0.20 mM ZnSO4 0,8 μM H3BO3 46 μM CuSO4 0,3 μM Citric acid 28.5 μM CoCl2 0,2 μM 41 For continuous light experiments, different light intensities according to the experimental needs were adjusted using a lightmeter (LI-250A quantum-Light Meter, LI-COR ). Cultures were either placed in non-bubbled Erlenmeyer flasks at 100 rpm or in conical flasks bubbled with a stream of 1% (v/v) CO2 in air. For experiments involving nitrogen deficiency, or inducer (NaAsO2) removal, cultures were washed twice by centrifugation (10 min at 7000 x g) and resuspended at the desired OD750 nm in BG110C or BG11C without inducer, respectively. For glucose removal experiments, 20 ml cultures were washed once by centrifugation (5 min at 7000 x g) and resuspended in the same volume of fresh media. Experiments under light/dark conditions were performed in Erlenmeyer flasks on a New BrunswickTM Innova® 43/43R shaker at 40 µE m-2 s-1 whit long-day (16 h/8 h light/dark intervals) or short-day (16 h/8 h dark/light intervals) programs. When required, media were supplemented with 1% w/v agar (Bacto-agar, Difco). Medium and agar were sterilized independently by standard autoclaving procedures and mixed before platting. Photoheterotrophic conditions were assayed on solid media adding 1 mM glucose and 20µM DCMU (3-(3,4dichlorofenil)-1,1-dimethylurea). Culture collection maintenance and segregation of strains were made on Petri plates with the appropriate antibiotics sterilized by filtration (50 µg ml-1 kanamycin, 20 µg ml-1 chloramphenicol, 2.5 µg ml-1 spectinomycin, 2.5 µg ml-1 streptomycin, 50 µg ml-1 nourseothricin and 5 µg ml-1 erythromycin). The plates were incubated in the same growth chambers as the liquid cultures under continuous 8 µE m-2 s-1 light intensity illumination. 1.2 Escherichia coli 1.2.1 Strains The different strains employed in this work and their genotype are described in Table 3. 42 Table 3: Escherichia coli strains used in this work Strains Genotype Reference DH5α F-, endA1, hsdR17 (mK+,rK- ) supE44, thi1 recA1 gyrA96 relA1 ΔlacU169 (φ80lacZΔM15) (Hanahan, 1983) BL21 (DE3) hsdS gal (λcIts857 ind1 Sam7 nin5 lac UV5-T7 gene1) (Studier & Moffatt, 1986) JW3392 (K-12 BW25113; Keio Collection) F-, Δ(araD-araB)567, ΔlacZ4787(::rrnB-3), λ-, ΔglgA762::kan, rph-1, Δ(rhaDrhaB)568, hsdR514 (Baba et al., 2006) In general terms, DH5α was used for DNA amplification, manipulation and cloning while BL21 and JW3392 were used for protein expression. 1.2.2 Media and culture conditions The culture medium for E. coli was the Luria-Bertani (LB) (Sambrook & W Russell, 2001) with a composition of 10 g/l NaCl; 10 g/l Tryptone and 5 g/l yeast extract. The medium was sterilized using standard autoclaving procedures. E. coli was inoculated in flasks and tubes at 1/5 of the maximum capacity of the recipient with constant shaking in an Innova®43 (New Brunswick Scientific). For solid media preparation, LB was supplemented with 1.5% w/v of agar (European Bacteriological Agar, Conda) before autoclaving. The liquid and solid cultures were incubated at 37 ºC. Antibiotics sterilized by filtration were provided when necessary (100 µg ml-1 ampicillin; 25 µg ml-1 kanamycin; 20 µg ml-1 chloramphenicol, 50 µg ml-1 spectinomycin, 50 µg ml-1 streptomycin and 50 µg ml-1 nourseothricin). For pSPARK insert detection by white/blue colony staining, plates were also supplemented with 40 µl of 2% w/v X-gal (5-bromo-4-chloro-3-indolyl-β-Dgalactopyranoside) and 4 µl of 100 mM IPTG (Isopropyl β-d-1thiogalactopyranoside) on the plate surface. To enhance glycogen synthesis, cultures growing on LB were supplemented with 20 mM filtered glucose. Culture conditions for protein expression are detailed in Section 3. 1 Protein expression in E. coli. 43 2 DNA analysis and manipulation 2.1 Plasmids Plasmids have been used to generate mutant strains in Synechocystis and for protein expression in E. coli. Complete list of plasmids both non-produced and generated in this work are included along with relevant features and descriptions in Table 4 and Table 5, respectively. pGEM-T ∆sll0726 Km pGEM-T ∆glgA1::Km pGEM-T ∆glgA2::Sp pGEMT ∆slr1334::Ery pSPARK ∆GGPS:Nat pSPARK ∆glgP2:Cm pSPARK ∆glgB::Nat pSPARK ∆Zwf::Nat 68 79 1656 bp Km_R KmR sll0746 1320 bp 2700 bp 1344 bp 111 1476 bp glgA1 1320 bp 1520 bp 1360 bp 107 111 107 KmR 36 1434 bp SpSt glgA2 1800 bp 2475 bp 3000 bp 39 36 39 149 1500 bp Nat_R NatR sll1556 1028 bp 2610 bp 2147 bp 152 165 149 152 950 bp 920 bp pSPARK ∆glgP1:Nat sll1356 153 2500 bp NatR 1028 bp 3660 bp 2122 bp 156 166 840 bp 941 bp Nat_R 153 156 160 slr1367 157 2500 bp CmR 1900 bp 3700 bp 167 3000 bp 2780 bp 157 160 3724 bp sll0158 112 2313 bp Nat_R NatR 1028 bp 2450 bp 168 115 1100 bp 1160 bp 115 112 slr1843 161 1530 bp Nat_R NatR 1028 bp 2660 bp 2170 bp 169 164 880 bp 835 bp 164 161 slr1334 1473 pb slr1219 577 bp 2200 bp Ery_R 1272 bp 72 EryR 72 184 80 pSPARk ∆ggpP::Cm slr0746 172 1269 bp CmR 1900 bp 2600 bp 3300 bp 920 bp 172 175 175 185 Cm_R 1800 bp pSPARk ∆glgX1::Ery pBS nrsD PGMc SpV3 pBS nsrD PcpcB GGPS:Cm 200 1338 bp nrsD slr0796 1473 bp nrsD UP nrsD DOWN SpSt 1800 bp ≈650 bp ≈650 bp g204 g204 g207 1240 bp 79 2520 bp 1656 bp sll0746 200 570 bp 1338 bp nrsD slr0796 nrsD UP p cpcB nrsD DOWN ≈650 bp ≈650 bp 191 g207 1875 bp 1500 bp sll1556 187 CmR 1900 bp pnrsD_PcpcB:PMM SpSt pnrsD_ParsB:PMM SpSt slr1334 570 bp 1338 bp nrsD slr0796 1473 bp nrsD UP p cpcB nrsD DOWN SpSt 1800 bp ≈650 bp ≈650 bp g204 g204 g207 1240 bp 184 2600 bp 81 1700 bp slr1334 450 bp 1338 bp nrsD 1473 bp nrsD UP p arsB nrsD DOWN SpSt 1800 bp ≈700 bp ≈700 bp g204 g204 g207 ≈1350 bp 184 1580 bp Arsenite 2480 bp 81 pSPARk ∆glgX2::Cm slr1857 125 2124 bp CmR 1900 bp 3350 bp 3130 bp 1180 bp 125 128 128 183 Cm_R 1700 bp slr0237 121 2241 bp 3595 bp 2579 bp 1223 bp 121 124 124 182 1860 bp Ery_R 1272 bp EryR Figure 16 Schematic representation of the strategy used for generating knock-out mutant strains. 44 All plasmids were verified by Sanger sequencing. Plasmid used in Synechocystis transformation are illustrated in Figure 16. 2.1.1 Commercial and laboratory collection plasmids. Plasmid from commercial origin or from the laboratory collection (generated by others members) are described in Table 4. Table 4 Plasmid from commercial origin or already present in the laboratory nº Name Ab R Reference Description pSPARK-II Amp Canvax Biotech Plasmid is utilized for cloning DNA fragments. When circularized without inserts, it expresses the α fragment of the β-galactosidase, enabling colony selection in the presence of X-gal and IPTG. p58 pGEM-T ∆slr1334#1 Amp (OrtegaMartínez et al., 2023) Plasmid similar to pSPARK-II. It has the slr1334 flanking regions separated by a BamHI restriction site to introduce a resistance gene. The C-terminal region of slr1334 was conserved to avoid disrupting the hypotetical protein Sll1219. p60 pGEM-T ∆sll0726 Km#2 (+) Amp, Km (OrtegaMartínez et al., 2023) Employed to generate ∆PGM strain. It has flanking regions of sll0726 separated by a kanamycin resistance gene cloned in BamHI restriction site. p22 pGEM-T ∆glgA1::Km# 1 + Amp, Km (Díaz-Troya et al., 2014) Employed to generate ∆glgA1 strains. It has flanking regions of sll0945 separated by a kanamycin resistance gene cloned in BamHI restriction site p24 pGEM-T ∆glgA2::Sp# 1 + Amp, SpSt (Díaz-Troya et al., 2014) Employed to generate ∆glgA2 strains. It has flanking regions of sll1393 separated by a spectinomycin resistance gene cloned in BamHI restriction site. p115 pBS nsrD PcpcB spV3 SpSt (OrtegaMartínez et al., 2023) Derived from the pBSII-KS(+) with the cpcB promoter, multi copy site for cloning and SpR, all flanked by nrsD regions. p116 pBS nsrD ParsB spV3 SpSt (OrtegaMartínez et al., 2023) Derived from the pBSII-KS(+) with the inducible by arsenite arsB promoter, multi copy site for cloning and SpR, all flanked by nrsD regions. p138 pBS plat glnN::cpcB:N at Nat Lab collection Derived from the pBSII-KS(+) with the cpcB promoter, multi copy site for cloning and NatR, all flanked by nrsD regions. 45 pQE 81L Amp Qiagen Plasmid from the pQE collection used for control protein expression in E.coli using IPTG as inducer. Protein purification by nickel affinity chromatography is possible due to the His-Tag in the N-terminal region. pET28a (+) Km Novagen Plasmid from the pET collection used for control protein expression in E. coli using IPTG as inducer. Protein purification by nickel affinity chromatography is possible due to the His-Tag in the N-terminal region. p67 pET28a sll0726#2 Km Lab collection Plasmid from the pET28 collection used for PGM expression in E. coli using IPTG as inducer. p68 pET28a slr1334#3 Km Lab collection Plasmid from the pET28 collection used for PGM/PMM expression in E. coli using IPTG as inducer. p26 pET45bglgA1 #1 Amp Lab collection Plasmid from the pET28 collection with the glgA1 CDS. Used for sub-cloning glgA1 CDS in pQE plasmid PRL cm Amp, Cm Lab collection Plasmid containing Chloramphenicol (Cm) resistance gene PRL Nat_term Amp, Nat Lab collection Plasmid containing Nourseothricin (Nat) resistance gene PRL Ery Amp, Ery Lab collection Plasmid containing Erythromycin (Ery) resistance gene 2.1.2 Plasmids generated in this work. Plasmids generated in this work are described in Table 5. Table 5 Plasmid generated in this work nº Name Ab R Description p91 pSPARK-II ∆GgpS Amp Plasmid derived from pSPARK-II with sll1566 (GgpS CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 149-150 for the UP region and 151-152 for the DOWN region. p92 pSPARK-II ∆glgP1 Amp Plasmid derived from pSPARK-II with sll1356 (GlgP1 CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 153-154 for the UP region and 155-156 for the DOWN region. p93 pSPARK-II ∆glgP2 Amp Plasmid derived from pSPARK-II with slr1367 (GlgP2 CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 157-158 for the UP region and 159-160 for the DOWN region. 46 p94 pSPARK-II ∆glgB Amp Plasmid derived from pSPARK-II with sll0158 (GlgB CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 112-113 for the UP region and 114-115 for the DOWN region. p95 pSPARK-II ∆G6PDH Amp Plasmid derived from pSPARK-II with slr1843 (Zwf/G6PDH CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 161-162 for the UP region and 163-164 for the DOWN region. p96 pSPARK-II ∆GgpS::Nat (+) Amp, Nat Plasmid used to generate ∆GgpS Synechocystis strain. It is derived from pSPARK ∆GgpS (p91) with NatR gene cloned in BamHI p98 pSPARK-II ∆glgP1::Nat (+) Amp, Nat Plasmid used to generate ∆glgP1 Synechocystis strain. It is derived from pSPARK ∆glgP1 (p92) with NatR gene cloned in BamHI p100 pSPARK ∆glgP2:Cm #1 (+) Amp, Cm Plasmid used to generate ∆glgP2 Synechocystis strain. It is derived from pSPARK ∆glgP2 (p93) with CmR gene cloned in BamHI p102 pnrsD_PcpcB:PMM SpSt Amp, SpSt Plasmid used to overexpress the PMM/PGM protein in Synechocystis. It is derived from pBS nsrD PcpcB spV3 (p115) with slr1334 from pET28 slr1334#3 (p68). Both parental plasmid were cut with XbaI/XhoI and appropriate fragments ligated. p103 pGEMT ∆slr1334::Ery Amp, Ery Plasmid used to generate ∆PMM/PGM Synechocystis strain. It is derived from pGEM-T ∆slr1334#1 (p58) with EryR ligated in BamHI p104 pSPARK ∆G6PDH::Nat(+) #3 Amp, Nat Plasmid used to generate ∆G6PDH Synechocystis strain. It is derived from pSPARK ∆G6PDH (p95) with NatR gene cloned in BamHI p106 pSPARK ∆glgB::Nat(+) #4 Amp, Nat Plasmid used to generate ∆glgB Synechocystis strain. It is derived from pSPARK ∆glgB (p94) with NatR gene cloned in BamHI p117 pnrsD_ParsB:PMM SpSt Amp, SpSt Plasmid used to induce the regulated expression of PMM/PGM protein by arsenite in Synechocystis. It is derived from pBS nsrD ParsB spV3 (p116) with slr1334 from pET28 slr1334#3 (p68). Both parental plasmids were cut with XbaI/XhoI and appropriate fragments ligated. p119 pSPARK-II ∆glgX2 Amp Plasmid derived from pSPARK-II with slr1857 (GlgX2 CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 125-126 for the UP region and 127-128 for the DOWN region. p120 pSPARK PGMc (sll0726+200 pb EcoRI/XhoI) Amp Plasmid derived from pSPARK-II with sll0726 plus 200 bp prior and after the gene. It was obtained by performing a PCR using Synechocystis genomic DNA and primers 180-181 carrying EcoRI and XhoI restriction site sequences at the 5' and 3’ ends, respectively. p121 pSPARK-II ∆glgX2::cm (-) Amp, Cm Plasmid used to generate ∆glgX2 Synechocystis strain. It is derived from pSPARK-II ∆glgX2 (p119) with CmR gene cloned in BamHI p122 pSPARK-II ∆glgX Amp Plasmid derived from pSPARK-II with slr0237 (GlgX1 CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 121-122 for the UP region and 123-124 for the DOWN region. 47 p123 pSPARK-II ∆glgX::Ery(+) Amp, Ery Plasmid used to generate ∆glgX1 Synechocystis strain. It is derived from pSPARK-II ∆glgX1 (p122) with EryR gene cloned in BamHI p127 pBS nrsD PGMc SpV3 #7 Amp, SpSt Plasmid used to express the PGM (sll0726) protein under its own promoter. Construction integrated in nrsD gene. Plasmid is derived from the pnrsD_PcpcB:PMM SpSt (p102) and pSPARK PGMc (sll0726+200 pb EcoRI/XhoI)(p120) both cut at EcoRI and XhoI restriction sites and appropriate fragments ligated. p129 pSPARK-II ∆ggpP slr0746 #2 Amp Plasmid derived from pSPARK-II with slr0746 (GgpP CDS) flanking regions separated by a BamHI restriction site. It was obtained by performing an overlapping PCR using Synechocystis genomic DNA and primers 172-173 for the UP region and 174-175b for the DOWN region. p131 pSPARK-II NdeI_GgpS_XhoI #8 Amp Plasmid derived from pSPARK-II with GgpS CDS flanked by NdeI and XhoI restriction site at the 5' and 3' ends, respectively. It was obtained by performing a PCR using Synechocystis genomic DNA and primers 186-187 carrying NdeI and XhoI restriction site sequences at the 5' and 3’ ends, respectively. p135 pSPARK-II ∆ggpP::Cm (+) Amp, Cm Plasmid used to generate ∆ggpP Synechocystis strain. It is derived from pSPARk-II ∆ggpP slr0746 #2 (p129) with CmR gene cloned in BamHI p141 pSPARK-II GgpS (BamHI/XhoI) Amp Plasmid derived from pSPARK-II with GgpS CDS flanked by BamHI and XhoI restriction site at the 5' and 3' ends, respectively. It was obtained by performing a PCR using Synechocystis genomic DNA and primers 192-187 carrying BamHI and XhoI restriction site sequences at the 5' end, respectively. p148 pBS nsrD PcpcB CmR #2 Amp, Cm Plasmid derived from pBS nsrD PcpcB spV3 (p115) changing the SpV3 resistance gene by CmR gene. Both the p115 and PRL CmR were cut with HindIII and appropriate fragments ligated. p149 pBS nsrD PcpcB GgpS Cm Amp, Cm Plasmid used to overexpress GgpS protein in Synechocystis. It is derived from ppBS nsrD PcpcB CmR (p1148) with ggpS from pSPARk-II NdeI_GgpS_XhoI (p131). Both parental plasmid were cut with NdeI/XhoI and appropriate fragments ligated. p152 pQE 81L ggpS Amp Plasmid derived from pQE 81L to express GgpS in E. coli. It is derived from pQE 81L with ggpS from pSPARk-II GgpS (BamHI/XhoI) (p141). Both parental plasmids were cut with BamHI/SalI and appropriate fragments ligated. p153 pQE 81L glgA1 Amp Plasmid derived from pQE 81L to express glgA1 in E. coli. It is derived from pQE 81L with glgA1 from pET45b glgA1 (p26). Both parental plasmid were cut with BamHI/HindIII and appropriate fragments ligated. 2.2 Primers All set of primers employed in this work with an use description are listed in Table 6. 48 Table 6: List of primers used in this work. nº Name Sequence (5'-->3') Use 32 glgA1 UP 5' GCCTTATTCTGTTGCTACGTCAATG Forward primer localized -500 pb before the 5’ end of the sll0945 ORF. Used to confirm complete segregation of glgA1 in Synechocystis by colony PCR (32-35). 