Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 30
Quantitation, Purification and Reconstitution of the Red Blood Cell Glucose Transporter GLUT1
SHUSHENG ZUO
ISSN 1651-6214ISBN 91-554-6196-4urn:nbn:se:uu:diva-5727
ACTAUNIVERSITATIS
UPSALIENSISUPPSALA
2005
To the memory of my supervisor professor Per Lundahl
List of Papers
This thesis is based on the following papers, which will be referred in the text by their Roman numerals
I Zuo, S. and Lundahl, P. (2000) A micro-Bradford membrane protein assay. Anal. Biochem. 284, 162-164.
II Gottschalk, I., Lundqvist, A., Zeng, C.-M., Lagerquist Hägglund, C., Zuo, S., Brekkan, E., Eaker, D. and Lundahl, P. (2000) Conver-sion between two cytochalasin B-binding states of the human GLUT1 glucose transporter. Eur. J. Biochem. 267, 6875-82.
III Zuo, S., Hellman, U. and Lundahl, P. (2003) On the oligomeric state of the red blood cell glucose transporter GLUT1. Biochim. Bio-phys. Acta 1618, 8-16.
IV Zuo, S. and Lundahl, P. The effects of cholesterol and polyethylene-glycol phospholipids on the activities of the reconstituted human red blood cell glucose transporter GLUT1, Manuscript
V Zuo, S., Lagerquist Hägglund, C. and Lundahl, P. (2002) Well-known sugar transporters: the GLUT1 glucose transporter of hu-man red blood cells and the glucose transporter and lactose per-mease of Escherichia coli. Recent Res. Devel. Biochem. 3, 527-546 (Review).
Reprints of the papers were made with permission from the publishers.
Contents
Introduction.....................................................................................................7A brief overview of membrane transport proteins .....................................7Self association of membrane transport proteins .......................................9Association of proteins in red blood cell..................................................10Lipids and detergents ...............................................................................11Substrate binding......................................................................................13
Present study .................................................................................................16Preparation of red blood cell membrane vesicles (Papers I, II and III)..................................................................................17Purification of GLUT1 (Papers I, II and III) ............................................18The modified micro-Bradford assay with CaPE precipitation (Paper I)....................................................................................................18Sulfhydryl affinity chromatography of GLUT1 (Paper III) .....................21In-gel digestion and mass spectrometry (Paper III) .................................23Amino acid analysis (Papers I, II, III and IV) ..........................................24Reconstitution of GLUT1 and immobilization (Papers II, III and IV).....25Quantitative frontal affinity chromatography (Papers II, III and IV).......26Glucose transport activity (Papers II and III) ...........................................31
Conclusions and future perspectives.............................................................32
Summary in Swedish ....................................................................................33Kvantifiering, rening och rekonstituering av GLUT1..............................33Min forskning...........................................................................................34
Acknowledgements.......................................................................................36
References.....................................................................................................37
Abbreviations
ATP adenosine triphosphate BSA bovine serum albumin C12E8 octaethylene glycol n-dodecyl etherCB cytochalasin B CaPE calcium phosphate adsorption and ethanol pre-
cipitationCMC critical micelle concentration pCMB para-chloromercuribenzoic acid DTE dithioerythritol EYP egg yolk phospholipids GLUT1 human red blood cell membrane glucose transporter MALDI-ToF-MS matrix-assisted laser desorption ionization-time of
flight-mass spectrometry NBTI nitrobenzylthioinosine NT nucleoside transporter OG octyl glucoside PEG polyethyleneglycol DSPE distearoylphosphatidylethanolamine SDS sodium dodecyl sulfate
7
Introduction
A brief overview of membrane transport proteins In contrast to the water-soluble proteins, membrane proteins are partly em-bedded in a membrane composed of phospholipids and other lipids, such as cholesterol. The membrane proteins mainly function as transport proteins or receptors. The transport proteins help to pump in the nutrient molecules or ions needed by the living cells and pump out the waste that needs to be re-moved. The receptors are involved in signal transduction.
Totally, 500 different membrane protein-mediated reactions have been identified (Maloney & Wilson 1993). For instance, in E. coli, approximately 205 genes have been predicted to be involved in membrane transport (Hig-gins 1992). There are three classified membrane transport proteins; Gram-negative bacterial outer membrane porins permeate carbohydrate and other molecules, the adenosine triphosphate (ATP)-binding cassette superfamily transports carbohydrate with ATP degradation and the major facilitator su-perfamily is largely involved in transport of sugars and their derivatives without energy consumption. The former superfamily is exemplified by ATP-driven Na+/K+ and Ca2+ pumps, the light-driven proton pump bacteri-orhodopsin, E. coli lactose/H+ cotransporter, Na+/sugar and Na+/amino acid cotransporters of mammalian membranes. The latter superfamily includes the mammalian glucose transporters, the nucleoside/H+ cotransporters and the anion transporter of human red blood cells (Saier 2000, Finean et al.1978).
The major facilitator superfamily includes 1000 members of 33 families (Saier 2000). The isoforms of the facilitative glucose transporters GLUTs exist in the brain tissues, but can also be found in other parts of our body (Table 1).
8
Table 1. Facilitative glucose transporter isoforms (Wood & Trayhurn 2003, Van-nucci et al. 1997)Transporter Sites of expression Transported substance GLUT1 Most cells, red blood cells, blood tissue
barriersBrain: Blood brain barrier, glia, neurons,choroid plexus, ependyma
Glucose
GLUT2 Liver, kidney, pancreas, intestine Brain: Astrocyte
Glucose (low affinity) Fructose
GLUT3 Placenta, spermatozoa, platelets Brain: Neuronal
Glucose (high affinity)
GLUT4 Heart, muscle, adipose Brain: Cerebellum, hippocampus neurons
Glucose (high affinity)Insulin sensitive
GLUT5 Small intestine, macrophage, spermatozoa Brain: Microglia
FructoseGlucose (very low affinity)
GLUT6 Brain, spleen, leucocytes Glucose GLUT7 Liver, kidney; G-6-phase component
Brain: Astrocytes Glucose
GLUT8 Testes, brain and other tissues Glucose GLUT9 Liver, kidney Not determined GLUT10 Liver, pancreas Glucose GLUT11+ Heart, muscle Glucose (low affinity);
FructoseGLUT12 Heart, prostate, muscle, small intestine,
white adipose tissue Not determined Insulin sensitive
HMIT# Brain, white adipose tissue H+-myo-inositol+GLUT11 has two splice variants: a short form (low-affinity glucose transport) and a long form (which may be a fructose transporter).#HMIT, H+-coupled myo-inositol transporter
The most abundant facilitative glucose transporter, GLUT1, is composed of 492 amino acids (Mr 54117) as determined from the cloned sequence of a human hepatoma cell G2 cDNA library (Mueckler et al. 1985) and it is pro-posed to comprise 12 -helical transmembrane domains. GLUTs 1 and 3 are the predominant ones detected in the brain in glucose transporter isoforms (Vannucci et al. 1997). GLUT1 serves the basal metabolic needs of prolifer-ating cells. It is connected with seizures and delayed development in infants with abnormally low concentrations of glucose in their cerebrospinal fluid. The glucose concentration in the blood is normal, but the red blood cells have a low glucose transport rate and a low cytochalasin B (CB) binding capacity (De Vivo et al. 1991). In Alzheimer´s disease, there is a large re-duction of GLUT1 and GLUT3 in the brain, but no decrease of GLUT1 in the red cell membranes (Kalaria & Harik 1989). It is suggested that the re-duction of GLUT1 in the brain may not be primary, but secondary, due to the decrease of brain glucose metabolism in severe dementia. The expression of GLUT1 is affected by oncogenic transformation, growth factors and acti-vators of protein kinase C (Massa et al. 1996, Murakami et al. 1992). In
9
vitro it is enhanced by the deprivation of glucose, probably by stabilization of the GLUT1 mRNA (Boado & Pardridge 1993).
GLUT4 and GLUT12 are insulin sensitive (Wood & Trayhurn 2003). GLUT4, and probably GLUT12 as well, have a central role in maintaining glucose homeostasis and are developmentally regulated. Insulin and gluca-gons, both produced by the endocrine pancreas, are the main regulating hor-mones of the glucose metabolism and of the blood glucose level, but some influence is also exerted by catecholamines, corticosteroids, growth hor-mone, and thyroid hormones. Insulin can increase the rate of glucose trans-port into muscle and fat due to a rapid translocation of the glucose trans-porter GLUT4 from intracellular membrane compartments to the plasma membrane (Cushman & Wardzala 1980).
Diabetes mellitus is one of the main diseases menacing human health nowadays. In type-1 diabetes mellitus, insulin production is reduced due to the destruction of -cells in the pancreas (Atkinson & Maclaren 1990). Dos-ing insulin to the body can somehow compensate this. In type-2 diabetes mellitus, so-called non-insulin-dependent diabetes mellitus, the tissues do not respond to elevated blood insulin levels, e.g., after meals. As a result, the pancreas increases insulin secretion to normalize the blood glucose level. However, an individual with exhausted -cells will develop non-insulin-dependent diabetes mellitus (Bailey 1992, Groop 1992).
Self association of membrane transport proteins The self association of membrane proteins is determined by inhibitor-binding analysis (Brekkan et al. 1996), ultracentrifugal sedimentation coeffi-cient determination (Hebert & Carruthers 1991), electron microscopy (Saseet al. 1982), size-exclusion chromatography (Lundahl et al. 1991), sulfhy-dryl affinity binding and dissociation (Casey et al. 1989) and crystallization (Abramson et al. 2003).
Lactose permease in E. coli is functionally monomeric (Sahin-Tóth et al.1994, Abramson et al. 2003) whereas the anion transporter of mammalian membranes is dimeric (Boodhoo & Reithmeier 1984, Dolder et al. 1993).The E. coli glucose transporter consists of subunits IIA and IICB and the IICB has trigonal unit cells of three dimeric units, as judged by electron mi-croscopy (Zhuang et al. 1999). GLUT1 in the human red blood cell mem-brane may be monomeric, dimeric or tetrameric according to electron mi-croscopy (Sase et al. 1982), electron beam inactivation analyses (Cuppolettiet al. 1981, Jarvis et al. 1986) and analyses of inhibitor binding and transport function (Gottschalk et al. 2000, Hebert & Carruthers 1991, Pessino et al.1991, Zottola et al. 1995, Coderre et al. 1995, Hamill et al. 1999, Zeng et al.1997).
