The cycle of day and night is one of the most recurrent and predictable environmental changes on our planet. Consequently, organisms have evolved mechanisms that allow them to measure time over 24 hours and prepare for the periodic changes between light and dark. These mechanisms, known as circadian clocks, alter the activity of some of the organism’s genes in a rhythmic way across the course of a day. This in turn causes certain behaviors and biological activities of the organism to follow a daily cycle.

The bacterium Synechococcus elongatus needs to be able to track the daily cycle of light and dark because it performs photosynthesis and depends on sunlight to form sugars, which can later be broken down to release energy. The time information encoded in the circadian clock of S. elongatus is transmitted to the protein RpaA, which drives the regular circadian changes in gene activity in the cell. If RpaA is removed from the cell or prevented from working, S. elongatus can no longer control rhythmic gene activity and is unable to survive the night.

To better understand how the circadian system schedules biological tasks to help an organism to survive, Puszynska and O'Shea studied S. elongatus cells. This revealed that the bacteria normally prepare for darkness by storing sugars during the day and activating several genes at dusk to make enzymes that are required to break down stored sugars. This provides the cells with energy that they need to survive the night. But mutant cells that lack the gene that produces RpaA do not prepare for darkness; they do not accumulate a store of sugars during the day or activate the vital genes at dusk. They have low internal energy levels in the dark and they cannot survive long periods of darkness.

Providing the mutant cells with sugar and restoring the activity of the genes responsible for breaking down sugar enabled the cells to maintain energy in darkness and survive the night. It therefore appears that one role of the circadian system of S. elongatus is to coordinate building up sugar reserves during the day with breaking down sugar stores to generate energy during the night.

Puszynska and O'Shea also found many other genes that are not activated at dusk in the mutant cells. It will therefore be important to study whether other pathways that help cells to survive and grow are defective in these mutant cells.

Organisms across kingdoms of life have evolved circadian clocks to temporally align biological activities with the diurnal changes in the environment. In the cyanobacterium Synechococcus elongatus PCC7942, the core circadian oscillator is comprised of the KaiA, KaiB and KaiC proteins (Nishiwaki et al., 2007; Rust et al., 2007) that generate oscillations in the phosphorylation state of KaiC. Time information encoded in the phosphorylation state of KaiC is transmitted to the transcription factor RpaA (Takai et al., 2006; Taniguchi et al., 2010; Markson et al., 2013) to generate circadian changes in gene expression (Markson et al., 2013). In continuous light, the expression of more than 60% of protein-coding genes in S. elongatus is regulated in a circadian manner (Vijayan et al., 2009) with two main phases of gene expression: genes peaking at subjective dawn or subjective dusk, where the term ‘subjective’ refers to an internal estimate of time in the absence of external cues. Deletion of rpaA disrupts these rhythms in mRNA abundance and arrests cells in a subjective dawn-like transcriptional state, rendering them unable to switch to a subjective dusk-like expression program (Markson et al., 2013).

While it is clear that the circadian transcriptional program schedules timing of gene expression when the clock is free-running in the absence of changes in external light, it is unclear how circadian control of gene expression is used under more physiologically relevant conditions when light and dark periods alternate. Exposure to darkness restricts energy availability, creating unique metabolic demands in cyanobacteria which depend on sunlight for energy production through photosynthesis. Strikingly, the rpaA^- strain exhibits defects in cell growth and viability in cyclic, but not in constant light environments (Takai et al., 2006), suggesting an important role for circadian regulation of gene expression in alternating light/dark cycles.

We find that both accumulation and utilization of the carbon reserves required for energy production during periods of darkness are defective in the rpaA^- strain, and that correction of these defects restores viability. Our results provide insight into the role of the circadian output pathway in enhancing fitness of cyanobacteria through coordination of central carbon metabolism with the metabolic demands imposed by periodic changes in the external environment.

As reported previously (Takai et al., 2006), in constant light wild type and the rpaA^- cells grow at the same rate (Figure 1A, top panel); however, the rpaA^- strain is not viable when cultured in alternating light/dark conditions (Figure 1A, bottom panel and Figure 1—figure supplement 1A). The rpaA^- strain rapidly loses viability when incubated in the dark (Figure 1—figure supplement 1B), suggesting that the defect is induced by darkness and not by repeated light-to-dark and dark-to-light transitions. Complementation of the rpaA^- strain with rpaA expressed from an ectopic site in the genome fully restores viability (Figure 1—figure supplement 1A and Figure 1—figure supplement 1B). In wild type cells, expression of the kaiBC genes depends on RpaA and thus deletion of rpaA abrogates the function of the KaiABC oscillator, altering the stoichiometry of the core clock proteins and abrogating the rhythms in KaiC phosphorylation (Takai et al., 2006). To establish whether the defect in viability during dark periods results from loss of Kai oscillator function or from loss of RpaA function, we performed viability experiments using the ‘clock rescue’ strain background in which kaiBC expression is made independent of RpaA (Teng et al., 2013). We observed that the rpaA^- ‘clock rescue’ strain phenocopies the rpaA^- strain and the isogenic rpaA⁺ ‘clock rescue’ strain phenocopies wild type strain, demonstrating that the viability defect stems from the loss of RpaA function and not from loss of Kai oscillator function (Figure 1B and Figure 1—figure supplement 1C).