35 glgA1 DO 3' TAGAATGAAGCTGGAAATCGGCTC Reverse primer localized +500 pb after the 3’ end of the sll0945 ORF. Used to confirm complete segregation of glgA1 in Synechocystis by colony PCR (32-35). 36 glgA2 UP 5' GCGGCGCTTGGTATTTGTGGAAG Forward primer localized -500 pb before the 5’ end of the sll1393 ORF. Used to confirm complete segregation of glgA2 in Synechocystis by colony PCR (36-39). 39 glgA2 DO 3' CCTGGGTAGGGTTATGGCTTTC Reverse primer localized +500 pb after the 3’ end of the sll1393 ORF. Used to confirm complete segregation of glgA2 in Synechocystis by colony PCR (36-39). 68 sll0726 UP 5' GGTCATTAACTGTGCCTTCGTCAC Forward primer localized at - 1000 bp of the sll0726 5' Used to confirm complete segregation of PGM in Synechocystis by colony PCR (68-79b-g159). 72 slr1334 UP 5' GATAGGAAGAACTCCCAACTCTACTG Forward primer located at - 1000 bp of the slr1334 5'. Used to confirm pmm/pgm substitution by EryR inserted in BamHI site (72-EryR) and complete segregation in Synechocystis by colony PCR (80-EryR). 79b sll0726 XhoI 3' gccctcgagTTAGCCCAAAGCCGAGGTAAC Reverse primer localized at the sll0726 3' end Used to confirm complete segregation of PGM segregation by colony PCR (6879b-g159) and integration of the PGM complementation in nrsD (g204-79). 80 slr1334 NdeI 5' ccggcatATGGTTTACACTCCTGCTCC Forward primer located at the 5’ end of the slr1334 ORF. Used to confirm complete segregation of pmm/pgm by colony PCR (80-184). 81b slr1334 XhoI 3' gccctcgagTCAGTGAGATGATTGTGCAG Reverse primer localized at the slr1334 Cterminal end with a XhoI restriction enzyme site. Used to clone slr1334 in p115 and p116 to overexpress and regulated expression, respectively, of PMM/PGM in Synechocystis. 107 glgA1 BamHI UP2 3' CAGTGGAGCGGGATccGGGTTCA Reverse primer localized after the 3’ end of the sll0945 ORF. Used to confirm complete ΔGlgA1 segregation by colony PCR (111107). 111 glgA1_1F CTCAATTAGTGGAATAACGACGGT Forward primer localized before the 5’ end of the sll0945 ORF. Used to confirm complete ΔGlgA1 segregation by colony PCR by colony PCR (111-107). 112 glgB UP 5' CGACCGGGTGGAAATCCTTG Used to generate a sequence of glgB ORF flanking regions by overlapping PCR to obtain a deletion mutant strain. 113 glgB BamHI UP3' TATCTAAGCggatccTCGGCTATGGTGATTTTTTG 114 glgB BamHI DO5' CATAGCCGAggatccGCTTAGATAACTGAGTTAGC 115 glgB DO 3' CTCCCAAACCTGGGGTCATC 121 glgX slr0237 UP 5' GAATGTATCAGTGATTCCAGCGAGAT Used to generate a sequence of glgX1 ORF (slr0237) flanking regions by overlapping PCR to obtain a deletion mutant strain. 122 glgX slr0237 BamHI UP 3' AAACTCAGTAAAggAtcCGGCAATTCCAAAAATTGT 123 glgX slr0237 BamHI DO 5' TTGGAATTGCCggatccTTTACTGAGTTTTTGCCCA 124 glgX slr0237 DO 3' TATTGAGCATGATTTTAGTGCGTTAA 49 125 glgX slr1857 UP 5' GGTAATGAACGGGCTAAGGGTAGTC Used to generate a sequence of glgX2 ORF (slr1857) flanking regions by overlapping PCR to obtain a deletion mutant strain. 126 glgX slr1857 BamHI UP 3' ATTTAAGACAATGgaTCCTGCTAAAATGCCTCTTT 127 glgX slr1857 BamHI DO 5' AGGCATTTTAGCAggatccATTGTCTTAAATTTCCCT 128 glgX slr1857 DO 3' TCAGACAAAGCAGGAGAAGTCCAACA 149 ggpS UP 5' AAGTCAAAAAGACTGGAAAATGGG Used to generate a sequence of ggpS ORF flanking regions by overlapping PCR to obtain a deletion mutant strain. 150 ggpS BamHI UP 3' GATCGCggatccATCAGCGGTCTCCAAAATC 151 ggpS BamHI DO 5' GCTGATggatccGCGATCGCCAATGCCAG 152 ggpS DO 3' CTTAACATCCTATGGGAAGGG 153 sll1356 UP 5' GCCCCATTCTCGAAATAATCGC Used to generate a sequence of glgP1 ORF (sll1356) flanking regions by overlapping PCR to obtain a deletion mutant strain. 154 sll1356 BamHI UP 3' GGTTAAGGaTCCCTTATAGTGTCGGACCAG 155 sll1356 BamHI DO 5' AGGGAtCCTTAACCTGCCTCTATTGCAC 156 sll1356 DO 3' CGGGACCAATGCCCTCAATGC 157 slr1367 UP 5' GCAGGCTTTCCATCCAAGTGC Used to generate a sequence of glgP2 ORF (slr1367) flanking regions by overlapping PCR to obtain a deletion mutant strain. 158 slr1367 BamHI UP 3' GATCTGGGAtCCAGGCTAGCCGACGC 159 slr1367 BamHI DO 5' TAGCCTGGaTCCCAGATCTGGTTTGACTGC 160 slr1367 DO 3' GGACGCCCTCCCTCAATGC 161 slr1843 UP 5' GGTGGGACTATTACCGGG Used to generate a sequence of G6PDH ORF (slr1843) flanking regions by overlapping PCR to obtain a deletion mutant strain. 162 slr1843 BamHI UP 3' TGCAGTggatccGGGGAATGCTCGTATT 163 slr1843 BamHI DO3' TTCCCCggatccACTGCAACGGTCATCG 164 slr1843 DO3' TGCCCTTTACCACCGGAGC 165 ggpS R TCAGAAGAATCGTAGGTG Reverse primer located inside ggpS. Used to confirm ΔggpS complete segregation by colony PCR (149-152-165). 166 sll1356 R AACGTAGCCAGGGAATC Reverse primer located inside sll1356. Used to confirm Δsll1356 complete segregation by colony PCR (153-156-166). 167 slr1367 R GACCTGGACTTAAACGA Forward primer located inside slr1367. Used to confirm Δslr1367 complete Synechocystis segregation by colony PCR (157-160-167). 168 glgB R CCGTCCCAGTTGTTGAA Reverse primer located inside glgB. Used to confirm ΔglgB complete segregation by colony PCR (112-115-168). 169 zwf R GAGATAGCTTTCTGGG Reverse primer located inside G6PDH. Used to confirm ΔG6PDH complete segregation by colony PCR (161-164-169). 172 slr0746 UP 5' AACTGGCTTTATCAGCG Used to generate a sequence of ggpP ORF (slr0746) flanking regions by overlapping PCR to generate complete deletion mutant strain. 173 slr0746 UP 3' TTCACCACCATAAggatccTATTAATCTGCTTAATT 174 slr0746 DO 5' TTAAGCAGATTAATAggatccTTATGGTGGTGAAG 175b slr0746 DO 3' GTCGCTATGGAGTTTACTGATTTC 56 incubation, the cells were centrifuged and resuspended in 100 µl. The resuspension was then plated on LB plates with the necessary antibiotics for transformant selection and incubated overnight at 37 ºC. 2.6.2.2 Cell preparation for heat shock The desired strains were inoculated in Luria-Bertani medium, with antibiotics if necessary, and incubated overnight at 37 ºC and 200 rpm. For the next step, a fresh Luria-Bertani medium (with antibiotics, if necessary) was used to inoculate 1/100 of the final volume of the preculture. The mixture was then incubated until the OD measured at 600 nm reached approximately 0.6 absorbance units. Flask was placed in an ice bath for 10 min and cells centrifuged 10 min 4 ºC 1500 g. The supernatant was discarded and the cells resuspended by gently pipetting in TJBI solution (100 mM RbCl, 50 mM MnCl2·4H2O, 10 mM CaCl2·2H2O, 15% glycerol, 5.8 pH, sterile and ice-cold) in 1/4 of the initial volume. Cells were harvested (10 min 4 ºC 1500 g) and gently resuspended in TJBII (10 mM MOPS, 10 mM RbCl, 75 mM CaCl2·2H2O, 15% glycerol, 7 pH, sterile and ice-cold), Next, cell suspension was incubated in cold water for 15 min, aliquoted, deep-frozen in N2 and stored at -70 ºC until use. 2.6.2.3 Electroporation E. coli BL21 (DE3) and JW3392 competent cells stored at -80 ºC were mixed with 0.2-2 µg of purified plasmids in a maximum of 4 µl and incubated for 5 min on ice. The mixture was transferred to a 2 mm Gene Pulser cuvette (BioRad) and thawed on ice for another 5 min. An electric pulse of 2.50 kV mode 2 was applied to the cuvette using a MicroPulser (Bio-Rad) device. After the pulse, the cells were recovered from the cuvette with 1 ml fresh LB and incubated at 37 ºC for at least 1 hour. After the incubation, 10 µl of the sample was inoculated into liquid LB with the appropriate antibiotics. Alternatively, cells were concentrated in 100 µl and plated on LB plates with the required antibiotic for transformant. The plates were then incubated overnight at 37 ºC. 57 2.6.2.4 Cell preparation for electroporation The desired strains were inoculated in Luria-Bertani medium, with antibiotics if necessary, and incubated overnight at 37 ºC and 200 rpm. For the next step, a fresh Luria-Bertani medium, also with antibiotics, if necessary, was used to inoculate 1/100 of the final volume of the preculture. The mixture was then incubated until the OD measured at 600 nm reached approximately 0.6 absorbance units. Flask was placed in an ice bath for 10 min. The culture was centrifuged at 1500 g for 5 min at 4 ºC. The supernatant was carefully removed, and the cells resuspended in ultra-pure cold sterile H2O at half of the initial volume. This step was repeated, and the cells were resuspended in ultra-pure cold sterile H2O at 1/100 of the initial volume with 10% glycerol. The cell suspensions were aliquoted and deep-frozen in N2 and stored at -70 ºC until use. 3 Biochemical methods 3.1 Protein expression in E. coli Recombinant protein expression in E. coli was performed according to the instructions of pET vectors (Novagen) and pQE vector (Qiagen) manufacturers. The plasmids derived from these chassis (p26, p68, p152, p153, or the empty pQE 81L vector as control in glycogen synthesis complementation in E. coli) were transformed into E. coli BL21 (DE3) or JW3392/BW25113 by electroporation (see section 2.6.2.3). The transformants were inoculated in LB medium supplemented with kanamycin (pET28 vectors) or ampicillin (pET45 and pQE 81L vectors). To propagate the culture, a desired amount of fresh LB with antibiotics was cultured by inoculating 1/100 of the total volume with the precultures. The cultures were grown until the OD measured at 600 nm reached approximately 0.6 absorbance units. Flasks were placed in an ice bath for 10 min and the expression of 6xHistagged recombinant proteins was induced adding 0.2 mM isopropyl-β-Dthiogalactoside (IPTG). For protein purification, cells were harvested by centrifugation after 24 hours of incubation at 20-25 ºC. To test glycogen synthesis complementation in E. coli, the cells were cultured at 37 ºC and aliquots were collected at two and 24 hours after IPTG induction. 58 In both cases, the protein pattern of the cultures was analyzed using SDS-PAGE and Coomassie staining to confirm protein induction visually. 3.2 Preparation of cell extracts 3.2.1 Glass bead lysis Pellets from harvested cultures were resuspended in 50 mM Tris-HCl (pH 8), 25 mM NaCl, and 1 mM phenylmethylsulfonyl fluoride (PMFS) with glass beads (equivalent volume to 150-300 µL). Cell lysates were obtained after mechanical disruption either by manual disruption (ten cycles of 1 min vortexing/resting on ice) or automatic disruption (at 6m/s for 30s in a FastPrep-24 5G; MP Biomedicals). Soluble and membrane protein fraction or just soluble proteins were collected in a new tube after centrifugation at 4 ºC for 5 min at 1500 g or 20 min at 15000 g, respectively. 3.2.2 Sonication lysis Cell pellets were resuspended in 50 mM Tris-HCl pH 8.0, 500 mM NaCl, 1 mM PMSF and 4 U DNase Turbo (Invitrogen). Cells were lysed by sonication (20 kHz, 75 W) on ice for 3 min (in 30-s periods). Lysates were centrifuged at 20000 g for 30 min and supernatants were filtered through 0.22 µM filters (Millipore). 3.2.3 Whole cell preparation The E. coli pellets (0.2-0.5OD/tube) were resuspended in 200 µL of 50 mM TrisHCl and 50 mM NaCl with 1x protein loading buffer. The tubes were then incubated for 5 min at 95 ºC. For Coomassie staining visualization, 20 µl of resuspension was loaded for SDS-PAGE. 59 3.3 Protein quantification 3.3.1 Bradford Soluble protein fractions were quantified with Bradford reagent (Biorad) using Bovine Serum Albumin (Sigma) as standard and following manufacturer indications. 3.3.2 Lowry Total protein content was determined by the method of Lowry as described by (Markwell et al., 1978). Briefly, samples up to 200 µl were mixed with reagents and after incubation absorbance was measured at 750 nm. As protein standard, known amounts of ovalbumin (Sigma) were employed. 3.4 Protein electrophoresis in denaturing polyacrylamide gels (1D SDSPAGE) Electrophoretic separation of proteins was performed on denaturing and reducing conditions based on the Laemmli system as described in (Sambrook & W Russell, 2001) using hand-made 10-12% acrylamide/bis-acrylamide gels. For membrane bound protein analysis, gels were supplemented with 6 M urea. Cell lysates from any of the preparation methods were mixed with 4x protein loading buffer (0.125 M Tris-HCl pH 6.8; β-mercaptoethanol 10% v/v, SDS 4% w/v, bromophenol blue 0.0025% w/v and glycerol 20% v/v) and denatured in a heat block at 100°C. Typically, 5-15 µg of protein was loaded per well and the electrophoresis was run in a Mini Protean III system (Bio-Rad) applying a voltage between 150 and 200 V for 45-80 min. As molecular mass standards, Precision Plus Protein™ Unstained Standards and Precision Plus Protein Dual Color Standards from BioRad were used. 3.4.1 Coomassie blue stainning For protein visualization in SDS-PAGE gels, they were embedded in a solution of Coomassie blue R-250 (Sigma) 0.1% w/v, 10% v/v acetic acid and 40% v/v methanol) until complete staining (10-20 min) and destained in Destaining 60 Solution (10% v/v acetic acid and 40% v/v methanol in water) for several rounds until background cleared. 3.4.2 Protein immunodetection (Western blotting) For protein immunodetection, the proteins separated in SDS-PAGE gels were transferred to either nitrocellulose (BioRad) or polyvinylidene difluoride (PVDF, Immobilon-P; Millipore) 0.45 µm pore size membranes during 1 h at 0.8 mA cm-2 in a TE 77 PWR Semi-Dry Transfer Unit (GE Healthcare). Membranes, gels and paper used were dampen in a Trans-Blot buffer (Tris-HCl 49,4 mM, glycine 39 mM, pH 8,3, SDS 1,3 mM and methanol 20% (v/v)). PVDF membranes were activated with pure methanol following manufacturer instructions prior protein transfer. When needed, nitrocellulose membranes were reversible stained with a Pounceau solution by incubating them for 5 min and subsequent washing steps to eliminate the excess of staining. Membranes were blocked during 1 h in blocking solution [PBS-T solution (phosphate buffered saline with 1% Tween-20) and 5% non-fat milk powder (Applichem)] and immunoblotted overnight at 4 °C at constant shaking with primary antibodies against specific proteins (Table 7) diluted in blocking solution. After incubation with primary antibody, membranes were washed four times for 8 min in PBS-T and then incubated for 1 h with horseradish peroxidase-conjugated secondary antibody (1:25000 in blocking solution, Sigma). Membranes were then rinsed four times with PBS-T prior to chemiluminescent signal detection using Clarity™ western ECL Substrate (Biorad) in an ImageQuant 800 imaging systems (Amersham). For signal quantification ImageQuant TL8.1 software was used. Table 7: Primary antibodies for inmmunoblotting used in this work. Antibody Reference Dilution α-AGP (Díaz-Troya et al., 2014) 1:20000 α-GlgA1 (Díaz-Troya et al., 2014) 1:5000 61 α-GlgA2 (Díaz-Troya et al., 2014) 1:5000 α-PGM (Ortega-Martínez et al., 2023) 1:50000 α-PMM/PGM (Ortega-Martínez et al., 2023) 1:20000 α FBP Lab collection 1:20000 α-SBP Lab collection 1:20000 α-GS I Lab collection 1:100000 α-D1 AS111786; Agrisera 1:10000 α-CP-47 AS04038; Agrisera 1:10000 α-PsaB AS10695; Agrisera 1:10000 α-ATPβ AS05085; Agrisera 1:10000 α-GroEL G6532; Sigma-Aldrich 1:75000 α-PC Lab collection 1:5000 α-CytC6 Lab collection 1:5000 3.5 Recombinant protein purification 3.5.1 Affinity chromatography Recombinant His-tag proteins expressed in E. coli were purified with the Ni-NTA Purification System (NeoBiotech) following manufacturer’s instructions. In brief, clarified supernatants obtained after sonication were supplemented with imidazole to a final concentration of 5 mM and loaded onto Ni-NTA agarose columns. After washing with 50 mM Tris-HCl, pH 8.0, 500 mM NaCl, 25 mM imidazole, bound recombinant proteins were eluted with an imidazole gradient (25-250 mM) in 50 mM Tris-HCl pH 8.0 buffer containing 500 mM NaCl. Next, purified proteins were desalted using PD-10 columns (GE Healthcare) pre- 62 equilibrated with 50 mM Tris-HCl pH 8.0, 150 mM NaCl. When needed, 30Kmolecular-weight cutoff Amicon Ultra centrifugal filter units (Millipore) were used to concentrate the protein. Protein stocks were supplemented with 20% glycerol and stored at -70 ºC. 3.6 Polyclonal antibodies production Polyclonal antibodies against PGM and PMM/PGM recombinant proteins were generated in rabbits following standard immunization protocols in the Centro de Experimentación Animal Oscar Pintado (CITIUS). Validation of the polyclonal antibody was performed with pre-immunization serum as negative control and immunoblotting Synechocystis protein extracts with different amounts of antibody. 