Hannelore Daniel

Molecular Nutrition Unit, Department of Food and Nutrition, Technical University of Munich, Germany

Introduction

Nutrient transport across the plasma membrane of cells is a critical step in metabolism and nutrient homeostasis since the cell membrane is a barrier for permeation and compartmentalizes metabolic processes. The phospholipid bilayer surrounding cells shows an intrinsically low permeability for hydrophilic low molecular weight compounds as well as for larger molecules. However, even lipophilic nutrients such as cholesterol or fatty acids - up until now believed to cross the cell membranes by passive diffusion - appear to require specialized membrane proteins for transport that also allow control of permeation and metabolic flow. The maintenance of an intracellular environment that is distinctly different from the extracellular milieu is essential for life, and therefore a large spectrum of membrane proteins with highly specialized functions has emerged during evolution that control the membrane permeability for ions, water, macro-and micronutrients, and metabolic intermediates, but also that for xenobiotics.

Our knowledge of the molecular architecture and functions of membrane proteins was sparse, as those proteins could in most cases not be isolated by means of classical protein purification techniques. Whereas extracellular and cytoplasmic proteins could be purified in quantities that allowed crystallization and determination of their three-dimensional structure and the structural changes associated with function, membrane transport function could not be related to distinct structural protein elements.

With the advent of molecular biology and new cloning strategies, this changed dramatically. In the 1980s, the first cDNAs encoding mammalian cell membrane proteins were isolated, and allowed predictions to be made about the structure of the proteins and their integration into the cell membrane compartment. The sequences obtained also allowed identification of related sequences and proteins by homology screening. In addition, expression cloning techniques were used to obtain novel sequences encoding nutrient transport proteins (Daniel, 2000). Here, a cDNA library produced from a mRNA pool comprising thousands of individual mRNAs isolated from the tissue ofinterest is screened for the encoded protein by measuring its function after a cDNA pool has been introduced into a model cell. This was done mainly in oocytes of the South African frog Xenopus laevis, as the cells of this organism are large (~1 mm) and cDNA pools can be introduced easily by micropipettes. Once the expected nutrient transporter function is obtained by transport measurements in the oocytes, the cDNA pool is divided, and the different pools are reinjected and screened again for the expression of function. Eventually, a single cDNA is identified that encodes the protein with the known function. Sequencing of this cDNA provides the coding region (open reading frame; ORF) and can be translated into the amino acid

┬ęCAB International 2003. Molecular Nutrition (eds J. Zempleni and H. Daniel)

sequence. A variety of algorithms can then be used to predict the secondary structure of the protein, based on the different polarities of the amino acid residues within the sequence and the assumption that stretches of 10-15 or more consecutive mainly hydrophobic amino acid residues are needed to form an a-helical domain that is inserted into the phosphobilayer of the cell. Accordingly, the more hydrophilic regions are predicted to connect these transmembrane domains and form the intra-cellular and extracellular linker regions. In many cases, the predicted topologies based on this hydropathy analysis of individual membrane transporters proved to be correct when verified experimentally. Analysis of membrane protein topology is done mainly by introducing flags or special epitopes (small stretches of amino acid sequences) that are accessible to specific antibodies or reagents either from the outside of the cell or after permeabilization of the cells for reaction from the inner side of the membrane. Alternatively, a cysteine scanning approach can be used. Here the protein transport function is used as a reporter system. Codons encoding a cysteine residue are incorporated into the cDNA by site-directed mutagenesis at any interesting position, and the chemical modification of the cysteine residues with special membrane-impermeable or -permeable reagents that eventually will impair protein function allows determination of the location of the cysteine residue as accessible from the inside or outside of the cell by functional analysis in a heterologous expression system.

