ABSTRACT Membrane transport proteins represent a large amount of cellular protein. There are 5 types of monosaccharide transport proteins that transport substrates such as glucose and fructose across the plasma membrane. GLUT-1 is a glucose transporter that is abundant in the heart and the red blood cell line. Under cellular stress, glucose uptake is increased by activation of GLUT-1 transporters pre-existing in the plasma and translocation of transporters from intracellular stores to the plasma membrane. 3T3-L1 fibroblasts copiously produce GLUT-1 transporters and exhibit GLUT-1 activation and translocation. Therefore, this cell line was used to investigate the activation and translocation of GLUT-1 under cellular stress caused by incubation with sodium azide and sorbitol. Our results showed that there was an increase in glucose uptake under cellular stress. Azide and sorbitol produced a 2.89 fold and 3.2 fold increase in [3H]-2-deoxy-D-glucose uptake respectively. A 1.48 fold increase in glucose uptake under sorbitol incubation was observed in preconfluent cells, but this was a single experiment. It was evident that translocation occurred in preconfluent fibroblasts and in confluent fibroblasts little or no translocation was apparent. Owing to lack of time, it was not clear that activation and translocation occur simultaneously and that there is a significant difference in the amount of glucose uptake between preconfluent and confluent fibroblasts. Therefore, further investigation will be required. ACKNOWLEDGEMENTS I would like to thank Dr Kay Barnes for helping me with my experiments, teaching me basic cell culture techniques and using her own time for my benefit. I would also like to thank Mrs Jean Ingram for great help with experiments and cell culture also. A big thanks to Prof. Stephen Baldwin for allocating this project to me and answering my queries. Lastly, I would like to thank Dr Zoë Booth for general help and moral support. ABBREVIATIONS AMP Adenosine monophosphate ATP Adenosine triphosphate DMEM Dulbecco’s modified Eagle’s medium FITC Fluorescein isothiocyanate GLUT Glucose transporter KRH Krebs-Heinseleit buffer PBS Phosphate buffered saline PI-3-KINASE Phosphatidylinositol-3-kinase PM Plasma membrane
1.0 INTRODUCTION Membrane transport proteins represent a large amount of cellular protein. They transport molecules such as ions, nucleosides and carbohydrates across the plasma membrane. Of the carbohydrates, monosaccharides such as glucose and fructose are transported by monosaccharide transport proteins. 1.1 Isoforms Monosaccharide transport proteins can either be constitutive or regulated. Constitutive transporters act as a kind of housekeeping transporter; while regulated transporters are induced under certain stimuli e.g. stress, insulin. Transporters can either transport monosaccharides by the passive process of facilitated diffusion (Figure 1) or by the active process of symport with Na+ (secondary active transport). All the passive monosaccharide transporters are 12 transmembrane (12TM) integral membrane proteins (Mueckler et al., 1985) and there are at least five isoforms in mammals: Table 1 Isoforms of monosaccharide transport proteins (GLUT) Isoform Location Substrate Reference GLUT-1 Intestinal epithelium, heart Glucose Ismail-Beigi et al., 1998 GLUT-2 b-cells of the islets of Langerhans in the pancreas Glucose Tomita,1999 GLUT-3 Brain, neurones Glucose Lucas et al., 1999 GLUT-4 Heart, muscle Glucose Fingar et al., 1999 GLUT-5 Kidney, small intestine Fructose Matzuzaki et al., 1999 There are also GLUT-6 and GLUT-7 transporters, but these are pseudogenes and are an artefact of cloning. The principle sugar transported is glucose. Glucose transporters can either be situated intracellularly (intracellular pools on nucleus and golgi) or reside at the cell surface. When the transporters are intracellular, they are usually located on vesicles. Under stimulation, vesicles translocate and dock with the plasma membrane e.g. GLUT-4 by insulin stimulation (Figure 2). SNAP23 is a protein involved in docking and consequently increases GLUT-4 mediated glucose uptake (Foster et al., 1999). GLUT-1 docks in a similar manner and is stimulated by cellular stress e.g. ischaemia and hypoxia. 