4 Biophysical methods 4.1 Oxygen evolution 4.1.1 Clark-type oxygen electrode A Clark-type oxygen electrode (Hansatech) was used to measure oxygen evolution in 10 min dark-adapted cultures in a 30 ºC chamber during 10 min at 50 µmol photon m-2 s-1 with 5 min dark periods before and after the illumination. To prevent carbon limitation, cultures were supplemented with 10mM of NaHCO3 just before measurements. 4.1.2 MIMS Gas fluxes in real-time were measured from intact cells at a concentration of 10 μg chl ml-1 under 50 μmol photons m−2 s−1 of white actinic light using Membrane Inlet Mass Spectrometry (MIMS) as described in (Solymosi et al., 2020). Before conducting the measurements, cultures were dark-adapted for 20 min, while purged with N2 and supplemented with 18O2 at an equivalent concentration to 16O2 enabling the differentiation of O2 uptake from O2 evolution. Additionally, 1.5 mM NaHCO3 was added. Gas exchange rates were calculated according to (Beckmann et al., 2009). After 15 min of recording, 0.5 mM 2,6-dichloro-1,4- 63 benzoquinone (DCBQ) in Dimethyl sulfoxide (DMSO) was introduced into the sample chamber. 4.2 Fluorescence measurements 4.2.1 Chlorophyll fluorescence analysis and Y(II) calculations Chlorophyll a fluorescence was measured by pulse-amplitude-modulation fluorometry with a Dual-PAM-100 (Walz) using intact cells at room temperature. Before recording, 2 ml culture was adapted in the dark for 10 min. A 250 ms saturation pulse (5000 µmol photons m-2 s-1) was set for determination of effective quantum yield of PSII [Y(II)]. Y(II) was determined with the formula (Fm'-Fs)/(Fm') where Fm’ is maximal fluorescence and Fs is basal fluorescence, both measured during exposure to actinic light at the same intensity used to culture the strains (40, 50, 100 or 200 µmol photons m-2 s-1). Other parameters analyzed were basal fluorescence in the dark (Fo) and maximum fluorescence measured in the presence of 20 µM of DCMU (Fm). These parameters were obtained from the recording of an induction curve, exposing the cultures to actinic light followed by a post-illumination recovery and multiple saturation pulses. 4.2.2 NAD(P)H fluorescence measurements Light-induced NADPH redox kinetics were measured on 7.5 µg chl ml-1 samples at 30 °C with by monitoring the changes in blue-green fluorescence (excitation at 365 nm and detection between 420 and 580 nm) using the NADPH/9-AA module of a DUAL-PAM (Walz, Germany) (Nikkanen et al., 2020; Mallen-Ponce et al., 2021). 4.2.3 Determination of P700 redox kinetics. P700 redox kinetics were measured by pulse-amplitude-modulation fluorometry with a Dual-PAM-100 (Walz) using intact cells at a concentration of 12.5 μg chl ml-1 at room temperature similarly to (Shimakawa et al., 2018) and based on the method of (Klughammer & Schreiber, 2008). Before recording, cells were adapted in the dark for 10 min. During an induction curve with actinic light at 80 64 µmol photons m-2 s-1, 300 ms saturation pulses (SP) at 5000 µmol photons m-2 s1 with peak emission at 635 nm were supplied at the indicated times to calculate P700 parameters: the PSI quantum yield of photochemical energy conversion [Y(I)= (Pm´-P)/Pm] and the non-photochemical energy dissipation due to donorside [Y(ND)= P/Pm] and acceptor-side [Y(NA)= (Pm -Pm´)/Pm] limitations. 4.2.4 Reoxidation kinetics of QA The relaxation kinetics of single-turnover flash-induced Chl fluorescence were monitored using a fluorometer (FL 3500; PSI Instruments) following the methodology described in (Solymosi et al., 2020). Cells were adjusted to a concentration of 7.5 μg chl ml-1 and dark adapted for 10 min before measurements. Measurements were performed in the absence or in the presence of 10 µM DCMU. Curves were baselined to the fluorescence prior to the flash and normalized to the maximum fluorescence after the flash. The fluorescence decay traces were fitted to a three-component exponential decay function (y=A1(-x/t1)+A2(-x/t2)+A3(-x/t3)+y0) using the OriginPro 2023b software for quantification of the fast, middle and slow components of QArelaxation kinetics (Vass et al., 1999). Figure 18: Example of a P700 redox kinetic. (a) A representative induction curve (IC) in a Dual-PAM-100 for the simultaneous measurements of Chl fluorescence and the oxidation–reduction state of P700. The induction curve starts with a SP in darkness (SP1), situation where P700 is reduced. That is followed by the obtention of maximum photooxidation (Pm), achieved by a SP under far-red illumination (SP2) which preferentially excites PSI. This value is used in combination with the P700 oxidation level under actinic light (P) and maximum P700 oxidation level (Pm´) under a SP during actinic light (SP3) to calculate P700 parameters. SP1-3 are the names designated to the saturation pulses analyzed in Figure 69 c and Figure 70b . (b) Detail of a saturation pulse during actinic light illumination (SP3) and description of P700 values, the PSI quantum yield Y(I), and the non-photochemical energy dissipation due to donor-side Y(ND) and acceptor-side Y(NA) limitations. The red arrow denotes the onset of the 300 ms saturation pulse at 5000 µmol photons m-2 s-1. 65 4.2.5 PC, P700 and Fd redox changes Redox changes from PC, P700 and Fd of cells adjusted to a concentration of 20 μg chl ml-1 were obtained by NIR absorption difference spectrometry using a DUAL-KLAS-NIR spectrophotometer (Walz) and a NIRMAX script (modified using specifically determined differential model spectra for cyanobacteria) to obtain maximal values of Fd reduction and PC and P700 oxidation (Schreiber & Klughammer, 2016). The model spectra for deconvolution of Synechocystis PC, P700, and Fd signals were measured as described in (Nikkanen et al., 2020). The script began with a period of darkness followed by 4 seconds exposure to red light, which preferentially excites PSII, with a multiple turnover flash (MT) for 200 ms within this light to fully reduce ferredoxin. Following a dark transition, a subsequent exposure to far-red light (lasting 10-20 seconds), which preferentially excites PSI, with a final MT pulse to achieve maximum P700 and PC oxidations. 4.3 Reactive oxygen species measurements ROS levels were measured following the protocol outlined in (Lee et al., 2018) with some modifications. Each sample of 1 OD750 nm was centrifuged at 10000 g for 5 min, and the supernatant was carefully removed. The resulting pellets were resuspended in 0.8 ml of 1x phosphate buffered saline (PBS, 0.137 M NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4) containing 50 µM of the fluorescent probe 2′,7′-dichlorofluorescein diacetate (DCFH-DA, D6883 Sigma-Aldrich). Samples were incubated for 1 h in darkness at 30 ºC in constant agitation in a wheel. To eliminate free DCFH-DA, the samples were centrifuged at 10000 g for 5 min and washed once with PBS. Finally, the cells were resuspended in 200 µl of PBS and transferred to a fluorescent 96-well microplate. Fluorescence of dichlorofluorescein (DCF), the oxidized form of DCFH, was measured at 525 nm after excitation at 488 nm in a Varioskan multiplate reader (Thermo Fisher Scientific). 72 electrode) using a CarboPac PA10 column in isocratic mode (18 mM NaOH as mobile phase 1 ml min−1). For quantification, standard/calibration curves of with known glucose and mannose concentrations were prepared. This method was performed by Dr. M. Teresa Ruiz (Plant Development Unit, IBVF). 6.5 Polysaccharide size characterization. For polysaccharide purification, 50 ml cultures were harvested and pellets were processed as described in Section 5.1 Glycogen quantification, scaling up volumes accordingly. Dried pellets were resuspended in distilled H2O, diluted to 2.5 mg ml-1, adjusted to 5mM NaOH and filtered (0.22 µm sterile syringe filters) prior liquid chromatography analysis. Size-exclusion chromatography was performed using a HiPrep 16/60 Sephacryl S-500 HR column (Cytiva, cat number 28935606) on an NGC Medium-Pressure Chromatography System (FPLC, BioRad). The column was operated in isocratic mode with 5mM NaOH as mobile phase (previously filtered in a vacuum system with a 0.45 µm pore nitrocellulose membrane) with 0.5 ml min-1 flow. Fractions (1.5 ml) were collected automatically. Total carbohydrate content in each fraction was determined using an anthronebased colorimetric assay. Briefly, 50 µl of each fraction was mixed with 100 µl of anthrone reagent (0.1 g anthrone per 50 ml H2SO4) in a multiwell plate. The plate was incubated in the dark at 85°C, and absorbance was measured at 620 nm using a Varioskan multiplate reader (Thermo Fisher Scientific). A standard curve was generated using glycogen (2-25 µg per well, bovine liver, Sigma). In addition, bovine liver glycogen and glycogen from E. coli were used in the analysis as control for chromatographic size comparison. To enhance glycogen accumulation in E. coli cells, precultures were supplemented with 20 mM glucose and harvested 24 hours later. 7 Other methods 7.1 pH measurements The pH of the different solution was measured and adjusted using a sensION™+ pH3 (Hach) pHmeter following manufacturer instructions. 73 7.2 Chlorophyll concentration determination The chlorophyll concentration was determined spectrophotometrically following the method described previously (Mackinney, 1941). Briefly, 1 ml of cells was collected and centrifuged at 12,000 g for 1 minute, and 900 μl of the supernatant was removed. Samples were frozen at -20 ºC until further use. The pellet was resuspended in the remaining supernatant, and 900 μl of absolute methanol was added. This mixture was vortexed until the complete mixture and centrifuged again at 12000 g for 1 minute. The chlorophyll concentration in the supernatant was measured at 665 nm, using an extinction coefficient of 74.46 mM-1 cm-1. 7.3 Cell density spectrophotometric measurements Visible light measurements were performed in a UV-Vis GENESYS™180 spectrophotometer (ThermoFisher) using cell suspensions in 10 mm polyestirene cuvettes (Kartell™ 0193800). 7.3.1 Whole cell spectra Absorbance of cultures adjusted to 1 OD750 was recorded using visible light ranging from 400-800 nm every 0.5 nm. 7.3.2 Graphic representation and statistical analysis. In general, visualization of continuous data with the associated standard error of the mean was performed using dplyr and ggplot2 (Tidyverse packages) in R. Figures containing discrete data and statistical analysis were obtained using the software GraphPad Prims 8.0.1 (244). 74 75 OBJECTIVES The general objective of this thesis is to elucidate the impact of glycogen synthesis on cyanobacterial response and adaptation to environmental changes, particularly under conditions of glucose availability or nitrogen deficiency, and its connection to central metabolism. To achieve this, we have established the following specific objectives: • Analysis of the role of phosphoglucomutases in glycogen and central carbon metabolism. • Determination of the role of glycogen synthesis in the transition to mixotrophy. • Study of the link between glycogen synthesis impairments and bleaching process blockage. • Characterization of a non-canonical glycogen synthesis pathway in cyanobacteria. Part of the results of the thesis have already been published: - Ortega-Martínez P, Roldán M, Díaz-Troya S, Florencio FJ. Stress response requires an efficient connection between glycogen and central carbon metabolism by phosphoglucomutases in cyanobacteria (2023). Journal of Experimental Botany 74: 1532–1550. (doi: 10.1093/jxb/erac474). - Ortega-Martínez P, Nikkanen L, Wey LT, Florencio FJ, Allahverdiyeva Y, DíazTroya S. Glycogen synthesis prevents metabolic imbalance and disruption of photosynthetic electron transport from photosystem II during transition to photomixotrophy in Synechocystis sp. PCC 6803. (2024) New Phytologist 243(1):162-179. (doi: 10.1111/nph.19793). 76 77 RESULTS As stated in the Introduction, the glycogen metabolic pathway is highly conserved and requires only a few steps from G6P to incorporate glucose into polysaccharide chains. As presented in the Objectives section, the aim of this study is to investigate the role of enzymes involved in the dynamics of this polysaccharide, and how cells utilize this carbon and energy storage to adapt to environmental changes and enhance the resilience of photosynthetic organisms. This will expand our understanding of how to regulate carbon flow towards desired compounds by diverting it from glycogen reserves. It is specially focused on the contribution of the phosphohexomutases to glycogen metabolism and taking into account the substantial work characterizing the phenotype of a glycogen-less strain by the disruption of the ADP-glucose pyrophosphorylase gene. In addition, the importance of glycogen in the transitions between phototrophic to photomixotrophic modes is addressed. On top of this, the metabolic and photosynthetic regulation in the adaptation to nitrogen depleted media was also analyzed. Finally, in this work, the functional diversity of the glycogen synthesis pathway is also explored opening the possibility to alternative polysaccharide formation without the canonical glycogen synthases. 1 Analysis of the connection between glycogen and central carbon metabolism by phosphoglucomutases in cyanobacteria. Substantial research has been conducted on glycogen synthesis using ∆AGP mutants due to the glycogen-less phenotype of this strain. However, the impact of limited interconversion between G1P and G6P on cell fitness, as well as the contribution of Synechocystis’ phosphohexomutases, PGM and PMM/PGM, in the connection between glycogen and central carbon metabolism have not been thoroughly explored. 