Figure 1 with 2 supplements see all

Download asset Open asset

Since phosphorylation of RpaA is required for binding to its target promoters and gene expression (Markson et al., 2013), we asked whether the rpaA deletion phenotype in light/dark conditions can be rescued by a specific phosphorylated form of RpaA: RpaA D53E mimicking the phosphorylated state of RpaA; and RpaA D53A mimicking the unphosphorylated state. We examined the viability of the OX-D53E and OX-D53A strains, which lack both kaiBC and rpaA and contain a phosphomimetic form of RpaA expressed from an IPTG-inducible Ptrc promoter (Markson et al., 2013). In these strains RpaA phosphorylation is not controlled by either the clock activity or by any other potential input factor, isolating RpaA activity from its regulators. We observed that expression of the RpaA D53E phosphomimetic restores viability in light/dark conditions, while expression of the RpaA D53A variant does not (Figure 1C). We also compared growth of the _cRpaA strain (subscript ‘c’ denotes constitutive, RpaA-independent expression; see Materials and methods) with the _cRpaA D53A strain, each made using the ‘clock rescue’ background, in light/dark cycles. We found that while the control _cRpaA strain grows under these conditions despite lower levels of RpaA driven by the constitutive promoter, the _cRpaA D53A strain is inviable (Figure 1—figure supplement 1D and E). We conclude that the active DNA-binding form of RpaA is necessary and sufficient for cell viability during exposure to light/dark conditions.

We next analyzed the role of the core KaiABC clock in sustaining viability of cyanobacterial cells in light/dark conditions. Deletion of the kaiBC operon leads to only a minor viability defect in light/dark conditions (Boyd et al., 2013 and Figure 1D) and substantial RpaA phosphorylation levels were observed in the kaiBC^- strain at the subjective dawn (CT = 2 hr) and afternoon (CT = 9 hr) in constant light conditions (Boyd et al., 2013). Since RpaA phosphorylation levels are regulated by the antagonistic activities of two clock-controlled histidine kinases, SasA and CikA, and complexes of KaiB and KaiC proteins activate the phosphatase activity of CikA in vitro (Gutu and O'Shea, 2013), we hypothesized that RpaA would remain highly phosphorylated throughout the light/dark cycle in the kaiBC^- strain. In agreement with this prediction, when we analyzed RpaA phosphorylation in the kaiBC^- mutant we detected high levels of RpaA phosphorylation both in light and in the dark (Figure 1E). Thus, the ability of the kaiBC^- mutant to survive alternating periods of light and dark likely stems from the high levels of phosphorylated RpaA in this strain. While the kaiBC operon is not essential for cell viability in light/dark cycles, specific KaiC phosphoforms, whose relative abundances vary during the circadian cycle (Rust et al., 2007), may have different effects on RpaA phosphorylation and, consequently, affect the ability of cells to survive light/dark cycling conditions. We tested this hypothesis by analyzing cell viability and RpaA phosphorylation in strains expressing KaiC point mutants: KaiC-AA (KaiC S431A T432A), representing the unphosphorylated phosphoform of KaiC, and KaiC-EE (KaiC S431E T432E), representing the hyperphosphorylated KaiC variant. We used the P0050 promoter, which releases KaiC and RpaA expression from circadian feedback, to ensure equal expression of KaiC variants as well as of RpaA between the phosphomimetic strains (Figure 1—figure supplement 2). We observed that while the _cKaiC-AA mutant is viable in light/dark conditions, the _cKaiC-EE strain is not able to grow (Figure 1F). Strikingly, the _cKaiC-AA strain maintains high RpaA phosphorylation levels during the diurnal cycle, while RpaA is phosphorylated in the _cKaiC-EE strain only at a low level (Figure 1G). This ability to control RpaA phosphorylation and cell viability through perturbations in KaiC phosphorylation status underscores the primacy of this regulatory axis in the diurnal control of RpaA phosphorylation, and suggests that any hypothetical clock-independent environmental factors play a lesser role in regulating RpaA phosphorylation under the conditions we tested.