10
The crystallization of GLUT1 is a big challenge to membrane protein re-searchers, since it probably occurs in mixed oligomeric states (Hebert & Carruthers 1991) and has a great tendency to aggregate irreversibly (Lundahl et al. 1991). The self-association state of membrane proteins in detergent solution depends both on the intrinsic properties of the protein and on the effects of the detergents. In solution, GLUT1 dimers or tetramers or larger oligomers (Hebert & Carruthers 1991, 1992, Acevedo et al. 1981) are ob-tained by the action of cholate on the membrane, whereas monomers and some dimers occur after solubilization of vesicles prepared in the presence of reductant, i.e. 0.2 mM dithioerythritol (DTE), with nonionic detergents such as octyl glucoside (OG) and octaethylene glycol n-dodecyl ether (C12E8)(Haneskog et al. 1996, Lundahl et al. 1991). Sodium dodecyl sulfate (SDS) provides a zone of monomeric GLUT1 upon SDS-PAGE of freshly prepared GLUT1 samples, but does not dissociate irreversible GLUT1 aggregates formed during sample storage. In the latter case, a mixture of monomers, dimers and oligomers is seen upon SDS-PAGE. Irreversible, temperature-dependent self association of anion transporter dimers occurred with a half-time of about 1 h at 37˚C, but did not occur at 0˚C after several days al-though with the loss of the inhibitor stilbene disulfonate binding site. The dimeric membrane domains were more stable (Vince et al. 1997).
Association of proteins in red blood cell The biconcave disklike shape of the red blood cell is supported by the cyto-plasmic cytoskeleton proteins spectrin (band 1 and 2) and actin (band 5) (Figure 1) and it assures the rapidity of oxygen diffusion. The cross-linked spectrin tetramers are associated with actin and band 4.1 to form a dense and irregular protein network under the rmembrane. Spectrin also binds to the anion transporter (band 3) via ankyrin (band 2.1) (Bennett & Stenbuck 1979a & b, Voet & Voet 1995). The skeleton proteins attached to the mem-brane proteins make the membrane easily deformable in the capillary blood vessels. Cytoplasmic proteins such as hemoglobin, band 4.1 and the glyco-lytic enzymes, including aldolase, phosphofructokinase and glyceralde-hydes-3-phosphate dehydrogenase (GAPDH, band 6) in red blood cells spe-cifically and reversibly bind to the anion transporter protein (Bruce et al.2003, Low 1986, Low et al. 1993). A GLUT1 C-terminal binding protein GLUT1CBP (Mr 39kd) that may link GLUT1 to the cytoskeleton has been detected in rat tissues (Bunn et al. 1999).
11
Figure 1. Coomassie brilliant blue-stained SDS-PAGE of human red blood cell ghost proteins. Reprinted from Voet & Voet 1995 with permission from John Wiley & Sons, Inc., NJ.
The GLUT1 transporter may bind to another membrane protein, stomatin (band 7) in red blood cell. Overexpression of stomatin results in a depression of glucose transport by decreasing the activity of GLUT1 (Zhang et al. 1999 & 2001).
Lipids and detergents The plasma membrane is a major structure in all cells, including microorgan-isms. Phospholipids, glycolipids and cholesterol are the major lipid compo-nents in the membranes of animal cells. In 1972, Singer and Nicolson devel-oped the ‘fluid mosaic model’ of the plasma membrane (Singer and Nicolson 1972). The lipid molecules serve as ‘surfactants’ for the massive and func-tional membrane proteins. Phospholipids have one or two hydrophobic chains and one hydrophilic head with phosphoryl and other groups.
Lipids are extensively used in pharmacological and food industry. Syn-thetic polyethyleneglycol (PEG) attached distearoylphosphatidyl-
12
ethanolamine (DSPE) induces an open and flat biomembrane due to the steric hindrance of its long flexible tail. The hydration of PEG(2000)-DSPE, cholesterol and distearoylphosphatidylcholine produces a new membrane structure, disks (Edwards et al. 1997). The disks have little possibility to fuse to a larger size upon freeze-thawing as liposomes do. Cholesterol increases the cohesive strength and reduces the membrane permeability of phosphol-ipid biomembranes (Needham and Nunn 1990). The disks produced by hy-dration of distearoylphosphatidylcholine, (PEG)5000-DSPE and cholesterol (Figure 2) show higher retention of neutral, positively and negatively charged drugs than do unilamellar and multilamellar liposomes composed of distearoylphosphatidylcholine and cholesterol (Johansson et al. 2004), which indicates that they might be a better medium for drug screening. Disk-like structure can also be induced by SDS (82 mol%) penetration into egg phos-phatidylcholine (Silvander et al. 1996).
Figure 2. The structure of cholesterol (above) and PEG(5000)-DSPE (below).
Detergents are also amphiphilic, which normally have a single alkyl or al-kenyl chain attached to a polar head group. They form micelles in aqueous solution at concentrations above their critical micelle concentration (CMC) (Table 2).
13
Table 2. The CMC values of detergents and lipids Detergents/lipids MW CMCC12E8 518 0.09 mM (0.01M TES, pH 7.5,
0.05 M NaCl, 0.1M CaCl2)n-Decyl -D-maltoside 482.6 1.8 mM (H2O)n-Dodecyl -D-maltoside 535.8 0.2 mM (H2O)OG 292.4 18-20 mM (H2O)
23.4 mM (0.1 M NaCl) SDS 288.4 2.6 mM (pH 7.5)
8.3 mM (H2O)Sodium cholate 430.6 14 mM (pH 7.5)
9.5 mM (pH 9.0) Triton X-100 647 (average) 0.01-0.016% (w/v) Distearoylphosphatidylcholine 790.15 Insoluble in H2ODSPE 748.1 Insoluble in H2OCholesterol 386.7 Insoluble in H2OEgg phosphatidylcholine 760.1 <0.001nM (H2O) CMCs of detergents are from the catalog of Anatrace (Maumee, Ohio) and those of lipids
are from the catalog of Avanti Polar lipids, Inc (Alabaster, Alabama).
Based on the nature of the head groups, detergents can be divided into (i) ionic detergents (ii) non-ionic detergents and (iii) zwitterionic detergents. The detergent molecules are more wedge-shaped and shorter than the more cylindrical bilayer-forming lipids. Therefore, they will disturb the lamellar packing of the lipid bilayer when they are inserted into it. Finally, the deter-gent molecules and a single membrane protein or a single lipid form a mi-celle and the membrane is disintegrated (Kragh-Hansen et al. 1993).
Substrate binding Glucose transport by GLUT1 is reversibly and competitively inhibitable by the fungal antibiotics cytochalasins A and B (on the cytoplasmic side) (Goto et al. 1998, Jung & Rampal 1977, Rampal et al. 1986), by genestein (Vera etal. 1996) and other tyrosine kinase inhibitors (Vera et al. 2001), by maltose (on the outside) (Hamill et al. 1999) and by the herbal diterpene forskolin (Lu et al. 1997, Shanahan et al. 1987, Lavis et al. 1987). The transport is modulated or regulated by, for example, ATP (Cloherty et al. 2001).
CB competes with glucose for binding to glucose transporters, although at very low concentrations it stimulates red blood cell sugar uptake (Cloherty etal. 2001). Photoaffinity coupling of CB to GLUT1 (Carter-Su et al. 1982) at Trp388 and Trp412 (Inukai et al. 1994) is used to identify GLUT1 polypep-tide.
Comparison between the crystallographic structures of CB and D-glucoseshows that N2, O7 and O23 in the inhibitor correspond to O6, O3 and O1, respectively, in the sugar (Figure 3) and may hydrogen-bond to GLUT1
14
(Inukai et al. 1994). On the other hand, the , -unsaturated lactone moiety at C20–C23–O24 of CB is probably important for the glucose-competing binding (Goto et al. 1998). There are three reported types of CB-binding sites with different apparent dissociation constants and competition proper-ties in human red blood cells (Jung & Rampal 1977). Only CB binding on site I is glucose sensitive. Site II of the red blood cell ghosts is removable by EDTA at low ionic strength, whereas site III displays glucose sensititivity of CB binding upon the removal of site II (Pinkofskin et al. 1978). Cyto-chalasin A also inhibits glucose transport (Jung & Rampal 1977). High-affinity cytochalasins A–E- and dihydrocytochalasin B-binding sites that are unrelated to sugar transport have been found in human, bovine, and rabbit red blood cells on the cytoplasmic side of the membrane. These sites repre-sent 10–30% of the total number of CB-binding sites in the cells (Jung & Rampal 1977, Lin & Snyder 1977). Dihydrocytochalasin B inhibits aromatic amino acid transport by System T in red blood cells, whereas the suggested tryptophan transport by GLUT1 (Widmer et al. 1990) has not been verified (Lagerquist Hägglund & Lundahl 2003). Genestein, a natural tyrosine kinase inhibitor, competitively inhibits hexose and dehydroascorbic acid transport of GLUT1. Its binding site may overlap the ATP-binding sites (Vera et al.1996 & 2001).
Figure 3. The structures of CB and D-glucose
ATP decreased the apparent efflux dissociation constant Kd of sugar-induced fluorescence quenching of red blood cell membrane proteins and increased the influx Kd, consistent with the D-glucose transport asymmetry and the existence of two efflux Kms (Carruthers 1986). However, later on no ATP effect was noted (Wheeler 1989). Nevertheless, ATP modulation of net sugar import has been suggested to result from substrate tunneling from transporter to bulk cytosol through a structure containing high-affinity, low-capacity sugar-binding sites (Cloherty et al. 1995). GLUT1 residues 301–364 were identified as the major ATP-binding domain (Levine et al. 1998)
15
and the cytoplasmic domains of the GLUT1 tetramer were suggested to form a glucose-binding cage in the presence of ATP (Heard et al. 2000). Intracel-lular ATP was reported to double the GLUT1 3-O-methyl glucose-binding capacity, from one to two sugar molecules per protein tetramer, and inhibit the release of bound sugar into the cytosol (Heard et al. 2000). Reduction with 10 mM dithiothritol eliminated binding of ATP or 3-O-methylglucose(Levine et al. 1998, Heard et al. 2000). In several papers from Carruther’s research group the glucose interactions with a cytoplasmic structure of GLUT1 have thus been held responsible, at least partially, for the observed asymmetry of transport. The oligomeric, or at least dimeric, structure of GLUT1 in the native membrane and in reconstituted systems rest also on several other lines of evidence (Gottschalk et al. 2000, Heard et al. 2000, Kevine et al. 2001).