78 1.1 Homology analysis and abundance of the phosphoglucomutases in Synechocystis. The Synechocystis PGM and PGM/PMM proteins have similar sizes, with lengths of 567 and 499 amino acids, respectively. Although certain regions, such as the catalytic motive, the magnesium/bivalent cation domain and the sugar-phosphate recognition domain, are highly conserved across kingdoms Figure 19 (Whitehouse et al., 1998), the global sequence protein alignment analysis of Synechocystis PGM and PGM/PMM proteins using the Basic Local Alignment Search Tool (BLASTp; NCBI) reflected little similarity between these two enzymes (Figure 19b). The overall percentage of matches for the given sequences was 21.4% with a similarity of 35.2%. However, the alignment of the PGM from Synechocystis and E. coli (P36938; UNIPROT), showed better coverage with higher identity (59%) and similarity (72%). Although the protein structure of the phosphohexomutases from Synechocystis has not been experimentally determined, a prediction of the three-dimensional structures of PGM and PGM/PMM was performed using the AlphaFold tool (Figure 19c). Despite the differences in their sequences, both proteins exhibited the characteristic ‘heart-shape’ of the phosphohexomutases superfamily, with the conserved domains located in close proximity, showing remarkable coincidence in the predicted superposition. Using specific antibodies raised against these two proteins, we were able to detect both by western blotting. Under our standard growth conditions (50 µE m2 s-1 constant illumination and bubbled with air supplemented with 1% CO2), the quantification of proteins in cell extracts through immunodetection, revealed that PGM was approximately 10-fold more abundant than PMM/PGM (0.75±0.03 ng PGM and 0.07±0.02 ng PMM/PGM per µg soluble cell extract, Figure 19 and Figure 20d). This result is in agreement with the quantitative analysis of the proteome of Synechocystis where PGM is reported to be up to 5 times more abundant than the PMM/PGM (Zavřel et al., 2019). 79 ES WT (µg) PMM (ng) 7,5102030 15 0,1250,510,252 7,5 152030 3,25 ES WT (µg) 2,55 7,5 10 PGM (ng) α-PGM α-PMM/PGM PGM PMM/PGM 168 191 353 376 424 447 (c)(a) (b) (d) Figure 19 Homology analysis and abundance of the phosphoglucomutases in Synechocystis. (a) Detail of conserved regions of MUSCLE aligned sequences of 5 phosphoglucomutases represented using WebLogo3: Rabbit muscle PGM (PDB: 3PMG);· E. coli PGM (Uniprot: P36938), P. aeruginosa PMM/PGM (PDB: 1P5G) and Synechocystis PGM and PGM/PMM. (b) Complete alignment analysis of PGM and PGM/PMM protein sequences from Synechocystis using the Basic Local Alignment Search Tool (BLASTp; NCBI). (c) Structural prediction of the PGM and PGM/PMM protein sequences using the AlphaFold tool. Visualization and overlapping were performed the using UCSF ChimeraX software (PGM displayed in brown and PMM/PGM in blue). In a-c, conserved regions are highlighted with colors: phosphoserine catalytic motive (red), the magnesium/bivalent cation domain (yellow) and the sugar-phosphate recognition domain (green), regulatory serine (blue, PGM sequence), other conserved regions (grey). (d) Detection of PGM and PGM/PMM protein in WT soluble protein extract and their quantification trough western blotting using specific polyclonal antibodies and comparison with known amounts of recombinant PGM and PGM/PMM. Pounceau straining of the membranes probed is shown below each blot. 80 1.2 Generation of a ∆PGM mutant strain To investigate the significance of this important metabolic valve, we aimed to generate and characterize knock-out mutant strains of the two enzymes with assigned phosphoglucomutase activity in Synechocystis. Our goal was to gain a better understanding of the proteins' roles in glycogen metabolism. The deletion of sll0726, which codes for PGM, was easily segregated, generating the ∆PGM strain (Figure 20a, and c). Consistent with previously described results (Liu et al., 2013), a mutant lacking the bifunctional enzyme PMM/PGM could not be segregated (∆PMM* strain, Figure 20b and d) suggesting that PMM/PGM plays an essential role in Synechocystis. ∆PGM strain exhibited no detectable PGM signal by Western blot analysis (Figure 20e). Notably, in the ∆PGM strain there were no compensations by changing the levels of the other phosphoglucomutase, PMM/PGM, nor did it affect the abundance of key enzymes involved in glycogen synthesis (AGP, GlgA1, GlgA2), or CBB cycle enzymes such as the FBP (Fructose-1,6-bisphosphatase) or the SBP (Sedoheptulose-1,7-bisphosphatase). Interestingly, PGM mutants generated in other laboratories have been described to accumulate moderate (Liu et al., 2013) or no (Doello et al., 2022) amounts of glycogen. We detected moderate amounts Figure 20: Generation and segregation of the ∆PGM and ∆PMM* mutant strains. (a) Scheme of the strategy for the deletion of sll0746 (coding for PGM) to generate the ∆PGM strain. (b) Scheme of the strategy for the deletion of slr1334 (coding for PMM/PGM) to generate the ∆PMM* strain. (c) PCR with primers depicted in (a) to analyze the grade of segregation of the ∆PGM strain. (d) PCR with primers depicted in (b) to analyze the grade of segregation of the ∆PMM* strain. Primers in lane 1: OL80/OL184; primers in lane 2: OL72/OL_EryR. ( e) Western blot analysis of the WT, ∆AGP and ∆PGM strains probed against glycogen synthesis enzymes (PGM, PMM/PGM, AGP, GlgA1, GlgA2), FBP (Fructose-1,6-bisphosphatase), SBP (Sedoheptulose-1,7bisphosphatase) and the GS-I (Glutamine synthetase I, used as loading control). 81 of glycogen in our PGM mutant in every growth condition tested, including varying concentrations of NaHCO3 in non-bubbled flask cultures (Figure 21) or under different stress conditions in cultures bubbled with CO2 supplemented air (Figure 22, Figure 23, Figure 24 and Figure 29). The glycogen levels in the ∆PGM mutant ranged between 10-50% of those found in the WT. This indicates that, in addition to PGM, there is another protein in Synechocystis with phosphoglucomutase activity, probably PMM/PGM, capable of providing G1P for glycogen synthesis, albeit less efficiently than PGM. Glycogen has been described to serve as a carbon and energy buffer and plays a crucial role in the plasticity of cell metabolism (Cano et al., 2018; Luan et al., 2019; Makowka et al., 2020). For this reason, in this work, the phenotype of the ∆PGM strain was characterized to analyze the effect of a limited carbon flux to and from glycogen under different growth conditions such as high light, dark-night cycles, nitrogen deprivation and salt acclimation. 1.3 Effect of high light exposure Under high light growth conditions, lack of glycogen synthesis negatively affects cell performance and causes organic acids overflow to dissipate excess energy Figure 21: Glycogen accumulation in the ∆PGM strain cultivated in media with different availabilities of HCO3. WT, ∆AGP and ∆PGM cells were inoculated at an OD750 of 0.5 in Erlenmeyer flasks with BG11 containing 12 mM, 5 mM o no HCO3 and glycogen content was measured after 7 days. Data are means ± SEM from three biological replicates. Significant differences were determined using unpaired two-tailed Student’s ttest: ∗P≤0.05, ∗∗P≤0.01, ∗∗∗P≤0.001. 88 (Figure 25). Y(II) values of the ∆AGP and ∆PGM strains in nitrogen-repleted medium were equivalent to that of the WT (Figure 24g). However, nitrogen deprivation induced a faster Y(II) decrease in the ∆AGP and ∆PGM strains than in the WT, with Y(II) values close to zero at 48 hours (Figure 24g). Furthermore, in contrast to the WT, Fo increased in the ∆PGM strain (Figure 25), suggesting an increase in free phycobilisomes (Acuña et al., 2016; Stirbet et al., 2018). 1.5.1 Metabolomic response to nitrogen depletion Glycogen synthesis is required for a transition to the dormant state when nitrogen is not available. However, the reasons why glycogen mutants cannot complete chlorosis upon nitrogen deprivation are not clear. They present an overflow of metabolites and photosynthetic switch off despite preserving the phycobiliproteins used to harvest photons (Figure 24). Studies analyzing the transcriptomic and proteomic response of the mutant strains lacking AGP conclude that the process is interrupted, and no major transcriptional or translational changes occur under nitrogen depletion (Carrieri et al., 2017). Figure 25: Fo and Fm photosynthetic parameters analysis of the ∆PGM strain during nitrogen deprivation. Maximum fluorescence yield ( Fm) and minimal fluorescence (Fo ) of WT and ∆PGM cultures during nitrogen deprivation and after nitrogen replenishment measured by PAM fluorometry. F m was determined in the presence of DCMU (20 µM) under growth light (50 µE m−2 s−1). Green to orange and orange to green bars represent cultivation in media without or with nitrogen, respectively. Data are means ±SEM from three biological replicates. Significant differences in Fo and Fm values of the WT and ∆PGM strains at time 144 hours after nitrogen removal compared with their respective values at time 0 hours were determined using paired two-tailed Student’s t-test. ∗ P<0.05, ∗∗ P<0.01, ∗∗∗P<0.001. 89 Synechococcus ∆glgC/AGP strain presents a retardation in phycobiliprotein degradation, although the sensing of nitrogen depletion and the transcription of nblA occurs, with several alterations in the metabolic profile compared to the WT (Hickman et al., 2013). The overflow metabolism and non-bleaching phenotype must be caused and regulated at the metabolite level, especially since most of the accumulated glycogen in the WT comes from the recycled phycobiliproteins (Hasunuma et al., 2013). To shed light on the halted bleaching presented by the glycogen mutant strains, we explored their metabolic profile and carbon partitioning in response to nitrogen depletion in comparison with the WT (Figure 26). Cultures were harvested before and at 6 and 24 hours after nitrogen removal and polar metabolites were quantified. In the WT strain, carbon intermediates remained at levels similar to the initial values, or decreased by half within 24 hours, with the exception of the slight increases in 2-OG, fumarate and malate pools. However, substantial changes were observed in the levels of most of the amino acids measured. Two distinct patterns of changes emerged: first, amino acids such as tyrosine, phenylalanine, serine, leucine/isoleucine, threonine or alanine exhibited transient fluctuations. Their concentration increased at 6 hours post-nitrogen depletion before returning to nitrogen repleted levels (tyrosine, phenylalanine, leucine/isoleucine) or becoming nearly undetectable (threonine, serine and alanine) (Figure 26). The second pattern involved amino acids synthetized from TCA intermediates, such as glutamate, glutamine, asparagine, and methionine, which exhibited a consistent decrease in their concentrations over the sampling period. Valine was the only amino acid that maintained a similar concentration before and 24 hours after nitrogen depletion (Figure 26). These modifications in the amino acid pools suggest that WT is recycling its phycobiliproteins and channeling amino acids as adaptative response without altering carbon intermediates, in agreement with (Hasunuma et al., 2013; Osanai et al., 2014). 90 Figure 26: Metabolite profile of WT, ∆AGP and ∆PGM strains in response to nitrogen deficiency. Cultures of WT, ∆PGM and ∆AGP strains were harvested and resuspended in BG110C twice before being adjusted to a final optical density (OD750 nm) of 1. Amounts of polar metabolites of strains before and after 6and 24-hours of nitrogen removal were measured by LC/MS. Metabolite names highlighted with colored rectangles refer to molecules shared with different pathways. Green background is used to highlight amino acids Abbreviations: Ru5P/Xu-5P (ribulose-5P/xylulose-5P pool), GAP (glyceraldehyde-3P), DHAP (dihydroxyacetone-P), 3PG (3Pglycerate), PEP (phosphoenolpyruvate). Values represent the mean±SEM of 3 independent biological replicates. 91 Figure 26 (Continued): Abbreviations: 3PG (3P-glycerate), PEP (phosphoenolpyruvate), AcCoA (Acetyl-CoA), 2-OG (2-oxoglutarate), SSAL (succinic semialdehyde), OAA (oxalacetate). 92 In contrast, the ∆AGP and ∆PGM strains diverged from the trend of the WT strain, although they exhibited similar behavior between them (Figure 26). Rather than showing gradual reductions or temporary accumulations in certain amino acids, mutant strains experienced rapid, and in some cases, complete depletion of amino acids such as tyrosine, phenylalanine, isoleucine-leucine, valine, asparagine, glutamate and notably methionine. Only alanine and serine, synthetized from pyruvate and 3P-glycerate, respectively, exhibited progressive accumulation during the sampling period. Furthermore, unlike the WT strain, these mutant strains displayed gradual increments in most of the analyzed carbon intermediates, including glucose-6P, glucose-1P, fructose-6P, sedoheptulose7P, erythrose-4P, GAP, DHAP, pyruvate, 2-oxoglutarate, fumarate, malate and notably a high increase in the fructose-1,6P within six hours while the WT depleted this pool. The metabolites AcCoA, isocitrate, succinate, 3PG, and the pentoses phosphate remained at initial levels. Notably, only PEP exhibited a reduction in concentration (Figure 26). Summarizing, upon transferring cultures to limited nitrogen media, WT cells began adapting, and metabolic profile changes were detected after 6 hours of nitrogen depletion, particularly in amino acid pools as reported in previous studies (Hasunuma et al., 2013; Osanai et al., 2014). On the contrary, glycogen mutant strains, despite sensing the nitrogen depletion (Hickman et al., 2013) failed to adapt and exhibited central carbon intermediates accumulations, in addition to serine and alanine, with depletion in almost all other measured amino acids. 93 1.5.2 Recovery from nitrogen deprivation After entering the dormancy state induced by nitrogen deprivation, Synechocystis cells are able to recover vegetative growth once nitrogen is available again. The recovery requires mobilization of glycogen to provide energy to restore the translational machinery, ATP synthesis, and nitrate assimilation, and finally for the synthesis of the photosynthetic apparatus (Neumann et al., 2021). However, mutants unable to accumulate glycogen not only have a non-bleaching phenotype, but also exhibit decreased viability when faced with nitrogen deprivation (Gründel et al., 2012; Doello et al., 2022). Under our growth conditions, after 15 days of nitrogen starvation ∆AGP and ∆PGM cultures had turned white, lost photosynthetic pigments and the viability of the cultures was severely compromised (Figure 27). Figure 27: Appearance and recovery of the WT, ∆AGP and ∆PGM strains after 15 days of nitrogen deprivation. WT, ∆AGP, and ∆PGM cultures were grown to mid-exponential phase in BG11C, washed in nitrogen-free medium (BG110C) and transferred to BG110C for 15 days. (a) Photographs of the WT, ∆AGP, and ∆PGM cultures immediately after and 15 days after nitrogen removal. (b) Whole cell spectra normalized to absorbance at 750 nm of WT, ∆AGP, and ∆PGM cultures after cultivation in BG110C medium for 15 days. (c) Viability assay of the WT, ∆AGP, and ∆PGM strains. Aliquots of the WT, ∆AGP, and ∆PGM cultures immediately after and 15 days after nitrogen removal were spotted on BG11C plates and cultivated under continuous light for 6 days. Dilutions are indicated above each panel. 