SEARCH

SEARCH BY CITATION

Keywords:

  • facilitated substrate transport;
  • transmembrane protein;
  • gut metabolism;
  • in situ localization;
  • yeast expression system

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Analysis of the pea aphid (Acyrthosiphon pisum) genome using signatures specific to the Major Facilitator Superfamily (Pfam Clan CL0015) and the Sugar_tr family (Pfam Family PF00083) has identified 54 genes encoding potential sugar transporters, of which 38 have corresponding ESTs. Twenty-nine genes contain the InterPro IPR003663 hexose transporter signature. The protein encoded by Ap_ST3, the most abundantly expressed sugar transporter gene, was functionally characterized by expression as a recombinant protein. Ap_ST3 acts as a low-affinity uniporter for fructose and glucose that does not depend on Na+ or H+ for activity. Ap_ST3 was expressed at elevated levels in distal gut tissue, consistent with a role in gut sugar transport. The A. pisum genome shows evidence of duplications of sugar transporter genes.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Sugar transport into and out of cells is a vital function for all organisms. It is particularly important for aphids which feed on phloem sap, and rely on sugars as their main source of carbon. The abundant supply of sugars is not wholly advantageous to aphids; sucrose concentrations in plant phloem are routinely high enough to result in a positive osmotic potential between ingested sap and the insect haemolymph, which would result in dehydration of the aphid and death (Karley et al., 2005). Various mechanisms which could enable aphids to deal with the high osmotic potential of their diet have been suggested, including accumulation of trehalose in their haemolymph to balance osmotic potential (Rhodes et al., 1997; Moriwaki et al., 2003), and conversion of ingested sugars to glucose oligomers to decrease osmotic potential (Ashford et al., 2000). For aphids, both utilization of ingested sugars as a nutritional resource, and maintenance of osmotic balance, depend on sugar transport across the gut epithelium, and into cells throughout the insect. In virtually all organisms studied so far, sugar transport across biological membranes is mediated by proteins belonging to one of two types: the sodium/solute symporter family (SSF; Pfam PF00474; Pfam protein family database http://pfam.janelia.org) and the major facilitator superfamily (MFS; Pfam CL0015). Sugar transporters in aphids are anticipated to have features in common with those of other organisms for physiological roles which are shared, but may also be expected to show some unique features that reflect the high sugar content of the aphid diet.

Proteins belonging to the MFS superfamily are found in all organisms, and are the most abundant type of small molecule transporter in animals (Pao et al., 1998). These proteins are integral to membranes, and mediate transport by facilitated diffusion, either as uniporters where transport occurs down a concentration gradient, or as symporters or antiporters where transport of a substrate against a concentration gradient is linked to transport of an ion down a concentration gradient. Structures of two MFS family transporters are available, and show an arrangement of 12 transmembrane helices connected by loops, with an inner ring of 8 amphipathic helices (1,2,4,5,7,8,10,11) and four hydrophobic outer helices (3,6,9,12) stabilising this arrangement in the membrane (Mueckler & Makepeace, 2006). The N-terminal and C-terminal regions of the proteins are cytoplasmic, and the molecules contain a cytoplasmic loop region between helices 6 and 7, dividing the structure into two domains, corresponding to a gene duplication event which is thought to have given rise to proteins of this type. This topological arrangement is common to MFS proteins in general, and prediction of transmembrane helices for novel sequences specifies their topology, and identifies regions of sequence similarity belonging to specific helices and intra- or extracellular loops.

In mammals, facilitated diffusion of glucose and related hexoses across membranes is mediated by MFS superfamily proteins encoded by the SLC2 family genes (Uldry & Thorens, 2004), which are referred to in humans as GLUT1-12, 14 (glucose transporters) and HMIT (or GLUT13; H+-myo-inositol co-transporter). The ‘glucose transporters’ are uniporters which show varying levels of specificity towards glucose or other hexoses (Manolescu et al., 2007), with ‘class I’ transporters favouring glucose over fructose, whereas ‘class II’ transporters favour fructose over glucose. However, these transporters, with the exception of HMIT, do not show absolute specificity for individual sugars, and galactose, glucosamine and/or 2-deoxyglucose have been shown to be substrates for different individual transporters. Insect sugar transporters have been little studied, beyond annotation of genes with similar sequences to mammalian sugar transporters. Examples of direct functional analysis of specific insect transporters are limited to a low-affinity glucose transporter in the hemipteran rice brown planthopper (Nilaparvata lugens; Price et al., 2007a) and a transporter for the disaccharide trehalose in the anhydrobiotic insect, Polypedilum vanderplanki (Kikawada et al., 2007). The latter is of particular interest because its specificity for a disaccharide was not predicted from mammalian models.