The inhibitor binding on biomembranes can be analysed by equilibrium dialysis (Hebert & Carruthers 1992), Hummel and Dreyer analysis (Hummel & Dreyer, 1962, Lundahl et al. 1999), and frontal affinity chromatography (Brekkan et al. 1996). Recent frontal affinity chromatography determina-tions of CB binding to GLUT1 have shown KD of 50–100 nM (Lu et al.1997, Gottschalk et al. 2000, Brekkan et al. 1996, Lundqvist et al. 1997, Lundquist & Lundahl 1997). The stoichiometry of CB binding to GLUT1, one or two binding sites per GLUT1 dimer, are determined by the chroma-tographic method.
Nitrobenzylthioinosine (NBTI) is not only a competitive inhibitor of nu-cleoside transporter (NT), the minor impurity component of GLUT1, with high affinity (Kd 0.04–0.4 nM), but also binds less strongly to reconstituted GLUT1 (Kd 80 nM) (Haneskog et al. 1998a & b). Stilbene disulfonates are potent inhibitors of anion transporter in red blood cells. They bind to a single site per anion transporter monomer that is accessible from the cell exterior (Casey et al. 1989).
16
Present study
GLUT1 looks like a cloudy ‘broadband’ on SDS-PAGE. Inhibitor-binding activity analyses also show diverse results depending on the sample prepara-tion. The aim of the thesis is to further characterize the inhibitor binding activities of GLUT1 reconstituted with natural lipids or with synthetic lipids.
Paper I introduces a new protein assay. Purified GLUT1 from anion-exchange chromatography is quantitated by an easy and efficient modified micro-Bradford assay with calcium phosphate adsorption and ethanol pre-cipitation.
Paper II describes the conversion of human red blood cell glucose trans-porter GLUT1 between two kinds of CB-binding states, i.e., one CB binding site per GLUT1 molecule and half a CB-binding site per GLUT1 molecule.
Paper III introduces a new preparation of GLUT1. From the partly solubi-lized human red blood cell membrane vesicles, the anion exchanger-purified GLUT1 retains a little CB binding activity. The combined elution with SDS and 2-mercaptoethanol of GLUT1 bound on para-chloromercuribenzoic acid (pCMB)-coupled Sepharose 4B indicates the presence of oligomeric GLUT1. This preparation may mainly have selected the non-CB-binding portion of GLUT1.
Paper IV preliminarily determines GLUT1 reconstituted in the presence of cholesterol or PEG (5000)-DSPE together with egg yolk phospholipids (EYP). The CB and glucose binding activities of GLUT1 in reconstituted bimembranes are affected.
Paper V reviews the proceedings of the facilitative sugar transporters, such as GLUT1 and the glucose transporter and lactose permease of E. coli. The substrate and inhibitor binding, the structural and function models and the transport activity are discussed.
17
Preparation of red blood cell membrane vesicles (Papers I, II and III) To purify GLUT1, the membrane peripheral proteins should be removed to produce membrane vesicles, which are much smaller than cell membranes and are mostly inside side out due to the endocytosis of the membranes (Steck et al. 1970). The vesicle preparation is a critical step in the study of GLUT1. There are two main ways to prepare red blood cell membrane vesi-cles. One is by low ionic alkaline stripping (Baldwin et al. 1982), mostly used to purify GLUT1, and the other is by low-ionic incubation and potas-sium iodide stripping (Casey et al. 1989), mostly used to purify anion trans-porter. The former is fast and cheap, but has a higher risk to proteolyze the anion transporter as compared to the latter one. In this work we used the former method with size-exclusion chromatography on Sepharose 4B and 6B to remove hemoglobin in the ghost membranes.
Red blood cell concentrates (240 to 300 ml) were purchased from the blood bank of Uppsala University Hospital, and had been stored at 4 ºC for 4-6 weeks after the donation. The red blood cells were separated from most of both serum and white blood cells by centrifugation. Five times of cen-trifugal washing with 5 mM phosphate buffer, pH, 8.0, containing 150 mM NaCl were followed by storage over the night at 4 C. The red blood cells were lysed by addition of 4 volumes of 5 mM phosphate buffer, pH 8.0. After storage for 5 min on ice, the lysed erythrocytes were transferred to a Sepharose 4B column (12 cm (i.d.) × 35 cm) equilibrated in the phosphate buffer and connected in tandem with a Sepharose 6B column (12 cm (i.d.) × 35 cm) equilibrated with 5 mM EDTA solution, pH 10.0. The eluted mem-brane was collected by centrifugation followed by 15-minute incubation with 5 volumes of 2 mM EDTA and 15 mM NaOH, pH 12, with (Papers I, II and III) or without 0.2 mM DTE (Paper III) to further remove cytoskeletal pro-teins. After centrifugation, the pellet was washed once more with the alka-line solution and then with 50 mM Tris-HCl, pH 6.8. The suspended vesicles were frozen in liquid nitrogen and stored at –70ºC.
In Papers II, III and IV, the solubilization extent of the prepared vesicles was different. In Paper II, the supernatant from solubilized vesicles stripped with DTE had a protein concentration similar to that in the original vesicle suspension. In Paper III, the supernatant from vesicles stripped without DTE showed a protein concentration only 65% of that in the vesicle suspension. Even for the vesicles stripped with DTE, the solubilized protein amount in the supernatant varied due to the individual preparation of vesicles (Paper III and IV). In addition to the effect of the reductant, this probably also accounts for the different contents of residual cytoskeletal proteins or other peripheral proteins. As described in introduction, spectrin exists as a tetramer complex and associates with part of anion transporter via ankyrin. The residual spec-trin molecules might be the ones attached to the anion transporter or to other
18
membrane proteins. As a result, this part of membrane proteins will be trapped in the insoluble debris in the presence of detergents.
Purification of GLUT1 (Papers I, II and III) In accordance with two ways of vesicle preparations, GLUT1 is mostly puri-fied by anion exchanger chromatography on DEAE cellulose and it can also be purified by other anion exchangers such as amino-ethyl Sepharose 4B (Casey et al. 1989), HiTrap Q (Boulter & Wang 2001) and Mono Q (Lun-dahl et al. 1984). The purified GLUT1 was not stable and started to aggre-gate at 4ºC, especially in OG solution and at high salt solution.
The red blood cell membrane vesicles were solubilized in 70 mM Tris-HCl buffer, pH 7.4, at 4ºC including 22 mg/ml C12E8 or 22–44 mg/ml OG with (Papers I, II and IV) or without 1 mM DTE (Paper III). After ultracen-trifugation (160K g), the supernatant was applied to the DEAE cellulose column (5 mg protein/1ml gel bed) and the breakthrough peak was collected as the purified GLUT1, A-GLUT1DTE (Paper I, II and IV), A-GLUT1 (Paper III) or A-GLUT1(DTE) (Paper III), depending on the presence of DTE in the vesicle preparation and anion exchange chromatography. The preparations contain 0.3-5% NT (Mr 50kD) which cannot be discerned from GLUT1 (Mr 54K) by SDS-PAGE.
As discussed above, the solubilization extent of vesicles differed among Papers II, III and IV. This resulted in the variation of protein yields in ultra-centrifugal supernatants and in sample solutions after anion exchange chro-matography. However, the purified GLUT1 showed a similar band 4.5 pat-tern on the gel of SDS-PAGE upon silver staining.
The modified micro-Bradford assay with CaPE precipitation (Paper I) Compared to the widespread Lowry method (Lowry et al. 1951), the use of the Bradford assay (Bradford 1976) takes very short waiting time and uses only routine reagents. The micro-Bradford assay is a smaller-scaled Brad-ford assay. The micro-Bradford assay with calcium phosphate adsorption and ethanol precipitation (CaPE) was first introduced by Pande and Murthy 1994 to eliminate interference from substances such as detergents and lipids. In Paper I we modified the micro-Bradford CaPE assay to make it more ver-satile in quantitation of membrane proteins with different intrinsic proper-ties. Equal volumes (20 µl) instead of different volumes (Pande and Murthy 1994) of 0.5 M potassium phosphate buffer, pH 7.4, and 0.5 M CaCl2 were added to 100 µl of sample solution. After mixing, 1 ml of 99.9% ethanol was
19
added. After centrifugation, the pellet was washed again with ethanol and finally, the pellet was resuspended with 200 µl 5 × Bradford reagent (105 mg of Coomassie brilliant blue G-250, 75 ml 95% ethanol, 150 ml of 85% phos-phoric acid and 75 ml H2O) and followed by 800 µl H2O for the spectropho-tometric determination. The modification provides a milder pH condition in the sample before ethanol precipitation and much improves the solubility of resuspended GLUT1 in the final dye reagent. The quantitation of GLUT1 by the modified micro-Bardford CaPE assay becomes possible. The method is also applicable to the proteins in immobilized biomembranes. Triton X-100 at 10% (v/v) is mixed with the gel bead suspension at the volume ratio of 7:3 to extract the proteins in the beads. The supernatant after a short spinning is taken as the sample.
The modified micro-Bradford CaPE assay proved to be a versatile method in its application to red blood cell membrane vesicles, GLUT1, bacteri-orhodopsin and glycophorin (Figure 4).
Figure 4. Calibration curves for bovine serum albumin (BSA), GLUT1, vesicles, bacteriorhodopsin and glycophorin in the modified micro-Bradford CaPE assay. The protein concentrations are those of the initial sample solutions. The absorbance val-ues are the differences between the average absorbance values for duplicate samples and duplicate controls. Reprinted from Paper I with permission from Elsevier.
20
The freshly purified monomeric GLUT1 gave a similar result as the aggre-gated GLUT1. Another method, which was not sensitive to the amino acid composition, such as amino acid analysis, was needed to make the standard curve for the protein. The modified Lowry assay with trichloroacetic acid and sodium deoxycholate precipitation (Peterson 1977) was compared (Fig-ure 5).
Figure 5. Calibration curves for BSA, GLUT1, bacteriorhodopsin, glycophorin and vesicles in the Lowry assay with trichloroacetic acid and sodium deoxycholate pre-cipitation (Peterson 1977). Protein concentrations and absorbance values as in Fig-ure 4.
The profiles of the direct microBradford assay and the modified microBrad-ford CaPE assay for bovine serum albumin (BSA) and GLUT1 were com-pared (Figure 6).
The modified micro-Bradford CaPE assay keep the merits of rapidity and simplicity of the Bradford assay. The use of calcium phosphate and ethanol is compatible with the contents of the Bradford dye reagent. This assay is applicable to samples containing sulfhydryl reagents, which are commonly used in membrane protein studies. Other methods, such as the Lowry method with precipitation, are not compatible with these reagents. This modified
21
micro-Bradford CaPE assay provided an efficient way to remove the inter-fering substances in a simple way. A limitation is that it has a narrow recti-linear range in the calibration curves for membrane proteins and relatively large differences in sensitivity for different membrane proteins.