Circadian clocks and their output synchronize the metabolic activity of cells with diurnal fluctuations (Tu and McKnight, 2006; Green et al., 2008; Graf et al., 2010; Dodd et al., 2015). To understand if the observed viability defect stems from an inability of the rpaA^- strain (which is arrested in a ‘dawn-like’ transcriptional state [Markson et al., 2013]) to meet the metabolic requirements of darkness, we assessed cellular energy charge in wild type and the rpaA^- strain over time in constant light conditions and in light/dark cycles. Energy charge describes the relative levels of ATP, ADP and AMP in the intracellular adenine nucleotide pool, and is a tightly controlled parameter that reflects the global metabolic energy status of the cell (Atkinson, 1968). In constant light, there was no difference in energy charge between wild type and the rpaA^- strain (Figure 2, top panel). Wild type cells subjected to darkness experience only a slight reduction in energy charge (Figure 2, bottom panel, Figure 2—figure supplement 1A). In contrast, rpaA^- cells experience a reduction in energy charge during the first dark period and an even more dramatic reduction during the second dark period, suggesting that this strain has a defect in energy metabolism affecting energy maintenance during the dark phase of the light/dark cycle (Figure 2, bottom panel, Figure 2—figure supplement 1B). The decrease in energy charge was accompanied by a reduction in the total adenine nucleotide pool size in the rpaA^- strain (Figure 2—figure supplement 2). A similar trend has been previously observed during starvation in carbon-limited cultures of E. coli (Chapman et al., 1971), prompting us to analyze changes in activity through pathways relevant for carbon metabolism in the strain lacking rpaA.

Figure 2 with 2 supplements see all

Download asset Open asset

In constant light conditions, the circadian program directs expression of anabolic and catabolic pathways, with genes relevant for carbon catabolism peaking in mRNA abundance at subjective dusk (Vijayan et al., 2009; Diamond et al., 2015). To investigate the role of RpaA in orchestrating diurnal transcription of metabolic genes, we performed RNA sequencing in the rpaA^- ‘clock rescue’ strain and an isogenic rpaA⁺ strain after exposure to darkness. The expression of circadian dusk-peaking genes was most affected in the rpaA deletion strain exposed to darkness (Figure 3—figure supplement 1A). Specifically, we observed that genes encoding enzymes involved in glycogen breakdown, glycolysis and the oxidative pentose phosphate pathway (Figure 3—figure supplement 1B and Supplementary file 1), which normally peak in expression at dusk, are among the genes with the greatest defect in expression in the rpaA^- cells (Figure 3A). These pathways are key for carbon utilization and energy production in the dark in cyanobacteria (Figure 3B), and their function is essential for survival of periods of darkness (Doolittle and Singer, 1974). Independent deletion of gnd as well as the glgP_gap1 and fbp_zwf_opcA operons, which exhibit severely reduced expression in the rpaA^- strain, results in impaired viability in light/dark cycles but not in constant light conditions (Figure 3—figure supplement 1C and Doolittle and Singer, 1974; Scanlan et al., 1995). We also measured expression of the key dusk metabolic genes, gnd, glgP and zwf, in the kaiBC^- strain at dusk and in the first hours of darkness (Figure 3—figure supplement 2). We observed that these genes are expressed at a high level, further emphasizing the role of phosphorylated RpaA in activation of the dusk transcriptional program and maintenance of cell viability in light/dark conditions. To confirm that the transcriptional defects in the rpaA deletion mutant affect carbon catabolism, we measured the enzymatic activities of glycogen phosphorylase (glgP), glucose-6-phosphate dehydrogenase (zwf) and 6-phosphogluconate dehydrogenase (gnd) in the dark, and observed that activity of each enzyme was strongly reduced in the rpaA^- strain compared to wild type (Figure 3C). Therefore, the rpaA^- strain appears to be unable to activate the carbon catabolic pathways upon the onset of darkness because it cannot induce transcription of the requisite enzymes at dusk.