The substrates of transporters in family MFS cannot be predicted with confidence from protein sequence. One major problem is that although novel MFS family proteins can be fitted readily to the transmembrane topological model for proteins of the clan, the presence of transmembrane helices does not establish the specific transport function, since structure–function relationships in these proteins are not well-characterized. Mutation studies used to identify residues important for substrate binding and/or transport in the human glucose transporter GLUT1 show that the transport channel is formed from the hydrophilic faces of the inner helices, and residues from all the inner helices apparently contribute to substrate binding (Mueckler & Makepeace, 2004, 2006). An extensive network of hydrogen bonds is involved in the transport process, along with a change in conformation of the transporter (Pascual et al., 2008), and thus a large number of residues affect transport if mutated. A motif in transmembrane helix 7, QLS, has been suggested to be involved in substrate recognition in mammalian GLUT family transporters (Manolescu et al., 2007), with evidence from mutational analysis suggesting that this motif confers selectivity towards glucose over fructose. However, detailed functional characterization has only been carried out for a limited number of MFS sugar transporters, and the results obtained with human glucose transporters are not necessarily valid for insect transporters.

The recent availability of the complete genome sequence for pea aphid provides the opportunity to screen for genes encoding proteins of the MFS superfamily. This paper reports the detection of 54 genes predicted to encode transporters of sugars and related molecules, identifies a sub-family of transporter genes containing a sugar transporter signature, and describes functional characterization of the most abundantly expressed sugar transporter gene, Ap_ST3. The protein encoded by Ap_ST3 is a low-affinity uniporter, which shows selectivity for fructose over glucose as a substrate. Ap_ST3 is expressed in all the major organs of the aphid, and is selectively upregulated in midgut tissue.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Identification of a sub-family of genes encoding sugar transporters in the genome of Acyrthosiphon pisum

The initial release of the pea aphid (Acyrthosiphon pisum) genome (v. 1.0), covering approx. 90% of the total sequence, includes a set of predicted proteins designated ‘reference sequences’, for which supporting evidence in the form of complete predicted open reading frames (ORFs), gene predictions by sequence similarity, and expressed sequence tag (EST) correspondence is available. It also contains further predicted protein sequences for which evidence of complete ORFs is not present. All predicted proteins belonging to the MFS superfamily (Pfam clan CL0015) were identified by searching for signatures diagnostic for individual protein families within this clan. Results are summarized in Table S1. This analysis gave a minimum of 204 genes encoding MFS superfamily proteins. Of these, 77 (including 69 reference sequences) contained the Sugar_tr (PF00083) signature, indicating that they could potentially encode sugar transporters. These sequences were selected for further analysis.

To obtain a non-redundant gene set with accurate predicted protein sequence, genomic scaffold sequences, EST sequences and intron boundary predictions were compared manually. This process identified two additional genes encoding potential sugar transporters and a number of genes in the family MFS_1 set that encoded proteins very similar to those encoded by Sugar_tr members. These were included for further analysis. The predicted proteins were classified into sub-families based on the more discriminating signatures utilized by the InterPro database, and sequence similarity. They were also compared to sequences in Drosophila melanogaster, to confirm the classification into sub-families. Results are given in Table S2. Four sub-families of genes were identified: 29 ‘true’ sugar transporters, encoding proteins containing the IPR003663 signature diagnostic for hexose transport (of which two also contained the IPR000803 glucose transporter signature); 25 potential sugar transporters, which encoded proteins containing either the IPR005828 or IPR011701 signatures, but not IPR003663; 14 organic cation transporters, which encoded proteins showing high levels of similarity to organic cation transporters in D. melanogaster and other insects; and 9 synaptic vesicle proteins, which encoded proteins showing high levels of similarity to synaptic vesicle proteins in D. melanogaster and other insects. Both ‘true’ sugar transporters and potential sugar transporters were similar to genes annotated as sugar transporters in D. melanogaster and other insects. Of the ‘true’ sugar transporter genes, 21/29 had corresponding ESTs; and for potential sugar transporter genes, 17/25 had corresponding ESTs. These genes were given temporary designations as numbered series in the form Ap_STxx.

The predicted amino acid sequences of all the A. pisum candidate sugar transporters were also compared by a Clustal analysis (result not shown). This analysis was largely consistent with the grouping based on signatures. The ‘lowest’ branch of the Clustal tree divided the Ap_ST sequences into two groupings, with most of the ‘true’ sugar transporters containing the IPR003663 signature in one grouping and most of the potential sugar transporters, not containing the IPR003663 signature, in the other, and a consistent separation of the ‘true’ sugar transporters and the potential sugar transporters into sub-groups at a low level of the tree. The only example where the signature groupings and the Clustal groupings were not wholly consistent was the subgroup of sugar transporters which contained both sequences lacking the IPR003663 signature (Ap_ST34, 36, 38, 39, 40, 46) and sequences containing the IPR003663 signature (Ap_ST21, 28, 35). The predicted organic cation transporters and synaptic vesicle proteins were assigned to different groups from all the sugar transporters by Clustal analysis.