At the Beginning of Molecular Mammalian Transport Biology: the GLUT Transporters

A landmark for the cloning of human nutrient transporters was the identification of the first GLUT family member from HepG2 cells (Mueckler et al., 1985). At the protein level, the cloned cDNA showed a high similarity and even amino acid identity to the red cell glucose transporter from which a partial sequence had been obtained by classical Edman sequencing. According to the proposed terminology based on HUGO data (human genome sequencing organization) the GLUT transporters are now designated as the SLC2A group with consecutive numbering. The GLUT family (see Table 3.1) currently contains the GLUT-1-GLUT-5 proteins, which are the primary transporters that mediate monosac-charide transport into and out of mammalian cells (Seatter and Gould, 1999). GLUT-6-12 are proteins with as yet not well defined functions, and GLUT-13 represents a proton-dependent myoinositol-transporter mainly expressed in the brain (Uldry et al., 2001). GLUTs generally are comprised of approximately 500 amino acids that are arranged most probably in 12 transmembrane (TM) segments that, by arrangement, form a pore. Both the N- and the C-terminus of the proteins face the intracellular compartment, a large extracellular loop near the N-terminus of the transporter is glycosylated in the mature protein and a large intracellular loop connecting TM segments

Table 3.1. The family of the mammalian GLUT transporters involved in transmembrane translocation of hexoses and related solutes and identified human diseases associated with their malfunction.

SLC2-a family Major sites of expression

HUGO code

Human diseases

GLUT-1 GLUT-2 GLUT-3 GLUT-4 GLUT-5 GLUT-6 GLUT-7 GLUT-8 GLUT-9 GLUT-X GLUT-13 (HMIT)

Erythrocytes, brain (vascular) SLC2A1

Liver, pancreas (islets) Brain (neuronal) Muscle, fat tissue, heart Intestine, kidney, testis

Novel isoforms, function not yet defined

Brain

SLC2A2 SLC2A3 SLC2A4 SLC2A5 SLC2A6 SLC2A7 SLC2A8 SLC2A9

SLC2A13

GLUT-1 deficiency syndrome

Fanconi-Bickel syndrome, NIDDM Fetal glucose toxicity Iatrogenic diabetes by HIV, NIDDM

(proton-dependent myoinositol transporter)

NIDDM, non-insulin dependent diabetes.

6 and 7 carries consensus sites for phosphorylation (Hruz and Mueckler, 2001). A three-dimensional model of GLUT-1 and its integration into the plasma cell membrane has been reported recently that shows a channel-like central structure that traverses the entire protein and that may form the substrate translocation pore (Zuniga et al., 2001). These models are important in understanding the structure and function as no crystal structures of any of the nutrient transporter proteins are available yet.

In terms of nutrition and glucose homeostasis, the GLUT-2 and GLUT-4 transporters are of central importance. GLUT-2 is responsible for influx and efflux of glucose in hepatocytes and therefore contributes to plasma glucose levels both in the post-prandial state with removal of glucose from portal blood, and in maintaining glucose levels during starvation by contributing to the efflux of glucose provided by gluconeogenesis and/or glycogenolysis. GLUT-2 expression in the basolateral membrane of epithelial cells such as those of the intestine allows glucose, galactose and fructose that enter the cells from the intestinal lumen mainly via the sodium-dependent glucose transporter SGLT-1 (glucose, galactose) or via the GLUT-5 transporter (fructose) to exit into the circulation. As the GLUT transporters (except GLUT-13) mediate monosaccharide transport following concentration gradients, metabolism acts as a modulator of the transmembrane concentration gradients and plays a crucial role in regulating overall transport capacity. GLUT-2 has a particularly important function in the b-cells of the pancreatic islets as the system that provides glucose to the cell that then in turn triggers insulin secretion. GLUT-2 in conjunction with glucokinase is central to the blood glucose-sensing pathway in b-cells. The other most important GLUT family member is GLUT-4 - a key target for modern therapy of non-insulin-dependent type II diabetes (NIDDM). Insulin-dependent incorporation of GLUT-4 transporters into the plasma membrane of adipocytes and muscle cells by translocation from intracellular vesicular stores can serve as a paradigm for the complexity of signalling pathways and the cellular machinery needed for membrane trafficking and membrane fusion. GLUT-4 synthesis and its rapid cycling between the plasma membrane and the storage vesicles is subject to regulation by a large number of factors including a variety of hormones and metabolites, and by exercise (Watson and Pessin, 2001). GLUT-4 is becoming a key player in the therapy of NIDDM with the goal ofsensitizing the system for enhanced performance. How critical the function of GLUT-4 is for glucose homeostasis was demonstrated recently in a very elegant study employing mice with a muscle-specific gene inactivation of glut-4 (cre/loxP gene targeting) that showed dramatic changes in glucose metabolism (Kim etal., 2001). Various types of malfunction of the glucose transporters of the GLUT family have been reported (see Table 3.1) and recently human immunodeficiency virus (HIV) protease inhibitors have been identified as inhibitors of GLUT-4, which may explain some of the metabolic impairments of energy and glucose metabolism in AIDS patients. GLUT-1 deficiency produces a seizure disorder with low glucose concentrations in the cerebrospinal fluid, and GLUT-2 deficiency is the basis of the Fanconi-Bickel syndrome, which resembles type I glycogen storage disease.