94 To investigate whether mobilization of the limited glycogen accumulated in the ∆PGM strain confers an advantage over the glycogen-less ∆AGP strain, we analyzed their recovery after the addition of nitrate as a nitrogen source (Figure 28). To avoid compromising culture viability we limited the length of the nitrogen starvation period to 6 days. After this time, the WT had fully degraded their phycobiliproteins while ∆PGM and ∆AGP maintained their blue-green color (Figure 28a and b) and no net oxygen evolution was detected in the WT or the mutant strains (Figure 24h and f). Under these conditions, transfer of the cells to nitrogen-repleted medium induced a fast mobilization of the glycogen in the WT cells, which recovered their blue-green pigmentation and resumed growth within 24 hours (Figure 28a-d). In contrast, nitrogen addition did not prevent Figure 28: Growth, glycogen content and photosynthetic characterization of the strain ∆PGM after nitrogen replenishment. Cells of the WT, ∆AGP, and ∆PGM strains cultured in nitrogen-free medium for 6 days were collected and transferred to nitrogen-repleted medium at an optical density at 750 nm of 1. (a) Photographs of the WT, ∆AGP, and ∆PGM cultures at different time points after nitrogen replenishment. (b) Whole cell spectra of WT, ∆AGP and ∆PGM cultures over 36 hours after nitrogen replenishment. Spectra were normalized to absorbance at 750 nm. (c) Growth curves of the WT, ∆AGP, and ∆PGM cultures after nitrogen replenishment. (d) Evolution of the glycogen content of the WT, ∆AGP and ∆PGM cultures over the course of three days after nitrogen replenishment. (e) PSII quantum yields [Y(II)] in WT and ∆PGM cultures measured by PAM fluorometry at the growth light intensity (50 µE m-2 s-1) at different times after nitrogen replenishment. (f) Oxygen evolution of the WT and ∆PGM cultures at different times after nitrogen replenishment measured by Clark-type electrode at the growth light intensity (50 µE m-2 s-1). Data are means ± SEM from four biological replicates. 95 whitening of the ∆AGP cells, which were unable to recover after 6 days of nitrogen deprivation. After addition of nitrate, the ∆PGM strain mobilized its glycogen reserves during the first 24 h, although to a lesser extent than the WT, both in total amounts (98 vs 19 µg OD750-1 for the WT and ∆PGM strains, respectively) and as a percentage of the initial glycogen amount (70% and 51 % for the WT and ∆PGM strains, respectively) (Figure 28d). However, this was enough to resume its growth at a similar rate to that of the WT (Figure 28c). Differences were also found in the photosynthetic performance of the WT and ∆PGM strains after nitrate addition, with a faster recovery of Y(II) and O2 evolution in the WT than in the ∆PGM strain (Figure 28e-f and Figure 25), although the ∆PGM cells conserved their photosynthetic pigments (Figure 28b). 1.6 Resistance to salt stress. Salt tolerance is one of the characteristics that could be affected by a limitation in the G6P-G1P interconversion, as G1P is required for the synthesis of osmolytes in Synechocystis (Kirsch et al., 2019). The synthesis and accumulation of the osmolytes sucrose and, in particular, glucosylglycerol (GG) allows Synechocystis cultures to tolerate moderate salinities (up to 1 M NaCl) (Kirsch et al., 2019). Sucrose is synthesized from fructose-6-phosphate and UDP-glucose, and GG from glycerol-3-phosphate and ADP-Glc (Kirsch et al., 2019) (Figure 29). The supply of both UDP-glucose and ADP-Glc is based on G1P and, thus, on phosphoglucomutase activity. In this regard, mutants lacking AGP present a saltsensitive phenotype due to their inability to synthesize ADP-Glc and therefore GG (Miao et al., 2003). To analyze the influence of limitation on G6P-G1P interconversion on salt tolerance, the WT, ∆AGP, and ∆PGM strains were cultured in NaCl-free medium for 24 hours and then NaCl was added to reach a final concentration of 250 or 500 mM (Figure 29). In the WT, this induced the accumulation of GG up to 37 nmol OD750-1 and 73 nmol OD750-1 for 250 and 500 mM NaCl, respectively, and a fast and transient accumulation of sucrose (Figure 29a-c). As expected, the ∆AGP strain did not accumulate GG. However, the accumulation of sucrose allowed the culture to grow similarly to the WT at 250 mM NaCl but not at 500 mM NaCl (Figure 29a-c). In the case of the ∆PGM strain, although the maximum accumulation of sucrose was lower than in the WT, GG 96 reached amounts similar to those of the WT, but with a slightly delayed kinetic at 500 mM NaCl. However, these alterations in the accumulation of osmolytes did not affect the growth of the ∆PGM strain at 250 or 500 mM NaCl (Figure 29a). Interestingly, the addition of NaCl induced the transient mobilization of glycogen in both the WT and ∆PGM strains (Figure 29d). Glycogen accumulation resumed in the WT 24-48 hours after salt addition, when the amount of GG stabilized at its maximum level in both strains (Figure 29b). To a lesser extent, the ∆PGM strains presented the same pattern of glycogen accumulation. Figure 29: Effect of NaCl addition on growth and osmolytes and glycogen content of the WT, ∆AGP and ∆PGM strains. WT, ∆AGP and ∆PGM cells were cultured in standard NaCl-free medium for 24 hours and then 250 or 500 mM NaCl was added. (a) Growth curves of the WT, ∆AGP, and ∆PGM cultures. Intracellular content of (b) glucosylglycerol and (c) sucrose of the WT, ∆AGP, and ∆PGM cultures along the experiment. (d) Glycogen content of the WT and ∆PGM cultures before and after the addition of 500 mM NaCl. Vertical arrows indicate NaCl addition. Data are means ±SEM from three biological replicates. 97 1.7 The overexpression of PMM/PGM compensates for the lack of PGM. As indicated above, PMM/PGM is ten times less abundant in the cell than PGM (Figure 19d). Furthermore, in vitro characterization of recombinant enzymes indicates that the phosphoglucomutase activity of PMM/PGM is also ten times lower than that of PGM (Liu et al., 2013). The evidence suggests that the relatively low phosphoglucomutase activity of PMM/PGM is sufficient to support the limited glycogen synthesis observed in the ∆PGM strain (Figure 21, Figure 22 and Figure 24). Consequently, we hypothesized that increased expression of PMM/PGM might compensate for the absence of PGM. To test this, we slr1334 570 bp 1338 bp nrsD slr0796 1473 bp nrsD UP p cpcB nrsD DOWN SpSt 1800 bp ≈650 bp ≈650 bp (a) (b) (c) (d) SE Figure 30: PMM/PGM overexpression in the WT, ∆PGM, OE:M, and ∆PGM_OE:M strains. (a) Schematic representation of the construct used to overexpress PMM/PGM ( slr1334 ). The expression of the slr1334 ORF is under the control of the promoter cpcB, and the construct integrated in the nrsD locus. (b) Analysis of PMM/PGM overexpression. Soluble protein extracts from the WT, ∆PGM , OE:M, and ∆PGM_OE:M cultures grown 24 hours under standard growth conditions were analyzed by Western blot with antibodies against PGM and PMM/PGM and GroEL, used as loading control. Ten µg protein were loaded in each lane. (c) Phosphoglucomutase activity measured in soluble protein extracts from the WT, ∆PGM , OE:M, and ∆PGM_OE:M cultures grown 24 hours under either standard growth conditions, 200 µE m-2 s-1 light intensity, nitrogen deprivation, or 500 mM NaCl in the media. Activity values were referenced to those of the WT strain grown under standard conditions (WT control). (d) Confirmation of the significant difference between low phosphoglucomutase activity of ∆PGM strain and background noise. Raw data from phosphoglucomutase activity assays with cell soluble extract (SE) from ∆PGM strain with and without the addition of the substrate G1P. Data in (c) are means ±SEM from three biological replicates. Significant differences compared with the WT strain at the same condition were determined using unpaired two-tailed Student’s t -test. In (d), significant difference was determined using paired two-tailed Student’s t-test. ns not significant, ∗P<0.05, ∗∗P<0.01, ∗∗∗P <0.001, ∗∗∗∗ P<0.0001. 104 and growth was followed (Figure 36a). In the presence of arsenite, the ∆PMM_Pars:M cultures showed dose dependent levels of PMM/PGM and growth rates similar to those of the WT (Figure 36a-c). However, in the absence of arsenite, were PMM/PGM remained absent, culture growth ceased and eventually died (Figure 36a and b). The observation of the cells by fluorescence microscopy revealed that 24 hours after the onset of the experiment, half of the population of the ∆PMM_Pars:M strain cultured without arsenite had lost chlorophyll fluorescence while the remaining population presented a fluorescence burst, typical of chlorophyll detachment and degradation (Figure 36d). These data confirmed that PMM/PGM is an essential protein in Synechocystis. Figure 35: Generation of the ∆PMM_Pars:M strain. (a) Scheme of the strategy for the deletion of slr1334 gene and insertion of the regulated copy of slr1334 in the nrsD locus. Primers used to analyze the segregation of mutants are included. (b) PCR with the primers shown in A to analyze the grade of segregation of the generated mutant strain. 105 To further test if the limited carbon flux to and from glycogen in the absence of PGM was provided by PMM/PGM, we designed a new mutant. The sll0726 gene (coding for PGM) was deleted in the ∆PMM_Pars:M strain, generating the ∆PGM∆PMM_Pars:M strain. This strain lacks PGM and has regulated expression of PMM/PGM. In the presence of the inducer arsenite, the ∆PGM∆PMM_Pars:M expressed the PMM/PGM and the culture was viable (Figure 36e). As expected, in the absence of the inducer, PMM/PGM was not expressed and the ∆PGM∆PMM_Pars:M strain was not viable (Figure 36e). WT M arsenite: PGMPMM_Pars:M PGM 0 5 0 5 0 0.5 1 2 5 GroEL PGM PMM/PGM 0 120 Time (hours) Arsenite 0 µM 10 µM 20 µM 20 µM 20 µM 20 µM (a) (b) (d) (e) (c) Figure 36: Characterization of the strains with regulated expression of PMM/PGM with different amounts of the inducer arsenite. Cultures of the WT, ∆PGM, ∆PMM_P ars:M and ∆PGM∆PMM_Pars :M strains were grown without arsenite for six days to ensure low levels of PMM/PGM in the PMM/PGM_regulated expression strains. Cultures were then used to inoculate fresh medium at an OD750 of 1 without arsenite or with increasing arsenite concentrations. Cultures were analyzed under standard growth conditions during the next five days. (a) Growth curves of the WT and ∆PMM_P ars :M cultures. (b) Photographs of the cultures 0 and 120 hours after the addition of the indicated arsenite concentrations. (c) Samples of the WT and ∆PMM_P ars :M cultures 0 and 72 hours after the addition of the different arsenite concentrations were analyzed by Western blot with specific antibodies against PGM and PMM/PGM. α GroEL was used as loading control. Ten µg of protein from soluble extracts were loaded in each lane. (d) Fluorescence microscopy images of ∆PMM_P ars :M cultured without or with 10 µM arsenite at 24 hours after the onset of the experiment. Two replicates for each condition are depicted vertically (e) Photographs of the cultures before and five days after arsenite addition (upper panel). PGM and PMM/PGM levels in WT, ∆PGM and ∆PGM∆PMM_P ars :M cells three days after arsenite addition (lower panel). Ten µg of soluble proteins were resolved by SDS-PAGE and probed with specific antibodies against PGM and PMM/PGM. Levels of GroEL was used as a control of equal loading. Data from a are means ±SEM three biological replicates. 106 Thus, ∆PMM_Pars:M and ∆PGM∆PMM_Pars:M strains allowed us to turn down the levels of PMM/PGM and evaluate its contribution to glycogen synthesis in the absence of PGM. We determined the glycogen content (Figure 37) of the strains in early and mid-exponential growth phases in the presence of increasing amounts of arsenite and hence increasing amounts of PMM/PGM in the regulated expression strains (Figure 36c and e). All strains harboring at least one endogenous copy coding for an enzyme with phosphoglucomutase activity (WT, ∆PGM and ∆PMM_Pars:M strains) increased their glycogen content independently of the arsenite concentration (Figure 37). On the contrary, the amount of glycogen had an inducer dose dependent accumulation in the ∆PGM∆PMM_Pars:M strain after five days, being almost undetectable in the absence of arsenite. When the inducer was added up to 2 µM, the amount of glycogen remained at the initial values while 5 µM of arsenite, caused glycogen accumulation (Figure 37). This strongly suggests that there is no other enzyme in addition to PGM and PMM/PGM catalyzing an efficient interconversion between G1P and G6P. Figure 37: Glycogen content of the strains with regulated expression of PMM/PGM with different amounts of arsenite/inducer. WT, ∆PGM, ∆PMM_Pars:M and ∆PGM∆PMM_Pars:M strains grown without inducer were inoculated in fresh medium at an OD750 of 1 without arsenite or with increasing arsenite concentrations. Glycogen in the cultures were measured at the time of inoculation (0 days) and after five days. Data are means ±SEM of four biological replicates. Significant differences were determined using paired two-tailed Student’s t-test. *P<0.05, **P<0.01, ***P<0.001 and “ns” non-significance. 107 1.8.1 In search of the lethality: Metabolic response to PMM/PGM depletion in the ∆PMM_Pars:M strain. The reason behind the essentiality of the PMM/PGM is not clear. Despite demonstrating that PMM/PGM possesses sufficient phosphoglucomutase activity to account for the glycogen content in the ∆PGM strain, this trait does not seem to be the reason behind the enzyme's requirement in Synechocystis, since phosphoglucomutase activity is fulfilled by PGM, and glycogen-deficient strains remain viable. Recently, PMM/PGM has been reported to present glucose-1,6bisphosphate synthase activity. G1,6BP is a key metabolite, acting as an enzymatic activator for enzymes within the phosphohexomutase superfamily serving as the phosphate donor required for the active catalytic (phospho)serine (Neumann et al., 2022a). To explore if the lack of the PMM/PGM induces metabolic alterations, potentially affecting G1,6BP levels or other metabolites, we conducted a time-course metabolic profiling of the ∆PMM_Pars:M strain. The preculture was grown in BG11C supplemented with 10 µM arsenite. Culture was centrifuged and resuspended twice in fresh BG11C for inducer removal and set to a final 0.5 OD750. Analysis of levels of PMM/PGM by western blotting and the metabolite quantification was performed before, 72and 144-hours post-inducer removal (Figure 38a). At the onset of the experiment, the amount of PMM/PGM in the ∆PMM_Pars:M strain was 4 times higher than in the WT strain. However, in the subsequent time points, PMM/PGM levels became undetectable by western blotting (Figure 38a). 108 Figure 38: Metabolite profile of ∆PMM_Pars:M after arsenite removal. (a) Western blot analysis of ∆PMM_Pars:M cultured in BG11C with 10 µM arsenite (0 hours) and after 72 and 144 hours of arsenite removal using antibodies targeting PMM/PGM. WT cultured in parallel was harvested at the initiation of the experiment as control. As loading controls, membrane was probed against GroEL and same amounts of soluble extracts were resolved electrophoretically and stained with Coomassie solution. Quantification of the PMM/PGM levels is shown relative to WT amounts, depicted as dotted line. (b) Amounts of polar metabolites measured by LC/MS of ∆PMM_Pars:M at 0, 72 and 144 hours after arsenite removal. Metabolites are grouped by families represented by colored backgrounds: amino acids in blue, shared metabolites of CCB cycle, OPP pathway and upper-glycolysis in green, down-glycolysis and trioses phosphate in white and TCA cycle in yellow. Data are means ±SEM from four biological replicates. 109 The metabolome analysis conducted at the specified time points revealed minimal changes in the majority of the analyzed metabolites (Figure 38b). The intermediate G16BP exhibited a decreasing trend throughout the time-course, although detected across all time points. Given that growth arrest was observed within the 144 hours after arsenite removal, it is unlikely that depletion of this metabolite solely contributes to the observed lethality. From the amino acids measured, most displayed a progressive reduction similar to that observed for G16BP. Exceptions included arginine, alanine, tryptophan, phenylalanine, and tyrosine, which did not exhibit notable decreases. Conversely, serine and particularly methionine demonstrated significantly reduced levels within 72 hours. Additionally, a progressive reduction in abundance was observed for 6PG, 2K3DPG, and AcCoA, while the remaining metabolites maintained consistent levels throughout the experiment. 