Stress response requires an efficient connection between glycogen and central carbon metabolism by phosphoglucomutases in cyanobacteria

Journal of Experimental Botany, Vol. 74, No. 5 pp. 1532–1550, 2023 https://doi.org/10.1093/jxb/erac474 Advance Access Publication 1 December 2022 © The Author(s) 2022. Published by Oxford University Press on behalf of the Society for Experimental Biology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. RESEARCH PAPER Stress response requires an efficient connection between glycogen and central carbon metabolism by phosphoglucomutases in cyanobacteria PabloOrtega-Martínez1,2,, MiguelRoldán2, SandraDíaz-Troya1,2,*, and Francisco J.Florencio1,2,*, 1 Instituto de Bioquímica Vegetal y Fotosíntesis, Universidad de Sevilla, Consejo Superior de Investigaciones Científicas, Américo Vespucio 49, Sevilla, 41092, Spain 2 Departamento de Bioquímica Vegetal y Biología Molecular, Facultad de Biología, Universidad de Sevilla, Profesor García González s/n, Sevilla, 41012, Spain * Correspondence: flor[email protected] or sdtr[email protected] Received 23 June 2022; Editorial decision 25 November 2022; Accepted 28 November 2022 Editor: Yves Gibon, INRAE-Bordeaux, France Abstract Glycogen and starch are the main storage polysaccharides, acting as a source of carbon and energy when necessary. Interconversion of glucose-1-phosphate and glucose-6-phosphate by phosphoglucomutases connects the metabolism of these polysaccharides with central carbon metabolism. However, knowledge about how this connection affects the ability of cells to cope with environmental stresses is still scarce. The cyanobacterium Synechocystis sp. PCC 6803 has two enzymes with phosphoglucomutase activity, PGM (phosphoglucomutase) and PMM/PGM (phosphomannomutase/phosphoglucomutase). In this work, we generated a null mutant of PGM (∆PGM) that exhibits very reduced phosphoglucomutase activity (1% of wild type activity). Although this mutant accumulates moderate amounts of glycogen, its phenotype resembles that of glycogen-less mutants, including high light sensitivity and altered response to nitrogen deprivation. Using an on/off arsenite promoter, we demonstrate that PMM/PGM is essential for growth and responsible for the remaining phosphoglucomutase activity in the ∆PGM strain. Furthermore, overexpression of PMM/PGM in the ∆PGM strain is enough to revoke the phenotype of this mutant. These results emphasize the importance of an adequate flux between glycogen and central carbon metabolism to maintain cellular fitness and indicate that although PGM is the main phosphoglucomutase activity, the phosphoglucomutase activity of PMM/PGM can substitute it when expressed in sufficient amounts. Keywords: Cyanobacteria, environmental stress, glycogen, high light, nitrogen metabolism, phosphoglucomutase, phosphohexomutase. Introduction Cyanobacteria are widespread photosynthetic microorganisms of enormous ecological, evolutionary, and biotechnological importance. They play an essential role in supporting the trophic chains through their oxyphototrophic metabolism, mainly in aquatic ecosystems, and are of great importance in carbon and nitrogen cycles. During recent years, these This paper is available online free of all access charges (see https://academic.oup.com/jxb/pages/openaccess for further details) Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 PMM/PGM overexpression compensates a PGM mutant | 1533 phototrophs have emerged as potential tools for biotechnological purposes, although the side effects of modifying their metabolic carbon flux are a drawback that requires further exploration (Luan et al., 2019). Photosynthesis supplies the energy and metabolites required to support the vast array of biosynthetic processes, development, and cell division. The ability to store part of the photosynthates as polysaccharide gives the cells a remarkable plasticity: its accumulation and mobilization, in addition to providing cells with carbon on demand, act as a buffer that helps maintain energy homeostasis and allows cells to cope with a changing environment (Cano et al., 2018). Glycogen is the main polysaccharide in cyanobacteria, but can be found in both prokaryotic (including archeobacteria) and eukaryotic cells (animal and fungi) (Ball and Morell, 2003; Cifuente et al., 2019). In cyanobacteria, glycogen plays an essential role to face some environmental stresses such as diel cycles, high irradiance, or nutrient-limited environments, especially nitrogen deprivation in non-diazotrophic cyanobacteria (Cano et al., 2018; Forchhammer and Schwarz, 2019; Welkie et al., 2019). Glycogen synthesis starts with the formation of ADP-glucose (ADPGlc) from glucose-1-phos- phate (G1P) and ATP by ADP-glucose pyrophosphorylase (AGP). Glycogen synthase (GlgA) employs ADP-glucose to elongate the growing polysaccharide chain. Finally, the branching enzyme (GlgB) introduces α-1,6 ramifications. For glycogen catabolism, glucose chains are shortened from their non-reducing ends by glycogen phosphorylases (GlgPs), liberating G1P. Once a branching point is reached, the α-1,6 glucosidic ramifications are removed by the hydrolase activity of the debranching enzyme (GlgX) (Fig. 1). The concerted action of these enzymes provides the characteristic architecture of glycogen granules (Ball et al., 2015; Cifuente et al., 2019). Since glycogen metabolism begins and ends with G1P, the flux between glycogen and central carbon metabolism relies on the reversible conversion of G1P and G6P by enzymes with phosphoglucomutase activity. These proteins belong to the phosphohexomutase superfamily. Phosphohexomutases are present in all kingdoms of life, playing a critical role in a vast array of metabolic processes. These proteins share a common catalytic mechanism and similar structural domains despite using different phosphosugar substrates such as glucose, mannose, glucosamine, or N-acetylglucosamine (Shackelford et al., 2004). In fact, the preferred substrate employed by these enzymes has traditionally been used to classify them into subgroups: phosphoglucomutase (PGM), phosphomannomutase/phosphoglucomutase (PMM/PGM), phosphoglucosamine mutase (PNGM), phoshoacetylglucosamine mutase (PAGM), phosphopentomutase, and glucose-1,6-biphosphate synthase (Stiers et al., 2017). These enzymes are phosphorylated on a conserved serine residue and their catalytic cycle occurs by phosphoryl exchanges between the enzyme and the substrate, allowing the transfer of the phosphate of the phosphosugar substrate from C6 to C1 or vice versa. Briefly, the process begins with the transfer of the phosphate from the catalytic phosphoserine to the substrate, generating a bisphosphorylated intermediate product. This dephospho-form of the enzyme acquires a more flexible conformation, allowing the intermediate product to reorient in a 180° flip. In that state, the serine residue reacquires the phosphate and a sugar monophosphate is released (Regni et al., 2002, 2004; Stiers et al., Glucose-6P Fructose-6P GMPP Glycogen ADP-Glucose Glucose-1P ADP AGP GlgP1/2 GlgB PPi GlgX1/2 Pi UDP-Glucose Sps Sucrose-6P Spp Sucrose CugP ATP PGM Mannose-6 PMannose-1P PMM/PGM GG-3P GgpS GgpP GG Fructose-6P GDP-Mannose UTP PPi PGI PMM/PGM Glycerol-3P PMI Fig. 1. Schematic representation of the connection between central carbon metabolism and the pathways for glycogen and glucosylglycerol synthesis in Synechocystis. AGP, ADP-glucose pyrophosphorylase; CugP, UTP-glucose-1-phosphate uridylyltransferase; GgpP, glucosylglycerolphosphate phosphatase; GgpS, glucosylglycerolphosphate synthase; GlgA1/2, glycogen synthases; GlgB, glycogen branching enzyme; GlgP1/2, glycogen phosphorylases; GlgX1/2, glycogen debranching enzyme; GG, glucosylglycerol; GG-3P, glucosylglycerol-3-phosphate; GMPP, GDP-mannose pyrophosphorylase; PGI, glucose-6-phosphate isomerase; PGM, phosphoglucomutase; PMI, mannose-6-phosphate isomerase; PMM/PGM, phosphomannomutase/phosphoglucomutase; Spp, sucrose-phosphate phosphatase; Sps, sucrose-phosphate synthase. Blue arrows, reactions for glycogen synthesis and degradation; red arrows, reactions for synthesis of osmolites glucosylglycerol and sucrose; black arrows, reactions connecting glucose-1P and mannose-1P with central carbon metabolism. Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 1534 | Ortega-Martínez et al. 2019). These enzymes need magnesium as a cofactor and present substrate inhibition by catalytic cleft occupancy of G1P in the dephospho-form of the enzyme, an effect alleviated by glucose-1,6-bisphosphate (G1,6BP) as an activator of autophosphorylation (Ray and Roscelli, 1964; Naught and Tipton, 2005). Synechocystis sp. PCC 6803 (hereafter Synechocystis) is a cyanobacterium that exclusively accumulates glycogen as storage polysaccharide. The genome of Synechocystis presents three genes coding for enzymes belonging to the phosphohexomutase superfamily: sll1758 coding for a predicted protein of the PNGM subgroup; and sll0726 and slr1334, coding for PGM and PMM/PGM, respectively, both enzymes with phosphoglucomutase activity (Liu et al., 2013). Recent findings in Synechocystis have highlighted the role of PGM as a key regulatory element in glycogen metabolism (Doello et al., 2022). When nitrogen starved, Synechocystis enters a dormantlike state (Neumann et al., 2021). Proteomic studies showed that PGM is highly phosphorylated at Ser47 (in the regulatory latch domain, a distinct serine from that in the catalytic cleft) under nitrogen deprivation (Spät et al., 2018). This phosphorylation maintains PGM in an inactive state and is essential for survival in these conditions (Doello et al., 2022). PMM/PGM has recently been described to have reversible in vitro activity as a G1,6BP synthase, using both fructose-1,6-biphosphate and either G1P or G6P. This makes PMM/PGM a potential key regulator of the enzymes from the phosphohexomutase superfamily (Neumann et al., 2022). As previously indicated, phosphoglucomutase activity is a key point in the connection between glycogen and central carbon metabolism. However, knowledge about the effects of limited G1P–G6P interconversion on cell fitness and the role of PGM and PMM/PGM in this process is scarce. Here, we generated and characterized mutant strains of the two enzymes with phosphoglucomutase activity in Synechocystis. We show that PGM is required for an adequate connection between glycogen and central carbon metabolism, and its absence causes a reduced fitness to stresses such as high light or nitrogen deprivation. However, the response to these stresses is recovered when PMM/PGM is overexpressed, indicating a partial metabolic redundancy of these enzymes. All of this emphasizes the importance of an optimal management of carbon reserves in cyanobacteria. Materials and methods Strain and culture conditions Synechocystis sp. PCC 6803 strains were cultured in BG11 medium (Rippka et al., 1979) supplemented with NaHCO3 to a final concentration of 12mM (BG11C) except for carbon-limited experiments where 5mM or no NaHCO3 was used. Cultures were grown at 30 °C under continuous light (4500k LED lights, 50 µE m−2 s−1) in non-bubbled Erlenmeyer flasks at 100rpm or in conical flasks bubbled with a stream of 1% (v/v) CO2 in air. For nitrogen deficiency experiments, cultures were washed twice by centrifugation (10min at 7000 g) and resuspension in nitrogen-free medium (BG110C), and finally resuspended at 1 OD750 nm in BG110C. For inducer (NaAsO2) removal experiments, cultures were washed twice by centrifugation (10min at 7000 g) and resuspension in media without inducer, and finally resuspended at 1 OD750 nm. Experiments under light/dark conditions were performed in Erlenmeyer flasks on a New Brunswick™ Innova® 43/43R shaker at 40 µE m−2 s−1.When required, media were supplemented with Bacto agar (Difco) [1% (w/v)] or the required antibiotics (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). Generation of mutant strains The ∆PGM strain lacking the sll0726 gene was obtained by transformation of the wild-type (WT) strain with plasmid pGT∆sll0726::Km, which allows for the deletion of the complete sll0726 ORF. To generate this plasmid, regions upstream and downstream of sll0726 were amplified with primers OL68/OL69 and OL70/OL71, respectively, which introduce a BamHI restriction site, and joined by overlapping PCR with primers OL68/OL71. This fragment was cloned into pGEM-T and a kanamycin resistance cassette was introduced in the generated BamHI site. The ∆AGP∆PGM strain was obtained by transformation of the ∆AGP strain with plasmid pGT∆sll0726::Km. Plasmid pGT∆slr1334::Ery was used to generate the ∆PMM* strain, a partially segregated ∆slr1334 mutant. This plasmid was also constructed with the insertion of a DNA fragment with a BamHI site between the upstream (primers OL72/OL73b) and downstream (primers OL74b/ OL75) regions of slr1334, obtained by overlapping PCR with primers OL72 and OL75. This fragment was cloned into pGEM-T and an erythromycin resistance cassette was introduced in BamHI. This construction does not delete the complete slr1334 ORF, as it partially overlaps with the hypothetical protein encoded by sll1219. To generate the PMM/PGM-overexpressing strains OE:M and ∆PGM_OE:M, cultures of the WT and ∆PGM strains, respectively, were transformed with plasmid pnrsD_PcpcB_slr1334_SpR. To obtain this plasmid, the slr1334 ORF was extracted from plasmid pET28_slr1334 (plasmid described in the next section) and cloned in XbaI/XhoI in plasmid pnrsD_PcpcB_SpR. The ∆PMM_Pars:M strain was obtained by introducing a regulated copy of slr1334 in the WT background with plasmid pnrsD_arsB_ slr1334_SpR (slr1334 ORF from plasmid pET28_slr1334 cloned in XbaI/XhoI in plasmid pnrsD_ParsB_SpR) and then deleting endogenous slr1334 with plasmid pGT∆slr1334::Ery in the presence of arsenite. For generation of the ∆PGM∆PMM_Pars:M strain, the ∆PMM_Pars:M strain was transformed with plasmid pGT∆sll0726::Km. Strains and primers used in this work can be found in Table 1 and Supplementary Table S1, respectively. All plasmids were verified by Sanger sequencing, and complete segregation of the mutants was confirmed by PCR. Purification of recombinant Synechocystis PGM and PMM/ PGM, and generation of specific antibodies The sequence of the complete ORFs encoding PGM and PMM/PGM were obtained from genomic Synechocystis DNA by PCR with primer pairs OL78/OL79 and OL80/OL81, respectively (sequences in Supplementary Table S1). Note that the fragment amplified for PGM did not include the sequence corresponding to the first 16 amino acids as depicted in Cyanobase, as they were not conserved in other cyanobacterial genomes. These sequences were cloned in NdeI/EcoRI in pET28, resulting in the plasmids pET28_sll0726 and pET28_slr1334. Escherichia coli transformation and recombinant protein expression were performed following the instructions of the manufacturer of the pET vectors (Novagen). In brief, E. coli BL21 (DE3) cells were transformed Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 PMM/PGM overexpression compensates a PGM mutant | 1535 with these plasmids with a standard heat shock protocol and inoculated in kanamycin-supplemented Luria–Bertani medium. The expression of 6×His-tagged recombinant proteins was induced adding 0.2 mM isopropyl-β-d-thiogalactopyranoside (IPTG). After 24h of incubation at 25 °C, cells were harvested by centrifugation and resuspended in 50mM Tris–HCl pH 8.0, 500mM NaCl, 1mM phenylmethylsulfonyl fluoride (PMSF). Cells were lysed by sonication (20kHz, 75 W) on ice for 3min (in 30s periods). Lysates were centrifuged at 20 000 g for 30min and supernatants were filtered through 0.22 µM filters (Millipore). Recombinant PGM and PMM/PGM were purified by affinity chromatography with the Ni-NTA Purification System (NeoBiotech) following the manufacturer’s instructions. In brief, clarified supernatants were supplemented with imidazole to a final concentration of 25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 increasing concentrations of imidazole in 50mM Tris– HCl pH 8.0, 500mM NaCl. Next, purified proteins were desalted using PD-10 columns (GE Healthcare) pre-equilibrated with 50mM Tris–HCl pH 8.0, 150mM NaCl, 20% glycerol, and concentrated with 30K molecular weight cut-off Amicon Ultra centrifugal filter units (Millipore). 