ReviewDifferences in cytosolic glucose dynamics in astrocytes and adipocytes measured by FRET-based nanosensors
Graphical abstract
Introduction
Glucose is the main source of energy and carbon used for metabolic and structural purposes in cells. Upon entering the cell, the first step in glucose utilization is phosphorylation by the enzyme hexokinase (HK) into glucose-6-phospate (G6P). G6P is a branching point leading to different metabolic pathways. Most commonly, G6P is degraded into pyruvate and lactate in the process of glycolysis. G6P can also be converted to acetyl coenzyme A and subsequently to fatty acids or directed to pentose phosphate pathway – an alternative route for glucose oxidation. In cells that can synthesize and accumulate glycogen, such as liver and muscle cells, adipocytes and also astrocytes [12], part of G6P is also converted to this form of short-term energy storage via glycogenesis. The fate of G6P depends on various factors on the cellular (eg energetic needs, oxygen availability, rate of glucose uptake, enzyme availability and activity) and on the whole body level (eg nutrient availability, hormonal signalling, physical activity). Importantly, G6P is an inhibitor of HK, so if other pathways are slow, eg, if phosphofructokinase, an enzyme committing glucose to glycolysis, is inhibited, G6P will increase and inhibit HK [94] and glucose will accumulate in the cytosol. Thus, the level of glucose in the cytosol mirrors the balance between the rate of glucose uptake across the plasma membrane and the efficiency of G6P utilization.
Real-time measurements of many cellular metabolites and signalling molecules with subcellular spatial resolution have only recently been introduced owing to the development of optophysiological techniques based on fluorescence resonance energy transfer (FRET) and the concomitant development of microscopy techniques and genetic engineering. The development of fluorescent indicator protein (FLIP) family sensors for various metabolites was pioneered by the Frommer's lab [29,41].
Standard techniques for studying cell glucose metabolism are based mostly on measurements of glucose uptake by cells using non-metabolizable glucose analogues, (eg, 3-O-methylglucose, 2-deoxyglucose), which are fluorescently tagged or radio-labelled, followed by enzymatic reactions and photometric or amperometric assays for detections [102]. With regard to glucose metabolism, these methods contribute significantly to the understanding of glucose homeostasis, but are still limited in some respects: first, they lack information on glucose utilization, providing only insight into the regulation of membrane transport; second, they are conducted on the cell population and thus do not provide information at the single cell level, thereby neglecting the influence of heterogeneity between cells, and third, the data obtained reflect static information. Some other techniques, such as nuclear magnetic resonance imaging and positron emission tomography, can provide dynamic data on glucose, but again with very limited spatial resolution [5]. On the other hand, FRET-based techniques also have specific limitations, especially regarding the sensitivity of fluorescent proteins to changes in the local environment (e.g. pH, temperature), refractive index and relatively low signal-to-noise ratio; but these drawbacks have been extensively discussed elsewhere [53].
FRET-based glucose nano-sensors, genetically encoded molecules, are reliably employed to monitor real-time changes in cytosolic glucose levels in single cells [41]. They usually consist of a binding protein for a target molecule (eg, glucose-binding protein) that is flanked with a FRET pair of green fluorescent protein variants (eg, variations of yellow and cyan fluorescent protein; YFP, CFP) - one donor and another acceptor fluorophore [28,80]. The donor: acceptor stoichiometry in such nano-sensors is 1:1, which is an important factor for proper FRET detection. FRET is a non-radiative energy transfer between two compatible FRET pair, thus the emission and excitation spectra of donor and acceptor fluorescence, respectively, must overlap significantly in order to observe the change. Furthermore, for FRET efficiency, it is important that the distance between the two fluorophores does not exceed 10 nm, [52] and that the fluorophores display suitable dipole orientation. Upon substrate binding (eg, glucose), the distance between the two fluorophores changes due to the allosteric change in the binding protein, resulting in FRET changes and, consequently, changes in the fluorescence intensities of both fluorophores in opposite directions (Fig. 1). Changes in cytosolic glucose levels are then readouts of a change in the ratio of two real-time fluorescence intensities (change in FRET signal) over a defined region of the imaged cell [22,52]. Importantly, glucose sensors measure free-glucose levels and are insensitive to phosphorylated glucose or other glucose metabolites [29,43], thus directly reflecting changes in glucose uptake and degradation.
The dynamics of cytosolic glucose in living cells (Fig. 2) was explored using FRET-based protein sensors on several different cell types and settings (Table 1), such as yeast [7], 3T3-L1 adipocytes [8,16,48], myotubes and myoblasts [43,104], pancreatic beta cells [45], astrocytes [8,9,65,76,77,98], neurons [4,8,23,75], COS-7 cells [29], HepG2 cells [91] and plant cells [22]. Interestingly, the concentration and dynamics of cytosolic glucose differ among different cell types.