1.2 Pathways involved in translocation/activation Cellular proteins are often regulated by phosphorylation/dephosphorylation of serine and threonine residues by kinases and phosphatases respectively. This is a simple mechanism of activation and deactivation of a protein, and can lead to many other activations and deactivations of proteins and thus causes a cascade. Signal transduction cascades are involved in translocation and activation of transporters. Insulin binds to a receptor which is a tyrosine kinase. It autophosphorylates tyrosine residues present on the receptor. After this event, many other proteins are involved in the translocation of GLUT-4 to the plasma membrane. One of these is insulin receptor substrate 1 (IRS-1), and another major protein is phosphatidylinositol-3-kinase (PI-3-kinase) (Oatey et al., 1999). PI-3-kinase is an enzyme that phosphorylates phosphatidylinositol bisphosphate (PIP2) to become PIP3. PIP3 can be further metabolised to produce many other cellular responses e.g. apoptosis (Downward et al., 1997). GLUT-4 is a widely studied transporter and results show that phosphatidylinositol-3-kinase (PI-3-kinase) is involved in the mechanism of translocation under insulin stimulation. Wortmannin is a fungal metabolite that is a PI-3- kinase inhibitor and when cells are exposed to this, glucose uptake is not increased in the presence of insulin (Fingar et al., 1999). GLUT-1 is less studied but recent results also suggest that PI-3-kinase is also involved in its translocation under insulin stimulation (Egert et al., 1999). The mechanism of translocation under stressful i.e. ischaemic conditions is still undefined. AMP-activated protein kinase (AMPK) is an enzyme that acts as a metabolic sensor and monitors AMP and ATP levels (Carling et al., 1998). If the AMP:ATP ratio rises, the enzyme switches off ATP consuming anabolic pathways and switches on ATP producing pathways, such as fatty acid oxidation. Since glucose transport and consequent oxidation ultimately produces ATP, this enzyme is fundamental the regulation of glucose transport. 1.3 Stress activated pathways Reactive oxygen species (ROS) such as O2- and hydrogen peroxide (H2O2) are produced during intracellular oxidation and reduction reactions such as oxidative phosphorylation. H2O2 is a stress compound that in synthesised in vivo and can stimulate GLUT-1mediated glucose transport, as well as other cellular processes. Phospholipase C (PLC) and protein tyrosine kinase are linked to an increase in glucose transport under cellular stress (Ismail-Beigi et al., 1999). The production of H2O2 will stimulate PLC and protein tyrosine kinase to increase glucose uptake, and consequently increase intracellular ATP levels. Stress activated protein kinase (SAPK), a mitogen activated protein kinase, is stimulated (Ismail-Beigi et al., 1999) under stress. It is not directly linked to an increase in glucose transport, so the exact pathophysiological role of this kinase is not yet established. Of the 5 transporter isoforms, GLUT-1 and GLUT-4 are particularly involved under cellular stress (Haddad et al., 1997). Inhibition of oxidative phosphorylation as a form of stress causes GLUT-1 and GLUT-4 to translocate to the cell surface, or activate GLUT-1 transporters that pre-exist in the plasma membrane (Elkhairi et al., 1999). Prolonged exposure to stress results in enhanced transcription of GLUT-1 (Behrooz et al., 1999). GLUT-1 is a glucose transporter isoform of 492 amino acids (Mueckler et al., 1985) which is expressed in the human red blood cell line (Ismail-Beigi et al., 1998) and is abundantly located in the heart (Egert et al., 1999). A large percentage of GLUT-1 transporters reside at the cell surface and can be activated by transduction pathways under stressful conditions. An increase in intracellular free calcium concentration has been reported to stimulate glucose uptake via GLUT-1 (Ismail-Beigi, 1993). It has been previously shown (Ismail-Beigi et al., 1998) that upon stimulation, there was no significant difference in the amount of GLUT-1 present on the plasma membranes of control or stimulated cells. This would suggest the activation of GLUT-1 already present in the plasma membrane. A GLUT-1 C-terminal binding protein (GLUT1CBP), which acts via a PDZ domain at the C-terminus, could provide a link to its activation (Bunn et al., 1999). There is also recent evidence to suggest its interaction with an integral membrane protein, stomatin (Band 7.2b), in red blood cells (Ebina et al., 1999). N-glycosylation of GLUT-1 plays a role in its function, so this could be a factor to its activation. In leukaemic cells, if this post-translational event is inhibited GLUT-1s affinity for glucose decreases rapidly (Ahmed, 1999). Though, this would not be a major factor in healthy cells. GLUT-1 also translocates from intracellular stores under stimulation by stress. It can be assumed that under stimulation, the transporters - present on vesicles - translocate to the cell surface via microtubules and a motor protein e.g. kinesin (DePina et al., 1999) 1.4 Cell lines The baby hamster kidney (BHK) cell line is good for studying the translocation of transporters (Baldwin et al., 1990) under stress or insulin stimulation. Glucose transporters in clone 9 cells, a nontransformed rat liver cell line, are always present at the plasma membrane and are therefore they are a good cell line for studying the activation of glucose transporters (Ismail-Beigi et al., 1998). 