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). Oxygen evolution A Clark-type oxygen electrode (Hansatech) was used to measure oxygen evolution in cultures in a 30 °C chamber during 10 min at 50 µmol photon m−2 s−1. To prevent carbon limitation, cultures were supplemented with 10mM NaHCO3 just before measurements. Photosynthetic measurements Chlorophyll fluorescence was measured by pulse amplitude modulation (PAM) fluorometry with a Dual-PAM-100 (Walz) using intact cells at room temperature. Before recording, cells were adapted in the dark for 10min. A 250ms saturation pulse (5000 µmol photons m−2 s−1) was set for determination of the effective quantum yield of PSII [Y(II)]. Y(II) was determined with the formula (Fmʹ–Fs)/(Fmʹ) where Fmʹ is maximal fluorescence and Fs is basal fluorescence, both measured during exposure to actinic light at the same intensity used to culture the strains (50, 100, or 200 µE m−2 s−1). Other parameters analyzed were basal fluorescence in the dark (Fo) and maximum fluorescence measured in the presence of 20 µM DCMU (Fm). Glycogen content determination Glycogen content was determined as described in Gründel et al. (2012) with minor modifications. A 2ml aliquot of culture was harvested by centrifugation (15 000 g for 10min at 4 °C) and stored at –20 °C after the supernatant was carefully removed. Frozen pellets were resuspended in 30% KOH, vigorously mixed by vortexing for 3min, and incubated for 2h at 95 °C. Cold ethanol was added for glycogen precipitation and incubated overnight at –20 °C. Glycogen was purified after two cycles of centrifugation (15 000 g for 10min at 4 °C) and washed with cold ethanol, ensuring total elimination of KOH traces. After drying the pellet, it was resuspended in 100mM sodium acetate (pH 5.2) and digested with 10 U of amyloglucosidase (form Aspergillus niger, Sigma) at 55 °C overnight in parallel with a calibration curve using glycogen (from bovine liver, Sigma). Glucose released in the process was quantified with a coupled enzyme reaction of glucose oxidase (from A. niger, Sigma) and peroxidase (from horseradish, Sigma) in the presence of O-dianisidine (Sigma) to a final concentration of 200mg ml–1, 25 U ml–1, and 5 U ml–1, respectively, for 30min at 30 °C. The reaction was stopped by adding H2SO4 to a final concentration of 4.8N, and absorbance was registered at 540nm on a Varioskan multiplate reader (Thermo Fisher Scientific). Determination of extracellular pyruvate and 2-oxoglutarate Determination of metabolites in the extracellular medium was performed as in Díaz-Troya et al. (2020) with minor modifications. A 6ml aliquot of culture samples was centrifuged (15 000 g for 10min at 4 °C) and supernatants were quick-frozen in liquid nitrogen, lyophilized (VirTis BenchTop Pro Freeze dryer, SP Scientific), and stored at –20 °C until analyzed. Samples were resuspended in 600 µl of H2O, and 10–100 µl aliquots were analyzed by enzyme-coupled assays in 375 mM Tris–HCl (pH 7.5) with 0.11 mM NADH (and 50mM NH4Cl for 2-oxoglutarate quantification). The reaction was triggered by the addition of 5 mU of lactate dehydrogenase (Sigma) or 12 mU of glutamate dehydrogenase (Sigma) for the determination of pyruvate or 2-oxoglutarate, respectively. When NADH oxidation was completed, the remaining NADH was measured spectrophotometrically at 340nm to calculate the concentration of the metabolite compared with standard curves of known amounts of pyruvate (Sigma) and 2-oxoglutarate (Sigma). Quantification of intracellular osmolytes (GG and sucrose) Frozen pellets from 4ml culture aliquots were incubated in 1ml of 80% ethanol for 4h at 65 °C and the supernatants were recovered by centrifugation (15 000 g at room temperature for 5min and dried in a vacuum centrifuge. Pellets were solubilized in 100 µl of ultrapure water and filtered through 0.20 µm filters (Millex-GN Nylon, Millipore). Samples (10 μl) were analyzed by HPLC in a Waters LC Module I Plus system equipped with a Waters 410 Differential Refractometer detector using an Aminex HPX87H Column (BioRad), operated in isocratic mode (5mM H2SO4, 0.6ml min−1). Table 1. Strains used in this work Name Genotype Reference WT Wild-type Synechocystis sp. PCC 6803 ∆AGP ∆agp::C.C1 Díaz-Troya et al. (2014) ∆PGM ∆pgm::CK1 This work ∆PMMa∆pmm/pgm::Ery This work OE:M nrsD::PcpcB:pmm/pgm:SpΩ This work ∆PGM_OE:M ∆pgm::CK1, nrsD::PcpcB:pmm/pgm:SpΩ This work ∆PMM_Pars:M ∆pgm/pmm::Ery, nrsD::ParsB:pmm/pgm:SpΩ This work ∆PGM∆PMM_Pars:M ∆pgm::CK1,∆pmm/pgm::Ery, nrsD::ParsB:pmm/pgm:SpΩ This work Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 1536 | Ortega-Martínez et al. Quantification of intracellular hexose-phosphate For intracellular metabolite quantification, 1–3 OD750 nm pellets culture samples were quickly collected by centrifugation (15 000 g for 30s at 4 °C) and immediately frozen in liquid N2 for storage at –80 °C until metabolite extraction. Frozen pellets were carefully resuspended in 500 µl of ice-cold 70:30 methanol–chloroform mixture supplemented with 10 µM acetaminophen as internal standard. The samples were processed by five cycles of a 30s vortex and cooling on liquid nitrogen. The vortexing treatment was repeated three times with 15 resting intervals at –20 °C between them. Next, 500 μl of ultrapure ice-cold water was added to the tubes and thoroughly mixed by vortexing. Samples were centrifuged (15 000 g for 5min at 4 °C) and the upper polar phase was transferred to a new tube. To ensure maximum extraction, the apolar phase was subsequently mixed with another 250 μl of ultrapure ice-cold water and the polar phase was again collected and combined with the previous extraction. Samples were evaporated using a vacuum concentrator in a 4 °C room. Dried pellets were resuspended in 100 μl of ultrapure water and analyzed by LC-MS. Chromatographic separation was performed with an XSELECT HSS XP 150 mm×2.1 mm×2.5 μm clumn (Waters) in an Exion HPLC (Sciex) connected to a QTrap 6500+ (Sciex) operating in negative mode. Sample data were acquired and processed with Analyst and SciexOS software. For quantification of the total amount of the metabolites, different known concentrations of each standard were used. Cell lysates Cell lysates were obtained after mechanical disruption with glass beads in 50mM Tris–HCl (pH 8), 5mM NaCl, and 1mM PMSF by 10 cycles of 1min vortexing/resting on ice. After centrifugation (15 000 g 20min at 4 °C), the soluble fraction was recovered, and the protein was quantified with Bradford reagent (Biorad) using BSA (Sigma) as standard. Western blot Equal amounts of protein from cell lysates were resolved by SDS–PAGE (10% acrylamide/bis-acrylamide) and transferred to nitrocellulose membranes (GE Healthcare) during 1h at 0.8 mA cm–2. 