110 2 Role of glycogen synthesis in the transition to mixotrophy. Besides photoautotrophic growth, some cyanobacteria such as Synechocystis can grow photomixotrophically by performing photosynthesis simultaneously with the utilization of organic carbon compounds from the environment, such as glucose. Despite their ecological and biotechnological significance (Muñoz-Marín et al., 2024), the understanding of the physiological mechanisms underlying the transition from photoautotrophy to photomixotrophy in cyanobacteria is scarce. In fact, the sensitivity of glycogen-less Synechocystis mutants to mixotrophic conditions (Gründel et al., 2012) suggests that the polysaccharide metabolism is crucial in the transition between trophic modes, but its role in that process has not been fully addressed. The synthesis of glycogen in Synechocystis involves not only the generation of intermediates by PGM and AGP, but also the elongation of granule chains by glycogen synthases (GlgA1 and GlgA2) and the introduction of ramifications facilitated by the branching enzyme (GlgB). Single mutant strains in GlgA1 and GlgA2 are viable and can accumulate glycogen levels similar to a WT strain (Yoo et al., 2014; Koch et al., 2019). GlgB mutant strain accumulates structurally altered and partially insoluble glycogen (Yoo et al., 2002; Welkie et al., 2016). To assess the impact of glucose in the physiology of cyanobacteria and the role of glycogen we evaluated the viability of WT and different glycogen synthesis mutant strains (∆AGP, ∆PGM, ∆glgA1, ∆glgA2 and ∆glgB strains) on solid BG11C under different conditions (Figure 39). All strains could grow under WT ∆AGP ∆PGM ∆glgA1 ∆glgA2 ∆glgB Control High Light 1 mM G250 mM NaCl 1mM G +250mM Nacl 1 mM G+DCMU 10010-1 10-2 10-3 10010-1 10-2 10-3 10010-1 10-2 10-3 10010-1 10-2 10-3 10010-1 10-2 10-3 10010-1 10-2 10-3 Figure 39: Spot assay of WT and single mutants of all the genes involved in glycogen synthesis. Cultures of WT, ∆AGP, ∆PGM, ∆glgA1, ∆glgA2 and ∆glgB in BG11C were diluted to 0.5 OD750, spotted on solid media and cultured in photoautotrophy (control), photomixotrophy (1 mM G), photoheterotrophy (1 mM G + DCMU), photoautotrophy with 250 mM NaCl, photomixotrophy with 250 mM NaCl and in high light (75 µmol photons m−2 s−1). Plates were cultivated under continuous light for 5 days. Dilutions are indicated above each panel. 111 photoautotrophic conditions as previously described (Xu et al., 2013; Koch et al., 2019; Díaz-Troya et al., 2020). However, under photomixotrophic conditions, the ∆glgA1 strain was the only mutant strain capable of growth (Figure 39). This suggests that disruptions in glycogen pathway due to restriction in its accumulation in the AGP and PGM strains or by the absence of GlgA2 or GlgB hindered the adequate metabolization of exogenous glucose. However, under photoheterotrophic conditions by culturing the strains in media with glucose and DCMU (inhibiting LET to prevent possible overexcitation of the in the PETC) both ∆glgA2 and ∆glgB regained viability (Figure 39). This indicated that in photomixotrophy, the combined carbon supply from glucose metabolism and CO2 fixation was unmanageable. However, this effect was not observed in the ∆AGP and ∆PGM strains, which exhibited null and limited glycogen synthesis, respectively (see Result Section: 1.Analysis of the connection between glycogen and central carbon metabolism by phosphoglucomutases in cyanobacteria.). The double glycogen synthases mutant (∆glgA) is not viable under standard growth conditions due to the toxic accumulation of ADP-glucose. Cultivation in salt supplemented media induces the redirection of ADP-glucose towards the synthesis of the osmolyte GG, which restores its viability (Díaz-Troya et al., 2020). To analyze if this metabolic redirection could also alleviate the lethal effect of glucose in the strains with altered glycogen synthesis, we cultured them in photomixotrophy with 250 mM NaCl (Figure 39). This combination of conditions enabled the survival of the ∆glgA2 strain, suggesting that the observed lethality in photomixotrophy of this strain could be attributed to the rapid influx of carbon to glycogen. Interestingly, this nonconcurrence in viability of the ∆glgA1 and ∆glgA2 mutant strains might be attributed to differences in kinetics properties of the enzymes, since GlgA2 is attributed with a more processive role compared with the distributive function of GlgA1 (Yoo et al., 2014). This could lead to stalling in the elongation process due to the absence of GlgA2 and thus behaving similarly to the ∆glgA strain. However, ∆glgB was unable to grow in media containing glucose and 250 mM NaCl (Figure 39). The same outcome was 112 observed in the ∆PGM and ∆AGP strains. This indicates that the inability to photomixotrophic growth of the ∆glgB strain might not be related to ADP-glucose toxicity, and it is not alleviated by the diversion of carbon to GG in salt supplemented media. This is further reinforced by the results of high light conditions on solid media, where the metabolic flux to glycogen would not be as pronounced as that imposed in photomixotrophy. Under increased light intensity the ∆glgB strain was unable to grow, similar to ∆AGP and ∆PGM strains, contrasting with the unimpeded growth of ∆glgA1 and ∆glgA2 strains (Figure 39). To gain insight into the role of glycogen in the transition to mixotrophy, we analyzed the effect of glucose addition on Synechocystis physiology, and the consequences of impaired glycogen synthesis in the photosynthetic and metabolic response to this trophic shift. First, we analyzed in depth the effect of glucose on the PGM and AGP strains, unable to growth in the presence of glucose even in the presence of DCMU or NaCl, which alleviates the lethality in other glycogen synthesis mutants. Second, we extended the analysis for the rest of the ∆glgA1, ∆glgA2 and ∆glgB strains to gain insight in the cause of the photomixotrophic impaired growth. 2.1 Impact of photomixotrophy on ΔPGM and ΔAGP strains 2.1.1 Physiological effects of glucose supplementation in WT, ΔPGM and ΔAGP strains In line with the previous results, to understand the role of glycogen metabolism during the transition to photomixotrophic conditions, we analyzed the effect of 2 mM glucose addition to photoautotrophically grown cultures of a WT Synechocystis strain and mutants lacking either PGM or AGP enzymes, which exhibit reduced or no glycogen synthesis, respectively (Figure 40). In contrast to previous experiments, where standard conditions were bubbling with 1% of CO2, the following photomixotrophic experiments were performed in Erlenmeyer flask with a constant shaking speed of 100 rpm at a light intensity of 40 µmol photons m−2 s−1. Prior to the addition of glucose, photoautotrophic cultures were inoculated in fresh medium at an OD750 of 1 and let adapt for two hours. 113 During the first four hours after the shift to photomixotrophic conditions, WT cells displayed six-fold greater increase in OD750 compared to cells kept under photoautotrophic conditions (from 1.0 OD750 to 1.6±0.19 OD750 and to 1.1±0.06 OD750, in photoautotrophy and photomixotrophy respectively, P=0.012 t-test twotailed (Figure 40b). Concomitantly, glycogen rapidly accumulated, reaching 66±6 µg OD750-1 within four hours of glucose addition. This amount of glycogen exceeded the levels found in the photoautotrophic culture by more than 50% within 24 hours (Figure 40c). Enhanced growth and glycogen synthesis were accompanied by complete consumption of the provided glucose within 24 hours (Figure 40d). (a) (b) (c) (d) Photoautotrophic growth Photomixotrophic growth WT ΔPGM ΔAGP WT ΔPGM ΔAGP 24 h 48 h 0 h Figure 40: Phenotypic characterization of WT Synechocystis and strains with deficient glycogen metabolism after 2mM glucose supplementation. (a) Photographs of WT, ∆PGM and ∆AGP strains cultivated in parallel in either photoautotrophic or photomixotrophic (addition of 2mM of glucose) conditions at 0, 24 and 48h of the experiment. (b) Growth curves of WT, ∆PGM and ∆AGP strains measured as optical density at 750 nm either in photoautotrophy (control, dash lines) or photomixotrophy (solid lines) at 0, 2, 4, 24 and 48 hours of experiment. (c) Time course glycogen content per optical density (OD750 nm) of WT, ∆PGM and ∆AGP strains both in the presence (solid lines) or absence (dashed lines) of glucose. (d) Glucose concentration profile in the media after the initial addition of 2 mM of glucose and 2, 4, and 24 hours thereafter. The data represent the mean ±SEM from six (b-d) biological replicates. 120 DHAP Glucose(out) Glucose-6P nmol OD750-1 E4P nmol OD750-1 S7P nmol OD750-1 PEP Xu5P S1,7BP Fructose-6P 6P-gluconate nmol OD750-1 RuBP Amount relative to WT0 h Fructose-1,6-BP nmol OD750-1 3PG nmol OD750-1 Fructose-6P nmol OD750-1 nmol OD750-1 Ru5P / Xu5P Glucose(in) Amount relative to WT0 h Serine Time (min) Time (min) Time (min) Time (min) R5P nmol OD750-1 Time (min) Time (min) Time (min) Glucose-1P nmol OD750-1 Time (min) Time (min) Time (min) nmol OD750-1 DHAP Time (min) Time (min) Time (min) GAP GAP Time (min) GAP Amount relative to WT0 h µg OD750-1 Glycogen WT ΔPGM ΔAGP Figure 45: Metabolic changes in upper metabolism of WT, ΔPGM and ΔAGP strains after glucose addition. Cultures of WT, ∆PGM and ∆AGP strains grown in BG11C were adjusted to a final optical density (OD750 nm) of 1 and harvested before (0) and 5, 15, 30 and 120 minutes post-glucose supplementation (2mM). The time course displays 14 metabolites involved in glucose assimilation, the oxidative pentose phosphate shunt (blue arrows) and Calvin-Benson-Bassham pathways (green arrows), glycolysis (black arrows) and glycogen metabolism (grey arrows). Metabolite names highlighted with coloured rectangles refer to molecules shared with different pathways. Abbreviations: R5P (ribose-5P), Ru5P/Xu5P (ribulose-5P/xylulose-5P pool), RuBP (ribulose-1,5P), S7P (sedoheptulose-7P), E4P (erythrose-4P), GAP (glyceraldehyde-3P), DHAP (dihydroxyacetone-P), 3PG (3Pglycerate). Values represent the mean±SEM or fold change ±SEM relative to WT value before glucose addition of 6 independent biological replicates. 121 Amount relative to WT0 h 3PG Citrate 2-OG Succinate Cis-aconitate Fumarate Malate Isocitrate OAA Aspartate SSAL PEP nmol OD750-1 Pyr nmol OD750-1 Alanine nmol OD750-1 AcCoA nmol OD750-1 nmol OD750-1 nmol OD750-1 nmol OD750-1 nmol OD750-1 nmol OD750-1 nmol OD750-1 nmol OD750-1 Time (min) Time (min) Time (min) Time (min) Time (min) Time (min) Time (min) Time (min) nmol OD750-1 Glutamate Time (min) Time (min) Time (min) Time (min) Time (min) Arginine nmol OD750-1 Time (min) Glutamine nmol OD750-1 Time (min) Lcitruline Time (min) Amount relative to WT0 h Asparragine Amount relative to WT0 h Time (min) WT ΔPGM ΔAGP Figure 46: Metabolic changes in downstream 3PG metabolism of WT, ΔPGM and ΔAGP strains after glucose addition. Cultures of WT, ∆PGM and ∆AGP strains were harvested before (0) and 5, 15, 30 and 120 minutes post-glucose supplementation (2mM). The time course display 17 metabolites involved in lower glycolysis, tricarboxylic acid cycle and ammonium assimilation . Abbreviations: PEP (phosphoenolpyruvate), Pyr (Pyruvate), AcCoA (Acetyl-CoA), 2OG (2-oxoglutarate), SSAL (succinic semialdehyde), OAA (oxalacetate). Values represent the mean±SEM or fold change ±SEM relative to WT value before glucose addition of 6 independent biological replicates. 122 In the WT strain, the observed metabolite patterns, as well as the glycogen measurements, indicated that under photomixotrophy excess carbon was stored as glycogen, with precursor intermediates building up upon the shift from photoautotrophy to photomixotrophy. Two distinct patterns of metabolic changes were observed (Figure 45, Figure 46 and Figure 47a). Firstly, levels of the glycolytic metabolites (e.g. fructose-1,6-BP (F1,6BP), dihydroxyacetone-P (DHAP), 3PG, phosphoenolpyruvate (PEP) or pyruvate), the RuBisCO substrate ribulose-1,5-BP (RuBP), and most TCA cycle intermediates, underwent transient fluctuations, peaking between 5 and 15 min after glucose addition. These metabolites typically reached more than double their initial amounts before returning to photoautotrophic values. Amino acids such as alanine and serine, derived from pyruvate and 3PG, respectively, increased more than 5-fold, along with similar trends for threonine and valine. Secondly, levels of metabolites from the first steps of the OPP pathway (e.g. G6P and 6PG), and those shared with the CBB cycle (e.g. glyceraldehyde-3P (GAP), fructose-6P (F6P), erythrose-4P (E4P) and sedoheptulose-7P (S7P)) increased linearly 3 to 5-fold until 120 min (the latest measurement after treatment) (Figure 45, Figure 46). In the glycogen mutants, glucose addition induced faster and more pronounced metabolic shifts than in WT (Figure 45, Figure 46 and Figure 47). There was a build-up of intermediates after glucose uptake. The first metabolites involved in glucose assimilation (G6P, 6PG, F6P) increased rapidly and dramatically (by 5 min more than 40, 10, and 25-fold, respectively). G1P, the first metabolite in glycogen synthesis, steadily increased by up to 10-fold in ∆AGP mutant but remained at low levels in ∆PGM, due to limited G6P/G1P interconversion in that mutant (Figure 41). Notably, F1,6BP was the only metabolite in this subgroup that showed subtle alterations and recovered to its initial values by the end of the experiment, similar to WT. The metabolites in the CBB cycle were also observed to change in the glycogen mutants upon the shift to photomixotrophy. There were moderate changes in the ribulose-5P (Ru5P)/xylulose-5P (Xu5P) pool as well as ribose-5P (R5P), both doubling their initial concentrations within 5 min after glucose addition and remaining stable throughout the experiment. RuBP, however, increased more 123 than 10-fold. The metabolites S7P, E4P, and GAP exhibited similar patterns to G6P or F6P accumulating to high levels (50, 30, and 10-fold for S7P, E4P and GAP, respectively) in glycogen mutants in the presence of glucose. In contrast, DHAP, an isomer of GAP, showed a similar trend to F1,6BP and correlated with WT values. Finally, 3PG, the product of RuBisCO, increased only transiently in WT, but accumulated steadily in glycogen mutants up to 2-fold, which constitutes a substantial increase due to the already high initial concentration (around 3 nmol OD750-1, more than the sum of the previously mentioned metabolites before glucose supplementation). Figure 47: Time course metabolic profile extension. (a) Details of the time-course metabolic response of the WT strain to glucose supplementation. Selected metabolites from Fig. 3 and Fig. 4 are displayed excluding data from ΔPGM and ΔAGP mutants to improve the visibility of changes in metabolite levels specifically in the WT strain (b) Fold-changes of some amino acids analyzed in the WT, ΔPGM and ΔAGP strains glucose time-course. This figure extends Figure 45and Figure 46 , and the values indicate the mean±SEM or fold-change±SEM in relation to the WT value before glucose addition, obtained from 6 independent biological replicates. 