Figure 6. Calibration curves for BSA in the modified micro-Bradford CaPE assay (open triangles) and the microBradford assay (open squares), GLUT1 in the modi-fied micro-Bradford CaPE assay (filled squares) and GLUT1 in the micro-Bradford assay (filled circles). Protein concentrations and absorbance values as in Figure 4.
Sulfhydryl affinity chromatography of GLUT1 (Paper III)For three-dimensional crystallization of the anion transporter (Lemieux et al.2002), only 5-13 lipids per protein molecule are required for the crystal for-mation. For two-dimensional crystallization of the anion transporter (Dolder et al. 1993), exogenous lipids are preferred for the incubation. To remove the endogenous lipids in the GLUT1 sample, a lectin column (Hebert & Car-ruthers 1991) was used to bind GLUT1 with 26 remaining lipids per GLUT1 molecule due to the glycosylation of GLUT1.
22
In Paper III, GLUT1 was applied to a sulfhydryl affinity matrix of pCMB-coupled Sepharose 4B. Most endogenous lipids were removed by further washing. The number of residual lipids bound per GLUT1 purified from vesicles prepared without DTE (S-GLUT1) are more than that for vesi-cles prepared with DTE (S-GLUT1(DTE)), i.e., 30 versus 7, which implied difference in the oligomeric states between two preparations. The sequential elutions with 5% (w/v) SDS and 2% 2-mercaptoethanol of S-GLUT1 bound on pCMB Sepharose 4B column indicated the presence of oligomers of S-GLUT1, on the average, pentamers (Figure 7).
Figure 7. Partial elution by SDS washing of A-GLUT1 bound on pCMB-Sepharose 4B. (A) a-b, application; b-c, washing with 70 mM Tris-HCl, pH 7.4, containing 1mg/ml C12E8 and 1 mg/ml dimyristoylphosphatidylcholine (buffer B); c-d, washing with buffer B containing 0.5% SDS; d-e, elution with buffer B containing 5% SDS warmed to 30 C to keep the SDS in solution; e-f, washing with 70 mM Tris-HCl, pH 7.4, containing 1mg/ml C12E8 (buffer A). (B) Quantitation of the amount of A-GLUT1 bound on p-CMB-Sepharose 4B and of the amount remaining bound after SDS washing. We applied A-GLUT1 until the protein began to elute, washed the column either with buffer A alone ( ) or with buffer B containing 5% SDS followed by buffer A ( ) as in Fig. 7A, sectioned the gel beds by Pasteur pipetting and re-leased the bound GLUT1 by 2% 2-mercaptoethanol in buffer A. The protein amounts were quantitated by the micro-Bradford assay (Bradford, 1976). The x-axis shows the distance in the column from the top of the gel bed. Reprinted from Paper III with permission from Elsevier.
The hypothetical association states of GLUT1 in red blood cells and red blood cell membrane vesicles stripped without reductant and in membrane
23
vesicles stripped in the presence of reductant are illustrated in Figure 8. The stripping of the membrane at pH 12 with DTE dissociates the oligomeric GLUT1 into monomeric GLUT1, but the newly produced GLUT1 mono-mers do not retain the CB-binding activity. The meaning of the variant oli-gomeric states and the conversion condition in vivo are not clear. They might be related to the length of the oligosaccharide covalently coupled to GLUT1.
Figure 8. Schematic illustration of hypothetical GLUT1 association states in red blood cells and red blood cell membrane vesicles stripped without reductant and in membrane vesicles stripped in the presence of reductant. The native monomeric GLUT1 represents 40% of the total GLUT1 molecules and the ratio of tetramers to octamers is 3:1. The monomers formed from the oligomeric GLUT1 do not bind CB significantly compared to the native GLUT1 monomers. Reprinted from Paper III with permission from Elsevier.
In-gel digestion and mass spectrometry (Paper III) Matrix-assisted laser desorption ionization-time of flight-mass spectrometry (MALDI-ToF-MS) provides information on the structure and end groups of big molecules like natural or synthetic polymers (Gies et al. 2005).
In Paper III, the stained area of band 4.5 of monomeric GLUT1 in SDS-PAGE gels was excised, washed with ammonium bicarbonate buffer in ace-tonitrile and dried. A trypsin solution was allowed to be absorbed into the reswelling gel. After overnight incubation and acidification, the generated peptides were extracted and analysed by Peptide Mass Fingerprinting in the MALDI-ToF-MS instrument (Hellman 1997). In the sample of S-GLUT1, the complete transmembrane helices 1 and 8 were found together with parts
24
from hydrophilic loops or the C terminus, which were also detected from the samples S-GLUT1(DTE) or A-GLUT1(DTE) (Table 3). No other proteins than GLUT1 were detected by this method in the sample gel, which indicated that the major content of the band 4.5 was GLUT1. However, NT was not de-tected, indicating the sensitivity limitation.
Table 3. Fragments in MALDI-ToF-MS of tryptic GLUT1a. Reprinted from Paper III with permission from Elsevier.
S-GLUT1(DTE) S-GLUT1
Residues Position Residues Position
118–126 L3–4 12–38 H1
256–264 L6–7 118–126 L3–4
257–264 L6–7 233–249 L6–7
459–468 C terminus 256–264 L6–7
478–492 C terminus 301–330 H8
459–492 C terminus aThe locations of the fragments in the amino acid sequence of GLUT1 are given. H1 is the transmembrane -helical domain of GLUT1 (Mueckler et al. 1985), etc. L3–4 is the hydro-philic loop between the two transmembrane domains 3 and 4, etc. The C-terminus is cyto-plasmic.
The detection of H1 and H8 in monomeic S-GLUT1, but not in monomeric S-GLUT1(DTE) implied conformation difference between S-GLUT1(DTE) and S-GLUT1.
Amino acid analysis (Papers I, II, III and IV) Amino acids, except for Trp and Cys, in purified GLUT1, immobilized pro-teoliposomes and membrane vesicles were determined by hydrolysis (24h, 6 M HCl, 110ºC) of the material followed by automatic ion-exchange chroma-tography. Val, Ile and Phe values were corrected by use of the factors 1.11, 1.13 and 1.05, respectively, as determined from three pairs of GLUT1 analy-ses following both 24h and 72h of hydrolysis. Trp, Cys and non-phosphatidylserine Ser values were obtained by use of the GLUT1 amino acid composition (SwissProt) or, for vesicles, as in Baldwin & Lienhard 1989.
25
In Paper I, the amino acid analysis was used to calibrate the standard curves for BSA, purified GLUT1, glycophorin and the proteins in membrane vesicles. In Papers II, III and IV, the amino acid analysis was used to quanti-tate proteins in immobilized biomembranes.
In Paper II, if GLUT1 purity was assumed to be 90% (Rampal et al.1986), the GLUT1 fraction in total protein amount in the membrane vesicles was determined to be 12.1%, similarly as reported by Mascher & Lundahl 1988. For the immobilized biotinylated red blood cells, the number of cells was determined by amino acid analysis of the hemoglobin eluted by water and the amount of GLUT1 in the cells was calculated using 5.1 × 105
GLUT1 molecules per cell, as in Gottschalk et al. 2000. In Paper III, the amino acid analysis indicated that A-GLUT1 or S-
GLUT1 were slightly impure. The minor contaminations are NT, hemoglo-bin and glyceraldehyde-3-phosphate dehydrogenase. Unknown protein may also be present in the preparations.
Reconstitution of GLUT1 and immobilization (Papers II, III and IV) EYP were extracted from egg yolks by methanol and chloroform (Yang & Lundahl 1994). EYP solution at 200 mM was prepared in 250 mM cholate, pH 8.4, at 22 C, together with 300 mM NaCl, 2 mM Na2EDTA and 20 mM Tris base. Insoluble material was removed by ultracentrifugation at 160K g for 90 min.
EYP contain approximately 70% phosphatidylcholine, 21% phosphati-dylethanolamine and 9% of other phospholipids, small amounts of choles-terol and other components (Yang & Lundahl 1994). Phosphatidylcholine and phosphatidylethanolamine were suitable to keep the activity of reconsti-tuted GLUT1 (Lundahl et al. 1991). Compared to distearoylphosphatidyl-choline, EYP are more fluid and have a better water solubility due to the presence of unsaturated fatty acids. PEG(5000)-DSPE prolongs the half time of liposomes in blood circulation (Needham et al. 1992) and induces the formation of disks. Cholesterol is a rigid molecule with weak amphiphilicity and the blood cholesterol level has great importance to control heart and blood vascular diseases.
In Papers II and III, the red blood cell glucose transporter GLUT1 was re-constituted with EYP or the endogenous phospholipids. Small unilamellar proteoliposomes were formed after size-exclusion chromatography. In Paper IV, GLUT1 was reconstituted with EYP alone, or together with cholesterol and PEG(5000)-DSPE, respectively. After size-exclusion chromatography, the reconstituted biomembranes were concentrated to approximately 200 mM lipids and mixed with Superdex 200 prep grade beads (75 mg/ml solu-
26
tion) for steric immobilization. Both cholesterol (Figure 9A) and PEG(5000)-DSPE (Figure 9B) provided efficient immobilization probably by increasing the rigidity and the viscosity, respectively, of the reconstituted biomembranes. The immobilizations were stable.
Figure 9. Effects of cholesterol (A) and PEG(5000)-DSPE (B) on the immobiliza-tion of artificial membrane structure.
The human red blood cells were coupled with sulfosuccinimidyl-6-(biotinamido) hexanoate and immobilized in a streptavidin-coupled agarose column for CB-binding activity by frontal affinity chromatography (Paper II). The CB binding capacity on immobilized cells was stable for up to 18 days at 4ºC, but less stable at 23ºC, similar as on lectin-adsorbed red blood cells in a Sepharose 4B column (Gottschalk et al. 2000).
Quantitative frontal affinity chromatography (Papers II, III and IV) Equation (1) was used to obtain Vmin, i.e., the elution volume (Vi) when the specific interactions were completely suppressed (1/[A]=0).
In quantitative frontal affinity chromatography, when one-site binding such as CB binding on immobilized biomembranes was analysed, Equation (2) was used resulting in a linear graph (Papers II, III and IV). When two-site binding such as nitrobenzylthioinosine (NBTI) binding on NT and GLUT1 was analyzed, Equation (4), which was induced from Equation (3), was used resulting in a double rectangular hyperbolic curve (Haneskog et al.1998a) (Paper III) with the assistance of Sigma Plot assuming that the two binding sites were independent and monovalent (Klotz 1983).
