Research review paperRegulation of glycolytic flux and overflow metabolism depending on the source of energy generation for energy demand
Introduction
The primary concern of living organisms is the uptake and breakdown of nutrients, together with their metabolism for energy generation (catabolism) and biomass synthesis (anabolism). The cell growth and catabolic rates are modulated by the multi-level regulation machinery consisting of gene expression (transcriptional regulation), post-transcriptional regulation, translation, post-translational regulation for ultimately modulating the metabolic fluxes (or enzymatic reaction rates). The metabolic fluxes are located on top of the hierarchical regulation systems, and represent the outcome of the integrated response of all levels of cellular regulation systems. It is thus important to understand how the metabolic fluxes are regulated from each level of regulation cascade. Adaptive laboratory evolution indicates that the inherent cell metabolism is robust, and new metabolic pathways are rarely created by evolution. Instead, the cell copes with the genetic and environmental perturbations by the combinatorial changes in capacity and activation of inherent metabolic pathways (Fong et al., 2006). Moreover, metabolic fluxes change little after evolution with some flux changes towards overflow metabolism for fast growth (Long et al., 2017).
Although the metabolic regulation mechanisms depend on the growth condition and the organism-specific evolution, some of them are well conserved in wide range of organisms from bacteria, fungi, mammalian cells to higher organisms. This gives us some hint for the in-depth understanding of the metabolic regulation mechanisms for the diverse species and evolutional changes (Smith and Morowitz, 2004). In particular, central carbon metabolism (CCM) is the hub from which the precursors of the cell constituents and energy are generated (Noor et al., 2010).
The energy efficient metabolism with full utilization of respiration is employed at low growth rate with low glucose uptake rate (GUR), while energy inefficient respiro-fermentative metabolism is employed at higher growth rate with higher GUR, showing overflow metabolism in Escherichia coli (Molenaar et al., 2009). This is rather common phenomenon observed also in other bacteria such as Bacillus subtilis (Sonneshein, 2007) and lactic acid bacteria (Teusink et al., 2006), as well as yeast such as Saccharomyces cerevisiae known as Crabtree effect (Crabtree, 1929), and tumor cells and cell lines including cancer cells known as Warburg effect (Warburg, 1956).
The transition from energetically efficient metabolism to inefficient metabolism has been recognized to be determined by a balance between costs of protein synthesis and benefits of enzyme activities, as shown to be a solution to the cost-benefit optimization problem for protein expression (Dekel and Alon, 2005), due to molecular crowding (Beg et al., 2007; Vazquez et al., 2008), or more specifically due to competition for the limited cytosolic membrane space between respiratory chain proteins and nutrient transporters (Zhuang et al., 2011, Szenk et al., 2017). The protein-to-lipid ratio must be kept below a critical level (less than about 50%) to maintain the cell integrity, and thus the available membrane space for proteins to express is limited (Molenaar et al., 2009). The fermentative metabolism is partly employed when the respiratory proteins reach their crowding limit, allowing further ATP production per unit of membrane area. Surface limitation thus forces the cell to modulate the metabolism based on the trade-off between membrane efficiency and ATP yield (Szenk et al., 2017).
The overflow metabolism can be analyzed by combining flux balance analysis (FBA) with molecular crowding (FBAwMC) imposing a global capacity constraint on the total cellular volume occupied by all metabolic enzymes (Beg et al., 2007) or on the total mass of enzymes (Shlomi et al., 2011). The acetate formation by overflow metabolism can be analyzed by a geneome-scale model (GEM) with FBAwMC by considering the efficiency of the oxidative phosphorylation in E. coli (Beg et al., 2007, Vazquez et al., 2008). The overflow metabolism has been also analyzed for other organisms such as S. cerevisiae (as Crabtree effect) (van Hoek and Marks, 2012, Nilsson and Nielsen 2016), and cancer cells (as Warburg effect) (Shlomi et al., 2011; Vasquez and Oltvai, 2011). The constraints may be imposed on the solvent capacity of cellular compartments (Vazquez and Oltavi, 2016) or cellular membrane (Zhuang et al., 2011, Szenk et al., 2017).