3T3-L1 preadipocytes (fibroblasts) are a rat cell line but exhibit GLUT-1 translocation (Yang et al., 1992). Before differentiation to adipocytes by insulin, they only synthesise the GLUT-1 isoform as a means of glucose uptake (Kaestner et al., 1989). After differentiation, they produce the acutely insulin sensitive isoform, GLUT-4. Preconfluent (non-contacting) 3T3-L1 fibroblasts show a high density of GLUT-1 transporters on the plasma membrane, whereas in confluent cells the density of transporters is greatly decreased. They are sequestered at the plasma membrane (Yang et al., 1992). GLUT-1 is the only transporter to be translocated or activated under stress (Young et al., 1997) and confluent cells have a high intracellular pool of transporters that can translocate. Therefore, 3T3-L1 fibroblasts are a good cell line to study the increase in glucose uptake under cellular stress by activation and translocation of GLUT-1 to the plasma membrane. Research into stress induced glucose uptake is fundamental to understanding the increase in glucose uptake under physiological stress i.e. hypoxia, ischaemia and post stroke conditions. Using 3T3-L1 fibroblasts to study increased glucose uptake has yet to be performed. Clone 9 or BHK cells are frequently used and are a popular choice for this type of research. 1.5 Aims This research project is a study into the increase in glucose uptake under cellular stress (incubation with sodium azide or sorbitol) using 3T3-L1 fibroblasts as a model cell line. It will be achieved by performing [3H]-2-deoxy glucose transport assays to determine the fold increase in glucose uptake. And secondly, using immunocytochemical techniques to study the translocation of GLUT-1. A specific GLUT-1 polyclonal antibody will be used and visualised with a fluorescein isothiocyanate (FITC) labelled antibody. //Figure 1 The facilitated diffusion of glucose across the plasma membrane Glucose is impermeable to the plasma membrane due to its size. It has to be carried across the membrane down its concentration gradient by a facilitative glucose transporter. (Source: D. Voet and J.G. Voet., 1995). Figure 2 The translocation of GLUT-4 under insulin stimulation Under high blood sugar, glucose is transported across the plasma membrane and membranous vesicles are sequestered (down regulation). Under low blood sugar, insulin stimulates the translocation of the vesicles to the plasma membrane, via a signal transduction pathway, to increase glucose uptake. //(Source: D. Voet and J.G. Voet., 1995). 2.0 MATERIALS AND METHODS 2.1 Materials All reagents were obtained from Sigma Chemical Company plc (Poole, Dorset, U.K.) and were all AnalR grade. All water used in procedures was obtained from a Milli Q ultra-filtration system. 2.2 Cell culture 3T3-L1 fibroblasts were cultured on an 80 cm2 Nunclon flask (Nunc laboratories) to approximately 106 cells/ml in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% (v/v) newborn serum, 1% (v/v) L-Glutamine and 1% (v/v) penicillin/streptomycin. They were incubated at 37?C in 5% CO2 and became confluent in approximately 5 days. Cells were passaged when confluent with 0.05% trypsin in EDTA (Ethylene diamine tetra-acetic acid) in PBS (phosphate buffered saline, pH 7.2) and diluted 1:10 (cells : medium). Before incubation of stress compounds cells were incubated for 2 h at 37?C in low serum DMEM (DMEM containing 0.5% Foetal calf serum). 2.3 2-deoxy D-glucose transport assay Confluent fibroblasts were passaged into 3 x 6 well plates (Nunc laboratories) and incubated at 37?C in 5% CO2. They were incubated for 3 or 5 days for preconfluent and confluent assays respectively. Cells were serum-starved for 2 h and preincubated at 37?C for 30 min in stress inducing compound with 25 mM glucose (2 ml per well): 5 mM sodium azide or 400 mM sorbitol in a KRH buffer (pH 7.4). Cells in 3 wells were then exposed to 1ml of 200 ?M radioactive deoxy-glucose (8?Ci) for 5 minutes at 37?C, and cells in 3 wells were exposed to same solution with 4mM cytochalasin B (inhibits GLUT-1 mediated glucose transport). After 3 washes with ice cold PBS, cells were permeabilized with 1% Triton-X 100 for 5 minutes and 250 ?l was counted for 10 minutes per vial in a liquid scintillation counter. Counts from uptake in presence of cytochalasin B were subtracted from that in its absence to yield a mediated uptake rate. 2.4 Immunocytochemistry 36 x 13 mm coverslips were washed in concentrated hydrochloric acid and then stored in absolute ethanol. The ethanol was flamed off 12 coverslips in a flowhood. 4 coverslips were then placed in the top 3 wells of a 6 well plate. This was repeated until 3 x 6 well plates were occupied by cover slips. Fibroblasts were passaged into the wells and incubated at 37?C until confluent (106 cells/ml). 