124 From 3PG to downstream metabolism, the first metabolite measured was PEP (Figure 46). PEP was reduced 5-fold in ∆PGM and ∆AGP a few min after glucose addition and remained at very low levels throughout the experiment. The decrease in PEP was coupled to an increase in pyruvate, the product of PEP dephosphorylation, reaching more than 0.25 nmol OD750-1 120 min after glucose addition. Levels of acetyl coenzyme A (AcCoA) remained steady upon transition of the glycogen mutants to photomixotrophy (Figure 46), similar to DHAP (Figure 45). AcCoA plays a central role in cellular metabolism, serving as a key intermediate compound into the TCA cycle. Levels of most intermediates in this cycle doubled, with citrate exhibiting the highest fold change and accounting for most of the carbon allocation in this cycle. Finally, amino acids pools underwent several changes in the glycogen mutants upon shift to photomixotrophy (Figure 45, Figure 46 and Figure 47b). Alanine and serine increased approximately 6-fold, and levels remained higher than 5 nmol OD750-1 in contrast to the transient accumulation observed in WT. Aspartate, threonine, valine and tryptophan showed transient accumulation peaking between 2and 4-fold photoautotrophic levels at approximately 30 min after glucose addition in both glycogen mutants, compared to unchanged levels in WT. Conversely, glutamate, an amino acid essential to nitrogen assimilation in the GS-GOGAT cycle, decreased to less than 2 nmol OD750-1 despite being one of the most abundant metabolites measured (more than 10 nmol OD750-1 under photoautotrophic conditions). Other amino acids, such as tyrosine, phenylalanine, methionine and leucine-isoleucine were considerably depleted in ∆PGM and ∆AGP mutants, while the WT showed similar levels or slight increases compared to photoautotrophic levels (Figure 47b). In summary, Synechocystis exhibits metabolic flexibility upon glucose addition, channeling carbon through PGI and OPP pathways to fuel the CBB cycle, coupled with transient fluctuations in some glycolytic and TCA cycle intermediates. The impact was more pronounced in ∆PGM and ∆AGP mutants, showing rapid and sustained increases in glucose assimilation metabolites and the CBB cycle within 5 min, highlighting the crucial role of glycogen synthesis as a metabolic sink in photomixotrophic acclimation. 125 2.1.2 Photosynthesis is severely impaired in the glycogen mutant strains cultured with glucose. Considering the strong growth penalty (Figure 40a and b) and the decreased level of photosynthetic proteins of the glycogen-deficient mutants under photomixotrophy (Figure 42), we next measured the functional status of the photosynthetic apparatus. We monitored the photosynthetic performance of the cells in both photoautotrophy and 30 min post-glucose addition, capturing the metabolic overflow in both glycogen deficient mutants (Figure 45, Figure 46 and Figure 47), but before photosynthetic protein levels decreased (Figure 42) and before long-term acclimation to photomixotrophy. Figure 48: Effect of glucose supplementation on the rate of CO2 fluxes in WT, ∆PGM and ∆AGP strains determined by MIMS. (a) CO2 exchange rates in photoautotrophy and 30 min after 2 mM glucose addition to 10 μg chl ml-1 WT, ∆PGM and ∆AGP cultures measured in a Membrane Inlet Mass Spectrometer (MIMS). The yellow rectangles represent an exposure to a light intensity of 50 μmol photons m−2 s−1 while the white rectangles represent dark periods. In photoautotrophic conditions, upon the onset of light, the carbon concentration mechanism (CCM) triggered an initial concentration step absorbing high amounts of CO2 (phase I). Subsequently, during the rest of the light period, consumption reaches a steady state (phase II). Conversely, when the lights are switched off, CCM activity ceases, leading to the release and efflux of the intracellular inorganic carbon pool back into the media (phase III). (b) Quantification of CO2 fluxes shown in (a) in the transition to light (maximum rate, phase I), during the steady-state in the light (phase II) and in the transition to dark (minimum rate, phase III) in photoautotrophy or 30 min after 2 mM glucose addition. The data represent the mean±SEM from four biological replicates. Statistical significance was denoted by * P<0.05; ** P<0.01, *** P<0.001 and “ns” non-significance (two-way ANOVA). 126 Real-time measurement of photosynthetic CO2 gas flux (Figure 48) by MIMS revealed similar rates and kinetics of CO2 uptake during photoautotrophy by all strains. While the WT maintained photoautotrophic CO2 uptake kinetics in the presence of glucose (Figure 48a), ∆PGM and ∆AGP mutants showed negligible steady-state CO2 uptake rates (Figure 48b), indicative of lack of CO2 fixation. Additionally, the glycogen mutants exhibited diminished maxima and minima of CO2 uptake rates during dark-to-light and light-to-dark transitions (phase I and III, in Figure 48b), respectively, associated with the import of CO2 by the lightactivated carbon concentration mechanism (CCM) and export of CO2 by diffusion. The decrease in carbon uptake (Figure 48) occurred already before ROS accumulation (Figure 43) or, as indicated by both immunoblotting (Figure 42) and 77K fluorescence emission spectra (Figure 49), before the loss of photosystems or any change in PSII/PSI ratio. This prompted us to investigate potential disturbance of photosynthetic electron transport by first conducting a saturating pulse analysis of Chl fluorescence. In photoautotrophic conditions, the Chl fluorescence traces (Figure 50a - black) and PSII effective yields [Y(II)] were similar in the WT, ∆PGM and ∆AGP strains (Figure 50b). After 30 min incubation with glucose, the WT strain showed unchanged fluorescence kinetics (Figure 50a - purple) and a slightly, albeit not statistically significant, lower Y(II) (Figure 50b). In contrast, the glycogen mutants exhibited higher Fs levels, resulting in an approximate halving of Y(II) (Figure 50b). Figure 49: Photosystems abundance ratio in photoautotrophy and 30 mins after glucose addition. 77 K fluorescence emission spectra of WT, ΔPGM and ΔAGP intact cultures grown under photoautotrophic conditions (black line) and 30 min after 2 mM glucose addition (purple line) excited at 440 nm. Spectra were normalized to their maximum PSI emission peak. The data represent the mean±SEM from four biological replicates. 127 The increase in flash-induced Chl fluorescence and its subsequent relaxation in darkness derive from the reduction and reoxidation of QA-, respectively, reflecting the status of the acceptor sides of the PSII complex (Vass et al., 1999). When the fluorescence curves are fitted with a three-component decay function, the initial, fast component corresponds to oxidation of QAby electron transfer to QB, the middle component corresponds to electron transfer from QAto PQ molecules from the PQ pool binding to empty QB sites, while the third, slow component derives from recombination reactions with the S2 state of the oxygen evolving complex (Vass et al., 1999). In the presence of glucose, WT exhibited only slightly, but non-significantly slower quenching of flash-induced fluorescence in darkness, in comparison to photoautotrophic cultures. In contrast, in both mutants reoxidation of QAwas slowed down significantly 30 min after treatment with glucose (Figure 51a). The time constant for the fluorescence decay in the fast phase increased from c.a. 0.35 ms to 1 ms in both glycogen mutants, while the time constants for the middle phase increased exponentially from c.a. 4 and 10 ms to 133 and 161 ms for the ΔPGM and ΔAGP mutants, respectively (Figure 51b). The amplitude of the fast component was roughly 6-7-fold lower and the amplitude of the slow component roughly 3-fold higher in the glycogen mutants in comparison to WT 30 min after introduction of glucose (Figure 51c). The Figure 50: Effect of glucose supplementation on chlorophyll fluorescence and PSII effective yield [Y(II)] in WT, ∆PGM, and ∆AGP strains. (a) Chlorophyll fluorescence induction curves were generated with a DUAL-PAM-100 fluorometer under 40 μmol photons m−2 s−1 actinic red-light using cultures of WT, ΔPGM and ΔAGP at 5 μg chl ml-1 grown in BG11C in photoautotrophy (black) or 30 min after 2 mM glucose addition (purple). The yellow rectangle in the background represents the light periods of the induction curve. Fluorescence parameters are indicated with black arrows: Fo represents the initial fluorescence in the dark; Fs represents the steady-state fluorescence under actinic red light, Fm represents the maximum fluorescence in the dark; Fm ’ represents the maximum fluorescence under the actinic light. (b) Quantification of the PSII effective yield [Y(II)] from chlorophyll fluorescence induction curves in (a). PSII effective yield is given by [Y(II) = ( Fm´- Fs)/F m´]. These parameters were determined from the third saturation pulse (250 ms at 5000 μmol photons m−2 s−1) applied during light exposure. The data represent the mean±SEM from four biological replicates. Statistical significance was denoted by *** P <0.001, otherwise indicate no significance (twoway ANOVA). 128 amplitude of the middle component, however, was only slightly affected. These results suggest that electron transfer from QAto QB is inhibited and recombination reactions are increased in a large sub-population of PSII centers in the glycogen mutants. The kinetics of QAreoxidation of photomixotrophic glycogen mutants were intermediate between photoautotrophic and DCMUtreated samples, where DCMU is an inhibitor that occupies the QB site in PSII and Figure 51: Effect of glucose supplementation on the relaxation of flash-induced fluorescence yield in WT, ∆PGM, and ∆AGP strains. (a) Reoxidation kinetics of QA after a single-turnover saturation pulse excitation to 7,5 μg chl ml-1 WT, ΔPGM and ΔAGP cultures grown in BG11 in photoautotrophy (black) or 30 min after 2mM glucose addition (purple) using a flash fluorimeter. Fluorescence relaxation was monitored both in the absence (solid lines, circles) or in the presence (dotted line, triangle, fainted colors) of 20 µM DCMU, which was added as a control to prevent QAreoxidation. (b) Analysis of the effect of glucose on the time constants ( t ) of flash-induced chlorophyll fluorescence yield in WT, ∆PGM and ∆AGP strains. Quantification of the time constants ( t ) of the three phases in the reoxidation kinetics of QAby analyzing the fluorescence relaxation traces from figure (a) with a three-component decay function. The initial, fast component corresponds to the oxidation of QAby electron transfer to QB. The middle component involves the binding of QAto PQ molecules from the PQ pool in the vacant QBsites. The third, slow component, derives from the recombination reactions with the oxidized S2 state of the oxygen-evolving complex. (c) Analysis of the effect of glucose on the amplitudes (A) of the fast, middle and slow components of the fluorescence decay kinetics in (a and b). All cultures were dark-adapted for 10 minutes before the measurements. The data represent the mean ±SEM from five biological replicates. Statistical significance was denoted by * P<0.05; ** P<0.01; *** P <0.001 and **** P<0.0001, otherwise indicate no significance (two-way ANOVA). 129 prevents QAoxidation (Figure 15). In the presence of DCMU, QAreoxidation occurs via charge recombination with donor side components (mainly the S2 state of the oxygen-evolving complex), thus reflecting the status of the PSII donor side. Therefore, we concluded that glucose induces substantial changes on the acceptor side of PSII in the glycogen mutants, whereas the PSII donor side remains unaffected. Subsequently, we employed MIMS to investigate how these alterations in electron transfer affect the photosynthetic oxygen fluxes (Figure 52). In photoautotrophy, all strains displayed similar oxygen fluxes, apart from slightly higher light-induced O2 uptake in the ΔAGP mutant. In the presence of glucose all strains expectedly showed increased rates of respiration (Figure 52b, O2 Figure 52: Effect of glucose supplementation on O2 fluxes in WT, ∆PGM, and ∆AGP strains measured in MIMS. (a) Oxygen flux dynamics in photoautotrophy and 30 min after 2mM glucose addition using cultures of WT, ΔPGM and ΔAGP at 10 μg chl ml-1 1 grown in BG11 in a Membrane Inlet Mass Spectrometer (MIMS). Dark periods are indicated by white rectangles, while yellow rectangles represent exposure to a light intensity of 50 μmol photons m−2 s−1. The arrows at 15 min time point represent the addition of 0,5 mM DCBQ. All cultures were dark-adapted for 10 minutes before the measurements. The data represent the mean ±SEM from four (a and e) and five (b, c and d) biological replicates. (b) Quantification of the steady state oxygen fluxes rates shown in (a). Gross and Net O2 production data were obtained from the steady-state during the light periods in the absence or presence of 0,5 mM DCBQ (plain or striped pattern, respectively). O2 consumption was calculated from the steady-state prior to light onset. The data represent the mean±SEM from four biological replicates. Statistical significance was denoted by * P <0.05; ** P<0.01; *** P<0.001 and “ns” non-significance (two-way ANOVA). 136 with the same OD750 under photoautotrophic conditions, the toxic effect of glucose probed irreversible at this stage, compromising culture viability and irreversibly leading to cell death. 2.2 Impact of photomixotrophy on glycogen synthases and branching enzymes mutant strains in Synechocystis. To further explore the phenotypes observed in photomixotrophic conditions for the ∆glgA1, ∆glgA2 and ∆glgB strains (Figure 39), and in light of the results obtained with the ∆PGM and ∆AGP strains (2.1 - Impact of photomixotrophy on ΔPGM and ΔAGP strains), we examined the response of these strains to glucose supplementation in liquid media. Since the phenotype was less severe compared to ∆AGP and ∆PGM strains, as ∆glgA2 and ∆glgB strains can recover the viability in the presence of DCMU (Figure 39), we focused on the mid-term adaptation to mixotrophy rather than immediately after glucose addition. The experimental setup was comparable to that used for the analysis of ∆AGP and Figure 58: Analysis of strains recovery after glucose is removed from the media. Cultures of WT, ∆PGM and ∆AGP strains were cultivated either in photoautotrophy (as control, No glucose) or in photomixotrophy (addition of 2mM of glucose). Aliquots of the cultures were harvested by centrifugation after 4 or 24 hours of glucose addition and resuspended in the same volume of fresh BG11C. Effect of the glucose removal was followed both (a) visually with photographs of WT, ∆PGM and ∆AGP strains at different time points and (b) time course of PSII effective quantum yields of energy conversion [Y(II)] to 5 μg chl ml-1 Synechocystis cells in a DUAL-PAM-100 fluorometer as explained in Figure 50 . Representative pictures (a) and mean±SEM (b) of four independent biological replicates are presented. 