A variety of modeling approaches have been considered so far for the analysis of overflow metabolism such as coarse-grained whole cell model (Basan et al., 2015, Molenaar et al., 2009, Weise et al., 2015), genome-scale model (Vazquez and Oltavi, 2016; van Hoek et al., 2012, Mori et al., 2016, Nilsson and Nielsen, 2016, Sanchez et al., 2017), and extended genome-scale model (O'Brien et al., 2015, Machado et al., 2016, Sanchez and Nielsen, 2015, Goelzer et al., 2015, Noor et al., 2016). Another important approach is to analyze the growth characteristics by elementary flux modes (EFMs) (Wortel et al., 2014). An EFM is a metabolic sub-network defined by stoichiometry (without enzyme kinetics), or a minimal set of thermodynamically feasible reactions that form a network from start node to terminal node (from substrate to biomass), where none of its pathway enzymes can be removed (Schuster et al., 2000). In general, any flux distribution can be decomposed into positive linear combination of EFMs (Schuster et al., 2000), and a cost vector can be assigned to each EFM in the constrained space, in which EFMs with lower enzyme costs produce a higher flux for the cell synthesis (de Groot et al., 2017).
Above approaches are certainly useful to interpret the overflow metabolism from the point of view of enzyme efficiencies and protein costs based on coarse-grained approach. However, it is preferred to clarify the metabolic changes for such phenomenon based on the metabolic regulation mechanisms (Millard et al., 2017; Shimizu, 2016, 2013). Moreover, it is not obvious to which extent the membrane-space limitation or the cellular volume capacity constrains the metabolism, since the cell size in terms of length and width changes with respect to the cell growth rate (Westfall and Levin, 2018). In the present article, therefore, the metabolic characteristics of overflow metabolism and modulation of the glycolytic flux are considered from the points of view of the metabolic regulation mechanisms. In particular, the metabolic changes are considered from the point of view of the source of energy generation. Moreover, the effects of culture condition such as oxygen limitation (that alters the metabolism towards fermentation), and genetic perturbation (specific pathway mutation) affecting the source of energy generation are also considered. Although attention is somewhat focused on the metabolic regulation mechanisms of E. coli, the similarities to and differences from other organisms will be also discussed.
Section snippets
Cell growth characteristics
The cell growth and nutrient uptake rates are of primal importance for the cell survival and for outcompeting with other organisms in the varying growth conditions in nature. The living organisms must cope with a variety of environmental changes by appropriately regulating the metabolisms for the efficient production of biomass precursors, redox cofactors such as NAD(P)H, and energy such as ATP with appropriate stoichiometry (Chubukov et al., 2014). Available nutrients are transported across
Enzyme level and transciptional regulations of the central metabolism
In order to understand the underlying mechanisms for the change in the metabolism, the fast (on the order of seconds to minute) enzyme level or allosteric regulation, and slow (on the order of minutes to hour) transcriptional regulation must be considered. The cell gains energy from substrates for biosynthesis via redox reactions in which electrons and/or hydrogen atoms (protons) are transferred for energy generation, where pyridine nucleotides such as NAD(P)H play essential roles for this. The
Characteristics of glycolysis and channeling of metabolites
Glycolysis is common to many organisms and plays essential roles for both catabolism and anabolism, where the glycolysis is meant here by the breakdown of carbon sources specifically via the Embden-Meyerhoff-Parnas (EMP) pathway. In the glycolysis, the estimated cumulative free energy change, ΔG is on average − 46 kJ/mol, where phosphofructo kinase (Pfk) is the most irreversible rate-limiting forward reaction, consuming nearly half of the available free energy (Park et al., 2016). Although Pyk
Glycolytic flux is affected by the source of energy generation
The glycolysis provides ATP by substrate level phosphorylation as well as reducing equivalent (NAD(P)H), and some of its intermediates are the precursors for the cell synthesis. In particular, the terminal metabolites such as PEP and pyruvate (and also AcCoA, one-step from pyruvate) are the starting metabolites for the fermentative pathways. Thus, modulation of the glycolytic flux is by far important for the cells to survive under varying growth conditions and for the metabolic engineering
Transcriptional regulation of the source of energy generation