Acyrthosiphon pisum sugar transporter gene; gene structure and gene clustering

Most A. pisum sugar transporter genes have a multi-exon structure, with 7–9 exons being typical. No global evidence for exon boundaries corresponding with functional units in the proteins, such as trans-membrane helices, was observed. Intron size was highly variable, although introns near the 5′ end of the coding sequence and in the 5′ untranslated region (UTR) were generally longer than others. There was unambiguous evidence from ESTs of alternate splicing at the 5′ ends of many well-expressed genes, and this affected the predicted sequence of the N-terminal cytoplasmic region of the protein in some cases. No introns were present in any 3′ UTR. Four genes had no introns in the coding sequences, and these were not represented by ESTs. The structure of Ap_ST3, the gene whose product is functionally characterized in this paper, is shown in Fig. 1A; this gene has 7 exons with alternative splicing giving two different exon 1 sequences in the 5′ UTR.

image

Figure 1. (A) Overall structure of Acyrthosiphon pisum Ap_ST3 gene. Ap_ST3 contains 7 exons; coding sequence is indicated by black bars, 5′ and 3′ untranslated regions (UTRs) are indicated by white boxes. The alternatively spliced 5′ UTR exons are shown. The 3′ UTR contains a single poly (A) signal sequence (AATAAA). (B) Predicted membrane topology model of Ap_ST3, with C-terminal ala-ala-ala linker region and c-myc-epitope from yeast expression construct (shaded black squares). Transmembrane a-helices (predicted by TMHMM v2.0) are numbered M1-12. Residues conserved across all members of MFS (region a) and conserved residues of functional significance, as determined by mutational analysis in mammalian GLUT1 (regions b–e) are highlighted as shaded black circles. Region (a) contains a GRR/K motif in a position characteristic of members of MFS. Regions (b) and (c) contain highly conserved glutamine which form part of the exofacial ligand binding site (Hashiramoto et al., 1992; Mueckler et al., 1994) and an extended ‘QLS’ motif which is present in mammalian glucose transporters (GLUT 1, 3, 4) and absent from glucose/fructose transporters (GLUT1, 2, 5) (Arbuckle et al., 1996; Seatter et al., 1998). Region (d) contains a proline residue that is necessary for conformational flexibility, allowing the ligand binding site to switch between outward- or inward-facing orientations (Tamori et al., 1994), and a tryptophan residue that is involved in binding to cytochalasin B (Garcia et al., 1992). Region (e) contains a tryptophan residue essential for transport activity (Garcia et al., 1991).

Download figure to PowerPoint

Clusters of sugar transporter genes are a conspicuous feature of the A. pisum genome. The extent of clustering cannot be fully determined due to the incomplete assembly of the genome sequence, but in terms of the existing scaffolds, 18/53 genes were present on scaffolds containing at least one other sugar transporter gene. The most dramatic example of clustering is shown by genomic scaffold 2518 (EQ113290), which contains 7 ‘true’ sugar transporter genes as a contiguous cluster in 270 kbp of sequence (see main genome paper). Five of these genes are expressed as judged by corresponding ESTs. Functional analysis of one of these genes, Ap_ST3, is considered later in this paper. Clusters of other MFS superfamily genes, for example those encoding synaptic vesicle proteins, are also present (data not shown).

Comparison of sugar transporter genes between Acyrthosiphon pisum and other organisms

A survey of sugar transporter genes from a selection of organisms with completely sequenced genomes (see Table S3) showed that A. pisum contains more genes encoding proteins with the IPR003663 signature than any other insect species except the flour beetle Tribolium confusum, and significantly more genes of this type than mammals or the nematode C. elegans. To examine phylogenetic relationships between predicted sugar transporters in different insects, which would indicate similar functionality, sequences containing the IPR003663 signature were subjected to a Clustal sequence comparison supplemented by the Bayesian inference method to produce a composite tree. The dataset used contained all the predicted genes from A. pisum, D. melanogaster and Apis mellifera (honey bee), along with further genes from other insects. The Bayesian tree is shown in Fig. 2; a similar tree was obtained when the Clustal comparison was supplemented by further analysis using the Maximum Likelihood method (Fig. S1).

image

Figure 2. Sequence similarity tree (Bayesian Inference method) of sugar transporters containing the IPR003663 signature in Acyrthosiphon pisum (Blue), Drosophila melanogaster (Red), Apis mellifera (Black) and other insects (Purple). For each branch, the posterior probability is shown. The scale bar indicates an evolutionary distance of 0.4 amino acid substitutions per position in the sequence.