Figure 3 with 2 supplements see all

Download asset Open asset

Glycogen is a crucial metabolic reserve used as a carbon and energy source during periods of darkness (Smith, 1983). In wild type cyanobacteria glycogen accumulates during the afternoon (Cervený and Nedbal, 2009; Knoop et al., 2010; Rügen et al., 2015) through the activity of a glycogen synthesis pathway composed of phosphoglucomutase (pgm1 and pgm2), ADP-glucose pyrophosphorylase (glgC) and glycogen synthase (glgA) (Figure 4A). To analyze whether the dawn-arrested rpaA^- strain can accumulate glycogen reserves, we measured expression and activity of the glycogen synthesis enzymes and found that neither the expression levels nor activity of the enzymes in the pathway were significantly reduced in the rpaA^- strain (Figure 4B and C). Surprisingly, we observed that while glycogen content oscillates over a period of 24 hr in wild type cells both in constant light and in light/dark cycles, glycogen is present at a very low level in the rpaA^- strain regardless of light conditions, despite high levels of the synthetic enzymes in the extracts (Figure 4D). The same glycogen accumulation defect occurs in the rpaA^- ‘clock rescue’ strain, demonstrating that it is a direct effect of deletion of rpaA (Figure 4E). Like the rpaA^- strain, glycogen synthesis mutants (glgA^- and glgC^-) exhibit reduced viability in alternating light/dark conditions (Gründel et al., 2012), Figure 4—figure supplement 1A, bottom panel) and in constant darkness (Figure 4—figure supplement 1B), but grow somewhat slower than the rpaA^- strain in constant light conditions (Figure 4—figure supplement 1A, top panel). We conclude that the deficiency in the preparation of a reserve carbon source – which occurs not at the level of transcription of the glycogen synthesis genes or regulation of the activity of the encoded enzymes – may contribute to the impaired viability of the rpaA^- strain in the dark.

Figure 4 with 1 supplement see all

Download asset Open asset

We hypothesized that the inability of the rpaA^- strain to maintain appropriate energy levels and cell viability in the dark stems from the defects in accumulation of carbon/energy stores during the day and their utilization at night. To assess whether restoration of the correct carbon catabolic route in darkness is sufficient to rescue the energy charge and viability of the rpaA^- strain in light/dark conditions, we reconstituted the activities of the carbon utilization pathways and provided a transporter for uptake of external carbon. We restored expression of the carbon utilization enzymes in the rpaA^- strain by placing their expression under an IPTG-inducible ectopic Ptrc promoter, and confirmed that induction with IPTG restored enzyme activity in cell extracts (Figure 5—figure supplement 1). To counterbalance the inability of the rpaA^- cells to accumulate internal carbon reserves, we expressed the GalP transporter to allow glucose uptake from the medium. Expression of either the missing enzymes or the GalP transporter alone with supplementation of glucose in the rpaA^- background was not sufficient to restore viability of this strain (Figure 5A and B). However, restoration of the carbon utilization pathways combined with the introduction of the glucose transporter in the presence of glucose rescued the viability defect of the rpaA^- strain (Figure 5A and B). Glucose feeding does not increase accumulation of internal carbon stores in this engineered strain (Figure 5C), nor in wild type expressing the GalP transporter (Figure 5—figure supplement 2). However, it is sufficient to restore high energy charge in the rpaA^- strain expressing both the key catabolic enzymes and GalP (Figure 5D). Our results strongly suggest that functional circadian output mediated by the activity of RpaA is required for the anticipatory accumulation of carbon reserves during the day, as well as metabolic switching to carbon catabolism at dusk. These metabolic adaptations allow cells to stimulate energy production and sustain cell viability during periods of darkness that challenge and constrain resources for cyanobacterial growth.

Figure 5 with 2 supplements see all

Download asset Open asset

Daily environmental changes in light availability give rise to periodic demand for metabolism of alternate energy sources in photosynthetic organisms (Doolittle, 1979). To maximize fitness, photosynthetic organisms need to accurately allocate resources by carrying out carbon assimilation during the day and utilizing stored carbohydrate reserves at night (Smith, 1983). Here we show that the phosphorylated form of the transcription factor RpaA, regulated by the circadian clock, is critical for the coordination of carbon anabolism and catabolism and thus cell viability in light/dark cycles. Deletion of rpaA prevents glycogen accumulation in light and renders cells unable to utilize existing glycogen at dusk (Figure 5E) – this leads to alterations in cellular energy charge and, as previously observed, to a severe reduction in fitness. It is yet to be determined whether RpaA is rewiring carbon metabolism solely through transcriptional control or through other mechanisms.

The mechanism underlying perturbed carbon accumulation in the rpaA^- strain is unclear. The activities of glycogen synthesizing enzymes in the rpaA^- strain are high in vitro, however they could be affected in vivo by the redox state of the cell or by metabolite levels. It is also possible that an RpaA-dependent transcript expressed in the afternoon is required to funnel assimilated carbon into glycogen synthesis. Alternatively, additional metabolic flux changes may occur in the rpaA^- strain that prevent formation of glycogen stores. Finally, it is formally possible that the rpaA^- cells perform a futile cycle, simultaneously synthesizing and breaking glycogen down. However, since the activity of the glycogen phosphorylase involved in glycogen degradation is low in the rpaA^- strain, we find this scenario less likely.