The SGLT Family of Electrogenic Transporters

The discovery of the sodium dependency of intestinal glucose transport in the early 1960s fostered vital research to understand the molecular basis of this coupling of glucose influx into enterocytes to sodium ion movement and the energetics of this uphill transport process. The huge daily quantities of glucose absorbed as the end-product of luminal carbohydrate digestion as well as from disaccharide hydrolysis suggested that the transport protein responsible would have a very high transport capacity. The energetics of glucose transport into enterocytes allows glucose to accumulate in the cell and provides a concentration gradient that enables downstream movement of glucose via GLUT-2 at the basolateral side as long as the concentration in the circulation is less than that in the cell. Phlorizin turned out to be a very helpful tool for analysis of glucose transport as it is a high-affinity uptake inhibitor.

The transport protein was identified almost 25 years after the initial proposal of the sodium gradient hypothesis of intestinal glucose transport. In 1987, expression cloning employing Xenopus oocytes and selecting mRNA pools that induced sodium-dependent and phlorizin-sensitive glucose transport in oocytes led to the first cDNA encoding an electrogenic glucose transporter named SGLT-1 (Hediger et al., 1987). HUGO data now show the large number of transport proteins that belong to the same family, with individual members not only in transport of hexoses but also, for example, in sodium-dependent transport of iodide, myo-inositol and water-soluble vitamins such as biotin and pantothenate. Proteins homologous to SGLT-1 are also found in prokaryotes and invertebrates, with transport functions for urea, proline, sugars or vitamins.

Various members of the SGLT family, when expressed in heterologous systems, showed remarkable similarities in function. The high-affinity Na+-glucose co-transporter SGLT-1 from three species (human, rabbit and rat), the low-affinity Na+-glucose co-transporter SGLT-2, the Na+-dependent myoinositol co-transporter SMIT1 and the Na+-dependent iodide co-transporter NIS have all been expressed in Xenopus oocytes and analysed by radioactive tracers and electrophysiological techniques (Turk and Wright, 1997). In the absence of substrate, they all show a limited Na+ permeability but, in the presence of substrates, voltage-dependent positive inward currents indicating Na+-substrate co-transport are recorded. The transport cycle is ordered, with Na+ binding to the protein in the first step, followed by substrate binding and the translocation of the loaded complex, and sequential release of the substrate and co-transported ion into the cytoplasm. The rate-

limiting step in the process appears to be the reorientation of the unloaded but charged transporter to the outer face of the membrane. More recently, an intrinsic water permeability has been proposed for SGLT-1, with a substantial number of water molecules co-transported in each cycle together with sodium ions and glucose (Loo et al, 1996). Based on the daily amount of glucose transported by SGLT-1, its water transport capacity could account for almost 5 l of fluid absorbed from the upper small intestine solely mediated by SGLT-1.