The question may arise on how the respiratory activity is regulated. As the glycolytic flux increases, the pool sizes of the upper glycolysis such as FBP/fructose 1-phosphate (F1P) increase, and inhibit the activity of catabolite repressor/activator (Cra), and thus Cra activity linearly decreases with respect to the glycolytic flux, enforcing the flux-sensing machinery (Kochanowki et al., 2013), where F1P and FBP both (inversely) correlate with Cra activity in E.coli (Gerosa et al., 2015,
Short- and long-term overflow metabolisms
The mechanism for a short-term overflow metabolism may be different from the long-term steady-state metabolism. The short-term overflow occurs as the immediate onset of respiro-fermentative metabolism on the order of several seconds to minutes, where the enzyme level regulation dominates, which is eventually replaced by the transcriptional regulation. Upon glucose supply, FBP concentration rapidly increases with the increase in the glycolytic flux (Link et al., 2013), and the increased NADH
Effect of genetic perturbation for glucose transport and energy generation via respiration on the glycolytic flux
Fig. 9 shows the specific GUR and acetate formation rate of several pts mutants including ΔptsG, and the related double and triple mutants with respect to the cell growth rate under aerobic condition (Fuentes et al., 2013, Steinsiek and Bettenbrock, 2012). Moreover, the specific GUR of the wild type strain cultivated under anaerobic condition is also shown in the same figure (Steinsiek and Bettenbrock, 2012). Comparison of Fig. 9 with Fig. 6 indicates that the characteristics of the specific
Increase in the glycolytic flux under oxygen limitation
The glycolytic flux (or equivalently the specific GUR) increases under oxygen limitation in E.coli (Emmerling et al., 2000), in S. cerevisiae (van der Brink et al., 2008), and in Lactococcus lactis (Neves et al., 2002) (as implied by Fig. 6). In the case of E.coli, cAMP level decreases under oxygen limitation (Steinsiek and Bettenbrock, 2011). The expression of ptsG gene is activated by cAMP-Crp (Shimada et al., 2011), which implies that ptsG gene expression is repressed under oxygen limtation.
The regulation of glucose uptake under sugar-phosphate stress and the stability of mRNA
The input to the glycolysis is the sugar molecules imported via transporters, and the glycolytic flux may be also regulated by the pool size of glucose 6-phosphate (G6P), starting metabolite for the glycolysis. Glucose molecule (and various PTS-sugar molecules) is transported and phosphorylated by EIICB, yielding a phosphate sugar such as G6P, which is in turn metabolized through glycolysis and PP pathway. Once the sugar was internalized with phosphorylation, the charged phosphate group
Regulation of PDHc and the effect of its mutation
In relation to the glycolytic flux regulation, pyruvate dehydrogenase complex (PDHc) plays an important gate role located between glycolysis and the TCA cycle. The aceE, aceF, and lpdA genes encoding each subunit of E1 (AceE: pyruvate dehydrogenase), E2 (AceF: dihydrolipoate acyltransferase), and E3 (LpdA: dihydrolipoate dehydrogenase) of PDHc together with pdhR gene form an operon as pdhR-aceE-aceF-lpdA in E.coli. The lpdA gene also encodes part of the 2-ketoglutarate dehydrogenase (KGDH) and
Concluding remarks
It is of great interest to understand the design principle that elucidates how the metabolism is coordinately regulated. In particular, the main concern is to understand why and how the glycolytic flux increases with overflow metabolism by changing the source of energy generation from energetically efficient respiration to inefficient respiro-fermentative metabolism as the cell growth rate (glycolytic flux) increases. Although not completely transparent, it may be useful to summarize the
Author contribution
Both KS and YM equally contributed to the present work.
References (244)
- et al.
Channeling in native microbial pathways: Implications and challenges for metabolic engineering
Biotechnol. Adv.
(2017) - et al.
Structural and biochemical evidence for an enzymatic quinine redox cycle in Escherichia coli: identification of a novel quinol monooxigenase
J. Biol Chem.
(2005) - et al.
Feedback control of ribosome function in Escherichia coli
Biochimie
(2008) - et al.
Endonucleolytic initiation of mRNA decay in Escherichia coli
Prog. Mol. Biol. Transl. Sci.
(2009) The Entner-Doudoroff pathway: history, physiology and molecular biology
FEMS Microbiol. Rev.
(1992)- et al.
Mitochondrial oxidative phosphorylation is regulated by fructose 1,6-bisphosphate
J. Biol. Chem.
(2008) - et al.
Pyruvate kinase of Bacillus subtilis
Biochim. Biophys. Acta
(1972) - et al.
Latent pathway activation and increased pathway capacity enable Escherichia coli adaptation to loss of key metabolic enzymes
J. Biol. Chem.
(2006) - et al.
Pseudo-transition analysis identifies the key regulators of dynamic metabolic adaptations from steady-state data
Cell Syst.
(2015) - et al.
Quantitative prediction of genome-wide resource allocation in bacteria
Metab. Eng.
(2015)