Our data support a model in which anticipation of periodic dark-induced resource limitation is one essential role of the cyanobacterial circadian system. The output of the core oscillator, through regulation of RpaA activity, temporally coordinates metabolism to prepare cells for the metabolic demands of periods of darkness. Correct scheduling of central carbon metabolic activities allows cells to survive the night. Recently, Lambert et al. observed that the circadian clock orchestrates a trade-off between rapid cell growth in the morning and robustness to starvation in the afternoon, suggesting preparation for the night as a plausible role for the circadian clock (Lambert et al., 2016). Our findings are consistent with this model and provide novel mechanistic insight into circadian regulation of cyanobacterial energy metabolism. It remains to be investigated whether the circadian-driven preparation for darkness is a shared strategy employed by other photosynthetic organisms.

Further, our results suggest a molecular explanation for the observed selective advantage of circadian resonance. When the periods of the endogenous circadian system and periods of the environmental fluctuations are equal, organisms show a fitness advantage and outcompete mutants whose clocks have an altered periodicity in the same environment (Ouyang et al., 1998; Woelfle et al., 2004). The mechanistic basis for this observation has been lacking. Based on our findings, we expect that the fitness defects in the strains with core oscillator periods unequal to the period of environmental variation result from mistiming of RpaA activation. The resultant mistiming of accumulation and mobilization of glycogen stores with the onset of the nightfall would lead to a carbon deficit at night, and impact growth as in the rpaA^- mutant during dark periods. One role of the KaiABC clock in our model, therefore, is to correctly phase the activity of RpaA and downstream carbon metabolism with the external environment, maximizing fitness and growth.

Though we show that RpaA activity is crucial for regulation of key aspects of carbohydrate metabolism, we identify ~90 genes whose expression is abrogated at dusk and throughout darkness in the rpaA^- strain and it is possible that additional physiological processes are defective in the absence of rpaA. Moreover, although our results indicate that the KaiABC clock is the main regulator of diurnal RpaA phosphorylation, other factors may influence the activity of RpaA or its regulon under fluctuating environmental conditions. RpaA is known to cooperate with RpaB (Espinosa et al., 2015), a transcription factor crucial for responses to high light conditions. CikA, the clock-controlled regulator of RpaA phosphorylation, has been proposed to directly sense redox state of the cell (Ivleva et al., 2006). Finally, additional RpaA regulators, LabA and CikB, have been proposed (Taniguchi et al., 2010; Boyd et al., 2016). Hence, it is possible that in a rapidly changing environment RpaA integrates circadian cues with environmental inputs important for viability. It will be interesting to analyze how deletion of rpaA affects the ability of cells to respond to other environmental challenges that cyanobacteria routinely face.

Finally, our data together with a recent study by Diamond and colleagues (Diamond et al., 2015) emphasize the role of the circadian output pathway in driving oscillations in metabolism. On the other hand, it has been demonstrated that metabolic processes provide feedback to the circadian clock by regulating its resetting, indicating an important role of the fluctuating metabolic status of a cell in driving circadian oscillations (Rust et al., 2011; Pattanayak et al., 2014, 2015). Together, a complex bidirectional relationship between the circadian program and metabolism emerges in which the circadian system both controls and is controlled by the metabolic rhythms to set and tune cell physiology with the environmental cycles. This tight coupling between circadian machinery and metabolism is coming to light as a universal feature across kingdoms of life (Green et al., 2008; Dodd et al., 2015).

1. 1. Puszynska AM
   2. O'Shea EK

   (2017) Analysis of gene expression in the "clock rescue" and rpaA- "clock rescue" strains of Synechococcus elongatus PCC7942 with RNA sequencing in light/dark conditions

   Publicly available at the NCBI Gene Expression Omnibus (accession no. GSE89999).

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE89999

1. 1. Atkinson DE

   (1968) The energy charge of the adenylate pool as a regulatory parameter. Interaction with feedback modifiers

   Biochemistry 7:4030–4034.

   https://doi.org/10.1021/bi00851a033

   • PubMed
   • Google Scholar
2. 1. Boyd JS
   2. Bordowitz JR
   3. Bree AC
   4. Golden SS

   (2013) An allele of the crm gene blocks cyanobacterial circadian rhythms

   PNAS 110:13950–13955.

   https://doi.org/10.1073/pnas.1312793110

   • PubMed
   • Google Scholar
3. 1. Boyd JS
   2. Cheng RR
   3. Paddock ML
   4. Sancar C
   5. Morcos F
   6. Golden SS

   (2016) A combined computational and genetic approach uncovers network interactions of the cyanobacterial circadian clock

   Journal of Bacteriology 198:2439–2447.

   https://doi.org/10.1128/JB.00235-16

   • PubMed
   • Google Scholar
4. 1. Cervený J
   2. Nedbal L

   (2009) Metabolic rhythms of the cyanobacterium Cyanothece sp. ATCC 51142 correlate with modeled dynamics of circadian clock

   Journal of Biological Rhythms 24:295–303.

   https://doi.org/10.1177/0748730409338367

   • PubMed
   • Google Scholar
5. 1. Chapman AG
   2. Fall L
   3. Atkinson DE

   (1971)

   Adenylate energy charge in Escherichia coli during growth and starvation

   Journal of Bacteriology 108:1072–1086.