At the structural level, the SGLT-1 protein has been studied extensively by functional analysis in combination with mutations, epitope tagging, immunological and other cell biological and biophysical techniques. Based on experimental data and computational analyses, a 14 membrane-spanning domain model of human SGLT-1 has been proposed (see Fig. 3.1). All TM domains are most likely to be a-helical, the N-terminus is extracellular and the large highly charged C-terminal domain is cytoplasmic (Turk et al., 1996, 1997). Human SGLT-1 has 664 amino acid residues; other members of the family have between 530 and 735 amino acids. Using truncated and modified SGLT-1-proteins and biophysical techniques, two regions have been identified in the protein that confer the functional domains, with respect to the binding pocket for sodium ions and the region that binds the substrate and thereafter mediates the translocation of sodium and glucose to the cell interior. The conformational change

Hannelore Daniel
Fig. 3.1. The predicted membrane topology of the electrogenic Na+-dependent glucose transporter SGLT-1 as found predominantly in the apical membranes of intestinal and renal epithelial cells.

necessary for the movement of sodium ions and substrate through the protein appears to require just the tilting of two TM helices within the plane of the membrane by a movement of 10-15 A that then generates a pore-like structure for permeation (Eskandari et al., 1998).

Glucose galactose malabsorption syndrome is a hereditary disease in which SGLT-1 does not function properly (Martin et al., 1996). The mutations found in humans have allowed the identification of critical amino acid residues in the SGLT-1 protein that cause the malabsorption of glucose and galactose in the gut leading to life-threatening diarrhoea. Surprisingly, almost all mutations identified in patients cause defects in the biosynthetic pathway of the SGLT-1 protein and a lack of integration into the cell membrane. This addresses one of the important areas in membrane transporter biology. How are proteins guided from the endoplasmic reticulum to the Golgi apparatus and to the specific cell membrane compartment (apical versus basolateral side)?

SGLT-1 expression in the gut is subject to regulation by a variety of hormones and luminal factors. Intestinal glucose transport also shows a significant circadian rhythm: kinetic studies suggested an altered SGLT-1 maximal velocity (Vmax) associated with changes in gene expression. SGLT-1 may be responsible for the fairly high absorption rate of some of the quercetin glycosides (and perhaps other glycosides) ofthe flavonoid class (Walgren et al., 2000; Ader et al., 2001). These secondary plant metabolites are good antioxidants and serve as modifiers of proteins in various signalling pathways.

The human kidneys filter approximately 180 g of D-glucose from the plasma each day, and glucose is reabsorbed very efficiently in the proximal tubules. Na+-glucose co-transport across the renal brush border membrane and facilitated diffusion across the basolateral membrane account for recycling ofglucose from primary urine into the circulation. The bulk of glucose is reabsorbed in the convoluted proximal tubule by a low-affinity, high-capacity transporter that might be SGLT-2 (Wright, 2001). Glucose escaping this first reabsorption step is then taken up in the straight proximal tubule by a high-affinity, low-capacity transporter that is most likely to be SGLT-1. Congenital renal defects in glucose reabsorption that cause severe glucosuria appear to be associated mainly with malfunction of the SGLT-2 protein, whereas patients suffering from malfunction of SGLT-1 only have a mild renal glucosuria (Sankarasubbaiyan et al., 2001).

Transport Processes for Nucleosides

Nucleosides are required in the synthesis of DNA and RNA, and as metabolic intermediates; various tissues cannot synthesize nucleosides but depend on an external nucleoside supply. Intertissue exchange of nucleosides requires specific nucleoside transporters. In addition, nucleosides are provided by the diet as the end-products of the gastrointestinal digestion of DNA, RNA and nucleotides. Intestinal epithelial cells express two types of Na+-dependent nucleoside transporters that couple uphill transport of nucleosides with movement of sodium ions down the electrochemical sodium gradient. These two electrogenic systems are the CNT proteins (concentrative nucleoside transporters). CNT1 transports nucleosides of pyrimidine bases, CNT2 those containing purine bases (Pastor-Anglada et al., 2001). The efflux of the nucleosides from the epithelial cells into the circulation is mediated by the equilibra-tive nucleoside transporter (the ENT subgroup). The members of this large family of proteins found throughout the human body are uniporters that operate bidirectionally along the transmembrane gradient for nucleosides (Hyde et al., 2001). Members of the ENT family also possess a certain specificity towards pyrimidine or purine bases (Ritzel et al., 2001). The concentrative Na+-dependent nucleoside transporter proteins have around 650 amino acid residues and most probably contain 13 membrane-spanning domains. Both, ENT and CNT proteins are important not only for metabolism of nucleosides but also for the uptake and delivery of antiviral and cytostatic drugs that structurally resemble nucleosides (Wang et al., 1997).