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 with antibodies against AGP (1:20 000), GlgA1 (1:5000), GlgA2 (1:5000) (Díaz-Troya et al., 2014), PGM (1:50 000), PMM/PGM (1:20 000), or GroEL (1:75 000, Sigma-Aldrich) diluted in blocking solution. After incubation with primary antibody, membranes were rinsed four times for 10min in PBS-T and then incubated for 1h with horseradish peroxidase-conjugated secondary antibody (1:25 000 in blocking solution, Sigma). Membranes were then washed four times with PBS-T prior to chemiluminescent signal detection using Clarity™ western ECL Substrate (Biorad) and ImageQuant 800 imaging systems (Amersham). ImageQuant TL8.1 software was used for analysis. PGM and PMM/PGM quantification in cell extracts Quantification of the absolute amount of PGM and PMM/PGM in WT soluble cell extracts was carried out by comparison with purified recombinant PGM and PMM/PGM. Serial known amounts of purified recombinant PGM or PMM/PGM and serial amounts of soluble cell extracts from the WT strain were resolved by SDS–PAGE, probed with antibodies against PGM or PMM/PGM, and analyzed with ImageQuant TL8.1 software as described above under ‘Western blot’. For relative quantification of PMM/PGM levels in PMM/PGM-over- expressing strains compared with WT strains, equal amounts of soluble extracts from the different strains were resolved by SDS–PAGE, probed with antibodies against PMM/PGM, and analyzed as described above. Phosphoglucomutase activity in cell lysates For phosphoglucomutase activity, quantification-coupled enzymatic assay following the reduction of NAD+ due to G6P oxidation was performed as described in Lindahl and Florencio (2003) with minor modifications. A total amount of 50–200 µg of extract proteins of the soluble fraction of Synechocystis strains was mixed in buffer with a final concentration of 100 mM Tris–HCl pH 8, 4 mM MgCl2, 2.5mM NAD+, and 0.2 U of glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides (Sigma) in 200 µl. Absorbance at 340nm was measured and the reaction was triggered by adding 4mM G1P and 40 µM G16BP, and followed for 30min at 30 °C. In parallel, as background control for every sample, the same reaction without G1P addition was performed and the changes in absorbance were subtracted from the reaction with G1P. Results PGM is 10 times more abundant than PMM/PGM in Synechocystis As previously indicated, Synechocystis has two enzymes with phosphoglucomutase activity, PGM and PMM/PGM. Using specific antibodies raised against these two proteins, we could detect both by western blot under our standard growth conditions (cultures under constant 50 µE m−2 s−1 illumination and bubbled with 1% CO2-supplemented air). Quantification of these proteins in cell extracts (as described in the Materials and methods) indicated that PGM was ~10-fold more abundant than PMM/PGM (0.75±0.03ng of PGM and 0.07±0.02ng of PMM/PGM per µg of soluble cell extract). To better understand the role of these two proteins in glycogen metabolism, we tried to generate mutants lacking PGM and PMM/PGM. Consistent with previously described results (Liu et al., 2013), a mutant lacking the bifunctional enzyme PMM/PGM could not be segregated (∆PMM* strain), suggesting that PMM/PGM has an essential role in Synechocystis. In contrast, the deletion of sll0726 (coding for PGM) was easily segregated, generating the ∆PGM strain (Supplementary Fig. S1). The ∆PGM mutant accumulated moderate amounts of glycogen in all growth conditions tested, including cultures bubbled with CO2-supplemented air or non-bubbled flask cultures in media with different concentrations of NaHCO3 (Fig. 2B; Supplementary Fig. S2). After 6 d, the glycogen accumulated in the ∆PGM strain was 37% of that in the WT under our standard growth conditions (constant 50 µE m−2 s−1 illumination and 1% CO2-supplemented air) (Fig. 2B). 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, although less efficiently than PGM. Glycogen has been proposed to act not only as carbon storage but also as an energy buffer to maintain energy homeostasis in the cell (Cano et al., 2018). For this reason, we characterized the phenotype of the ∆PGM strain to analyze the effect of a limited carbon flux to and from glycogen under different growth conditions. We have chosen contexts in which the buffer role of glycogen could be more relevant, which requires Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 PMM/PGM overexpression compensates a PGM mutant | 1537 Fig. 2. Characterization of the phenotype of the ∆PGM strain in response to high light exposure. WT, ∆AGP, and ∆PGM cultures bubbled with 1% CO2-supplemented air at 50 µE m−2 s−1 light intensity were used to inoculate cultures at an initial OD750 nm of 0.5. These cultures were grown at different light intensities (50, 100, and 200 µE m−2 s−1) for 6 d. (A) Growth curves of the WT, ∆AGP, and ∆PGM strains at 50, 100, and 200 µE m−2 s−1 measured as optical density at 750nm. (B) Glycogen content of the WT, ∆AGP, and ∆PGM strains at 50, 100, and 200 µE m−2 s−1. Experiments at 200 µE m−2 s−1 were terminated after 5 d because ∆AGP and ∆PGM had whitened. (C) Effective quantum yields of energy conversion in PSII [Y(II)] of the WT, ∆AGP, and ∆PGM strains at 50, 100, and 200 µE m−2 s−1 measured by PAM fluorometry with an actinic light of 50, 100, and 200 µE m−2 s−1, respectively, during the first 48h of the experiment. (D) Photographs of WT, ∆AGP, and ∆PGM cultures after 0, 6, 24, and 48h growing at 50 µE m−2 s−1 or 200 µE m−2 s−1. (E) Concentration of pyruvate and α-ketoglutarate released to the culture media by the WT, ∆AGP, and ∆PGM cells after 0, 24, and 120h growing at 100 µE m−2 s−1. Data are means ±SD from four (A–C) or three (E) biological replicates. In (A–C), significant differences in the last time point compared with the value of the WT strain were determined using paired two-tailed Student’s t-test. The color of the asterisks matches the color of the corresponding strain. In (E), significant differences compared with the value of the WT strain were determined using unpaired two-tailed Student’s t-test: *P<0.05, **P<0.01, ***P<0.001. Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 1538 | Ortega-Martínez et al. increased accumulation or mobilization of carbon storage, or increased availability of ADP-glucose, such as high light, diel cycles, nitrogen deprivation, or salt stress. PGM is required to survive under high light exposure Under high light growth conditions, lack of glycogen synthesis causes overflow of organic acids to dissipate excess energy and negatively affects cell performance (Carrieri et al., 2012; Gründel et al., 2012; Cano et al., 2018). This suggests that limitations in carbon flux to glycogen in the ∆PGM strain could also negatively affect cell performance at higher light intensities. To test this, we analyzed the phenotype of the ∆PGM strain grown under different light intensities and compared it with the WT and the glycogen-less ∆AGP strain, which lacks the enzyme ADP-glucose pyrophosphorylase (Fig. 2). Cells grown under constant 50 µE m−2 s−1 illumination were maintained in this situation or transferred to 100 µE m−2 s−1 or 200 µE m−2 s−1. The WT and mutant strains showed similar growth rates at 50 µE m−2 s−1. However, the growth of the ∆AGP and ∆PGM strains was negatively affected at 100 µE m−2 s−1. When transferred to 200 µE m−2 s−1, the growth of the mutant strains was completely suppressed and the cultures turned white and died after 48–72h, while the growth of the WT strain was not affected (Fig. 2A, D). Glycogen content was also affected by light intensity (Fig. 2B). The WT strain accumulated up to 90 µg/OD750 nm independently of the light conditions, but this accumulation was faster at 100 µE m−2 s−1 and 200 µE m−2 s−1 than at 50 µE m−2 s−1 (Fig. 2B). After 6 d, the accumulation of glycogen in the ∆PGM strain at 100 µE m−2 s−1 was 