Download figure to PowerPoint

This analysis showed that the characterized trehalose transporter from P. vanderplanki (Kikawada et al., 2007) was located in a distinct subgroup of sequences containing Ap_ST11; this also contained the D. melanogaster gene products CG8234 and CG30035 (annotated as trehalose transporters), and transporters from bee, mosquito spp. and locust. Functional analysis will be required to confirm that Ap_ST11 is also a trehalose transporter. A simple Clustal analysis suggested that the two aphid sugar transporters containing the IPR000803 signature characteristic of glucose transporters, Ap_ST4 and Ap_ST43, grouped together with a D. melanogaster gene product, CG1086, annotated as a glucose transporter on this basis (result not shown). The Bayesian Inference and Maximum Likelihood trees did not group these gene products together, although they agreed in separating them from other transporters, except for sequences from the bee and the wasp Nasonia vitripennis. The characterized low-affinity glucose transporter from N. lugens grouped tightly with three pea aphid sequences (Ap_ST1, 16, 17) suggesting that these might be true orthologues (both species are hemipterans). The four sequences grouped more loosely with sequences from D. melanogaster and bee. The transporter selected for functional characterization in this paper, Ap_ST3, grouped together with potential orthologues from bee, wasp and D. melanogaster (CG1213).

The more general features of the sequence tree were a distinct trend for aphid sequences to have orthologues in bee, but not in D. melanogaster (Ap_ST4, 5, 6, 9, 13, 21, 23, 28, 30, 35, 41), although the (non-expressed) Ap_ST24/25/26 group were more similar to D. melanogaster CG1208 than other sequences. Although the major branches of the tree contained sequences from all three organisms where a complete survey of the genome had been carried out, there was a distinct tendency for ‘clusters’ of very similar genes from the same organism to be present, particularly in the case of pea aphid (clusters contained Ap_ST21/28/35, 9/23, 24/25/26, 1/16/17, 2/10/19). The ‘clustering’ suggests that gene duplication has taken place in the aphid after species divergence.

A comparison between potential sugar transporters in A. pisum and D. melanogaster was also carried out by manual analysis, which allowed all potential sugar transporter genes to be identified. Drosophila is the only insect genome in which the full gene complement has been fitted to a physical map, and thus accurate gene predictions can be made. This analysis confirmed that the dipteran species contained fewer sugar transporter genes than A. pisum, with 17 ‘true’ sugar transporter genes containing sequence signature IPR003663 (28 in aphid), and 16 additional genes encoding sugar transporter-like proteins which lacked the IPR003663 signature (25 in aphid). Most (>80%) of the Drosophila genes are expressed, as shown by the presence of corresponding ESTs.

Expression of Ap_ST3 in Acyrthosiphon pisum

The global sequence databases contain 32 ESTs corresponding in sequence to Ap_ST3, where 5′ and 3′ reads derived from a single clone are counted as one EST. This sugar transporter gene has the largest number of corresponding ESTs of any sugar transporter gene detected in the genome analysis, and is represented in cDNA libraries derived from whole insects, guts and other tissues. The putative D. melanogaster orthologue of this gene, CG1213, is expressed at very high levels in the adult salivary gland, at high levels in crop, midgut and spermatheca, and at lower levels in other adult tissues, but is only expressed at low levels in larval stages (http://www.flyatlas.org; Chintapalli et al., 2007). Tissue-specificity of expression of Ap_ST3 was investigated by real time PCR, carried out on cDNA prepared from RNA templates isolated from whole insects, embryos dissected from adults, heads, guts and bacteriocytes. Results are shown in Fig. 3A. Compared to the level in whole insects, Ap_ST3 mRNA was present at the same level in embryos; it was downregulated approx. 3-fold in heads, but upregulated over 2-fold in gut and 1.4-fold in bacteriocyte tissue.

image

Figure 3. (A) Quantitative analysis of Ap_ST3 mRNA in whole mature adult insects and head, embryos, gut and bacteriocytes dissected from mature adult insects. Ap_ST3 expression levels were quantified using comparative Ct methodology, with GAPDH as endogenous amplification control. Each bar is mean ± 95% confidence intervals, n= 3. (B) In situ localization of Ap_ST3 mRNA in dissected Acyrthosiphon pisum gut tissue. Guts were probed with either antisense or sense DIG-labelled Ap_ST3 RNA probes and probe hybridization was detected colorimetrically. S, stomach (anterior or proximal end of gut); P, posterior or distal end of gut. Scale bar = 10 micrometers. (C) Quantitative analysis of Ap_ST3 mRNA levels in aphids reared on artificial diets containing 250 mM, 500 mM and 1000 mM sucrose and aphids reared on a diet containing 500 mM sucrose for 2, 4 and 6 days.