Monocarboxylate Transporters

The exchange of monocarboxylates among tissues and organs is essential for a variety of metabolic processes including the release of lactate from muscle tissue and other cells followed by its re-uptake into liver for gluconeogenesis (Cori cycle), and the transport of ketone bodies during starvation and ketoacidosis. Other monocarboxy-lates such as pyruvate, butyrate, propionate and acetate or the keto-derivatives of the branched chain amino acids also show significant interorgan fluxes. Figure 3.2 summarizes some of the cellular functions of the identified transporters for the various monocarboxylates that belong to the MCT family. Monocaboxylate transporters MCT-1-MCT-9 have been identified in humans (HUGO: SLC16A1-SLC16A9), with a large number of related genes in other pro- and eukaryotic organisms (Halestrap and Price, 1999). MCT-1 is expressed ubiquitously in mammals. Very strong expression is found in heart and red muscle, where MCT-1 is up-regulated in response to increased work, suggesting its special role in lactic acid delivery for oxidation (Bonen et al., 1997). It is interesting to note that MCT-1 in striated muscle is also found in mitochondrial membranes based on immunostaining, suggesting that it can be targeted to plasma and mitochon-drial membranes (Brooks et al., 1999). MCT-2 is a high-affinity-type transporter expressed in renal proximal tubules, sperm tails and neurons. MCT-3 is expressed uniquely in the retinal pigment epithelium (Halestrap and Price, 1999).

The transporter MCT-4 is most evident in white muscle and other cells with a high rate of glycolysis, including tumour cells and white blood cells, with a particularly high demand for efficient

Glycolysis The Red Blood Cell
Fig. 3.2. The proposed cellular functions of identified MCT carriers and their role in cellular fluxes of monocarboxylates that link a variety of metabolic pathways.

lactic acid efflux (Bonen, 2000). Functional analysis of heterologously expressed MCT-4 as the major transporter isoform present in white skeletal muscle shows Km values for L-lactate, D-lactate and pyruvate of 28, 519 and 153 mM, respectively, characterizing MCT-4 as a low-affinity-type transport system. For MCT-1, affinities range from 2.1 mM for pyruvate, to 4.4 mM for L-lactate and to >60 mM for D-lactate (Juel and Halestrap, 1999). Monocarboxylate transport occurs in most cases electroneutrally by symport ofthe anion with a proton. The kinetics of pyruvate and lactate transport into red blood cells have been characterized thoroughly and show that the characteristics of MCT-1 are consistent with it being the only MCT present in the plasma membrane of red blood cells (Poole and Halestrap, 1993). Transport involves proton binding to the transporter as the initial step, followed by binding oflactate and the translocation of the complex with the sequential release of the solutes into cytoplasm. The transporter can operate in both directions dependent on the concentrations of substrate and co-transported protons; equilibrium is reached when [lactate] in/[lactate]out = [H+]out/[H+]in (Halestrap and Price, 1999).

Regulation of MCT gene expression is an important part of the adaptation of metabolism to various conditions. Advanced physical activity can improve lactate/H+ transport capacity in muscle by up-regulation of MCT-1 expression. In contrast, in denervated rat muscles, the rate of lactate transport is reduced and expression of MCT-1 and MCT-4 is diminished. Hypoxia is known to increase the expression of lactate dehydrogenase-M and other glycolytic enzymes, as well as GLUT-1. This adaptation involves a variety of transcription factors and response elements including the hypoxia-inducible factor 1, cAMP response element and the erythropoietin hypoxic enhancer. Since the distribution of MCT-4 within muscle fibres is similar to that oflactate dehydrogenase-M, hypoxia could also increase MCT-4 expression by similar molecular mechanisms. Up-regulation of the cardiac monocarboxylate transporter MCT-1 in a rat model of congestive heart failure suggests that lactate may be used as an important respiratory substrate for cardiac metabolism, which becomes more dependent on carbohydrates under these disease conditions (Halestrap and Price, 1999).

