Here, we present a reliable and easy assay to measure the glycogen content in cyanobacterial cells. The procedure entails precipitation, selectable depolymerization, and the detection of glucose residues. This method is suitable for both wildtype and genetically engineered strains and can facilitate the metabolic engineering of cyanobacteria.
Cyanobacteria accumulate glycogen as a major intracellular carbon and energy storage during photosynthesis. Recent developments in research have highlighted complex mechanisms of glycogen metabolism, including the diel cycle of biosynthesis and catabolism, redox regulation, and the involvement of non-coding RNA. At the same time, efforts are being made to redirect carbon from glycogen to desirable products in genetically engineered cyanobacteria to enhance product yields. Several methods are used to determine the glycogen contents in cyanobacteria, with variable accuracies and technical complexities. Here, we provide a detailed protocol for the reliable determination of the glycogen content in cyanobacteria that can be performed in a standard life science laboratory. The protocol entails the selective precipitation of glycogen from the cell lysate and the enzymatic depolymerization of glycogen to generate glucose monomers, which are detected by a glucose oxidase-peroxidase (GOD-POD) enzyme coupled assay. The method has been applied to Synechocystis sp. PCC 6803 and Synechococcus sp. PCC 7002, two model cyanobacterial species that are widely used in metabolic engineering. Moreover, the method successfully showed differences in the glycogen contents between the wildtype and mutants defective in regulatory elements or glycogen biosynthetic genes.
Cyanobacteria accumulate glycogen as the major carbohydrate store of carbon from CO2 fixed in light through photosynthesis. Glycogen is a glycan consisting of linear α-1,4-linked glucan with branches created by α-1,6-linked glucosyl linkages. Glycogen biosynthesis in cyanobacteria starts with the conversion of glucose-6-phosphate into ADP-glucose through the sequential action of phosphoglucomutase and ADP-glucose pyrophosphorylase. The glucose moiety in ADP-glucose is transferred to the non-reducing end of the α-1,4-glucan backbone of glycogen by one or more glycogen synthases (GlgA). Subsequently, a branching enzymes introduce the α-1,6-linked glucosyl linkage, which is further extended to generate the glycogen particle. In the dark, glycogen is broken down by glycogen phosphorylase, glycogen debranching enzymes, α-glucanotransferase, and malto-dextrin phosphorylase into phosphorylated glucose and free glucose. These feed into catabolic pathways, including the oxidative pentose phosphate pathway, the Embden-Meyerhof-Parnas pathway (glycolysis), and the Entner-Doudoroff pathway1,2,3,4.
Glycogen metabolism in cyanobacteria has garnered increasing interest in recent years because of the potential for cyanobacteria to develop into microbial cell factories driven by sunlight to produce chemicals and fuels. Glycogen metabolism could be modified to increase the yield of the products, because glycogen is the largest flexible carbon pool in these bacteria. An example is the cyanobacterium Synechococcus sp. PCC 7002, which has been genetically engineered to produce mannitol; the genetic disruption of glycogen synthesis increases the mannitol yield 3-fold5. Another example is the production of bioethanol from glycogen-loaded wildtype Synechococcus sp. PCC 70026. The wildtype cell glycogen content may be up to 60% of the dry weight of the cell during nitrogen starvation6.
Our understanding of glycogen metabolism and regulation has also expanded in recent years. While glycogen is known to accumulate in the light and to be catabolized in the dark, detailed kinetics of glycogen metabolism during the diel cycle was only recently revealed in Synechocystis sp. PCC 68037. Moreover, several genes affecting the accumulation of glycogen have been identified. A notable example is the discovery that the putative histidine kinase PmgA and the non-coding RNA PmgR1 form a regulatory cascade and control the accumulation of glycogen. Interestingly, the pmgA and pmgR1 deletion mutants accumulate twice as much glycogen as the wildtype strain of Synechocystis sp. PCC 68038,9. Other regulatory elements are also known to affect the accumulation of glycogen, including the alternative sigma factor E and the transcriptional factor CyAbrB210,11.
As interest in glycogen regulation and metabolism grow, a detailed protocol describing the determination of the glycogen content is warranted. Several methods are used in the literature. Acid hydrolysis followed by the determination of the monosaccharide content through high-pressure anion exchange liquid chromatography coupled with a pulsed amperometric detector or spectrometric determination following treatments with acid and phenol are widely used methods to approximate the glycogen content9,10,12,13. However, a high-pressure anion exchange liquid chromatographic instrument is very expensive and does not discriminate glucose derived from glycogen from that derived from other glucose-containing glycoconjugates, such as sucrose14, glucosylglycerol15, and cellulose16,17,18, which are known to accumulate in some cyanobacterial species. The acid-phenol method can be performed using standard laboratory equipment. However, it uses highly toxic reagents and does not distinguish glucose derived from different glycoconjugates, nor does it distinguish glucose from other monosaccharides that constitute cellular materials, such as glycolipids, lipopolysaccharides, and extracellular matrices12. Notably, the hot acid-phenol assay is often used for the determination of total carbohydrate content rather than for the specific determination of glucose content12. Enzymatic hydrolysis of glycogen to glucose by α-amyloglucosidase followed by the detection of glucose through an enzyme-coupled assay generates a colorimetric readout that is highly sensitive and specific to glucose derived from glycogen. The specificity can be enhanced further with the preferential precipitation of glycogen from cell lysates by ethanol5,8,19.