   • PubMed
   • Google Scholar
6. 1. Clerico EM
   2. Ditty JL
   3. Golden SS

   (2007) Specialized techniques for site-directed mutagenesis in cyanobacteria

   Methods in Molecular Biology 362:155–171.

   https://doi.org/10.1007/978-1-59745-257-1_11

   • PubMed
   • Google Scholar
7. 1. Díaz-Troya S
   2. López-Maury L
   3. Sánchez-Riego AM
   4. Roldán M
   5. 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.

   https://doi.org/10.1093/mp/sst137

   • PubMed
   • Google Scholar
8. 1. de Marsac NT
   2. Houmard J

   (1988) Complementary chromatic adaptation: physiological conditions and action spectra

   Methods in Enzymology 167:318–328.

   https://doi.org/10.1016/0076-6879(88)67037-6

   • Google Scholar
9. 1. Diamond S
   2. Jun D
   3. Rubin BE
   4. Golden SS

   (2015) The circadian oscillator in Synechococcus elongatus controls metabolite partitioning during diurnal growth

   PNAS 112:E1916–E1925.

   https://doi.org/10.1073/pnas.1504576112

   • PubMed
   • Google Scholar
10. 1. Dodd AN
    2. Belbin FE
    3. Frank A
    4. Webb AA

    (2015) Interactions between circadian clocks and photosynthesis for the temporal and spatial coordination of metabolism

    Frontiers in Plant Science 6:245.

    https://doi.org/10.3389/fpls.2015.00245

    • PubMed
    • Google Scholar
11. 1. Doolittle WF
    2. Singer RA

    (1974)

    Mutational analysis of dark endogenous metabolism in the blue-green bacterium Anacystis nidulans

    Journal of Bacteriology 119:677–683.

    • PubMed
    • Google Scholar
12. 1. Doolittle WF

    (1979) The cyanobacterial genome, its expression, and the control of that expression

    Advances in Microbial Physiology 20:1–102.

    https://doi.org/10.1016/S0065-2911(08)60206-4

    • PubMed
    • Google Scholar
13. 1. Espinosa J
    2. Boyd JS
    3. Cantos R
    4. Salinas P
    5. Golden SS
    6. Contreras A

    (2015) Cross-talk and regulatory interactions between the essential response regulator RpaB and cyanobacterial circadian clock output

    PNAS 112:2198–2203.

    https://doi.org/10.1073/pnas.1424632112

    • PubMed
    • Google Scholar
14. 1. Fu J
    2. Xu X

    (2006) The functional divergence of two glgP homologues in Synechocystis sp. PCC 6803

    FEMS Microbiology Letters 260:201–209.

    https://doi.org/10.1111/j.1574-6968.2006.00312.x

    • PubMed
    • Google Scholar
15. 1. Gibson DG
    2. Young L
    3. Chuang RY
    4. Venter JC
    5. Hutchison CA
    6. Smith HO

    (2009) Enzymatic assembly of DNA molecules up to several hundred kilobases

    Nature Methods 6:343–345.

    https://doi.org/10.1038/nmeth.1318

    • PubMed
    • Google Scholar
16. 1. Graf A
    2. Schlereth A
    3. Stitt M
    4. Smith AM

    (2010) Circadian control of carbohydrate availability for growth in Arabidopsis plants at night

    PNAS 107:9458–9463.

    https://doi.org/10.1073/pnas.0914299107

    • PubMed
    • Google Scholar
17. 1. Green CB
    2. Takahashi JS
    3. Bass J

    (2008) The meter of metabolism

    Cell 134:728–742.

    https://doi.org/10.1016/j.cell.2008.08.022

    • PubMed
    • Google Scholar
18. 1. Gründel M
    2. Scheunemann R
    3. Lockau W
    4. 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.

    https://doi.org/10.1099/mic.0.062950-0

    • PubMed
    • Google Scholar
19. 1. Gutu A
    2. O'Shea EK

    (2013) Two antagonistic clock-regulated histidine kinases time the activation of circadian gene expression

    Molecular Cell 50:288–294.

    https://doi.org/10.1016/j.molcel.2013.02.022

    • PubMed
    • Google Scholar
20. 1. Ivleva NB
    2. Gao T
    3. LiWang AC
    4. Golden SS

    (2006) Quinone sensing by the circadian input kinase of the cyanobacterial circadian clock