27
[ A ]1
[ P ]V
)[ B ]( 1
[ P ]V
)[ B ]( 1
VV1
A PB Pm i n
2
B P
B Pm i n
B P
i KKK
KK (1)
[ B ]
[ P ]V1
[ P ]V
1
VV1
m i nB Pm i nm i n K
(2)
B P
m i n 1[ B ]P [ B ]) [ B ]V( V
K
(3)
B P 2
2
B P 1
1
m i n 1[ B ][ B ]P
1[ B ][ B ]P) [ B ]V( V
KK
(4)
In the application of Equations (1) and (2) for CB binding on GLUT1, A and B represent glucose and CB, respectively. P is the number of CB binding sites. In the application of Equations (1) and (4) for NBTI binding on NT and GLUT1, A and B represent adenosine and NBTI, respectively. P1 and P2
are the numbers of NBTI binding sites on NT and GLUT1, respectively. In all equations, V is equal to the elution volume of B in the absence of A,
Vi is the elution volume of B in the presence of A, KAP and KBP are the asso-ciation constants (reciprocal to dissociation constants Kd) for the interaction of P with A and B, and [P] is the concentration of P in the volume Vmin. Vmin
was obtained by plotting 1/(V-Vi) versus 1/(A) and extrapolating to infinite [A] (Winzor 1985) as indicated in Equation 1.
Except for quantitative frontal affinity chromatography, Hummel and Dreyer analysis (Hummel & Dreyer, 1962, Lundahl et al. 1999) was also used to analyse CB-binding activity on free membrane vesicles or proteoli-posomes according to Equation (6), which was related with Equation (5). Vmin was obtained by running CB solution containing 250 mM glucose.
[ B ]
P [ B ]
) [ B ]V( V
d
m i n K
(5)
[ B ]
P [ B ]
B
dK
(6)
In Paper III, the reconstituted GLUT1 from the limited extraction of red blood cell membrane vesicles stripped without DTE retained a little CB
28
binding capacity compared to that in Paper II, whereas approximately half of NBTI-binding sites remained (Table 4).
Table 4. The number of binding sites of CB and NBTI per GLUT1 monomer in membrane vesicles and GLUT1 proteoliposomes. Estimated relative error limit is
15%. Reprinted from Paper III with permission from Elsevier.
Preparation NT-Ta CB-Tb NBTI-Tc
A-GLUT1 0.0035 0.066 0.31
S-GLUT1 0.0005 0.028 0.19
A-GLUT1DTEd 0.018 0.40 0.59
Ve 0.018 0.41 –f
VDTEg 0.029 0.51 –f
aNT-T, the molar ratio of NBTI sites on NT over the total amount of protein molecules in the column when the molecular weight of all of the protein was taken as 54117 (GLUT1);bCB-T, the ratio of CB sites on GLUT1 over the total amount of protein molecules in the column when the molecular weight of all of the protein was taken as 54117; cNBTI-T, the ratio of NBTI sites on GLUT1 over the total amount of protein molecules in the column when the molecular weight of all of the protein was taken as 54117;dA-GLUT1DTE, data from Haneskog et al. 1998a;eV, red blood cell membrane vesicles prepared without reductant;f–, no measurable interaction;gVDTE, red blood cell membrane vesicles prepared with reductant, data of NT-T from Hane-skog et al. 1998b.
In Paper II, the values for CB-binding sites per GLUT1 and Kd for CB-binding in biotinylated red blood cells both fell between those for lectin-coupled agarose bound red blood cells with or without polylysine coating (Gottschalk et al. 2000). The CB-binding sites on free membrane vesicles were doubled by Hummel and Dreyer analysis (Hummel & Dreyer, 1962, Lundahl et al. 1999) compared to the immobilized membrane vesicles by frontal affinity chromatography, whereas those on proteoliposomes of GLUT1 did not differ. The results implied the conversion between two CB-binding states of GLUT1 (Figure 10).
29
Figure 10. Stoichiometry of CB binding to GLUT1 (binding sites per monomer, r)versus the dissociation constant for the CB–GLUT1 interaction. The materials were biotinylated red blood cells ( ), red cells bound to wheat germ lectin-agarose ( ),polylysine-coated red blood cells bound to wheat germ lectin-agarose ( ), sterically immobilized ( ) and free ( ) membrane vesicles, and sterically immobilized ( )and free ( ) GLUT1 proteoliposomes. Some error bars fall within the data symbol. Reprinted from Paper II with permission from Blackwell Sciences.
In Paper IV, in the presence of cholesterol, Kd of CB was relatively consis-tent, but Kd of D-glucose increased gradually with the concentration of cho-lesterol (Figure 11). The number of CB binding sites on GLUT1 was initially decreased and then recovered although with decreased glucose affinity with cholesterol concentration (Figure 12A). This was probably caused by the increased rigidity of the reconstituted biomembranes due to the inclusion of cholesterol, whereas fluidity favored the activity of GLUT1.
In the presence of PEG(5000)-DSPE, the dissociation constants of CB and D-glucose were relatively constant. However, the CB binding sites on GLUT1 started to decrease at 12 mol% PEG(5000)-DSPE (Figure 12B). This was probably caused by steric obstruction of the long and flexible tail of PEG(5000)-DSPE. The result supported the application prospects of PEG(5000)-DSPE in pharmacology.
30
Figure 11. The effect of cholesterol on dissociation constants (Kd) of CB ( , left y-axis) and D-glucose ( , right y-axis) bindings of GLUT1 in the immobilized biomembranes.
Figure 12. The effects of cholesterol (A) and PEG(5000)-DSPE (B) on the CB bind-ing sites of GLUT1 in immobilized biomembranes.
The effect of cholesterol suggests the importance of controlled cholesterol level for efficient glucose uptake in the human body. The effect of
31
PEG(5000)-DSPE allows prediction of the influence of large polymer mole-cules on protein activity.
Glucose transport activity (Papers II and III) The glucose transport activity of proteoliposomes and red blood cells is de-termined by equilibrium exchange, zero-trans influx or efflux.
In Paper II, zero-trans influx (30s, 23°C) of 10 mM 2-deoxy-D-glucoseinto cells biotinylated with 0–5 mM sulfosuccinimidyl-6-(biotinamido)hexanoate was determined. Packed cells were diluted 3-fold for the incubation. The transport was stopped by three fold washing in ice-cold 2 mM HgCl2. Aliquots of the final pellet (homogenized in 3% perchloric acid) and the pooled supernatants were analyzed by liquid scintillation counting. Biotinylation with up to 1 mM reagent had little or no effect on 2-deoxy-D-glucose transport.
The zero-trans influx (60s, 23 C) of D- and L-glucose (at five concentra-tions in the range of 0.1–2 mM) into freeze-thawed GLUT1 proteoliposomes was determined by separation of the free glucose from the proteoliposomes on Sephadex G-50 in the presence of 0.1 mM HgCl2 with flow-scintillation detection (Radiomatic FLO-ONE/Beta 525TR, Packard, Meriden, CT, USA). The Km value determined by nonlinear regression analysis for zero-trans influx of D-glucose was 5.4 mM, higher than reported earlier, 1.6 mM (Wheeler et al. 1998). For human red blood cells, the zero-trans Km value was 1.6 mM for influx and 4.6 mM for efflux (Lowe & Walmsley 1986).
In Paper III, for determination of the equilibrium-exchange rate of D-glucose, proteoliposomes with 50 mM D-glucose were incubated at 22 Cwith trace amount of D- 14C glucose and L- 3H glucose followed by size-exclusion chromatography in the presence of 0.1 mM HgCl2. Leakage indi-cated by L- 3H glucose was compensated as described in Andersson & Lun-dahl 1990. The equilibrium exchange rate for S-GLUT1 eluted without ex-ogenous lipids at 50 mM glucose was 0.80 0.02 mmol g–1 s–1 for 1 minute incubation, similar to the reported values of 1.1 mmol g–1 s–1 for GLUT1 with a 20 second incubation (Lundquist et al. 1997) and 0.43 0.03 and 0.35
0.04 mmol g–1 s–1 with and without exogenous lipids, respectively, for a 2 minute incubation (Haneskog et al. 1996). This indicates that the sulfhydryl-affinity purified GLUT1 is active to glucose transport.
32
Conclusions and future perspectives
In this thesis, I analysed the CB-binding activity of GLUT1 in the human red blood cell membrane vesicles stripped with or without DTE, the purified GLUT1 with or without DTE and the human red blood cells immobilized by the interaction of streptavidin and biotin on gel beads. We arrived at the hy-pothesis that the GLUT1 indicated by band 4.5 on SDS-PAGE contains ap-proximately two equal portions, one portion is monomeric with high-affinity CB-binding activity and the other portion is oligomeric without high-affinity CB-binding activity. In the partial solubilization of the membrane vesicles, GLUT1, which does not have high-affinity CB-binding activity is pooled. This might be a standing step to prepare purified GLUT1 for crystallization and for pharmaceutical interests.
A useful modified micro-Bradford assay with CaPE precipitation was de-veloped by use of BSA, GLUT1, bacteriorhodopsin, glycophorin and human red blood cell membrane vesicles to achieve a routine quantitation method for membrane proteins.
The effects of cholesterol and PEG(5000)-DSPE on reconstituted GLUT1 were preliminarily determined. GLUT1 reconstituted in disk-producing lip-ids, i.e., distearoylphosphatidylcholine, PEG(5000)-DSPE and cholesterol, shall become a useful attempt of proteodisks.
33
Summary in Swedish
Kvantifiering, rening och rekonstituering av GLUT1 Det finns tre klasser av transportproteiner i cellmembran; poriner i gramne-gativa bakteriers yttermembran, adenosintrifosfat (ATP)-bindande kassett proteiner och faciliterad-transportproteiner. Poriner bildar porer i cellmem-bran för passage av substanser. ATP-bindande kassett proteiner kräver ener-gi för att transportera substanser. Faciliterad-transportproteiner påskyndar substansdiffusion, vilken sker med hjälp av koncentrationsgradienten. Iso-former av den facilitativa glukostransportören (GLUT) transporterar glukos och fruktos med olika affiniteter och de föredrar olika vävnader. Bland dem är GLUT4 och GLUT12 insulinkänsliga, vilket påvisas av att antalet trans-portör i cellmembranen i de celler som dessa transportörer finns, ökar efter en måltid.