36% higher than at 50 µE m−2 s−1, representing 49% of the glycogen content in the WT under the same conditions. However, the loss of viability in the ∆PGM strain after 5 d at 200 µE m−2 s−1 correlated with a decrease in glycogen content compared with its content at 50 µE m−2 s−1 (12.7 µg/OD750 nm versus 31.6 µg/OD750 nm, P<0.05) or 100 µE m−2 s−1 (43.0 µg/OD750 nm, P<0.01; unpaired two-tailed Student’s t-test) (Fig. 2B). As expected, no glycogen was detected in the ∆AGP strain in any condition (Fig. 2B). The shift to higher light intensity affected the photosynthetic performance of the WT and mutant strains. We used PAM to measure effective quantum yields of energy conversion in PSII [Y(II)] of the cultures during the first 48h after being transferred to the different growth light intensities (Fig. 2C). Measurements were carried out using the same light intensity used to culture the strains (50, 100, or 200 µE m−2 s−1). Y(II) at 50 µE m−2 s−1 was similar for the three strains. An increase in the light intensity immediately caused a similar decrease in the Y(II) in the WT and mutant strains. However, only the WT strain was able to adapt to the higher light intensities. After 24h at 100 µE m−2 s−1 and 200 µE m−2 s−1, the WT strain recovered Y(II) values that were 80% and 51%, respectively, of the value at 50 µE m−2 s−1 (Fig. 2C). However, after 24h at 100 µE m−2 s−1, Y(II) of the ∆AGP and ∆PGM strains was still 42% and 65%, respectively, of Y(II) at 50 µE m−2 s−1. The effect of high light on photosynthetic performance was more dramatic at 200 µE m−2 s−1, with a Y(II) value of zero for both mutant strains after 24h (Fig. 2C). Furthermore, unlike the WT, the ∆AGP and ∆PGM strains secreted similar levels of the organic acids pyruvate and α-ketoglutarate at 100 µE m−2 s−1 (Fig. 2E). All this indicates that the reduced carbon flux between G1P and G6P present in the ∆PGM strain, which only allows the accumulation of limited levels of glycogen, was not sufficient to buffer the excess energy at higher light intensities. Lack of PGM causes growth retardation under shortday conditions Glycogen also plays a significant role during light–dark periods, and the growth and viability of glycogen-less mutants are affected under these conditions (Gründel et al., 2012; Welkie et al., 2019). We analyzed the effect of limited synthesis of glycogen in the ∆PGM strain under long-day (16h light/8h dark) and short-day (8h light/16h dark) regimes (Fig. 3). Similarly to the WT, glycogen content in the ∆PGM strain increased during light and decreased in the dark in both regimes, although to a lesser extent than in the WT. Despite reduced amounts of glycogen in the ∆PGM strain compared with the WT, both grew quite similarly under a long-day regime (Fig. 3A). Surprisingly, under these conditions, the ∆AGP strain also showed a growth rate close to that of the WT and ∆PGM strains in the absence of detectable glycogen (Fig. 3A). Thus, after 200h under a long-day regime, the three strains reached a similar final OD750 (OD 750 nm=2.2±0.1) (Fig. 3A). In contrast, after 200 h growth in a short-day regime, the ∆PGM culture reached a lower final OD750 nm than the WT, while the growth of the ∆AGP strain was completely abolished (Fig. 3B), indicating the importance of glycogen storage during longer dark periods. PGM is required for an adequate response to nitrogen deprivation In addition to light-dependent phenotypes, the response to nitrogen deficiency is also altered in glycogen-less mutants (Gründel et al., 2012; Jackson et al., 2015; Díaz-Troya et al., 2020). Thus, WT cells during nitrogen deficiency conditions accumulate large amounts of glycogen and degrade their phycobiliproteins to recycle the nitrogen present in them and adjust to the new nutritional situation. This causes the loss of their characteristic blue-green color, that turns into a yellow color, a process known as bleaching or chlorosis (Forchhammer and Schwarz, 2019). Mutants unable to accumulate glycogen do not degrade their phycobiliproteins and compromise their viability in nitrogen deprivation conditions (Gründel et al., 2012). As shown in Fig. 4, the lack of PGM clearly impaired the adequate adaptation to nitrogen deficiency. Thus, similarly to the ∆AGP strain, when the ∆PGM strain was transferred to a medium without nitrogen the growth of the culture ceased immediately Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 PMM/PGM overexpression compensates a PGM mutant | 1539 and degradation of phycobiliproteins did not occur (Fig. 4A, B; Supplementary Fig. S3A). This happened even though the ∆PGM strain accumulated glycogen, although in much lower amounts than the WT (32 µg/OD750 versus 160 µg/OD750) (Fig. 4C). Under these conditions, carbon flux was clearly limited in the ∆PGM strain, causing excretion of organic acids at levels similar to those of the ∆AGP strain (Fig. 4D, E). One of the effects of nitrogen deficiency is the decrease in photosynthetic activity (Krasikov et al., 2012; Ogawa and Sonoike, 2016). We observed this decrease in both the WT and mutant strains, despite the presence of phycobiliproteins in the latter (Fig. 4F). Therefore, net oxygen evolution was residual in the WT, ∆AGP, or ∆PGM strains 48 h after nitrogen removal (Fig. 4F). To investigate how photosynthesis is affected in the ∆PGM strain in this situation, we tested its photosynthetic performance by using PAM analysis (Fig. 4G; Supplementary Fig. S3C). In the WT, nitrogen deprivation produced a 45% decrease in Y(II) after 48h compared with nitrogen-replete conditions, and after 6 d this value was zero (Fig. 4G). The WT showed a decrease in both basal fluorescence (Fo) and maximum fluorescence (Fm) due to degradation of the phycobilisomes (Supplementary Fig. S3C). Y(II) values of the ∆AGP and ∆PGM strains in nitrogen-repleted medium were equivalent to that of the WT (Fig. 4G). 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h (Fig. 4G). Furthermore, in contrast to the WT, Fo increased in the ∆PGM strain (Supplementary Fig. S3C), suggesting an increase in free phycobilisomes (Acuña et al., 2016; Stirbet et al., 2019). Fig. 3. Growth curves and glycogen content of the ∆PGM strain cultivated under long-day and short-day diel regimes. WT, ∆AGP, and ∆PGM cultures inoculated at an OD750 nm of 0.5 in Erlenmeyer flasks were cultivated under (A) long-day (16h light/8h dark) or (B) short-day (8h light/16h dark) regimes at 40 µE m−2 s−1 for 9 d. White and gray areas represent light and dark periods, respectively. Growth (measured as OD750) and glycogen content were determined at the indicated time points. Data are means ±SD from three biological replicates. Significant differences in OD750 nm or glycogen content at the end of the experiment compared with the WT strain were determined using paired two-tailed Student’s t-test: *P<0.05, **P<0.01, ***P<0.001. The color of the asterisks matches the color of the corresponding strain. Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 1540 | Ortega-Martínez et al. An adequate response to nitrogen deprivation is essential for survival. Under these conditions, Synechocystis cells enter a dormant-like state in which they can survive for at least 1 month. They are able to recover vegetative growth once nitrogen is available again (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 d of nitrogen starvation, ∆AGP and ∆PGM cultures had turned white, lost photosynthetic pigments, and the viability of the cultures was severely compromised (Supplementary Fig. S4). Recovery from this dormancy state induced by nitrogen deprivation requires mobilization of glycogen to provide energy to restore the translational machinery, ATP synthesis, and nitrate assimilation, and finally for synthesis of the photosynthetic apparatus (Neumann et al., 2021). 