2.4.1 Preparation of plasma membrane (PM) lawns by freeze-thawing 3 mls of ice-cold PBS was added to a 6 well plate containing confluent fibroblasts and left on ice for 15 minutes. Cells were then freeze thawed by adding 1 ml of ice-cold distilled water to each well and floating the plate in ethanol - which was incubated in dry ice - for 5 minutes. The plate was then thawed in water kept at 37?C. Using a Gilson pipetteman P1000 (Gilson laboratories), the coverslips were flushed approximately 10 times to remove cytoplasm. Coverslips were fixed with 4% paraformaldehyde in PBS for 20 minutes at room temperature. The paraformaldehyde was quenched with 100mM glycine in PBS for 15 minutes at room temperature. Coverslips were stored in PBS/0.2% azide at 4?C until needed. 2.4.2 Permeabilization and staining Coverslips were blocked for 45 minutes with 20 ?l of 6% normal goat serum (NGS) /PBS. They were then incubated for 2 hours at room temperature in 20 ?l of polyclonal primary antibody, GLUT-1 in 1% NGS/0.1% saponin (1/50 dilution). For PM lawns 1% NGS in PBS was used. For background staining, control Rabbit IgG was used at a dilution of 1/200 (same diluent as for GLUT-1 antibody). Coverslips were incubated for 30 minutes in a fluorescently labelled secondary antibody, Goat anti Rabbit FITC (Fluorescein isothiocyanate). Coverslips were mounted in 20 ?l Vectashield (Vector laboratories) in the middle of a microscope slide. Duplicates were placed on the same slide approximately 10 mm apart. Images were obtained using a Nikon CCCD camera linked microscope and Open Lab deconvolution software. Images were obtained at a wavelength of 485 nm and were not deconvolved. An FITC colour was added for easier visualisation. 2.5 BCA Assay 250 ?l of duplicate samples from transport assay were incubated for 30 minutes at 37?C with 200 ?l of 5% copper sulphate/bicinchoninic acid (400 ?l CuSO4/BCA 20 ml) and absorbance read at 280 nm using a 96 well plate spectrophotometer. Sample concentrations were calculated from a standard curve of 10 concentrations (10% solution - 100% solution) of bovine serum albumin (BSA). 3.0 RESULTS 3.1 2-deoxy D-glucose transport assay We aimed to determine whether chemical (azide) or osmotic (sorbitol) stress had any effect on glucose transport in 3T3-L1 fibroblasts by activation of GLUT-1. The assay was repeated three times with confluent fibroblasts, and once with preconfluent. The data for each assay were plotted separately and shown as histograms. Also a graph to show non-mediated and mediated glucose transport was plotted i.e. glucose transport ??cytochalasin B (figure 6). Cytochalasin B inhibits GLUT-1 mediated transport of glucose, so it was used to observe if any non-mediated glucose transport was apparent i.e. simple rather than facilitated diffusion across the plasma membrane. Non-mediated transport was a small fraction of total glucose transport. Cytochalasin B efficiently inhibits GLUT-1 mediated transport by 59 % of the control value, 92.4 % for azide and 81.6% for sorbitol in assay with amended protocol (figure 6). Therefore, a correction could be made to cancel out any non-mediated glucose transport. The data obtained from cells in the presence of cytochalasin B was subtracted from data obtained from cells in the absence of cytochalasin B to acquire a realistic result of GLUT-1 mediated transport. In two initial experiments on confluent fibroblasts (figure 3 and 4), sodium azide was found to have a considerable effect on glucose uptake: 1.36 fold increase (figure 3) and 2.23 fold increase (figure 4) respectively. Sorbitol did not have a great effect on the stimulation of glucose uptake from these two experiments (0.84 and 1.14 fold respectively). To show if this was due to methodology, the protocol was amended with startling results. Adding 5 ?l of cytochalasin B to a dry monolayer of cells would produce inaccurate results due its distribution. Some cells would be inhibited before others and the duration of inhibition would be uneven. Therefore, 15 ?l of cytochalasin B was added to the 3 ml [3H]-2-deoxy glucose aliquot and consequently all cells received an even distribution. Figure 5 shows sorbitol (3.2 fold increase) had a greater effect than azide (2.89 fold increase), and they were both higher values than from previous experiments; sorbitol’s effect was more than double. From the first two assays, azide caused an increase in glucose uptake but the results were not greatly reproducible i.e. 1.36 fold and 2.28 fold increase; a 0.92 fold difference. If the assays were to be duplicated, the results would most likely be dissimilar due the uneven distribution of cytochalasin B on its addition. With the amended assay, it is more likely that the results would be very similar and more reproducible. It can be clearly seen from figure 5 that sorbitol had a large effect on glucose transport, but even azide had a larger effect on uptake compared to previous assays. Comparatively, this single assay performed with the amended protocol (figure 5) had a vast effect on glucose uptake in the presence of azide and sorbitol. To test if glucose uptake differed between confluent and preconfluent cells, a transport assay was