GLUT1 är ett membranprotein som finns i de flesta djurceller, såsom röda blodkroppar och epitelceller i blod-hjärnbarriären. Från den klonade sekven-sen av cDNA från humana levercancerceller har man fått fram att GLUT1 är ett protein bestående av 492 aminosyror med en glykosylerad oligosackarid-kedja på asparagin 45 i en extracellulär loop. Substraten som transporteras är glukos och dehydroaskorbinsyra. Cytochalasin B (CB), forskolin och gene-stein är hämmare av GLUT1 och tävlar med substraten om att transporteras. ATP ändrar substratens bindning till GLUT1. Uppmärksamhet dras till lipid-effekten på aktiviteten hos rekonstituerade GLUT1. Fosfatidylkolin och fos-fatidyletanolamin från biologiska källor är bra för att bibehålla GLUT1s aktivitet. Närvaron av omättade fosfolipider är troligen en viktig bidragande faktor till denna effekt.
Perifiera proteiner i röda blodkroppars cellmembran tas ofta bort från membranet för att producera membranvesiklar, så kallad stripping. Metoder-na som används för att ta bort de perifiera proteinerna är låg jonstyrkeinku-bation med antingen alkalinstripping eller kaliumjodidstripping. Den första metoden används ofta vid rening av GLUT1 från röda blodkroppars cell-membran. Det oligomera stadiet av GLUT1 i röda blodkroppar är ovisst. GLUT1 blir främst monomerisk när den löses upp med hjälp av de icke-joniska detergenterna oktyl glukosid (OG) eller oktaetylen glykol n-dodecyleter (C12E8). Om GLUT1 istället solubiliseras med hjälp av kolat blir det en mix av dimerer och tetramerer. Närvaro av reduktanten ditioerytritol (DTE) under reningen ökar halten dimerer. Är det detergenterna eller reduktanten
34
som är den kritiska faktorn för att bestämma det oligomera stadiet av solubi-liserat GLUT1? Vad gäller för det oligomera stadiet av GLUT1 i röda blod-kroppar? Är det detsamma som för renat GLUT1?
Aminosyraanalys är oberoende av aminosyrakompositionen i proteinet och av närvaron av interfererande substanser som detergenter och lipider, men den tar några dagar att göra och den är dyr. Analysmetoder för att kvan-tifiera membranproteiner är den modifierade Lowry-metoden med triklorät-tiksyrafällning i närvaro av natriumdeoxykolat och mikro-Bradford kalcium-fosfatadsorption med etanolfällning (CaPE) metoden. Den modifierade Low-ry-metoden är tidsödande. Den tar en halv dag att göra och passar inte ihop med sulfhydrylreagens, vilka ofta används i membranproteinstudier. Mikro-Bradford CaPE-metoden är en enkel, snabb och robust metod för rutinkvan-tifiering av membranproteiner.
Min forskning I artikel I så används den modifierade mikro-Bradford CaPE-metoden för att kvantifiera GLUT1. Mängden kaliumfosfat i förhållande till mängden kalci-umklorid modifierades för att anpassa metoden till GLUT1, vilken har en stark tendens att klumpa ihop sig irreversibelt i annat fall. Den modifierade mikro-Bradford CaPE-metoden är lämplig för kvantifiering av flera olika membranproteiner, till exempel proteiner i humana röda blodkroppars cell-membranvesiklar, GLUT1, bakterierodopsin och glykoforin och för diverse annat, såsom ihopklumpade GLUT1 och återupplösta GLUT1 från proteoli-posomer immobiliserade i gelkulor. Den modifierade metoden behåller för-delarna från den vanliga mikro-Bradford CaPE-metoden, men begränsas av sitt smala linjära område i kalibreringskurvorna och sin relativt stora skillnad i känslighet för olika membranproteiner.
I artikel II och III så omvandlades humana röda blodkroppars cellmem-bran till vesiklar med låg jonstyrkeinkubation med alkalinlösning med (arti-kel II och III) eller utan (artikel III) DTE. Supernatanten av delvis solubilise-rade membranvesiklar gav cirka 65% av vesikelproteinerna. GLUT1 som renats från supernatanten visade liten hög affinitets CB-bindande aktivitet. Vesiklar som strippades i närvaro av DTE gav mest monomert GLUT1 och de som strippades i frånvaro av DTE gav monomert och oligomert GLUT1. MALDI-ToF-MS användes för att se de olika GLUT1 fragmenten från de två preparationsvarianterna. Antalet endogena fosfolipider som fanns kvar på GLUT1 var också olika för de två preparationsvarianterna, men jämviktsut-bytet av glukos var detsamma. GLUT1s CB-bindande aktivitet i streptavi-din-biotin-immobiliserade röda blodkroppar visade att både dissociations-konstanten och antalet bindningssäten per GLUT1 hamnade mellan värdena i vetegroddslecitin-immobiliserade röda blodkroppar med eller utan polylysin, vilket visar på växling mellan två CB-bindande stadier hos GLUT1. Slutsat-
35
sen blir att GLUT1-stadiet i röda blodkroppar består av två liknande stora delar; monomeriskt med hög affinitets CB-bindande aktivitet och oligome-riskt utan hög affinitets CB-bindande aktivitet. I delvis solubilisering av membranvesiklar så förs de GLUT1 som inte har hög affinitets CB-bindande aktivitet ihop. Detta kan kanske vara lösningen för att välja ut olika oligome-ra och funktionella GLUT1 för kristallisering.
I artikel IV så undersöktes preliminära effekter av kolesterol och polyety-lenglykol 5000-distearoylfosfatidyletanolamin (PEG(5000)-DSPE) på re-konstituerade GLUT1. I närvaro av kolesterol var CBs Kd-värde relativt kon-sekvent, emedan glukos Kd-värde ökade med kolesterolkoncentrationen. Antalet CB-bindande säten på GLUT1 minskade i början men ökade sedan till sitt normala antal, fastän med minskad affinitet, när koncentrationen ko-lesterol ökade. Detta berodde troligen på minskad rörelseförmåga i de re-konstituerade membranen till följd av kolesterol, eftersom fluiditeten är vik-tig för GLUT1s aktivitet. I närvaro av PEG(5000)-DSPE så var dissocia-tionskonstanterna för CB och glukos relativt konstanta, men antalet CB-bindande säten på GLUT1 började minska vid 12 mol% PEG(5000)-DSPE. Detta berodde troligen på steriska hinder orsakade av de långa flexibla svan-sarna på PEG(5000)-DSPE-molekylerna.
36
Acknowledgements
At this moment I would like to acknowledge financial support from Uppsala University, The Swedish Natural Science Research Council, the Swedish Research Council and the O. E and Edla Johansson Science Foundation.
I would like to thank my late supervisor professor Per Lundahl for introduc-ing me to the area of biomembrane. Your encouragement to the work and help in writing the papers are especially appreciated. Without you there would not have been the present outcome of the work.
Thanks to professor Bengt Mannervik for being my examiner and supervisor with interesting discussions. Thanks to professor Ulf Hellman for analysis of MALDI ToF MS in our collaboration and to professor Katarina Edwards for guidance of the proteodisk project. Thanks to Emma Johansson for discus-sions and cryo-TEM work.
I would also like to thank the past and present biomembrane members to create an active and nice environment: Eva Greijer for the vesicle prepara-tion in the beginning of the work. Chengming Zeng for comments on my projects. Christine Lagerquist Hägglund for help in the lab and being my vice supervisor. Ingo Gottschalk for assistance with computer programs and collaboration; Farideh Beigi for chats. Andreas Lundquist for collaboration. Anna Lundquist for help with lab work. Caroline Engvall and Elisabet Boija for Swedish translations. Sanela Kurtovic for interest in the GLUT1 work. Lars Andersson, Lars Haneskog and Eggert Brekkan for discussions.
Thanks to David Eaker, Karin Elenbring and Marie Sundquist for the amino acid analysis and David also for linguistic revision of the papers. Thanks to Birgitta Evremar and Ulla Lidberg for efficient help and Lilian Forsberg for assistance in lab teaching. Thanks to Per-Axel Lidström for fixing equip-ments and computers. Thanks to the other stuff in the department for your help.
Finally, I would like to thank my parents, parents-in-law, my sisters, Xiang-sheng and Yisheng, my brother, Zhipeng, and my other relatives for support; my husband Xingwu for sharing your life with me, and our sons Shuai and Viktor for your inspirations.
37
References
Abramson, J., Smirnova, I., Kasho, V., Verner, G., Kaback, H.R. and Iwata, S. (2003) Structure and mechanism of the lactose permease of Escherichia coli.Science 301, 610–615.
Acevedo, F., Lundahl, P. and Fröman, G. (1981) The stereospecific D-glucose trans-port protein in cholate extracts of human erythrocyte membranes. Molecular sieve chromatography and estimation of molecular weight. Biochim. Biophys. Acta 648, 254–262.
Andersson, L. and Lundahl, P. (1990) Substantial glucose leakage from liposomes on filters and upon molecular-sieve chromatography in determinations of recon-stituted glucose-transport activity and liposome volumes. Biochim. Biophys. Acta 1030, 258–268.
Atkinson, M.A. and Maclaren N.K. (1990) What causes diabetes? Sci. Amer. July, 42–49.
Bailey, C.J. (1992) Biguanides and NIDDM. Diabetes Care 15, 755-772. Baldwin, S.A., Baldwin, J.M. and Lienhard, G.E. (1982) Monosaccharide trans-
porter of the human erythrocyte. Characterization of an improved preparation. Biochemistry 21, 3836–3842.
Baldwin, S.A. and Lienhard, G.E. (1989) Purification and reconstitution of glucose transporter from human erythrocytes. Methods Enzymol. 174, 39 50.
Bennett, V. and Stenbuck, P.J. (1979a) Identification and partial purification of Ankyrin, the high affinity membrane attachment site for human erythrocyte spectrin. J. Biol. Chem. 254, 2533–2541.
Bennett, V. and Stenbuck, P.J. (1979b) The membrane attachment protein for spec-trin is associated with band 3 in human erythrocyte membranes. Nature 280, 468–473.
Boado, R.J. and Pardridge, W.M. (1993) Glucose deprivation causes posttranscrip-tional enhancement of brain capillary endothelial glucose transporter gene ex-pression via GLUT1 mRNA stabilization. J. Neurochem. 60, 2290–2296.
Boodhoo, A. and Reithmeier, R.A.F. (1984) Characterization of matrix-bound band 3, the anion transport protein from human erythrocyte membranes. J. Biol. Chem. 259, 785–790.
Boulter, J.M. and Wang, D.N. (2001) Purification and characterization of human erythrocyte glucose transporter in decylmaltoside detergent solution, ProteinExpr. Purif. 22, 337–348.
Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of micro-gram quantities of protein utilizing the principle of protein–dye binding. Anal.Biochem. 72, 248–254.
Brekkan, E., Lundqvist, A. and Lundahl, P. (1996) Immobilized membrane vesicle or proteoliposome affinity chromatography. Frontal analysis of interactions of cytochalasin B and D-glucose with the human red cell glucose transporter. Bio-chemistry 35, 12141–12145.
38
Bruce, L.J., Beckmann, R., Ribeiro, M.L., Peters, L.L., Chasis, J.A., Delaunay, J., Mohandas, N., Anstee, D.J. and Tanner, M.J. (2003) A band 3-based macro-complex of integral and peripheral proteins in the red cell membrane. Blood101, 4180–4188.
Bunn, R.C., Jensen, M.A. and Reed, B.C. (1999) Protein interactions with the glu-cose transporter binding protein GLUT1CBP that provide a link between GLUT1 and the cytoskeleton. Mol. Biol. Cell 10, 819–832.
Carruthers, A. (1986) Anomalous asymmetric kinetics of human red cell hexose transfer: role of cytosolic adenosine 5'-triphosphate. Biochemistry 25, 3592–3602.
Carter-Su, C., Pessin, J.E., Mora, R., Gitomer, W. and Czech, M.P. (1982) Photoaf-finity labeling of the human erythrocyte D-glucose transporter. J. Biol. Chem.257, 5419–5425.
Casey, J.R., Lieberman, D.M. and Reithmeier, R.A.F. (1989) Purification and char-acterization of Band 3 protein. Methods Enzymol. 173, 491–512.
Cloherty E.K., Levine K.B. and Carruthers A. (2001) The red blood cell glucose transporter presents multiple, nucleotide-sensitive sugar exit sites. Biochemistry40, 15549–15561.
Cloherty E.K., Sultzman, L.A., Zottola, R.J. and Carruthers, A. (1995) Net sugar transport is a multistep process. Evidence for cytosolic sugar binding sites in erythrocytes, Biochemistry 34, 15395–15406.
Coderre, P.E., Cloherty, E.K., Zottola, R.J. and Carruthers, A. (1995) Rapid sub-strate translocation by the multisubunit, erythroid glucose transporter requires subunit associations but not cooperative ligand binding. Biochemistry 34, 9762–9773.
Cuppoletti, J., Jung, C.Y. and Green, F.A. (1981) Glucose transport carrier of human erythrocytes. Radiation target size measurement based on flux inactivation. J.Biol. Chem. 256, 1305–1306.
Cushman, S.W. and Wardzala, L.J. (1980) Potential mechanism of insulin action on glucose transport in the isolated rat adipose cell. J. Biol. Chem. 255, 4758-4762.
De Vivo, D.C., Trifiletti, R.R., Jacobson, R.I., Ronen, G.M., Behmand, R.A. and Harik, S.I. (1991) Defective glucose transport across the blood-brain barrier as a cause of persistent hypoglycorrhachia, seizures, and developmental delay. NewEngl. J. Med., , 325: 703–709.
Dolder, M., Walz, T., Hefti, A. and Engel, A. (1993) Human erythrocyte band 3 solubilization and reconstitution into two-dimensional crystals. J. Mol. Biol.231, 119–132.
Edwards, K., Johansson, M., Karlsson, G. and Silvander, M. (1997) Effect of poly-ethyleneglycol-phospholipids on aggregate structure in preparations of small unilamellar liposomes. Biophys. J. 73, 258–266.
Finean J.B., Coleman R. and Michell R.H., Membranes and their cellular functions. 1978, 2nd ed., Blackwell Scientific Publications, Oxford.
Gies, A.P., Nonidez, W.K., Ellison, S.T., Ji, H. and Mays, J.W. (2005) A MALDI-TOF MS study of oligomeric poly(m-phenyleneisophthalamide). Anal. Chem.,77, 780–784.
Goto, S., Cheng, K.-X., Oda, S., Hori, H. and Terada, H. (1998) Dynamic structural features of macrocyclic cytochalasin analoges responsible for their hexose trans-port inhibition, J. Am. Chem. Soc., 120, 2457–.
Gottschalk, I., Li, Y.-M. and Lundahl, P. (2000) Chromatography on cells: Analyses of solute interactions with the glucose transporter Glut1 in human red cells ad-sorbed on lectin-gel beads. J. Chromatogr. B 739, 55–62.
Groop, L.C. (1992) Sulfonylureas in NIDDM. Diabetes Care 15:737-754.
39
Hamill, S., Cloherty, E.K. and Carruthers, A. (1999), The human erythrocyte sugar transporter presents two sugar import sites. Biochemistry, 38, 16974–16983.
Haneskog, L., Andersson, L., Brekkan, E., Englund, A.-K., Kameyama, K., Liljas, L., Greijer, E., Fischbarg, J. and Lundahl, P. (1996), Monomeric human red cell glucose transporter (Glut1) in non-ionic detergent solution and a semi-elliptical torus model for detergent binding to membrane proteins. Biochim. Biophys. Acta1282, 39–47.
Haneskog, L., Lundqvist, A. and Lundahl, P. (1998a) Biomembrane affinity chro-matographic analysis of nitrobenzylthioinosine binding to the reconstituted hu-man red cell nucleoside transporter, J. Mol. Recogn. 11, 58–61.
Haneskog, L., Zeng, C.-M., Lundqvist, A. and Lundahl, P. (1998b) Biomembrane affinity chromatographic analysis of inhibitor binding to the human red cell nu-cleoside transporter in immobilized cells, vesicles and proteoliposomes. Bio-chim. Biophys. Acta 1371, 1 4.
Heard, K.S., Fidyk, N. and Carruthers, A. (2000) ATP-dependent substrate occlu-sion by the human erythrocyte sugar transporter. Biochemistry 39, 3005–3014.
Hebert, D.N. and Carruthers, A. (1991) Cholate-solubilized erythrocyte glucose transporters exist as a mixture of homodimers and homotetramers. Biochemistry30, 4654–4658.
Hebert, D.N. and Carruthers, A. (1992) Glucose transporter oligomeric structure determines transporter function. J. Biol. Chem. 267, 23829–23838.
Hellman, U. (1997) Isolation of peptides for microsequencing by in-gel proteolytic digestion, in: Protein structure analysis. Preparation, characterization, and mi-crosequencing (Kamp, R.M. Choli-Padopoulou, T. Wittmann-Liebold, B. Eds.). Springer, Heidelberg, pp. 97–104.
Higgins C.F. (1992) ABC transporters: From microorganisms to man. Annu. Rev. Cell Biol. 8, 67-113.
Hinkle, P.C., Sogin, D.C., Wheeler, T.J. and Telford, J.N. (1979) Studies of the glucose transporter from human erythrocytes reconstituted in liposomes. In Function and molecular Aspects of Biomembrane Transport (Quagliarello, E. Palmieri, F. Papa, S. and Klingenberg, M., Eds.), Elsevier/North Holland, NY, 487–492.
Hummel, J.P. and Dreyer, W.J. (1962) Measurement of protein-binding phenomena by gel filtration. Biochim. Biophys. Acta 63, 530–532.
Inukai, K., Asano, T., Katagiri, H., Anai, M., Funaki, M., Ishihara, H., Tsukuda, K., Kikuchi, M., Yazaki, Y. and Oka, Y. (1994) Replacement of both tryptophan vesicles at 388 and 412 completely abolished cytochalasin B photolabelling of the GLUT1 glucose transporter. Biochem. J. 302, 355–361.
Jarvis, S.M., Ellory, J.C. and Young, J.D. (1986) Radiation inactivation of the hu-man erythrocyte nucleoside and glucose transporters. Biochim. Biophys. Acta855, 312–315.
Johansson, E., Engvall, C., Arfvidsson, M., Lundahl, P. and Edwards, K. (2004) Development and initial evaluation of PEG-stabilized disks as novel model membranes. Biophys. Chem. 113, 183–192.
Jung, C.Y. and Rampal, A.L. (1977) Cytochalasin B binding sites and glucose trans-port carrier in human erythrocyte ghosts. J. Biol. Chem. 252, 5456–5463.
Kalaria, R.N. and Harik, S.I. (1989) Reduced glucose transporter at the blood-brain barrier and in cerebral cortex in Alzheimer disease. J. Neurochem. 53, 1083-1088.
Kevine, K.B., Hamill, S., Cloherty, E.K. and Carruthers, A. (2001) Alanine scanning mutagenesis of the human erythrocyte glucose transporter putative ATP binding domain. Blood cells, Molecules, and Diseases 27, 139–142.
40
Klotz, I.M. (1983) Ligand-receptor interactions: what we can and cannot learn from binding measurements. Trends Pharmacol. Sci. 4, 253–255.
Kragh-Hansen U., Le Maire M., Nöel J.P., Gulik-Krzywicki T. and Møller J.V. (1993) Transitional steps in the solubilization of protein-containing membranes and liposomes nonionicdetergent. Biochemistry 32, 1648–1656.
Lagerquist Hägglund C. and Lundahl, P. (2003) Centrifugal and chromatographic analyses of tryptophan and tyrosine uptake by red blood cells and GLUT1 pro-teoliposomes with permeability estimates and observations on dihydrocytoch-lasin B. J. Biochem. Biophys. Methods 55, 127–140.
Lavis, V.R., Lee, D.P. and Shenolikar, S. (1987) Evidence that forskolin binds to the glucose transporter of human erythrocytes. J. Biol. Chem. 262, 14571–14575.
Lemieux, M.J., Reithmeier, R.A.F. and Wang, D.N. (2002) Importance of detergent and phospholipid in the crystallization of the human erythrocyte anion-exchanger membrane domain. J. Struc. Biol. 137, 322–332.
Levine, K.B., Cloherty, E.K., Fidyk, N.J. and Carruthers, A. (1998) Structural and physiologic determinants of human erythrocyte sugar transport regulation by adenosine triphosphate. Biochemistry 37, 12221–12232.
Lin, S. and Snyder, C.E. (1977) High affinity cytochalasin B binding to red cell membrane proteins which are unrelated to sugar transport. J Biol. Chem., 252, 5464–5471.
Low, P.S. (1986) Structure and function of the cytoplasmic domain of band 3: Cen-ter of erythrocyte membrane-peripheral protein interactions. Biochim. Biophys. Acta 864, 145–167.
Low, P.S., Rathinavelu, P. and Harrison, M.L. (1993) Regulation of glycolysis via reversible enzyme binding to the membrane protein, band 3. J. Biol. Chem. 268, 14627–14631.