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 (Fig. 5). As mentioned above, both strains showed Fig. 4. Characterization of the phenotype of the ∆PGM strain in response to nitrogen deprivation. WT, ∆AGP, and ∆PGM cells were grown in BG11C to mid-exponential growth phase, washed twice in nitrogen-free medium, and transferred to nitrogen-free medium at an OD750 nm of 1. (A) Photographs of the WT, ∆AGP, and ∆PGM cultures 0, 24, 48, and 144h after nitrogen removal. (B) Growth curves of the WT, ∆AGP, and ∆PGM strains cultivated in nitrogen-free medium measured as optical density at 750nm. (C) Glycogen content of the WT, ∆AGP, and ∆PGM strains at different time points after nitrogen removal. Concentration of extracellular (D) pyruvate and (E) α-ketoglutarate in WT, ∆AGP, and ∆PGM cultures 0, 24, 48, and 144h after nitrogen removal. (F) Oxygen evolution of the WT, ∆AGP, and ∆PGM cultures at different times after nitrogen removal measured by a Clark-type electrode at the growth light intensity (50 µE m−2 s−1). (G) PSII quantum yields [Y(II)] in WT, ∆AGP, and ∆PGM cultures measured by PAM fluorometry at the growth light intensity (50 µE m−2 s−1) 0, 24, 48, and 144h after nitrogen removal. Data are means ±SD from four (B, D–G) or three (C) biological replicates. In (B and C), significant differences in OD750 nm or glycogen content at 144h after nitrogen removal compared with the WT strain were determined using paired two-tailed Student’s t-test. The color of the asterisks matches the color of the corresponding strain. In (D and E), significant differences compared with the value of the WT strain at the same temporal point were determined using unpaired two-tailed Student’s t-test: *P<0.05, **P<0.01, ***P<0.001. Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 PMM/PGM overexpression compensates a PGM mutant | 1547 et al., 2022), emphasized its role as a key element in metabolism. This is particularly evident as mutants carrying a nonphosphorylatable version of PGM are not able to survive the dormant state induced by nitrogen deprivation (Doello et al., 2022). In this work, we have focused on the role of PGM in the connection between central carbon metabolism and glycogen. We have always detected glycogen in our mutant lacking PGM in Synechocystis, with levels ranging between ~50% and 15% with respect to the WT depending on the culture conditions (Figs 2–6; Supplementary Fig. S2). There are discrepancies in this aspect in previous work on Synechocystis. While Liu et al. (2013) also detected glycogen accumulation in a Synechocystis mutant lacking PGM, Doello et al. (2022) reported no glycogen in their mutant. These discrepancies in glycogen content may not be attributed to different culture conditions, as we have tried a wide variety of growth conditions and glycogen was detected in all of them (Figs 2–6; Supplementary Fig. S2). However, we cannot rule out an effect of the genetic background, as several WT backgrounds are used across different laboratories, or even the presence of spontaneous mutations acquired during the segregation process. Reports in the literature on the storage polysaccharide content of mutants lacking PGM in other organisms range from a 50% reduction in glycogen levels in the double pgm1/2 mutant in yeast (Boles et al., 1994; Daran et al., 1997) to near starchless in the plastidial PGM (pPGM) mutant in Arabidopsis (Bahaji et al., 2014) or complete glycogen less in E. coli (Eydallin et al., 2007). In the case of Arabidopsis, the residual starch content in the absence of pPGM has been proposed to be due to the existence of an alternative source of ADP-Glc that would not require the action of pPGM or AGP (Bahaji et al., 2014). This would not be the case in Synechocystis, as mutants lacking AGP are completely devoid of glycogen. In contrast, it is more plausible that, due to its phosphoglucomutase activity, PMM/ PGM would be the main candidate for G1P synthesis in glycogen metabolism in the absence of PGM. As PMM/PGM is essential, it is not easy to ascribe glycogen synthesis detected in the ∆PGM strain to PMM/PGM, because suppression of PMM/PGM expression affects cell viability. However, our data show that down-regulation of PMM/PGM to non-lethal levels causes decreased glycogen levels, indicating that PMM/ PGM is involved in glycogen synthesis at least in the absence of PGM. Phosphoglucomutase activity has been described in enzymes belonging to other subgroups of the phosphohexomutase superfamily due to some degree of substrate promiscuity (Mitra et al., 2010; Bandini et al., 2012; Patel et al., 2019). Although Synechocystis has an enzyme assigned to the PNGM subgroup, our data strongly suggest a negligible effect, if any, of any other enzyme in addition to PGM and PMM/PGM in the interconversion of G1P and G6P in glycogen metabolism (Fig. 8D). Furthermore, overexpression of PMM/PGM in a strain lacking PGM increases the glycogen content and recovers the response to different types of stresses (Fig. 7; Supplementary Fig. S6) which reinforces the secondary role of PMM/PGM in glycogen metabolism. Due to the presence of PMM/PGM in Synechocystis, in the ∆PGM strain the carbon flux between central carbon metabolism and carbon storage is not completely blocked, as in extensively characterized mutants lacking agp or glgA genes, but is constricted by the limited capacity of PMM/PGM. This provides additional information on the degree of dependence on glycogen metabolism for an adequate response to environmental stresses. The stresses examined can be classified as low/ moderate and highly dependent on glycogen metabolism. The low/moderate dependency category includes salt stress and long-day conditions. The synthesis of the osmolytes GG and sucrose to face salt stress is based on G1P (Fig. 1). This metabolite can be generated from photosynthetic fixed carbon through PGM (and/or PMM/PGM) or by glycogen mobilization by glycogen phosphorylases. In this regard, it is also interesting to note that the addition of NaCl to the WT or ∆PGM strains induced the mobilization of glycogen, which was greater in the WT than in the ∆PGM strain (Fig. 6D). Consistent with previous results in Synechococcus elongatus PCC 7942 (Qiao et al., 2018), this suggests that at least part of the G1P required for osmolyte synthesis could be derived from glycogen mobilization to G1P by glycogen phosphorylases. This would also explain the slower kinetics of GG accumulation in the ∆PGM strain (with reduced glycogen levels) (Fig. 6B) and, mainly, in sucrose accumulation in the glycogen-less ∆AGP strain (Fig. 6C), although G1P levels seem higher in the ∆PGM than in the WT strain (Fig. 7G). Glycogen does not appear to play a critical role in survival of Synechocystis cells in long-day light–dark regimes (Fig. 3A). However, exposure to cycles with extended dark periods (16h dark/8h light) had a negative impact on the ∆AGP and ∆PGM strains (Fig. 3B). In this case, the existence of a flux, although limited, between glycogen and central carbon metabolism in the ∆PGM strain represents a clear advantage over the glycogen-less ∆AGP strain, which is not viable under these conditions (Fig. 3B). Glycogen would act here as a source of carbon and reducing power through its respiration in the dark (Preiss, 1984; Gründel et al., 2012) and to replenish the intermediates of the Calvin–Benson cycle depleted during the dark period (Makowka et al., 2020; Shinde et al., 2020). On the other hand, both exposure to high light intensity and response to nitrogen deficiency require a fully functional glycogen metabolism. In these situations, the phosphoglucomutase activity of the endogenous PMM/PGM is clearly insufficient