Here, we describe a detailed protocol for an enzyme-based assay of the glycogen content in two of the most widely studied cyanobacterial species, Synechocystis sp. PCC 6803 and Synechococcus sp. PCC 7002, in the wildtype and mutant strains. In order to ensure efficient hydrolysis, a cocktail of α-amylase and α-amyloglucosidase is used8. The endo-acting α-amylase hydrolyzes the α-1,4-linkages in various glucans into dextrins, which are further hydrolyzed to glucose by exo-acting α-amyloglucosidase20. The synergistic effects of these enzymes are well known, and these enzymes are routinely used for the selective hydrolysis of starch, which is an α-linked glucan like glycogen, without affecting other glycoconjugants, such as cellulose, in the plant biomass21. The released glucose is quantitatively detected following an enzyme-coupled assay consisting of glucose oxidase-which catalyzes the reduction of oxygen to hydrogen peroxide and the oxidation of glucose to a lactone-and peroxidase-which produces a pink-colored quinoneimine dye from hydrogen peroxide, a phenolic compound, and 4-aminoantipyrine22.
1. Preparation
2. Determination of the Cell Dry Weight (Optional)
3. Lysis of Cyanobacterial Cells
4. Glycogen Precipitation
5. Enzymatic Hydrolysis and Glycogen Determination
6. Determination of the Total Glucose Content Using the GOD-POD Reagent
10 mL of wildtype Synechocystis sp. PCC 6803 were grown under photoautotrophic conditions until the OD730nm value reached approximately 0.8. The cells were harvested and resuspended in 50 mM Tris-HCl, pH 8. The OD730nm value was adjusted to 2-3. The glycogen content was analyzed following the protocol described above. The glycogen content per the OD730nm was 13 ± 1.8 µg/mL/OD730nm (N = 12). The glycogen content relative to the protein content was 0.24 ± 0.03 µg/µg (N = 12), and the glycogen content relative to the chlorophyll a content was 5.7 ± 0.6 µg/µg (N = 12). Due to the small amount of material available, the measurement of the cell dry weight was omitted. Given that the protein content in cyanobacteria cultivated under comparable conditions is about 50% of the cell dry weight25, the glycogen content is estimated to be 12% of the cell dry weight, which is consistent with previous studies26.
The detection limits of the GOD-POD assay showed a linear correlation between the glucose concentration and absorbance at 510 nm in the glucose concentration range between 10 and 100 µg/mL, corresponding to absorbance values of 0.08 and 0.7 at 510 nm. The minimum limit is likely due to the instrumental detection limit. When glucose concentrations higher than 150 µg/mL were used, dark green precipitates formed, causing a large variation in the absorbance readout. Concerning the amount of cell materials used, we routinely obtained reproducible glycogen contents when the OD730nm value of cell resuspension prior to cell lysis was between 2 and 10. Cell resuspensions with an OD730nm value of 1 or below gave rise to signals close to the minimum detection limit, leading to highly variable results. Cell resuspension with an OD730nm value higher than 20 was not suitable because cell lysis was incomplete, and either extended lysis or dilutions were required.
Figure 1 shows representative results of the glycogen contents in Synechocystis sp. PCC 6803 wildtype and two mutant strains (ΔpmgA and ΔpmgR1). The cells grown at the exponential growth phase were used. The glycogen contents were first normalized by the total protein contents and were subsequently expressed relative to the value for the wildtype. The results show that the mutant strains have glycogen contents that are at least two-fold higher than the wildtype strain.
Next, the glycogen content was analyzed in strains of Synechococcus sp. PCC 7002 engineered to produce mannitol5. The first strain (Glg+) contains the wildtype glgA1 and glgA2 genes encoding two functional glycogen synthases, whereas the second strain (Glg–) lacks functional copies of these genes5. The glycogen and mannitol contents were then measured in both strains (Figure 2). The results show that the Glg– lacked a detectable level of glycogen, while it produced more mannitol than the Glg+ strain. This suggests that the carbohydrate synthesized by photosynthesis is redirected to mannitol in the mutant strain that lacks the ability to synthesize glycogen. The OD730nm values of the cultures were approximately 10, providing sufficient cell materials for cell dry weight analysis. The glycogen contents were normalized using the cell dry weight.
Figure 1: The glycogen contents measured in different lines of Synechocystis sp. PCC 6803. Relative glycogen contents in the WT and in two mutants (ΔpmgA9 and ΔpmgR18) are shown. The glycogen levels were normalized by the total protein contents. Means of three biological replicates are shown, with error bars representing standard deviations. Please click here to view a larger version of this figure.