    PNAS 103:17468–17473.

    https://doi.org/10.1073/pnas.0606639103

    • PubMed
    • Google Scholar
21. 1. Knoop H
    2. Zilliges Y
    3. Lockau W
    4. Steuer R

    (2010) The metabolic network of Synechocystis sp. PCC 6803: systemic properties of autotrophic growth

    Plant Physiology 154:410–422.

    https://doi.org/10.1104/pp.110.157198

    • PubMed
    • Google Scholar
22. 1. Lambert G
    2. Chew J
    3. Rust MJ

    (2016) Costs of Clock-Environment misalignment in individual cyanobacterial cells

    Biophysical Journal 111:883–891.

    https://doi.org/10.1016/j.bpj.2016.07.008

    • PubMed
    • Google Scholar
23. 1. Liu L
    2. Hu H
    3. Gao H
    4. Xu X
    5. Hh H
    6. Xd X

    (2013) Role of two phosphohexomutase genes in glycogen synthesis in Synechocystis sp. PCC6803

    Chinese Science Bulletin 58:4616–4621.

    https://doi.org/10.1007/s11434-013-5958-0

    • Google Scholar
24. 1. Markson JS
    2. Piechura JR
    3. Puszynska AM
    4. O'Shea EK

    (2013) Circadian control of global gene expression by the cyanobacterial master regulator RpaA

    Cell 155:1396–1408.

    https://doi.org/10.1016/j.cell.2013.11.005

    • PubMed
    • Google Scholar
25. 1. Nishiwaki T
    2. Satomi Y
    3. Kitayama Y
    4. Terauchi K
    5. Kiyohara R
    6. Takao T
    7. Kondo T

    (2007) A sequential program of dual phosphorylation of KaiC as a basis for circadian rhythm in cyanobacteria

    The EMBO Journal 26:4029–4037.

    https://doi.org/10.1038/sj.emboj.7601832

    • PubMed
    • Google Scholar
26. 1. Ouyang Y
    2. Andersson CR
    3. Kondo T
    4. Golden SS
    5. Johnson CH

    (1998) Resonating circadian clocks enhance fitness in cyanobacteria

    PNAS 95:8660–8664.

    https://doi.org/10.1073/pnas.95.15.8660

    • PubMed
    • Google Scholar
27. 1. Pattanayak GK
    2. Lambert G
    3. Bernat K
    4. Rust MJ

    (2015) Controlling the cyanobacterial clock by synthetically rewiring metabolism

    Cell Reports 13:2362–2367.

    https://doi.org/10.1016/j.celrep.2015.11.031

    • PubMed
    • Google Scholar
28. 1. Pattanayak GK
    2. Phong C
    3. Rust MJ

    (2014) Rhythms in energy storage control the ability of the cyanobacterial circadian clock to reset

    Current Biology 24:1934–1938.

    https://doi.org/10.1016/j.cub.2014.07.022

    • PubMed
    • Google Scholar
29. 1. Rust MJ
    2. Golden SS
    3. O'Shea EK

    (2011) Light-driven changes in energy metabolism directly entrain the cyanobacterial circadian oscillator

    Science 331:220–223.

    https://doi.org/10.1126/science.1197243

    • Google Scholar
30. 1. Rust MJ
    2. Markson JS
    3. Lane WS
    4. Fisher DS
    5. O'Shea EK

    (2007) Ordered phosphorylation governs oscillation of a three-protein circadian clock

    Science 318:809–812.

    https://doi.org/10.1126/science.1148596

    • PubMed
    • Google Scholar
31. 1. Rügen M
    2. Bockmayr A
    3. Steuer R

    (2015) Elucidating temporal resource allocation and diurnal dynamics in phototrophic metabolism using conditional FBA

    Scientific Reports 5:15247.

    https://doi.org/10.1038/srep15247

    • PubMed
    • Google Scholar
32. 1. Scanlan DJ
    2. Sundaram S
    3. Newman J
    4. Mann NH
    5. Carr NG

    (1995) Characterization of a zwf mutant of Synechococcus sp. strain PCC 7942

    Journal of Bacteriology 177:2550–2553.

    https://doi.org/10.1128/jb.177.9.2550-2553.1995

    • PubMed
    • Google Scholar
33. 1. Schaeffer F
    2. Stanier RY

    (1978) Glucose-6-phosphate dehydrogenase of Anabaena sp. kinetic and molecular properties

    Archives of Microbiology 116:9–19.

    https://doi.org/10.1007/bf00408728

    • PubMed
    • Google Scholar
34. 1. Smith AJ

    (1983) Modes of cyanobacterial carbon metabolism

    Annales De l'Institut Pasteur / Microbiologie 134B:93–113.