and the ∆PGM strain has a phenotype like the ∆AGP strain (Figs 2, 4). In the case of the response to nitrogen deficiency, one of the main characteristics of adaptation to this nutritional condition is a fast and prominent accumulation of glycogen that reaches its maximum ~48 h after nitrogen removal under our conditions (Fig. 4C). During this period, carbon is relocated from the recycling of phycobiliproteins, which will be partially accumulated as glycogen (Forchhammer Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 1548 | Ortega-Martínez et al. and Schwarz, 2019). This process involves an intense flux of carbon into glycogen that requires high levels of interconversion between G6P and G1P provided by PGM that cannot be supplied by endogenous PMM/PGM and results in an accumulation of G6P in the ∆PGM strain (Fig. 7G). Nitrogen starvation also induces in the ∆PGM strain a metabolic overflow and a blockage of the bleaching process as occurs with the ∆AGP strain (Fig. 4; Supplementary Fig. S3), even when the ∆PGM strain accumulates reduced glycogen levels (Fig. 4C). In any case, these reduced glycogen levels represent an advantage for the ∆PGM strain, which survives longer periods under nitrogen deficiency than the ∆AGP strain (Supplementary Fig. S4). It is interesting to note that after nitrogen replenishment, the ∆PGM strain can resume vegetative growth at rates comparable with those of the WT (Fig. 5B) despite the reduced glycogen levels (Fig. 5C). This could be due to a lower energy requirement for the awakening program, as the phycobilisomes were still present (Fig. 5A; Supplementary Fig. S3A). Finally, one of the classic strategies in metabolic modification of cyanobacteria for biotechnological purposes has been to eliminate carbon reserves, mainly glycogen, to redirect the fixed carbon toward the production of compounds of interest. This strategy has generally been carried out by deleting agp or glgA genes. However, these modifications generally lead to a reduction in fitness and sensitivity to various stresses (reviewed in Luan et al., 2019), including those analyzed in this work, which ultimately reduce the productivity of the modified strains. Reducing, but not completely abolishing, glycogen storage by limiting phosphoglucomutase activity, as described here, is an alternative that improves stress resistance and deserves further exploration. Supplementary data The following supplementary data are available at JXB online. Table S1. Oligonucleotides used in this work. Fig. S1. Generation and segregation of the ∆PGM and ∆PMM* mutant strains. Fig. S2. Glycogen accumulation in the ∆PGM strain cultivated in media with different availabilities of HCO3. Fig. S3. Degradation of the phycobilisomes and Fo and Fm photosynthetic parameters of the ∆PGM strain during nitrogen deprivation. Fig. S4. Appearance and recovery of the WT, ∆AGP, and ∆PGM strains after 15 d of nitrogen deprivation. Fig. S5. Phosphoglucomutase activity of WT, OE:M, ∆PGM, and ∆PGM_OE:M after 24h exposure to high light, nitrogen deprivation, or salt stress, and levels of proteins involved in glycogen synthesis. Fig. S6. Phenotype of PMM/PGM-overexpressing strains grown under diel regimes. Fig. S7. Phenotype of OE:M under high light. Fig. S8. Generation of the ∆PMM_Pars:M strain. Fig. S9. Growth and PMM/PGM levels of the ∆PGM∆PMM_ Pars:M strain cultivated with different amounts of arsenite. Acknowledgements The authors thank Dr Luis López-Maury for plasmids pnrsD_PcpcB_ SpR and pnrsD_ParsB_SpR. Author contributions PO-M, MR, and SD-T: performing experiments; PO-M, SD-T, and FJF: data analysis and writing the manuscript; SD-T and FJF: design and supervision of the overall study; FJF: funding acquisition. Conflict of interest The authors declare no conflict of interest. Funding This work was supported by grants PID2019-104513GB-I00/ AEI/10.13039/501100011033 from Agencia Estatal de Investigación and by Junta de Andalucia Group BIO-284, co-financed by European Regional Funds to FJF. PO-M is a recipient of grant FPU18/06580 from the Ministerio de Universidades. Data availability All data supporting the findings of this study are available within the paper and within its supplementary data published online. References Acuña AM, Snellenburg JJ, Gwizdala M, Kirilovsky D, Van Grondelle R, Van Stokkum IHM. 2016. Resolving the contribution of the uncoupled phycobilisomes to cyanobacterial pulse-amplitude modulated (PAM) fluorometry signals. Photosynthesis Research 127, 91–102. Bahaji A, Baroja-Fernández E, Sánchez-López AM, et al. 2014. HPLC-MS/MS analyses show that the near-Starchless aps1 and pgm leaves accumulate wild type levels of ADPglucose: further evidence for the occurrence of important ADPglucose biosynthetic pathway(s) alternative to the pPGI–pPGM–AGP pathway. PLoS One 9, e104997. Ball S, Colleoni C, Arias MC. 2015. The transition from glycogen to starch metabolism in cyanobacteria and eukaryotes. In: Nakamura Y, ed. Starch: metabolism and structure. Tokyo: Springer Japan, 93–158. Ball SG, Morell MK. 2003. From bacterial glycogen to starch: understanding the biogenesis of the plant starch granule. Annual Review of Plant Biology 54, 207–233. Bandini G, Mariño K, Güther MLS, Wernimont AK, Kuettel S, Qiu W, Afzal S, Kelner A, Hui R, Ferguson MAJ. 2012. Phosphoglucomutase is absent in Trypanosoma brucei and redundantly substituted by phosphomannomutase and phospho-N-acetylglucosamine mutase. Molecular Microbiology 85, 513–534. Boles E, Liebetrau W, Hofmann M, Zimmermann FK. 1994. A family of hexosephosphate mutases in Saccharomyces cerevisiae. European Journal of Biochemistry 220, 83–96. Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 PMM/PGM overexpression compensates a PGM mutant | 1549 Cano M, Holland SC, Artier J, Burnap RL, Ghirardi M, Morgan JA, Yu J. 2018. Glycogen synthesis and metabolite overflow contribute to energy balancing in cyanobacteria. Cell Reports 23, 667–672. Carreras J, Bartrons R, Climent F, Cusso R. 1986. Bisphosphorylated metabolites of glycerate, glucose, and fructose: functions, metabolism and molecular pathology. Clinical Biochemistry 19, 348–358. Carrieri D, Paddock T, Maness PC, Seibert M, Yu J. 2012. Photocatalytic conversion of carbon dioxide to organic acids by a recombinant cyanobacterium incapable of glycogen storage. Energy and Environmental Science 5, 9457–9461. Cifuente JO, Comino N, Trastoy B, D’Angelo C, Guerin ME. 2019. Structural basis of glycogen metabolism in bacteria. The Biochemical Journal 476, 2059–2092. Daran JM, Bell W, François J. 1997. Physiological and morphological effects of genetic alterations leading to a reduced synthesis of UDP-glucose in Saccharomyces cerevisiae. FEMS Microbiology Letters 153, 89–96. Díaz-Troya S, López-Maury L, Sánchez-Riego AM, Roldán M, Florencio FJ. 2014. Redox regulation of glycogen biosynthesis in the cyanobacterium Synechocystis sp. PCC 6803: analysis of the AGP and glycogen synthases. Molecular Plant 7, 87–100. Díaz-Troya S, Roldán M, Mallén-Ponce MJ, Ortega-Martínez P, Florencio FJ. 2020. Lethality caused by ADP-glucose accumulation is suppressed by salt-induced carbon flux redirection in cyanobacteria. Journal of Experimental Botany 71, 2005–2017. Doello S, Neumann N, Forchhammer K. 2022. Regulatory phosphorylation event of phosphoglucomutase 1 tunes its activity to regulate glycogen metabolism. The FEBS Journal 289, 6005–6020. Egli B, Kölling K, Köhler C, Zeeman SC, Streb S. 2010. Loss of cytosolic phosphoglucomutase compromises gametophyte development in Arabidopsis. Plant Physiology 154, 1659–1671. Eydallin G, Viale AM, Morán-Zorzano MT, Muñoz FJ, Montero M, Baroja-Fernández E, Pozueta-Romero J. 2007. Genome-wide screening of genes affecting glycogen metabolism