Experiments with FRET-based glucose sensors are mainly conducted on cells in cell cultures, however some studies have also used in vivo (eg, small mammals, insects) and ex vivo (acute tissue slices, small organs) models [61,96,100]. Regarding physiology, each approach has certain advantages not found in the other. Cell culture models are usually very practical and easy to implement and the experimental conditions are well-defined, easier to control, and allow easier experimental manipulation of cells. On the other hand, in ex vivo/in vivo models tissue architecture and connectivity are maintained and cells retain their functional morphological specialization. Thus, we can study cells in their physiological environment, capture interactions between different cell types, and test the influence of more complex cellular communication.
Assessment of cytosolic glucose concentration is of particular interest in insulin-sensitive cells (muscle and fat), to study glucose metabolism under different conditions. Glucose is also an important fuel for the brain, where 20% of the total glucose in the human body is utilized. Glial cells, particularly astrocytes, are important modulators of neuronal metabolism and function [70]. Thus, recently, interest in monitoring cytosolic glucose in astrocytes has accelerated, as it is imperative for understanding brain metabolism in health and neurodegeneration.
Section snippets
Cytosolic glucose levels are dependent on glucose availability
In living organisms, glucose availability for cells refers mostly to the concentration of glucose in the blood and in the interstitial space, and varies frequently due to nutritional status. In in vitro experimental settings, glucose availability depends on the glucose concentration in the culture medium, determined by varying culture conditions. In addition, in specialized cell types such as hepatocytes, kidney cells and astrocytes, the source of cytosolic glucose is also endogenous
Membrane permeability to glucose is cell-type specific and developmentally regulated
Glucose enters all cells through facilitated transport, i.e. through glucose transporters or GLUTs that are located in the plasma membrane (see Reviews in [50,67,85]. This ATP-independent, bi-directional transport of glucose across the cell membrane is carried out down its concentration gradient [58,60,93]. In most cells, cytoplasmic glucose levels are lower than plasma- and interstitial fluid glucose levels; hence, glucose flux is directed to the cytosol. To date, fourteen subtypes of this
Cytosolic glucose levels reflect the efficacy of heksokinases
Upon entering the cytosol, glucose is phosphorylated by the enzyme hexokinase (HK) to glucose-6-phosphate (G6P). This step allows the trapping of glucose in the cell, because G6P transport is not supported by GLUTs. This reaction simultaneously serves to maintain a low intracellular concentration of glucose, and thereby a concentration gradient of glucose across the plasma membrane, in a direction that allows a constant uptake of glucose into the cell.
There are four types of HKs, isoforms I-IV.
Changes in cytosolic glucose levels upon insulin action are cell-type specific
The low membrane permeability to glucose in adipocytes at rest is not surprising, considering that a major part of postprandial plasma glucose rise is normalized by glucose uptake into three major insulin-responsive cell types: muscle cells, hepatocytes [64] and adipocytes. Adipocyte glucose uptake represents 10% of whole-body insulin-stimulated glucose uptake [18]. In addition to the lipid storage compartment, the role of adipocytes is to buffer a part of excess plasma glucose, which is
Concluding remarks
FRET–based experiments on cytosolic glucose dynamics have often reported that a variable percentage of cells do not respond with an increase in cytosolic glucose upon exposure to certain stimuli [48,65,98]. It is unclear whether this phenomenon is due to cell heterogeneity [6] or due to cell cycle dependent plasma membrane organization variability [15,21], or both. Yet, these aspects cannot be observed in various biochemical approaches, where the cell population rather than the individual cell
Declaration of Competing Interest
The author declares no conflicts of interest.
Acknowledgement
This work was supported by Slovenian Research Agency (grants P3 310, J3-6790, J3-9266).
References (104)
- et al.
Why glucose transport in the brain matters for PET
Trends Neurosci.
(2005) - et al.
Carbohydrate metabolism
- et al.
Transport and transporters in the endoplasmic reticulum
Biochim. Biophys. Acta
(2007) - et al.
Potential mechanism of insulin action on glucose transport in the isolated rat adipose cell. Apparent translocation of intracellular transport systems to the plasma membrane
J. Biol. Chem.
(1980) - et al.
Cell cycle dependent changes in the plasma membrane organization of mammalian cells
Biochim. Biophys. Acta Biomembr.
(2017) - et al.
In vivo imaging of the dynamics of glucose uptake in the cytosol of COS-7 cells by fluorescent nanosensors
J. Biol. Chem.
(2003) - et al.
Arachidonic acid stimulates the intrinsic activity of ubiquitous glucose transporter (GLUT1) in 3T3-L1 adipocytes by a protein kinase C-independent mechanism
Cell. Signal.
(1996) - et al.
Astrocytic insulin Signaling couples brain glucose uptake with nutrient availability
Cell
(2016) - et al.
The regulation by insulin of glucose transporter gene expression in 3T3 adipocytes
J. Biol. Chem.
(1989) - et al.
Evidence that erythroid-type glucose transporter intrinsic activity is modulated by cadmium treatment of mouse 3T3-L1 cells
J. Biol. Chem.
(1991)