performed to test this (Figure 7). In preconfluent cells, the effect of cellular stress on GLUT-1 mediated glucose transport is lessened compared to confluent cells. Azide did not produce any increase in glucose transport, it only produced a 0.68 fold increase. Sorbitol only produced a minor effect on glucose transport, and from the error bars of figure 7B (74.39 pmol to 110.45 pmol/min/well (SEM ± 19.83)) it can be seen that the increase is barely significant. Whether this is due to experimental error or is an artefact of preconfluent cells is unclear. Subsequent experiments would have to be performed to elucidate this. The difference in fold increase in glucose uptake between preconfluent and confluent cells in the presence of sorbitol is 2.16 fold (3.2/1.48). From the experiments performed, this indicates there is over double the increase in glucose transport from preconfluent to confluent cells. More experiments with preconfluent cells would have to be performed to confirm whether this is the case. //Figure 8 Standard curve for BCA assay achieved using bovine serum albumin (BSA) This standard curve was used to calculate the protein concentrations below. Table 2 Protein concentrations (mg/ml) for BCA assay in preconfluent (A) and confluent fibroblasts (B) A B 1.20 3.82 2.00 8.81 0.32 6.93 2.17 4.64 6.62 9.02 2.48 5.52 8.23 8.47 3.74 7.01 Average: 3.34 Average: 6.78 per µl: 0.07 Per µl: 0.27 per well: 33.40 Per well: 135.60 Absorbance readings were obtained from standard curve to determine protein amount (mg) per well of the 6 well plate. Average absorbance readings were converted to per ?l by dividing by 50 (50 ?l was the volume in each separate well of the 96 well plate), and to per well by multiplying by 500. Therefore, units for glucose uptake transport assays could be read as pmol/min/mg rather than pmol/min/well. Due to a large variation in absorbance readings, these concentrations could not be used. //Table 1 shows the protein concentrations calculated for preconfluent and confluent fibroblasts (data obtained from amended assay). The data was calculated so the glucose uptake assay values were in pmol/min/mg rather than pmol/min/well. The protein concentrations show too much variation for the results to be viable. The standard curve used to calculate the concentrations from optical densities of the samples is shown in figure 9. This curve shows little variation and a good regression line so is, in itself, viable. 3.2 Immunofluorescence assay Preconfluent and confluent fibroblasts were incubated for 30 minutes in stress inducing compound. They were then either permeabilized (in 0.1% saponin) and stained immediately or freeze thawed to produce plasma membrane lawns and stained. From analysis of GLUT-1 by fluorescence microscopy, its distribution in the cell can be seen quite clearly (figure 9). In preconfluent fibroblasts it can be seen that a small number of transporters are present at the cell surface but most are localised at the nuclear membrane. Under azide stress it can be seen that the distribution of transporters that were at the nuclear surface are now more widely distributed. Under sorbitol stress the transporters also seem to be more disperse, but any observation is difficult due to the image quality. The amount of translocation probably resembles that of azide. In confluent fibroblasts it can be clearly seen that many transporters are located intracellulary. Under azide stress, there seems to be little if no difference in transporter location compared to the basal level, although any observation of this image is also difficult due to its quality. There also seems to be very little translocation under sorbitol stress compared with the basal level. To further test if translocation of GLUT-1 to the cell surface had occurred, plasma membrane lawns were prepared for preconfluent and confluent cells. It can be clearly seen that transporters in the preconfluent fibroblasts are present at the cell surface under azide stress (figure 10a). This confirms the observation in figure 9c that translocation to the plasma membrane has occurred. In confluent cells it also appears that there are transporters on the cell surface. But owing to this being a single experiment, no clear conclusion can be drawn that translocation has occurred. Or indeed these were transporters pre-existing in the plasma membrane. 