in Escherichia coli K-12. FEBS Letters 581, 2947–2953. Forchhammer K, Schwarz R. 2019. Nitrogen chlorosis in unicellular cyanobacteria—a developmental program for surviving nitrogen deprivation. Environmental Microbiology 21, 1173–1184. García-Cañas R, Giner-Lamia J, Florencio FJ, López-Maury L. 2021. A protease-mediated mechanism regulates the cytochrome c6/plastocyanin switch in Synechocystis sp. PCC 6803. Proceedings of the National Academy of Sciences, USA 118, e2017898118. García-Domínguez M, Lopez-Maury L, Florencio FJ, Reyes JC. 2000. A gene cluster involved in metal homeostasis in the cyanobacterium Synechocystis sp. strain PCC 6803. Journal of Bacteriology 182, 1507–1514. Gründel M, Scheunemann R, Lockau W, Zilliges Y. 2012. Impaired glycogen synthesis causes metabolic overflow reactions and affects stress responses in the cyanobacterium Synechocystis sp. PCC 6803. Microbiology 158, 3032–3043. Jackson SA, Eaton-Rye JJ, Bryant DA, Posewitz MC, Davies FK. 2015. Dynamics of photosynthesis in a glycogen-deficient glgC mutant of Synechococcus sp. strain PCC 7002. Applied and Environmental Microbiology 81, 6210–6222. Kirsch F, Klähn S, Hagemann M. 2019. Salt-regulated accumulation of the compatible solutes sucrose and glucosylglycerol in cyanobacteria and its biotechnological potential. Frontiers in Microbiology 10, 2139. Krasikov V, Aguirre von Wobeser E, Dekker HL, Huisman J, Matthijs HCP. 2012. Time-series resolution of gradual nitrogen starvation and its impact on photosynthesis in the cyanobacterium Synechocystis PCC 6803. Physiologia Plantarum 145, 426–439. Lindahl M, Florencio FJ. 2003. Thioredoxin-linked processes in cyanobacteria are as numerous as in chloroplasts, but targets are different. Proceedings of the National Academy of Sciences, USA 100, 16107–16112. Liu L, Hu HH, Gao H, Xu XD. 2013. Role of two phosphohexomutase genes in glycogen synthesis in Synechocystis sp. PCC6803. Chinese Science Bulletin 58, 4616–4621. López-Maury L, Florencio FJ, Reyes JC. 2003. Arsenic sensing and resistance system in the cyanobacterium Synechocystis sp. strain PCC 6803. Journal of Bacteriology 185, 5363–5371. Luan G, Zhang S, Wang M, Lu X. 2019. Progress and perspective on cyanobacterial glycogen metabolism engineering. Biotechnology Advances 37, 771–786. Mackay MA, Norton RS, Borowitzka LJ. 1984. Organic osmoregulatory solutes in cyanobacteria. Journal of General Microbiology 130, 2177–2191. Makowka A, Nichelmann L, Schulze D, Spengler K, Wittmann C, Forchhammer K, Gutekunst K. 2020. Glycolytic shunts replenish the Calvin–Benson–Bassham cycle as anaplerotic reactions in cyanobacteria. Molecular Plant 13, 471–482. Mallen-Ponce MJ, Huertas MJ, Sanchez-Riego AM, Florencio FJ. 2021. Depletion of m-type thioredoxin impairs photosynthesis, carbon fixation, and oxidative stress in cyanobacteria. Plant Physiology 187, 1325–1340. McFarland AL, Bhattarai N, Joseph M, Winkler ME, Martin JE. 2021. Cellular Mn/Zn ratio influences phosphoglucomutase activity and capsule production in Streptococcus pneumoniae D39. Journal of Bacteriology 203, e0060220. Miao X, Wu Q, Wu G, Zhao N. 2003. Sucrose accumulation in salt-stressed cells of agp gene deletion-mutant in cyanobacterium Synechocystis sp PCC 6803. FEMS Microbiology Letters 218, 71–77. Mills LA, McCormick AJ, Lea-Smith DJ. 2020. Current knowledge and recent advances in understanding metabolism of the model cyanobacterium Synechocystis sp. PCC 6803. Bioscience Reports 40, BSR20193325. Mitra S, Cui J, Robbins PW, Samuelson J. 2010. A deeply divergent phosphoglucomutase (PGM) of Giardia lamblia has both PGM and phosphomannomutase activities. Glycobiology 20, 1233–1240. Naught LE, Tipton PA. 2005. Formation and reorientation of glucose 1,6-bisphosphate in the PMM/PGM reaction: transient-state kinetic studies. Biochemistry 44, 6831–6836. Neumann N, Doello S, Forchhammer K. 2021. Recovery of unicellular cyanobacteria from nitrogen chlorosis: a model for resuscitation of dormant bacteria. Microbial Physiology 31, 78–87. Neumann N, Friz S, Forchhammer K. 2022. Glucose-1,6-bisphosphate, a key metabolic regulator, is synthesized by a distinct family of α-phosphohexomutases widely distributed in prokaryotes. mBio 13, e0146922. Ogawa T, Sonoike K. 2016. Effects of bleaching by nitrogen deficiency on the quantum yield of photosystem II in Synechocystis sp. PCC 6803 revealed by Chl fluorescence measurements. Plant and Cell Physiology 57, 558–567. Patel V, Black KA, Rhee KY, Helmann JD. 2019. Bacillus subtilis PgcA moonlights as a phosphoglucosamine mutase in support of peptidoglycan synthesis. PLoS Genetics 15, e1008434. Preiss J. 1984. Bacterial glycogen synthesis and its regulation. Annual Review of Microbiology 38, 419–458. Qiao C, Duan Y, Zhang M, Hagemann M, Luo Q, Lu X. 2018. Effects of reduced and enhanced glycogen pools on salt-induced sucrose production in a sucrose-secreting strain of Synechococcus elongatus PCC 7942. Applied and Environmental Microbiology 84, e02023-17. Ray WJ, Roscelli GA. 1964. A kinetic study of the phosphoglucomutase pathway. Journal of Biological Chemistry 239, 1228–1236. Regni C, Naught L, Tipton PA, Beamer LJ. 2004. Structural basis of diverse substrate recognition by the enzyme PMM/PGM from P. aeruginosa. Structure 12, 55–63. Regni C, Tipton PA, Beamer LJ. 2002. Crystal structure of PMM/PGM: an enzyme in the biosynthetic pathway of P. aeruginosa virulence factors. Structure 10, 269–279. Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY. 1979. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Microbiology 111, 1–61. Shackelford GS, Regni CA, Beamer LJ. 2004. Evolutionary trace analysis of the alpha-d-phosphohexomutase superfamily. Protein Science 13, 2130–2138. Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024 1550 | Ortega-Martínez et al. Shinde S, Zhang X, Singapuri SP, Kalra I, Liu X, Morgan-Kiss RM, Wang X. 2020. Glycogen metabolism supports photosynthesis start through the oxidative pentose phosphate pathway in cyanobacteria. Plant Physiology 182, 507–517. Spät P, Klotz A, Rexroth S, Maček B, Forchhammer K. 2018. Chlorosis as a developmental program in cyanobacteria: the proteomic fundament for survival and awakening. Molecular and Cellular Proteomics 17, 1650–1669. Stiers KM, Graham AC, Zhu JS, Jakeman DL, Nix JC, Beamer LJ. 2019. Structural and dynamical description of the enzymatic reaction of a phosphohexomutase. Structural Dynamics 6, 024703. Stiers KM, Muenks AG, Beamer LJ. 2017. Biology, mechanism, and structure of enzymes in the α-d-phosphohexomutase superfamily. Advances in Protein Chemistry and Structural Biology 109, 265–304. Stirbet A, Lazár D, Papageorgiou GC, Govindjee. 2019. Chlorophyll a fluorescence in cyanobacteria: relation to photosynthesis. In: Mishra AK, Tiwari DN, Rai AN, eds. Cyanobacteria. From basic science to applications. London: Academic Press, 79–130. Tegtmeyer LC, Rust S, van Scherpenzeel M, et al. 2014. Multiple phenotypes in phosphoglucomutase 1 deficiency. New England Journal of Medicine 370, 533–542. Ugalde JE, Czibener C, Feldman MF, Ugalde RA. 2000. Identification and characterization of the Brucella abortus phosphoglucomutase gene: role of lipopolysaccharide in virulence and intracellular multiplication. Infection and Immunity 68, 5716–5723. Welkie DG, Rubin BE, Diamond S, Hood RD, Savage DF, Golden SS. 2019. A hard day’s night: cyanobacteria in diel cycles. Trends in Microbiology 27, 231–242. Zhou J, Zhang H, Meng H, Zhu Y, Bao G, Zhang Y, Li Y, Ma Y. 2014. Discovery of a super-strong promoter enables efficient production of heterologous proteins in cyanobacteria. Scientific Reports 4, 4500. Downloaded from https://academic.oup.com/jxb/article/74/5/1532/6862049 by Sevilla Psych Lib B21 user on 26 February 2024