Download figure to PowerPoint

The presence and localization of Ap_ST3 expression in the aphid gut was confirmed by in situ hybridization, carried out on fixed dissected guts. Results are shown in Fig. 3B. Staining due to hybridization of the labelled anti-sense RNA probe was observed only in the posterior part of the mid-gut, with the strongest staining at the distal end. No staining was observed in this region with the negative control sense probe. Both sense and anti-sense probes gave very low levels of non-specific staining in the ‘stomach’ region in some guts. The expression pattern for Ap_ST3 is consistent with a role for its encoded protein in transporting products of sucrose hydrolysis out of the gut lumen, since hydrolysis products will be at higher concentrations towards the distal end of the gut, allowing transport into the haemolymph down a concentration gradient.

To determine whether Ap_ST3 expression responded to sugar concentration in diet, quantitation by real time PCR was carried out with RNA templates from whole insects exposed to diets containing varying concentrations of sucrose (range 0.25–1 M). Ap_ST3 expression was slightly increased (1.4x) in insects fed the low sucrose concentration (250 mM) compared to the standard diet of 500 mM sucrose, but unaffected in insects fed the high sucrose concentration. In addition, Ap_ST3 levels were assayed in insects fed on standard diet at different stages of development, from 2 to day-old-larvae to 8-day-old teneral adults. A consistent but small increase in Ap_ST3 levels over this period (approx. 1.5x) was observed (Fig. 3C).

Functional analysis of the Acyrthosiphon pisum sugar transporter gene Ap_ST3

Comparison of genomic scaffold sequence with the sequences of ESTs corresponding to the Ap_ST3 gene shows that alternate splicing is predicted to occur at the 5′ end of the Ap_ST3 mRNA. 5′-RACE PCR confirmed this prediction in that two products with different 5′ exons were observed. However, the alternate splicing is in the predicted 5′ UTR, and does not affect the sequence containing the ORF. Consequently, a single complete coding sequence for the protein encoded by Ap_ST3 was amplified from cDNA by PCR using primers specific for the N-terminal and C-terminal ends of the predicted coding sequence. The product was identical in sequence to the prediction from the RefSeq, XM_001950662.

The Ap_ST3 coding sequence was introduced into the yeast shuttle vector pDR195 immediately downstream of the pma1 (plasma membrane ATPase 1) constitutive promoter. To provide optimal levels of protein, expression the 5′ end of the Ap_ST3 coding sequence was modified to provide a Kozak initiation sequence (Kozak, 1991). To allow protein detection, the 3′ end of the coding sequence was modified to encode a C-terminal Ala-Ala-Ala linker region and a c-myc epitope tag (Fig. 1B). The molecular weight of the modified polypeptide is predicted to be 59 kDa. The resulting recombinant plasmid (Ap_ST3-pDR195) was transformed into the Saccharomyces cerevisiae hexose transport mutant EBY.VW4000, in which genes encoding all known hexose transporters have been deleted. Yeast mutant transformed with pDR195-empty vector was used as a negative control. Transformants were grown on maltose-containing plates, and were selected by uracil auxotrophy and screened by colony PCR. Western blotting of protein extracts from membrane fractions of selected transformants, using anti-myc antibodies as a probe, showed that Ap_ST3 protein accumulated in cells transformed with Ap_ST3-pDR195 as a polypeptide of mol. wt. 47 kDa (Fig. 4A). The lower molecular weight determined for this protein on SDS-PAGE compared to the predicted value (47 vs. 59 kDa) is a result of the hydrophobic nature of the protein, and is in agreement with results obtained for other transmembrane transporters (Ward et al., 2000). Immunofluorescence was used to confirm expression of the aphid sugar transporter; fixed yeast cells were incubated with anti-myc antibodies and labelled secondary antibodies. In the resulting cells, fluorescence was observed in all regions of the plasma membrane, with little fluoresence elsewhere in the cells, showing that the expressed Ap_ST3 had been efficiently inserted in the cell membrane (Fig. 4B). Control cells gave only background fluorescence with no localization.

image

Figure 4. (A) Western blot of yeast total membranes (5 µg membrane protein/lane) from Saccharomyces cerevisiae strain EBY.VW4000 transformed with empty pDR195 vector (EBY-vec cells, lane 1) and pDR195 + Ap_ST3 expression construct (EBY-Ap_ST3 cells, lane 2). Membrane proteins were resolved on a 12.5% polyacrylamide gel transferred to nitrocellulose and probed with anti-myc antibody. (B) Subcellular localization of Ap_ST3 in EBY-Ap_ST3 cells (Ap_ST3 panel) and EBY-vec cells (vec panel); yeast protoplasts were probed with anti-myc primary antibodies and fluorescent-labelled secondary antibodies (green). Cell nuclei are stained with DAPI (blue). Scale bar = 2 µm.