137 ∆PGM strains (Figure 40): precultures of WT, ∆AGP, ∆PGM, ∆glgA1, ∆glgA2 and ∆glgB in BG11C were adjusted to a final OD750 of 1 and supplemented with 2 mM glucose. We measured the time course of OD, glycogen content, glucose consumption and PSII quantum yield [Y(II)] before and after two, four and 24 hours of glucose addition (Figure 59). The behavior of ∆AGP and ∆PGM strains served as control for strains with impaired photomixotrophic growth and aligned with previous observations, including growth arrest, minimal or no glycogen accumulation, reduced glucose consumption, and a rapid drop in Y(II) (Figure 40, Figure 50 and Figure 58b). Consistent with our findings on solid BG11C media (Figure 39), both ∆glgA2 and ∆glgB were unable to grow in photomixotrophic conditions, resulting in a noticeable whitening of the culture within 48 hours, particularly pronounced in the ∆glgB mutant (Figure 59a). Despite this pigment loss like that in the ∆AGP and 24 h 0 h 48 h WT ∆AGP ∆PGM ∆glgA1 ∆glgA2 ∆glgB Photomixotrophic growth WT (a) (b) (c) (d) (e) Figure 59: Phenotypic characterization of single mutants in all the genes involved in glycogen synthesis after 2mM glucose supplementation. Precultures of WT, ∆AGP, ∆PGM, ∆glgA1, ∆glgA2 and ∆glgB grown in BG11C were adjusted to 1 OD750 and supplemented with 2 mM glucose. (a) Photographs of the strains cultivated under photomixotrophic conditions at 0, 24 and 48h of the experiment. (b) Growth curves at 0, 2, 4, 24 hours of experiment. (c) Glycogen content time course (d) Glucose concentration profile in the media after the initial addition of 2 mM of glucose and 2, 4, and 24 hours thereafter. (e) Effective quantum yields of energy conversion in PSII [Y(II)] of the strains measured by PAM fluorometry. The data represent the mean ±SEM from four biological replicates. 138 ∆PGM strains, the OD750 values rapidly increased after glucose addition, reaching levels comparable to those in WT and ∆glgA1 for the ∆glgB strain and halfway for the ∆glgA2 strain (Figure 59b). The glycogen content mirrored the OD750 pattern (Figure 59c). Within the first 4 hours after glucose addition, the glycogen content of the ∆glgB strain resembled that of the WT, but by the end of the experiment, ∆glgB exhibited higher levels, possibly due to WT started consuming the accumulated glycogen (Figure 59c). In contrast, 4 hours after glucose addition, the ∆glgA2 strain accumulated higher glycogen amounts than the ∆PGM, reaching up to 40% of WT levels and maintaining similar levels by the end of the time-course (Figure 59c). The glycogen accumulation was reflected in the glucose consumption, since the ∆glgB consumed glucose at a similar pace to the WT, while the ∆glgA2 strain exhibited consumption rates between the WT and the ∆AGP and ∆PGM strains within the first 4 hours after glucose addition (Figure 59d). However, glucose was not fully consumed by ∆glgB and ∆glgA2 strains within 24 hours, indicating that growth impediments and stalling of glucose metabolism occur between four and 24 hours. In fact, the ∆glgA2 strain presented a delayed rate of glucose consumption since the beginning of the photomixotrophic growth. For the ∆AGP and ∆PGM strains, in contrast, that growth impediment was almost immediate and glucose consumption was minimal (Figure 59d). Finally, we evaluated the PSII quantum yield Y(II) (Figure 59e). Within 4 hours, all strains experienced a sequential reduction in Y(II) values, with ∆AGP, ∆glgB and particularly ∆PGM being the most affected. Both WT and ∆glgA1 recovered photoautotrophic values 24 hours after glucose addition, while the remaining strains exhibited a complete loss of fluorescence signal and thus, Y(II) decrease (Figure 59e). In summary, the ∆glgB initially exhibited a behavior similar to WT during the first four hours but ultimately displayed a glycogen-deficient phenotype. Regarding glycogen synthases, the absence of GlgA1 did not affect photomixotrophic growth, while the absence of the GlgA2 isoform resulted in a phenotype intermediate between WT and the glycogen-deficient strains (∆AGP and ∆PGM), emphasizing the functional differences between glycogen synthase isoforms. 139 We further evaluated the metabolic profile of these strains following glucose addition (Figure 60). As anticipated, the metabolic profile of the ∆glgA1 strain closely resembled that of the WT, while ∆AGP and ∆PGM exhibited alterations consistent with those previously described (Figure 45, Figure 46 and Figure 47). The ∆glgA2 strain displayed metabolic alterations similar to those in the mutants with null or limited glycogen synthesis. These alterations included the accumulation of serine, pyruvate, 3PG, and malate, but without depletion of methionine and PEP (Figure 60). Notably, there was a significant accumulation of glucose-1P and ADP-glucose, intermediates preceding glycogen synthesis, indicating a bottleneck in glycogen synthase activity due to the absence of GlgA2, which corroborates the earlier findings (Figure 39 and Figure 59). In fact, the increase in ADP-glucose levels in the ∆glgA2 strain was over 300-fold higher. compared to the WT, approximately 25-fold higher than the next highest level observed in the ∆glgB and ∆glgA1 mutant strains. These data suggest that GlgA2 is required for efficient glycogen synthesis under mixotrophic conditions, and its absence cannot be compensated by the activity of GlgA1. In essence, the behavior of the ∆glgA2 strain resembles that of a double glycogen synthase knockout mutant (∆glgA) (Díaz-Troya et al., 2020). Conversely, the metabolic profile of the ∆glgB strain, which is also severely affected under mixotrophic conditions, did not exhibit major changes (Figure 60). Only a few of the measured amino acids, including glutamine, methionine, aspartate, glutamate, and histidine, showed slight decreases compared to the WT. Minor increases were observed in glucose-1P and ADP-glucose, but these changes were far less pronounced than those in the ∆glgA2 strain. The changes detected in the ∆glgB mutant during the first 4 h after glucose addition do not strongly support a metabolic cause for the inability to survive under mixotrophic conditions. While changes in other undetected metabolites cannot be ruled out, we propose that the lethality of the glgB mutant in mixotrophy may be due to a non-metabolic issue. One possibility, supported by the observation that NaCl failed to improve viability in photomixotrophy and that high light conditions resulted in ∆glgB lethality while ∆glgA2 remained viable (Figure 39), could be 140 Figure 60: Metabolite profile of single mutants in all the genes involved in glycogen synthesis after 2mM glucose supplementation. Amounts of polar metabolites measured by LC/MS of WT, ∆AGP, ∆PGM, ∆glgA1, ∆glgA2 and ∆glgB at 0, 2 and 4 hours after 2 mM glucose supplementation. Metabolites are grouped by families represented by colored backgrounds: amino acids in blue, shared metabolites of CCB cycle, OPP pathway and upper-glycolysis in green, osmolytes pathways in red, lower glycolysis and trioses phosphate in white and TCA cycle in yellow. Data are means ±SEM from four biological replicates. 141 related to structural problems in synthesized glycogen. In the absence of the branching enzyme in the ∆glgB strain, the inability to insert branches into glycogen most likely results in the synthesis of long chains that may lead to physical impediments and disruption of thylakoid membranes. As glycogen predominantly accumulates between them (Figure 44), this could potentially explain the observed changes in PSII quantum yield (Y(II)). Further characterization of the glycogen from the ∆glgB strain and microscopy visualization are required to confirm this hypothesis. 142 3 Integration of the metabolic and photosynthetic regulation of the bleaching process. Non-diazotrophic cyanobacteria exhibit a complex regulatory response to nitrogen starvation (Krauspe et al., 2021). This response allows cells to enter a dormant state and survive this nutrient limitation. The central regulator NtcA, orchestrates the coordinated steps of this adaptation program. NtcA senses the cellular carbon-to-nitrogen (C/N) balance through fluctuations in 2-oxoglutarate levels. This molecule belonging to the TCA cycle plays a crucial role as substrate for nitrogen assimilation in the GS-GOGAT cycle. When nitrogen availability diminishes, 2-oxoglutarate accumulates, triggering a response from NtcA. Several scenarios, including limitations on cell growth or photosynthesis imposed by inhibitors (DCMU, DBMIB, carbenicillin, cerulenin), restrictive conditions like low light (Yoshihara & Kobayashi, 2022) or the incorporation of 2-oxoglutarate through a transporter (Hickman et al., 2013) prevent the complete bleaching process. Additionally, chlorosis upon nitrogen depletion is hindered in specific mutant strains. The most straightforward example is the absence of nblA, nblB and, nblD phycobiliprotein adaptor proteins essential for their degradation (Baier et al., 2014; Levi et al., 2018). Consequently, mutations in ntcA, a key component regulating nblA and nblB expression also affect phycobiliprotein degradation (Sauer et al., 1999). In Synechochoccus elongatus PCC 7942, mutations in Hik33 (also known as NblS) (Van Waasbergen et al., 2002) or in another response regulator, NblR, prevents bleaching (Schwarz & Grossman, 1998). However, mutations in Synechocystis such as dspA, a NblS/Hik33 homolog, or in the nblRlike gene does not affect NblA expression, implying regulatory differences in the acclimation to nitrogen deprivation between these two species (Zabulon et al., 2007). However, inactivation in Synechocystis of the phosphotransacetylase in combination with dspA causes bleaching abrogation (Morrison et al., 2005). Mutations in other enzymes, such as alanine dehydrogenase, also cause a delay in phycobiliprotein degradation in Synechochoccus elongatus PCC 7942 (Lahmi et al., 2006). 143 Intriguingly, glycogen mutant strains are unable to complete the chlorosis program during nitrogen deprivation (Gründel et al., 2012; Hickman et al., 2013). This suggests a yet unresolved connection between chlorosis and glycogen synthesis. As shown in Figure 24, when deprived of nitrogen, glycogen mutants exhibited growth arrest, lack of phycobiliprotein degradation, and metabolic overflow. This overflow was evident from the excretion of organic compounds and a distinct metabolic profile compared to WT strain (Figure 26). Notably, several fold change accumulation or depletion were observed in several intracellular metabolites, including 2-oxoglutarate, although on a less dramatic scale. Additionally, these mutants displayed a progressive and fast closure of PSII reaction centers, limiting O2 evolution despite retaining the phycobiliproteins complex (see Results Section: 1.5 Response to nitrogen deprivation). These findings suggest that both photosynthetic and metabolic regulations contribute to chlorosis. Interestingly, the metabolic and photosynthetic response of glycogen mutants in nitrogen deprivation resembles the phenotype observed when glucose is added to the medium (see Results Section: 2.1 Impact of photomixotrophy on ΔPGM and ΔAGP strains). Under these photomixotrophic conditions, ∆AGP and ∆PGM strains exhibited rapid overflow of metabolites, compromising photosynthetic capacity in a minute timescale. Additionally, although WT cells can perform bleaching in nitrogen deprived media and also shift to a photomixotrophic growth mode in the presence of glucose, both conditions deal with metabolic adaptations (Figure 26, Figure 45, Figure 46 and Figure 47) and photosynthetic restructuration (Figure 24f and g, Figure 25 and Figure 58). This parallelism suggests that the successful chlorosis program might be directly influenced and disturbed by alterations in metabolism and photosynthesis, alongside the established role of nitrogen sensing and C/N imbalance. 144 3.1 Effect of photosynthesis inhibition and exogenous hexose supplementation during nitrogen deprivation. To test the hypothesis that alterations in photosynthesis and metabolism due to unmanageable carbon partitioning might disrupt the chlorosis program, we cultured WT cells under various conditions in nitrogen-depleted media. We compared photoautotrophic conditions (80 µmol photons m−2 s−1) with conditions altering photosynthesis (darkness and photoautotrophy with DCMU) or carbon metabolism (supplementation with an additional carbon input such as glucose or its epimer mannose, thereby forcing the cells to manage an increased pool of intermediary metabolites). As control, the same conditions were analyzed in nitrogen-repleted media (BG11C). After 24 hours, dark-grown cultures were Figure 61: Analysis of the bleaching process in WT cells under several conditions. WT culture was harvested and resuspended in nitrogen repleted and depleted media at 1 OD750. Each culture was divided in six and cultured in photoautotrophy (as control, 80 µmol photons m−2 s−1), dark, DCMU (20 µM), glucose (4 mM), and mannose (4 mM). (a) Photographs of the cultures at 0, 24 and 48 hours after the onset of the experiment. After 24 hours cultures in Dark were illuminated at the same conditions as control (denoted with an asterisk). (b) Cultures growth and (c) cultures glycogen content after 24 hours of experiment. A dashed line in yaxis denotes the initial value (0 h). (d) Whole cell spectra of WT cells before (0 h) and 24 hours after nitrogen deprivation with either dark, DCMU, glucose or mannose treatment to analyze phycobilisomes degradation. Spectra were normalized to absorbance at 750nm. The data in (b) and (c) represent the mean ±SEM from three biological replicates. 145 switched to photoautotrophic conditions (denoted with an asterisk). Hexoses were added at 4 mM to prevent their total consumption and avoid the return to autotrophic conditions before metabolomics analysis (Figure 61). As expected from its autotrophic metabolism, Synechocystis cells cultured in the dark or in the presence of DCMU were unable to grow, regardless of the nitrogen availability, and their initial glycogen reserves were depleted (Figure 61). Interestingly, after 24h, cells cultivated in BG11C supplemented with hexoses showed a slight increase in their glycogen levels compared with the control condition, suggesting that both glucose and mannose are incorporated into the metabolism and eventually stored as glycogen. However, only minor changes were observed in growth in both BG11C and BG110C, as well as in glycogen content in BG110C compared with the control condition (Figure 61b-d). However, in nitrogen-depleted conditions, cultures grown in the dark or treated with DCMU, both of which inhibit photosynthesis, did not perform bleaching. This was also observed in the glucose supplemented culture, despite the increase in OD750 (Figure 61a and d). Mannose supplemented culture exhibited an unexpected behavior. Although glycogen accumulation in BG11C conditions suggests that both glucose and mannose were incorporated similarly, cultures in BG110C supplemented with mannose were able to undergo bleaching (Figure 61d). These findings strongly support that bleaching in response to nitrogen deprivation is a light-driven process. Both dark conditions and the presence of DCMU, which inhibits photosynthetic electron flow, blocked the process, as previously described (Yoshihara & Kobayashi, 2022). In fact, re-introducing light to darkgrown cultures after 24 hours in nitrogen-depleted media resulted in the resume of bleaching (Figure 61 - dark 48 h, denoted with an asterisk). Moreover, the similar behavior observed in dark and DCMU-treated cells suggests that bleaching requires a fully functional photosynthetic apparatus powered by an actinic light source. Interestingly, excitation of photosystems alone is not sufficient, as blocking electron flow with DCMU in the presence of light prevented bleaching. In addition, the bleaching process requires the metabolic flexibility