Another important tissue in which transport of weak organic acids such as butyrate, lactate, propionate and acetate is mediated by MCT family members is the colonic epithelium. Monocarboxy-lates are generated in the colon in huge quantities by the microorganisms that metabolize soluble fibres (i.e. inulin and oligofructoses) and non-digested starch. Luminal concentrations of the short chain fatty acids may reach 100-150 mM, and uptake across the brush border membrane provides the fuel for colonocytes (butyrate) and delivers others to the circulation (propionate, lactate and acetate). In addition to its role as an energy substrate, butyrate promotes cell growth in normal colon but inhibits proliferation in established colon adenomas and cancers. The kinetic properties of butyrate transport into colonocytes are consistent with the properties of MCT-1 and therefore this transporter appears to be very important in tissue homeostasis and control of intestinal adaptation (Stein et al., 2000).

The importance of specialized transporters for short chain fatty acids was discussed controversially in view of the high permeability of cell membranes to the non-charged species of these low molecular weight compounds. pH-dependent partition between the two sides of the membrane with unrestricted diffusion of the non-polar species was considered to be the sole mechanism for transport. Of course, this process represents a background route for permeation ofmonocarboxylates as it is determined by the physicochemical characteristics of the weak acids. However, transport proteins such as MCT-1 and the modulation of the expression level of these carriers provide a level of control for the flux of various organic weak acids into and out of a cell that is mandatory for metabolic homeostasis.

Fatty Acid Transporters

Long chain fatty acids (LCFAs) were also believed to permeate cell membranes without the need for specialized transport proteins as they are highly hydrophobic and insert rapidly into phospholipid bilayers. However, homeostasis demands regulation of membrane transport for a coordinated coupling to metabolism. An increasing volume of evidence suggests that not only the plasma membrane-associated and cytoplasmic fatty acid-binding proteins - known for quite some time -but also specialized transporters for the membrane permeation step are involved in cellular fatty acid uptake. LCFAs not only serve as energy storage substrates and metabolic fuel but are also implicated in the regulation of cell growth and various other cellular functions. Several membrane-associated fatty acid-binding/transport proteins, such as the 43 kDa plasma membrane fatty acid-binding protein (FABPpm), the 88 kDa fatty acid translocase (FAT) and a variety of fatty acid transporter proteins (FATP class), have been identified (Bonen et al., 1998). There are six known proteins of the FATP/SLC27A family that are expressed in tissues and that allow transmembrane movement of LCFAs in mammals (Hirsch et al., 1998). Whereas the saturated fatty acids are very good substrates, polyunsaturated fatty acids are transported very poorly. With the exception of retinoic acid, fat-soluble vitamins are not accepted as substrates. Whereas the coenzyme A (CoA) derivatives of fatty acids are not transported by FATP, overexpression of most FATPs in cells also increases CoA synthase activity 2-5 times, suggesting an intrinsic or associated enzymatic activity of the FATP (Herrmann et al, 2001). The SLC27A family shows a conserved signature motif and a conserved domain for AMP binding, which is essential for transport function. Neither the mechanism nor the exact membrane topology of the FATPs is known yet. However, it may be speculated that LCFAs are co-transported with protons to allow their electroneutral movement into a cell against the inside negative membrane potential.

Expression of FATPs varies with tissue type (Hirsch et al., 1998). Whereas liver expresses various isoforms (designated 2-5) with prominent expression of FATP-5, muscle tissue possesses mainly FATP-1. FATP-2 is found in kidney and FATP-3 predominantly in lung. Heart expresses the FATP-6 isoform. Expression of FATP-4 in the brush border membrane of the intestinal epithelial cells suggests that this protein is involved in intestinal absorption of LCFAs. FAT/CD36 shows a close co-regulation with the expression of FAB. Knock-out mice deficient of CD36 show impaired uptake and use of fatty acids, increased cholesterol levels, decreased body weight, lower than normal plasma glucose and insulin levels, and a reduced metabolic rate (Coburn et al., 2000). In addition, their endurance is reduced. A CD36 deficiency in humans - occurring with an incidence of 0.3-11% in different populations - frequently is associated with defective cardiac fatty acid uptake and hypertension, as well as impaired glucose tolerance. In this context, it should be mentioned that insulin-sensitizing drugs have been shown to regulate CD36 (Yamamoto et al., 1990).