4.0 DISCUSSION This project was an investigation into the translocation and activation of the glucose transporter isoform GLUT-1 in 3T3-L1 fibroblasts. We have shown that cellular stress can have a profound effect on glucose transport. Moreover, the increase in glucose transport had to GLUT mediated due to the addition of cytochalasin B. Cytochalasin B efficiently inhibits glucose transporters e.g. GLUT-1 and 4. GLUT-1 transporters on the plasma membrane were activated and translocation of GLUT-1 from intracellular pools was also apparent. Sodium azide causes a chemical stress, and in solution it causes the release of hydrazoic acid (HN3) which is highly toxic. It is a specific (relatively) inhibitor of oxidative phosphorylation so this stresses the cell to increase glucose uptake (Elkhairi et al., 1999). Under a high concentration it is presumed that all cellular processes would cease. Sorbitol causes an osmotic stress. If fibroblasts are in a hypertonic solution, the membranes will be under a greater pressure (stress) than in an isotonic solution. Water will be trying to diffuse out of the cell. 4.1 Activation of GLUT-1 transporters From the early transport experiments it was observed that azide had a vast effect – more than double (2.28 fold) on glucose uptake. After the amendment of the assay, sorbitol had a more rearkable effect on the amount of glucose uptake (3.2 fold). If time persisted, these assays would have been repeated at least three more times to achieve data of at least a three fold increase in glucose uptake. Three transport assays were performed for confluent cells and averaging these data would give a more realistic value to the fold increase from the two stress compounds. Unfortunately the first two assays were poorly correlated with the third amended assay (azide having a greater stimulation than sorbitol), but a definite stimulation of glucose transport under stress could be inferred from the data of these three assays. Only one assay was performed with preconfluent fibroblasts. Therefore a comparison with confluent fibroblasts would only be a crude estimation of how glucose uptake differs between these cells. It has been previously reported (Yang et al., 1992) that in preconfluent fibroblasts glucose transport was ~5 times higher than in confluent fibroblasts. We have demonstrated that this is not the case and confluent fibroblasts exhibit higher glucose transport characteristics. This is most likely due to a large number of GLUT-1 transporters already present on the plasma membrane and they may not sequester as the fibroblasts reach confluence as demonstrated by Yang et al. Activation of GLUT-1 already present in the plasma membrane is most likely a response to increase the cell’s survival. If the cell is under stress, it will take in as much glucose as it can to produce more ATP by oxidative phosphorylation. ATP is then used for cellular processes such as maintaining ion gradients and cytoskeletal integrity. If the stress continues, the cell will start to undergo apoptosis which an energy consuming process. If the stress is too great for the cell to increase its glucose uptake, it will undergo necrosis. The cell will be closely neighboured by other cells in its tissue i.e. kidney, which will be acting in the same way. If many cells undergo apoptosis or necrosis this will consequently lead to severe kidney damage depending on the localisation of the stress i.e. pars recta (S3 segment) of nephron under ischaemic conditions for approximately 20 minutes. 4.2 Translocation of GLUT-1 transporters in preconfluent fibroblasts Translocation is also involved in increasing glucose uptake to consequently increase intracellular ATP levels. From translocation experiments there is definitive evidence to suggest that translocation of GLUT-1 has occurred under cellular stress. It has been clearly demonstrated that in preconfluent cells, azide stress causes a large translocation of an intracellular pool of GLUT-1 transporters that seemed to be localised near the nucleus. There is a large difference in the distribution of transporters compared to the basal level (Figure 9a). This is consistent with the findings (Baldwin et al., 1990) that translocation of GLUT-1 from the nucleus to the membrane occurs under cellular stress. No conclusions can be drawn from the image of preconfluent fibroblasts under sorbitol stress as a result of its quality. Due to the vast increase in glucose uptake under sorbitol incubation from the transport experiments, it is presumed that translocation has occurred and the image would look similar to stimulation by azide. In confluent cells, there seems to be no great difference in translocation under stress, so the increase in uptake is most likely to be increased activation of transporters pre-existing in the plasma membrane. Although this is contradictory to the findings of Yang et al i.e. GLUT-1 transporters are sequestered in confluent fibroblasts. 