Lowe, A.G. and Walmsley, A.R. (1986) The kinetics of glucose transport in human red blood cells. Biochim. Biophys. Acta 857, 146–154.
Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein meas-urement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275.
Lundahl, P., Greijer, E., Lindblom, H. and Fägerstam, L.G. (1984) Fractionation of human red cell membrane proteins by ion exchange chromatography in deter-gent on Mono Q, with special reference to the glucose transporter. J. Chroma-togr. 297, 129–137.
Lundahl, P., Mascher, E., Andersson, L., Englund, A.-K., Greijer, E., Kameyama, K. and Takagi, T. (1991) Active and monomeric human red cell glucose transporter after high performance molecular-sieve chromatography in the presence of octyl glucoside and phosphatidylserine or phosphatidylcholine. Biochim. Biophys. Acta 1067, 177–186.
Lu, L., Lundqvist, A., Zeng, C.-M., Lagerquist, C. and Lundahl, P. (1997) D-Glucose, forskolin and cytochalasin B affinities for the glucose transporter Glut1. Study of pH and reconstitution effects by biomembrane affinity chroma-tography. J. Chromatogr. A, 776, 81–86.
Lundahl, P., Zeng, C.-M., Lagerquist Hägglund, C., Gottschalk, I. and Greijer, E. (1999) Chromatographic approaches to liposomes, proteoliposomes and biomembrane vesicles. J. Chromatogr. B 722, 103–120.
Lundqvist, A., Brekkan, E., Lagerquist, C., Haneskog, L. and Lundahl, P. (1997) Frontal affinity chromatographic analysis of membrane protein reconstitution. Mater. Sci. Eng. C 4, 221–226.
Lundqvist, A. and Lundahl, P. (1997) Glucose affinity for the glucose transporter Glut1 in native or reconstituted lipid bilayers. Temperature-dependence study by biomembrane affinity chromatography. J. Chromatogr. A, 776, 87–91.
41
Maloney P.C. and Wilson T.H. (1993) The evolution of membrane carriers. In Mo-lecular biology and function of carrier proteins. (Reuss, L., Russell, J.M., & Jennings, M.L., Eds.), Rockefeller University Press, NY, pp. 147-160.
Mascher, E. and Lundahl, P. (1988) The human red cell glucose transporter in octyl glucoside. High specific activity of monomers in the presence of membrane lip-ids. Biochim. Biophys. Acta 945, 350 359.
Massa, S.M., Swanson, R.A. and Sharp, F.R. (1996) The stress gene response in brain. Cerebrovascular Brain Metabolism Rev. 8, 95–158.
Mueckler, M., Caruso, C., Baldwin, S.A., Panico, M., Blench, I., Morris, H.R., Allard, W.J., Lienhard, G.E. and Lodish, H.F. (1985) Sequence and structure of a human glucose transporter. Science 229, 941–945.
Murakami, T., Nishiyama, T., Shirotani, T., Shinoraha, Y., Kan, M., Ishii, K., Kanai, F., Nakazuru, S. and Ebina, Y. (1992) Identification of two enhancer elements in the gene encoding the type 1 glucose transporter from the mouse which are responsive to serum, growth factor and oncogenes. J. Biol. Chem. 267, 9300–9306.
Needham, D., McIntosh, T.J. and Lasic, D.D. (1992) Repulsive interactions and mechanical stability of polymer-grafted lipid membranes, Biochim Biophys Acta., 1108, 40–48
Needham, D. and Nunn, R.S. (1990) Elastic deformation and failure of lipid mem-branes containing cholesterol. Biophys. J. 58, 997–1009.
Pande, S.V. and Murthy, M.S.R. (1994) A modified micro-Bradford procedure for elimination of interference from sodium dodecyl sulfate, other detergents, and lipids. Anal. Biochem. 220, 424–426.
Pessino, A., Hebert, D.N., Woon, C.W., Harrison, S.A., Clancy, B.M., Buxton, J.M., Carruthers, A. and Czech, M.P. (1991) Evidence that functional erythrocyte-type glucose transporters are oligomers. J. Biol. Chem. 266, 20213–20217.
Peterson, G.L. (1977) A simplification of the protein assay method of Lowry et al.which is more generally applicable. Anal. Biochem. 83, 346–356.
Pinkofsky, H.B., Rampal, A.L., Cowden, M.A. and Jung, C.Y. (1978) Cytochalasin B binding proteins in human erythrocyte membranes. J. Biol. Chem. 253, 4930–4937.
Rampal, A.L., Jung, E.K.Y., Chin, J.J., Deziel, M.R., Pinkofsky, H.B. and Jung, C.Y. (1986) Further characterization and chemical purity assessment of the hu-man erythrocyte glucose transporter preparation. Biochim. Biophys. Acta 859, 135–142.
Sahin-Tóth, M., Lawrence, M.C. and Kaback, H.R. (1994) Properties of permease dimmer, a fusion protein containing two lactose permease molecules from Es-cherichia coli. Proc. Natl. Acad. Sci. USA, 91, 5421–5425.
Saier, M.H. (2000) Families of transmembrane sugar transport proteins. 35, 699–710.
Sase, S., Anraku, Y., Nagano, M., Osumi, M. and Kasahara, M. (1982) Random distribution of the glucose transporter of the erythrocytes in reconstituted lipo-somes. J. Biol. Chem. 257, 11100–11105.
Shanahan, M.F., Morris, D.P. and Edwards, B.M. (1987) [3H]Forskolin direct photoaffinity labeling of the erythrocyte D-glucose transporter, J. Biol. Chem.262, 5978–5984.
Silvander, M., Karlsson, G. and Edwards, K. (1996) Vesicle solubilization by alkyl sulfate surfactants: a Cryo-Tem study of the vesicle to micelle transition. J. Col-loid Interface Science 179, 104–112.
Singer, S.J. and Nicolson, G.L. (1972) The fluid mosaic model of the structure of cell membranes. Science 175, 720–731.
42
Steck, T.L., Weinstein, R.S., Straus, J.H. and Wallach, D.F. (1970) Inside-out red cell membrane vesicles: preparation and purification. Science168,255–257.
Vannucci, S.J. Maher, F. and Simpson I.A. (1997) Glucose transporter proteins in brain: Delivery of glucose to neurons and glia. GLIA 21, 2–21.
Vera, J.C., Reyes, A.M., Cárcamo, J.G., Velásques, F.V., Rivas, C.I., Zhang, R.H., Strobel, P., Iribarren, R., Scher, H.I., Slebe, J.C. and Golde, D.W. (1996) Gen-istein is a natural inhibitor of hexose and dehydroascorbic acid transport through the glucose transporter, GLUT1. J. Biol. Chem. 271, 8719–8724.
Vera, J.C., Reyes, A.M., Velásques, F.V., Rivas, C.I., Zhang, R.H., Strobel, P., Slebe, J.C., Núñez-Alarcón, J. and Golde, D.W. (2001) Direct inhibition of the hexose transporter GLUT1 by tyrosine kinase inhibitors. Biochemistry 40, 777–790.
Vince, J.W., Sarabia, V.E. and Reithmeier, R.A.F. (1997) Self-association of Band 3, the human erythrocyte anion exchanger, in detergent solution. 1326, 295–306.
Voet, D. and Voet, J.G. (1995) Lipids and membranes. In: Biochemistry (2st edition, Voet, D. and Voet, J.G. eds.). John Wiley & Sons, Inc, NJ, pp 271–329.
Wheeler, T.J. (1989) ATP does not regulate the reconstituted glucose transporter, Biochemistry 28, 3413–3420.
Wheeler, T.J., Cole, D. and Hauck, M.A. (1998) Characterization of glucose trans-port activity reconstituted from heart and other tissues. Biochim. Biophys. Acta1414, 217–230.
Widmer, J., Raffin, Y., Gaillard, J.-M., Bovier, P. and Tissot, R. (1990) Reversible in vitro decrease of L-tyrosine and L-tryptophan influx across the human eryth-rocyte membrane induced by cytochalasin B, the specific inhibitor of D-glucose transport, Neuropsychobiol. 24, 67–73.
Winzor, D.J. (1985) Quantitative characterization ofinteractions by affinity chroma-tography. In Affinity Chromatography, A Practical Approach (Dean P.D.G., Johnson W. S. and Middle F.A., eds.), IRL Press, Oxford, pp. 253–298.
Wood, I.S. and Trayhurn, P. (2003) Horizons in nutritional science glucose trans-porters (GLUT and SGLT): expanded families of sugar transport proteins. Br.J. Nutr. 89, 3–9.
Yang, Q. and Lundahl, P. (1994) Steric immobilization of liposomes in chroma-tographic gel beds and incorporation of integral membrane proteins into their lipid bilayers, Anal. Biochem. 218 210–221.
Zeng, C.-M., Zhang, Y., Lu, L., Brekkan, E., Lundqvist, A. and Lundahl, P. (1997) Immobilization of human red cells in gel particles for chromatographic activity studies of the glucose transporter Glut1. Biochim. Biophys. Acta 1325, 91–98.
Zhang, J.Z., Hayashi, H., Ebina, Y., Prohaska, R. and Ismail-Beigi, F. (1999) Asso-ciation of stomatin (band 7.2b) with Glut1 glucose transporter. Arch. Biochem. Biophys. 372, 173–178.
Zhang, J.Z, Abbud, W., Prohaska, R. and Ismail-Beigi, F. (2001) Overexpression of stomatin depresses GLUT-1 glucose transporter activity. Am. J. Physiol. Cell Physiol. 280, C1277–83.
Zhuang, J., Gutknecht, R., Flükiger, K., Hasler, L., Erni, B. and Engel, A. (1999) Purification and electron microscopic characterization of the membrane subunit (IICB(Glc)) of the Escherichia coli glucose transporter. Arch. Biochem. Biophys.372, 89–96.
Zottola, R.J., Cloherty, E.K., Coderre, P.E., Hansen, A., Hebert, D.N. and Car-ruthers, A. (1995) Glucose transporter function is controlled by transporter oli-gomeric structure. A single, intramolecular disulfide promotes GLUT1 tetramerization. Biochemistry 34, 9734–9747.
Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 30
Editor: The Dean of the Faculty of Science and Technology
A doctoral dissertation from the Faculty of Science and Technology, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology. (Prior to January, 2005, the series was published under the title "Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology".)
Distribution: publications.uu.seurn:nbn:se:uu:diva-5727
ACTAUNIVERSITATIS
UPSALIENSISUPPSALA
2005