Transmembrane Transport of Amino Acids and Short Chain Peptides

The 20 proteinogenic amino acids and their derivatives are a heterogenous group of compounds that differ in polarity, net charge and molecular mass. They serve as building blocks for protein synthesis, as carriers for nitrogen and carbon units in interorgan metabolism, and as energy substrates, but also as precursors of biologically active compounds such as neurotransmitters. Some are neurotransmitters themselves and others are used for conjugation and excretion of compounds. Amino acid transport proteins mediate and regulate the flow of these nutrients across the plasma membrane for influx and efflux, and thus play a central role in interorgan metabolism. A multitude of membrane transporters with different specificity for the various classes of amino acids mediate transport, due to the different physicochemical characteristics of the amino acids.

The substrate specificity ofthe different transport pathways was determined in pioneering work by Halvor N. Christensen's group in the 1960s mainly based on analysis of transport in erythro-cytes, hepatocytes and fibroblasts (Christensen, 1990). Some general principles were also identified in this early work, such as the stereoselectivity with faster transport of the L-isomers for almost all transport systems and a rather broad substrate specificity of some of the carriers. Some transporters were relatively specific for either acidic or basic amino acids or for amino acids with an aromatic side chain. In addition, differences in the thermo-dynamic properties of the transport steps were observed with identification of equilibrative systems and systems that were ion-dependent and showed uphill transport capability. Figure 3.3 summarizes the different mechanisms and modes of amino acid transporting systems and representative carrier types.

The first cDNA encoding a mammalian amino acid transporter was identified in 1991 as an ecotropic murine retrovirus receptor that turned

Fig. 3.3. Amino acid transport activities of representative members of the various transporter families and their proposed mode of function, participating ions and flux coupling ratios.

out to transport cationic amino acids (Kim et al., 1991; Wang et al., 1991). Preceding this discovery, the first transporter for the neutrotransmitter g-aminobutyric acid (GABA) was cloned (Guastella et al., 1990). In the last few years, a rapidly growing number of transporters have been isolated by means of expression cloning and homology screening. Currently, about 30 cDNAs have been identified (not including splice variants) that encode proteins with different amino acid transport activities. Table 3.2 summarizes the main transport pathways for amino acids in mammalian cells, subdivided into Na+-dependent and Na+-independent pathways and cDNAs that have been cloned that show transport activity when expressed in a target cell.

Transport pathways for zwitterionic amino acids

The zwitterionic amino acid transport systems A, ASC, N and L are present in almost all cell types. Systems A, ASC and N mediate the influx of amino acids with a small side chain by co-transport with sodium ions. System L transporters

Table 3.2. Classified transport activities of the various amino acid-transporting systems found in the plasma membrane of mammalian cells and cDNAs that encode the proteins mediating this activity.

Isolated cDNA(s) encoding this activity

Table 3.2. Classified transport activities of the various amino acid-transporting systems found in the plasma membrane of mammalian cells and cDNAs that encode the proteins mediating this activity.

Na+ dependent

A

ATA1-3

N

SN1-3

GLY

GlyT1-2

ASC

ASC1-2

BETA

GAT1-3, BGT-1

IMINO

Not identified yet

Bo

ATBo

Bo +

ATBo +

X-AG

EAAT1-5

Na+ independent

L

4F2hc/LAT(X)