4.3 GLUT-1 antibody binding The GLUT-1 polyclonal antibody binds to the C-terminus of GLUT-1 at residues 477-492 (figure 11). //Figure 11 GLUT-1 is a 12 transmembrane glucose transporter with a cytoplasmic N and C-terminus which can bind a GLUT-1 specific polyclonal antibody A specific GLUT-1 polyclonal antibody binds to the cytoplasmic C-terminus at residues 477-492 and can be visualised with a GLUT-1 FITC labelled antibody. (Source: D. Voet and J.G. Voet., 1995). In freeze thawing fibroblasts ready for plasma membrane lawn experiments, the cytoplasm is flushed off to expose the cytoplasmic portions of the transporter. Since the antibody binds to the cytoplasmic C-terminus, the transporter is visible. If this was not the case and the cytoplasmic C-terminus was extracellular, the FITC labelled antibody would be hidden under the plasma membrane and poorly visualised. In this way it was easier to see that under cellular stress, GLUT-1 glucose transporters were present on the plasma membrane in azide incubated preconfluent fibroblasts (figure 10a). From the images in figure 9 it seems apparent that translocation from the nucleus had occurred. But due to only one plasma membrane lawn experiment being performed, it can not be confidently said that GLUT-1 transporters had definitely translocated to the plasma membrane. It could be staining of transporters pre-existing in the plasma membrane; as is the case with confluent cells. More plasma membrane lawns would have to be performed to confirm this. 4.4 Summary In summary, it is evident that in 3T3-L1 fibroblasts glucose transport is greatly increased under cellular stress due to activation and translocation of GLUT-1. It is still unclear whether these two processes occur simultaneously and further research would have to be performed to confirm this. 5.0 FUTURE STUDIES 5.1 Transport assays Now an efficient protocol for transport assays has been established I would like to perform three assays with preconfluent and confluent fibroblasts. This would give me sufficient evidence to prove that glucose uptake in confluent cells was higher than in preconfluent. This would be done before coming to any firm conclusions about the difference in transport between preconfluent and confluent fibroblasts. I could then plot both sets data on histograms and could critically compare if there was actually a difference or whether this was due to experimental error. I could also improve the BCA assay to obtain values in pmol/min/mg. Triton X-100 caused too many bubbles for accurate reading of optical density. Instead of permeabilizing the cells with this detergent, sodium hydroxide could be used. This would be a quite harsh permeabilization agent but no bubbles would be formed. If sodium hydroxide proved to be too harsh an agent other agents could be tested i.e. Triton X-114 or octyl glucoside, although these a very likely to form bubbles also. If these resemble the same characteristics as sodium hydroxide, Triton X-100 could be used again and pipetting into wells could be carefully monitored so bubble formation is reduced. 5.2 Immunofluorescence experiments I would have greatly liked to spend more time on these experiments to achieve acceptable images of a good quality. From this I could then make educated comparisons between azide and sorbitol stimulation in preconfluent cells. Also to see whether translocation was actually apparent in confluent cells or whether a large density of transporters were present on the plasma membrane. I would have liked to have had greater visualisation of the GLUT-1 transporters in both preconfluent and confluent cells, so vesicles containing transporters could be more distinguishable. To amplify the FITC on the GLUT-1, donkey anti-Rabbit biotinylated IgG would be used as a secondary antibody instead of G anti-Rabbit FITC. After washing of this antibody, the cells would be incubated in Streptavidin FITC. This would amplify the signal four times (figure 12). //Figure 12 A biotinylated GLUT-1 antibody subsequently incubated in Streptavidin FITC amplifies fluorescence four fold Streptavidin binds tightly to biotin in four regions, so labelling the antibody with biotin and labelling Streptavidin with FITC gives a four fold increase in fluorescence. Incubating fibroblasts in biotinylated GLUT-1 antibody and Streptavidin FITC, would give greater visualisation of GLUT-1 in whole cells and PM lawns. 5.3 PM lawns With plasma membrane lawns stained with biotin-Streptavidin FITC, it would be significantly easier to distinguish GLUT-1 to non-specific tissue staining. Consequently, this would give enough evidence to see if GLUT-1 transporters are actually present on the plasma membrane in confluent fibroblasts. If no transporters were observed on confluent but only in preconfluent, more experiments with intact fibroblasts would have to be performed to see if GLUT-1 actually translocates to the plasma membrane. Especially for preconfluent cells, actually searching for plasma membranes under the microscope was a difficult task. So as well as using biotin-Streptavidin FITC, a dye could also be added that stains phospholipids i.e. membrane. This would mean that in preconfluent cells, the membranes could be found more easily and the detection of GLUT-1 transporters on the membrane would be brighter. 6.0 CITATIONS AHMED N., BERRIDGE M.V., (1999). N-glycosylation of glucose transporter-1 (GLUT-1) is associated with increased transporter affinity for glucose in human leukemic cells. Leukemia Research. 