Download figure to PowerPoint

To perform an initial functional assay, cells were plated on minimal media containing hexoses as the sole sugar source. Transformation with Ap_ST3-pDR195 rescued the hexose transport mutant EBY.VW4000, and either fully or partially restored growth on minimal media plates with 60 mM fructose, galactose, glucose and mannose, presumably as a result of transport of the hexoses into yeast cells. As expected, no recovery of growth was observed in EBY.VW4000 cells transformed with pDR195-empty vector (negative control) (Fig. 5). Growth of EBY.VW4000 cells transformed with Ap_ST3-pDR195 was best on media containing fructose or galactose, less good on media containing glucose, and poor on media containing mannose. No growth was observed on plates containing arabinose, xylose, sorbitol, or mannitol (data not shown). Rescue of growth is limited by the ability of the yeast to utilize the transported sugar as a carbon source, and therefore transport specificity cannot be unambiguously determined by these assays.

image

Figure 5. Functional analysis of Saccharomyces cerevisiae hexose transport mutant (EBY.VW4000) expressing the Ap_ST3 (EBY-Ap_ST3 cells) or cells transformed with empty vector (EBY-vec cells). Plates show growth of EBY-Ap_ST3 and EBY-vec cells on minimal media (SC media without uracil) supplemented with 60 mM hexose sugars, as indicated. 1, 1/5, 1/25, 1/125 indicate dilutions of yeast cell suspension (applied to plate in 10 microlitres droplets) from OD660 nm = 1.0.

Download figure to PowerPoint

For more detailed analysis of transport specificity, and determination of kinetic parameters, uptake of 14C glucose and 14C fructose by yeast cells was analysed under zero-trans conditions (i.e. where all transport was assumed to be into the cells) using rapid filtration. Assays were carried out over an interval in which uptake was linear, and corrected for uptake into cells transformed with an empty vector control (<10% of experimental values). Assays carried out over a range of substrate concentrations showed that transport by Ap_ST3 could be saturated, giving hyperbolic curves of rate against glucose or fructose concentrations (Fig. 6A, C). Analysis by Michaelis-Menten kinetics gave estimates for Km of 66 ± 12 mM for glucose and 47 ± 6 mM for fructose. Varying the extracellular pH from 5.5 to 8 led to only a small increase in glucose uptake (approx. 10%) by cells transformed with pDR195-Ap_ST3 (Fig. 7A), suggesting that transport was not dependent on a pH gradient across the yeast plasma membrane. In confirmation, pre-treating cells with the proton gradient uncouplers CCCP and DNP decreased transport by <25% and <10%, respectively (Fig. 7B), whereas a proton-coupled transporter would have shown high levels of inhibition under these conditions. Addition of NaCl to the transport buffer had no significant effect on transport activity (Fig. 7B), suggesting that the transport process was not dependent on a concentration gradient in Na+. Ap_ST3 appears to behave as a true uniporter in mediating facilitated transport, with no requirement for a gradient of pH or Na+ across the plasma membrane.

image

Figure 6. Kinetic analysis and sugar specificity of Ap_ST3 in intact EBY.VW4000 yeast cells for 14C-glucose and 14C-fructose. (A) Glucose uptake analysis Ap_ST3 in intact yeast, EBY-Ap_ST3 cells were incubated with indicated concentrations of 14C glucose for 2 h at 30 °C and uptake determined by liquid scintillation counting. Uptake data were fitted to Michaelis-Menten equations and Km was determined by nonlinear regression using PRISM software. (B) 14C-Glucose uptake (50 mM outside cell concentration) in the presence of 100 mM potential competitors (as indicated). (C) Fructose uptake analysis Ap_ST3 in intact yeast, EBY-Ap_ST3 cells were incubated with indicated concentrations of 14C fructose for 10 min at 30 °C and uptake determined by liquid scintillation counting. (D) 14C-fructose uptake (50 mM outside cell concentration) in the presence of 100 mM potential competitors (as indicated). Each value is the mean ± SEM, n= 3.

Download figure to PowerPoint

image

Figure 7. (A) pH dependence of Ap_ST3; 14C fructose uptake in EBY.VW4000 using indicated extracellular pH. (B) NaCl dependence of Ap_ST3; uptake was performed in the presence of either 50 mM NaCl (+NaCl) or in a buffer where NaCl was replaced with 50 mM choline chloride (−NaCl). Inhibitors of membrane potential CCCP (Carbonyl cyanide 3-chlorophenylhydrazone) and DNP (2,4-Dinitrophenol) were preincubated with EBY-Ap_ST3 cells. All experiments were performed at 30 °C and cells were collected by rapid-filtration after 10 min of uptake. Each value is the mean ± SEM, n= 3.