y+

CAT1-3

bo +

rBAT/bo+AT

PAT

LYAAT-1

operate independently of sodium and mediate the flux of amino acids with bulky side chains (i.e. branched and aromatic groups). System L has been postulated to serve in many tissues as an amino acid efflux system rather than for influx. Systems A and N show a preference for glutamine, alanine and serine; system ASC prefers alanine, serine and cysteine. In contrast, amino acids with a small, non-branched side chain are poor substrates for system L, for which the analogue BCH (2-aminoendobycyclo-2,2,1-heptane-2-carboxy-lic acid) is a model substrate for sodium-independent transport. Zwitterionic amino acids with bulky side chains are, in most cases and in most non-epithelial cells, transported via sodium-independent exchange processes (Palacin et al., 1998). In many epithelia, sodium-dependent transport of alanine can be inhibited by N-methylaminoisobutyric acid (MeAIB), and sodium-dependent transport of MeAIB is used for determining system A-specific transport activity (Reimer et al., 2000). System A is highly pH sensitive and electrogenic, whereas the sodium-dependent system ASC is relatively pH insensitive and operates electroneutrally, suggesting that it may be an antiporter of amino acids associated with the movement of sodium in both directions. An important feature of system A is that its activity in many cell types is highly regulated. This includes an up-regulation during cell cycle progression and cell growth in many cells and tissues, as well as hormonal control by insulin, glucagon, catecholamines, glucocorticoids, various growth factors and mitogens by quite different signalling pathways (McGivan and Pastor-Anglada, 1994). Whereas glucagon and epidermal growth factor induce an immediate increase in system A activity in hepatocytes, insulin up-regulates system A in a gene transcription-dependent manner in hepatocytes and additionally by a rapid pathway -possibly by recruiting preformed transporters to the plasma membrane (also in skeletal muscle). However, insulin deficiency or insulin resistance are also associated with an adaptative up-regulation of system A activity in liver and skeletal muscle, suggesting that the system A transporters are targets of a complex network of regulatory factors for short-term, intermediate and chronic adaptation.

The most interesting amino acid-transporting systems, at least in view oftheir molecular architecture and the pathophysiology of malfunctions, are the heterodimeric amino acid exchangers that represent the y+L and bo,+ transport activities (see Table 3.2), which are found in a variety of cells

(Chillaron et al., 2001; Wagner et al., 2001). The novelty of their structural aspects is that they consist of two separate proteins: one heavy chain (out of two heavy chains that have been identified) and one light chain (out of seven). The two monomers oligomerize via an extracellular disulphate bridge; transport activities and membrane locations of the various complexes depend on the nature of its two subunits. In general terms, the two subunits that form the active amino acid transporters consist of a glycoprotein heavy chain designated as 4F2hc or rBAT that combines with one of the seven light chains (LAT1, LAT2, y+LAT1, y+LAT2, ascATl, xCT and bo,+AT) that have been identified (Chillaron et al., 2001). The heavy chain rBAT is expressed mainly in epithelial cells; association with a light chain is followed by translocation ofthe complex to the apical membrane, resulting in a sodium-independent amino acid exchange process (Palacin etal., 2001a). The other heavy chain 4F2hc associates with various light chains that then mediate amino acid exchange mechanisms in basolateral membranes of epithelial cells as well as in non-epithelial cells. The heavy chains have a glycosidase-like extracellular domain attached to a single TM domain (Verrey et al., 2000). The light chains vary in size, but all possess 12 membrane-spanning domains of a polytopic membrane protein with the N- and C-termini facing the cytoplasm. Figure 3.4 depicts models of how the various light chains associate with either rBAT or 4F2hc to form the heterodimeric complexes that possess different amino acid-transporting selectivities. It appears that the main function of the heavy chain is to serve as an internal carrier that enables the complex to be translocated and integrated into the target cell membrane. The heterodimeric transporters can transport a variety of neutral amino acids in an obligatory exchange mode, which means they mediate influx of certain amino acids at the expense of intracellular amino acids. The bo,+ activity of the rBAT-associated complex can, in addition, transport cationic amino acids in exchange for neutral amino acids, which results in transport currents, whereas the y+LAT1-4F2hc complex exchanges neutral amino acids with sodium co-transport for intracellular cationic amino acids. A variety of mutations in the transporter complexes have been identified that cause malfunctions and severe metabolic disturbances. Defects in bo,+AT cause non-type I cysteinuria with clinical manifestations in the kidney (Palacin et al.,

Diabetes Sustenance

Diabetes Sustenance

Get All The Support And Guidance You Need To Be A Success At Dealing With Diabetes The Healthy Way. This Book Is One Of The Most Valuable Resources In The World When It Comes To Learning How Nutritional Supplements Can Control Sugar Levels.

Get My Free Ebook


Post a comment