Figure 2: Glycogen and mannitol production in genetically manipulated Synechococcus sp. PCC 70025. Glg+, the strain synthesizing mannitol and glycogen. Glg–, the strain synthesizing mannitol but no glycogen. The values depicted are the average of three biological replicates, with error bars representing the standard deviations. CDW: cell dry weight, N.D.: not detected. Please click here to view a larger version of this figure.
Critical steps within the protocol are glycogen precipitation and resuspension. After centrifugation following ethanol precipitation, glycogen forms a translucent pellet that loosely adheres to the walls of the microcentrifuge tubes. Therefore, when removing the supernatant, special attention needs to be given so as not to remove the pellet. The glycogen pellet is sticky, and solubilization can be difficult if it dries out. Note that the complete solubilization of the glycogen pellet is important because incomplete solubilization will lead to inefficient enzymatic digestion and will therefore give rise to large variations between technical replicates. The application of sonication prior to solubilization by vortexing may facilitate the process.
Concerning the choice of normalization method, the cell dry weight is a widely-used reference. It is more laborious and requires more cell biomass than determinations of protein and chlorophyll a. The total protein content provides an alternative reference to the cell biomass and can be determined on a small scale, as described in the present protocol. The total protein content per dry cell biomass is typically in the range between 40% and 50%, although the exact value depends upon growth conditions25,28. The chlorophyll a content can be determined readily and can be used as a proxy to the cell amount present in the sample. However, it is well known that the amount of chlorophyll a per cell varies significantly depending upon the growth conditions, particularly in response to changing nutrient concentrations and light intensities29.
One of the significant advantages of the presented technique with respect to other methods is that it is highly selective. The hot acid and phenol protocol and the monosaccharide composition analysis following acid hydrolysis are relatively simple methods and have been used in previous studies9,10,12,13. However, these methods can overestimate the glycogen content because non-glycogen glucose and additional sugars can contribute to the measurement, depending on the carbohydrate detection technique12. The described technique selectively detects glucose in glycogen. It can also discriminate glucose from cellulose, because α-amylase and α-amyloglucosidase do not hydrolyze the β-linked glucosyl linkages present in cellulose. In previous studies, the enzymatic hydrolysis of glycogen was performed solely by α-amyloglucosidase7,27. Inclusion of the endo-acting α-amylase together with the exo-acting α-amyloglucosidase, as presented in this protocol, can ensure the efficient hydrolysis of glycogen. A similar selective hydrolysis is routinely applied to treatments of plant biomass, wherein the combination of α-amylase α-amyloglucosidase is used to hydrolyze starch without affecting other glucose-containing polysaccharides, such as cellulose21.
The main limitation of the technique is that the procedure is low-throughput because individual samples are processed separately using microcentrifuge tubes. The glycogen precipitation step is the primary factor preventing higher throughput. Implementation as a high-throughput procedure would require the use of deep-well plates and the ability to centrifuge these at a high speed (20,000 x g). While centrifuges that can centrifuge 96-well plates at the necessary speed are available, most deep-well plates cannot tolerate force larger than 6,000 x g. Hence, the careful choice of materials is required to adapt the protocol for high-throughput analysis.
The authors have nothing to disclose.
The authors acknowledge Nordic Energy Research (AquaFEED, project no. 24), Innovationfonden Denmark (Pant Power, project no. 12-131844), and Villum Fonden (project no. 13363)
QSonica Sonicators Q700 | Qsonica, LLC | NA | QSonica |
SpectraMax 190 Microplate Reader | Molecular Devices | NA | Eliza plate reader |
Bullet Blender Storm | Next Advance | BBY24M-CE | Beads beater |
Ultrospec 3100 pro UV/Visible Spectrophotometer | Amersham Biosciences | NA | Spectrophotometer |
Tris | Sigma-Aldrich | T1503 | Buffer |
HCl | Merck | 1-00317 | pH adjutment |
Sodium acetate | Sigma-Aldrich | 32319 | Buffer |
Amyloglycosidase (Rhizopus sp.) | Megazyme | E-AMGPU | Enzyme for glycogen depolymerization |
α-Amylase, thermostable (Bacillus licheniformis) | Sigma-Aldrich | A3176 | Enzyme for glycogen depolymerization |
D-Glucose | Merch | 8337 | Standard for the glucose assay |
Pierce BCA Protein assay kit | Thermo Fisher scientific | 23225 | For determination of protein concentrations |
Aluminum drying trays, disposable | VWR | 611-1362 | For determination of cell dry weights |
D-Glucose assay kit (GODPOD format) | Megazyme | K-GLUC | For determination of glucose concentrations |
Zirconium oxide breads, 0.15 mm | Next Advance | ZrOB015 | Beads for cell lysis in a Bullet Blendar Storm |
RINO tubes | Next Advance | NA | Tubes for cell lysis in a Bullet Blendar Storm |