Download figure to PowerPoint

The specificity of the transporter for a variety of sugars was determined by competition experiments, in which transport of labelled glucose or fructose at a concentration similar to the determined Km value (50 mM) was compared in the presence and absence of a 2-fold molar excess of unlabelled sugar. Glucose and fructose, used as a positive controls, decreased transport of label by 45 ± 2%, and 51 ± 0.1%, respectively, in good agreement with the theoretical values calculated from the determined Km values (due to the increase in transport at higher sugar concentration being offset by the lower specific activity). Under the conditions used, fructose inhibited glucose transport to a greater extent than any other sugar tested, decreasing measured glucose uptake by 64 ± 2%, while mannose gave a decrease of 30 ± 3%. Trehalose and galactose did not decrease (<5%) glucose transport (Fig. 6B). No sugar was as effective as fructose itself in competing with fructose transport, with only glucose (28 ± 0.3% decrease), xylose (27 ± 0.4% decrease), mannose (17 ± 1% decrease), sucrose (14 ± 2% decrease) and myo-inositol (11 ± 2% decrease) giving >10% decrease in fructose transport (Fig. 6D). Other sugars tested (arabinose, trehalose, mannitol, galactose and sorbitol) all gave <10% decrease in fructose transport. The results indicate a specificity for transport of fructose over glucose in this protein, with mannose and xylose also accepted as substrates.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The analysis of the available genomic sequence for A. pisum has identified a set of genes encoding potential sugar transporters of the MFS superfamily, most of which have been validated by corresponding ESTs. The availability of an extensive EST database (>120 000 sequences as of mid-2008) for this species has allowed gene identification to be made secure, and has identified additional genes not detected by the gene prediction algorithms due to gaps in the scaffold sequences. It is unlikely that any substantially expressed sugar transporter genes have been missed by the analyses, although genes that are expressed at low levels, or under conditions not represented in the EST libraries (Zhou et al., 2006), may have been omitted if they have not been recognized by the gene prediction algorithms. The existing genome assembly also contains several predicted genes which appear to be partial duplicates of others, and which have been excluded from the analysis presented here. The status of these genes will need clarification; at least some of them may be true duplications in the genome. However, version 1.0 of the A. pisum genome contains unresolved sequence redundancy between different scaffolds, and the status of these duplicated genes remains uncertain. Full assembly of the genome will allow an accurate estimate for the number of potential sugar transporter genes to be made, and will clarify possible allelic variation.

For reasons discussed in the Introduction to this paper, annotation of genes encoding proteins in the Major Facilitator Superfamily requires manual analysis of sequence comparisons. The set of proteins which are predicted to belong to the Pfam Sugar_tr family on the basis of comparison by software algorithms includes proteins which can be assigned other functions by additional manual annotation, on the basis of a high level of sequence similarity to proteins whose function has been characterized (e.g. organic cation transporters, synaptic vesicle proteins). The remaining predicted Sugar_tr proteins may also show transport activity towards substrates other than simple sugars. Analysis of sequences for diagnostic signatures from the Interpro database cannot be used as a certain predictor of function for proteins of this family; for example, the sequence of the trehalose transporter from P. vanderplanki contains the IPR003663 signature, although it is not a hexose transporter, but a disaccharide transporter (Kikawada et al., 2007). Nevertheless, the presence or absence of the IPR003663 signature does correlate loosely with overall sequence similarity in the A. pisum sugar transporters, and is a useful first indication of potential function when designing assays. The Ap_ST3 transporter functionally characterized in this paper is correctly predicted to be a sugar transporter on this basis. The potential usefulness of more specific signatures, such as the IPR000803 glucose transporter signature, remains to be evaluated; preliminary data suggest that Ap_ST4, which contains this signature, is a hexose transporter active in transporting several different sugars (data not presented).

Annotation of the A. pisum sugar transporter genes will be greatly aided by functional characterization of more sugar transporter proteins; either from the aphid itself, or from other organisms. Where sufficient sequence similarity is present between aphid transporters and characterized transporters in related organisms, as in the case of Ap_ST11 and the P. vanderplanki trehalose transporter, then a well-founded prediction of function can be made. Where species are more distant, and sequence similarity is weaker, the prediction becomes insecure; for example, sequence comparisons between the well-characterized human GLUT transporters and the A. pisum sugar transporters did not identify functional equivalents, since differences in predicted protein sequences between species are sufficient to obscure similarities based on functional activity. (All the aphid transporters grouped as being most similar to two human sugar transporters, GLUT6/8; result not shown.)