23:395-401. BALDWIN S.A., DAVIES A., MARTIN S., PASTERNAK C.A., WIDNELL C.C., (1990). Cellular stress causes a redistribution of the glucose transporter. FASEB Journal. 4:1634-1637. BEHROOZ A., ISMAIL-BEIGI F., (1998). Induction of GLUT-1 mRNA in response to azide and inhibition of protein synthesis. Molecular And Cellular Biochemistry. 187:33-40. BEHROOZ A., ISMAIL-BEIGI F., PRASAD RK., (1999). LY-83583 stimulates glucose transporter-1-mediated glucose transport independent of changes in cGMP levels. European Journal of Pharmacology. 366:101-9 BEHROOZ A., ISMAIL-BEIGI F., ZHANG J.Z., (1999). Regulation of glucose transport by hypoxia. American Journal Of Kidney Diseases. 34:189-202. BUNN R.C., JENSEN M.A., REED B.C., (1999). Protein interactions with the glucose transporter binding protein GLUT1CBP that provide a link between GLUT-1 and the cytoskeleton. Molecular Biology Of The Cell. 10:819-832. CARLING D., CARLSON M., HARDIE D.G., (1998). The AMP-activated/SNF1 protein kinase subfamily: Metabolic sensors of the eukaryotic cell? Annual Review Of Biochemistry. 67:821-855. DePINA A.S., LANGFORD G.M., (1999). The role of actin filaments and myosin motors. Microscopy Research And Technique. 47:93-106. DOWNWARD J., RODRIGUEZ-VICIANA P., WARNE P.H., KHWAJA A., MARTE B.M., PAPPIN D., DAS P., WATERFIELD M.D., RIDLEY A., (1997). Role of Phosphoinositide 3-OH Kinase in Cell Transformation and Control of the Actin Cytoskeleton by Ras. Cell. 89:457-467. EBINA Y., HAYASHI H., ISMAIL-BEIGI F., PROHASKA R., ZHANG J.Z., (1999). Association of stomatin (Band 7.2b) with GLUT-1 glucose transporter. Archives Of Biochemistry And Biophysics. 372:173-178. EGERT S., NGUYEN N., SCHWAIGER M., (1999). Myocardial glucose transporter GLUT-1: translocation induced by insulin and ischaemia. Journal of Molecular & Cellular Cardiology. 31:1337-44 ELKHAIRI F.S., PRASAD R., ISMAIL-BEIGI F., HAMRAHIAN A.H., (1999). Activation of GLUT-1 glucose transporter in response to inhibition of oxidative phosphorylation - Role of sites of mitochondrial inhibition and mechanism of GLUT-1 activation. Archives Of Biochemistry And Biophysics. 368:375-379. FOSTER L.J., YAWORSK Y.K., TRIMBLE W.S., KLIP A., (1999). SNAP23 promotes insulin-dependent glucose uptake in 3T3-L1 adipocytes: possible interaction with cytoskeleton. American Journal of Physiology. 276:C1108-14. FINGAR D.C., GARZA L.A., HAUSDORFF S.F., MORIOKA K., WHITEMAN E.L., SUMMERS S.A., BIRNBAUM M.J., (1999). Identification of wortmannin-sensitive targets in 3T3-L1 adipocytes. Dissociation of insulin-stimulated glucose uptake and GLUT-4 translocation. Journal of Biological Chemistry. 274:24677-84. HADDAD G.G., WARSHAW J.B., XIA Y., (1997). Effect of chronic hypoxia on glucose transporters in heart and skeletal muscle of immature and adult rats. American Journal of Physiology. 273:R1734-41. ISMAIL-BEIGI F., KOSEOGLU M.H., (1998). Role of GLUT-1 glucose transporter activation in stimulation of glucose transport by A231876 and TPA. Biochemical Society Transactions. 26:289 ISMAIL-BEIGI F., ZHANG J.Z., (1998). Activation of GLUT-1 glucose transporter in human erythrocytes. Arch.Boichem.Biophys. 356:86-92. KAESTNER K.H., CHRISTY R.J., MCLENITHIN J.C., BRAITERMAN L.T., CORNILIUS P., PEKALA P.H., LANE D.M., (1989). Proc.Natl.Acad.Sci. U.S.A. 86:3150-3154. LUCAS L.R., MAGARINOS A.M., MCCALL A.L., MCEWEN B.S., REAGAN L.P., VAN BUEREN A, (1999). Regulation of GLUT-3 glucose transporter in the hippocampus of diabetic rats subjected to stress. American Journal Of Physiology-Endocrinology And Metabolism. 39:E879-886. MATSUZAKI T., NARUSE T., SUGAWARA-YOKOO M., SUZUKI T., TAKATA K., (1999). Presence of fructose transporter GLUT-5 in the S3 proximal tubules in the rat kidney. Kidney International. 56:1022-8. MUECKLER M., CARUSO C., BALDWIN S.A., PANICO M., BLENCH I., MORRIS H.R., ALLARD W.J., LIENHARD G.E., LODISH H.F., (1985). Sequence and structure of a human glucose transporter. Science. 229:941-945. OATEY P.B., VENKATESWARLU K., WILLIAMS A.G., FLETCHER L.M., FOULSTONE E.J., CULLEN P.J., TAVARE J.M., (1999). Confocal imaging of the subcellular distribution of phosphatidyl 3,4,5-triphosphate in insulin- and PDGF-stimulated 3T3-L1 adipocytes. Biochemical Journal. 344:511-518. TOMITA T., (1999). Immunocytochemical localization of glucose transporter-2 (GLUT-2) in pancreatic islets and islet cell tumors. Endocrine Pathology. 10: 213-221. VOET. D AND VOET J.G., (1995). Biochemistry. 2nd edition. Wiley press. YANG J., CLARK A.E., CUSHMAN S.W., KOZKA I.J., HOLMAN G.D., (1992). Development of an intracellular pool of glucose transporters in 3T3-L1 cells. Journal of Biological Chemistry. 267:10393-10399. YOUNG L.H., RENFU Y., RUSSELL R., HU X., CAPLAN M., REN J., SHULMAN G.I., SINUSAS A.J., (1997). Low flow ischemia leads to translocation of canine heart GLUT-4 and GLUT-1 glucose transporters to the sarcolemma in vivo. Circulation. 95:415-422.