    https://doi.org/10.1016/S0769-2609(83)80099-4

    • PubMed
    • Google Scholar
35. 1. Suzuki E
    2. Ohkawa H
    3. Moriya K
    4. Matsubara T
    5. Nagaike Y
    6. Iwasaki I
    7. Fujiwara S
    8. Tsuzuki M
    9. Nakamura Y

    (2010) Carbohydrate metabolism in mutants of the cyanobacterium Synechococcus elongatus PCC 7942 defective in glycogen synthesis

    Applied and Environmental Microbiology 76:3153–3159.

    https://doi.org/10.1128/AEM.00397-08

    • PubMed
    • Google Scholar
36. 1. Takai N
    2. Nakajima M
    3. Oyama T
    4. Kito R
    5. Sugita C
    6. Sugita M
    7. Kondo T
    8. Iwasaki H

    (2006) A KaiC-associating SasA-RpaA two-component regulatory system as a major circadian timing mediator in cyanobacteria

    PNAS 103:12109–12114.

    https://doi.org/10.1073/pnas.0602955103

    • PubMed
    • Google Scholar
37. 1. Taniguchi Y
    2. Takai N
    3. Katayama M
    4. Kondo T
    5. Oyama T

    (2010) Three major output pathways from the KaiABC-based oscillator cooperate to generate robust circadian kaiBC expression in cyanobacteria

    PNAS 107:3263–3268.

    https://doi.org/10.1073/pnas.0909924107

    • PubMed
    • Google Scholar
38. 1. Taton A
    2. Unglaub F
    3. Wright NE
    4. Zeng WY
    5. Paz-Yepes J
    6. Brahamsha B
    7. Palenik B
    8. Peterson TC
    9. Haerizadeh F
    10. Golden SS
    11. Golden JW

    (2014) Broad-host-range vector system for synthetic biology and biotechnology in cyanobacteria

    Nucleic Acids Research 42:e136.

    https://doi.org/10.1093/nar/gku673

    • PubMed
    • Google Scholar
39. 1. Teng SW
    2. Mukherji S
    3. Moffitt JR
    4. de Buyl S
    5. O'Shea EK

    (2013) Robust circadian oscillations in growing cyanobacteria require transcriptional feedback

    Science 340:737–740.

    https://doi.org/10.1126/science.1230996

    • PubMed
    • Google Scholar
40. 1. Tu BP
    2. McKnight SL

    (2006) Metabolic cycles as an underlying basis of biological oscillations

    Nature Reviews Molecular Cell Biology 7:696–701.

    https://doi.org/10.1038/nrm1980

    • PubMed
    • Google Scholar
41. 1. Vijayan V
    2. Zuzow R
    3. O'Shea EK

    (2009) Oscillations in supercoiling drive circadian gene expression in cyanobacteria

    PNAS 106:22564–22568.

    https://doi.org/10.1073/pnas.0912673106

    • PubMed
    • Google Scholar
42. 1. Woelfle MA
    2. Ouyang Y
    3. Phanvijhitsiri K
    4. Johnson CH

    (2004) The adaptive value of circadian clocks: an experimental assessment in cyanobacteria

    Current Biology 14:1481–1486.

    https://doi.org/10.1016/j.cub.2004.08.023

    • PubMed
    • Google Scholar

Author details

1. Anna M Puszynska

   1. Department of Molecular and Cellular Biology, Harvard University, Cambridge, United States
   2. Faculty of Arts and Sciences Center for Systems Biology, Harvard University, Cambridge, United States
   3. Howard Hughes Medical Institute, Harvard University, Cambridge, United States

   Contribution

   AMP, Conceptualization, Formal analysis, Investigation, Writing—original draft, Writing—review and editing

   For correspondence

   apuszyns@fas.harvard.edu

   Competing interests

   No competing interests declared.

   "This ORCID iD identifies the author of this article:" 0000-0002-8751-5335

2. Erin K O'Shea

   1. Department of Molecular and Cellular Biology, Harvard University, Cambridge, United States
   2. Faculty of Arts and Sciences Center for Systems Biology, Harvard University, Cambridge, United States
   3. Howard Hughes Medical Institute, Harvard University, Cambridge, United States
   4. Department of Chemistry and Chemical Biology, Harvard University, Cambridge, United States

   Contribution

   EKO, Conceptualization, Formal analysis, Writing—original draft, Writing—review and editing

   For correspondence

   erin_oshea@harvard.edu

   Competing interests

   EKO: Chief Scientific Officer and a Vice President at the Howard Hughes Medical Institute, one of the three founding funders of eLife.

   "This ORCID iD identifies the author of this article:" 0000-0002-2649-1018

• 2,909

  views

• 366

  downloads

• 43

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.