Expression of Ap_ST3 as a recombinant protein in yeast results in efficient targeting to the plasma membrane, with little protein retained in intracellular compartments, suggesting that that this protein localizes to the cell surface in insect gut cells. Retention of proteins within cells has been observed for some human glucose transporters (e.g. GLUT4) and is a mechanism to regulate sugar uptake, since the transporter can be relocated to the plasma membrane following insulin signalling (Uldry & Thorens, 2004). The functional analysis of Ap_ST3 shows that this protein is a uniporter able to transport hexose sugars down a concentration gradient, with a preference for fructose and glucose. The binding affinity of the protein for its substrates is low in comparison with human GLUT transporters, which have Km values approx. 10-fold lower for their preferred substrates (Uldry & Thorens, 2004). Ap_ST3 can also transport other sugars, but binds them less well than glucose or fructose, as shown by the competition assay. The low level of inhibition of glucose and fructose uptake in the presence of galactose in this assay is apparently inconsistent with the growth of the yeast EBY.VW4000 mutant transformed with pDR195-Ap_ST3 on galactose-containing media. However, growth of the yeast cells on galactose is concentration dependent between 50 mM and 500 mM sugar, with better growth at higher sugar concentration, whereas growth on glucose, fructose or mannose is not concentration dependent (result not presented). This indicates that the Km value for galactose transport is high, and whereas this sugar can be utilized efficiently once transported to support growth, it is not transported well. Consequently, it does not compete effectively in the short term uptake assay. Complete characterization of the sugar transport capacity of Ap_ST3 by complementation assays in this mutant is not possible, since the assay cannot be used to assess transport of some physiologically relevant compounds such as polyol sugars which are not readily metabolized (myo-inositol, sorbitol, mannitol), or disaccharides (trehalose and sucrose). The preference for fructose shown by Ap_ST3 is not consistent with the presence of a QLS motif in transmembrane helix 7 of the protein, since this motif has been suggested to confer a preference for glucose over fructose in mammalian transporters (Manolescu et al., 2007). The difficulty in defining clear structure–function relationships in MFS sugar transporters is well illustrated by this discrepancy. The P. vanderplanki trehalose transporter also contains this motif, although in this case a glycoside of glucose is transported.

The high levels of expression of Ap_ST3 throughout the aphid, with elevated levels in gut tissue, and the lack of change in expression level in response to diet or development, suggests that Ap_ST3 is the major constitutive facilitated transporter in this species. The putative D. melanogaster orthologue, CG1213, is also expressed at a relatively high level compared to other sugar transporters in the fly (although less than CG10960 and CG30035; http://www.flyatlas.org), but has a different tissue-specificity in expression, suggesting differences in physiological roles. The elevated levels of expression of Ap_ST3 in aphid gut tissue compared to whole insects is consistent with it playing a major role in transport of the sugars formed by sucrose hydrolysis, and to be important in both nutrition and maintenance of osmotic balance. In this context, the relatively high Km values for the preferred substrates of Ap_ST3 may not be disadvantageous, since gut sucrose concentration is an order of magnitude greater than the Km value, and hydrolysis could readily generate glucose and fructose concentrations which exceed the Km values. Sucrose is localized to the entire midgut region in pea aphid, with higher levels of expression towards the anterior end of the gut (Price et al., 2007b), so that glucose and fructose produced by sucrose hydrolysis accumulate towards the distal end of the midgut, where Ap_ST3 is expressed. Fructose is preferred over glucose for transport from the aphid gut to the haemolymph (Ashford et al., 2000), and this is in agreement with the relative Km values of Ap_ST3 for the two sugars. Metabolism, and conversion of the transported sugars to trehalose will reduce concentrations of glucose and fructose in the haemolymph, and thus a concentration gradient from gut to haemolymph can easily be maintained to allow transport in the desired direction. This is different from the situation in mammals, where sugar transport into cells from the blood can occur down a concentration gradient, but sugar uptake into gut cells from the gut lumen takes place against a concentration gradient. This initial step in sugar transport must involve active transport, and is mediated by a Na ± linked symporter (Wright & Loo, 2000). Glucose is accumulated in gut cells and is subsequently exported down a concentration gradient into the circulatory system by a uniporter (GLUT2). It is not known whether glucose and fructose transport against a concentration gradient occurs in aphid guts (or other tissues), and if so, what proteins mediate it. The aphid genome contains 5 predicted genes encoding proteins of family SSF (Pfam PF00474), which are sodium-solute cotransporters of the type used by mammals; however, sequence similarities clearly suggest that these proteins are all transporters of short-chain fatty acids, lactate or choline, and not sugars (data not presented). The presence and nature of active sugar transport in aphids thus remains to be characterized, but a mechanism similar to that operating in mammals is unlikely.

The A. pisum genome contains a relatively large number of genes encoding predicted sugar transporters; the only comparable insect is the flour weevil Tribolium confusum. T. confusum is a storage pest which survives on a diet with very low water levels, and thus, similar to A. pisum, has to deal with high sugar concentrations in the gut. A hypothesis can be advanced that the large number of sugar transporters in aphids results from relatively recent gene duplications as an adaptive response to diet composition. While such a hypothesis is necessarily speculative, it may contribute to developing models for how insects are able to adapt to survive on extreme dietary compositions, and in turn lead to new strategies for producing insecticidal compounds.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Bioinformatics

The 1.0 release of the A. pisum, pea aphid, genome was used as a basis for the bioinformatic analysis, supplemented by the pea aphid EST collection. GLEAN4 predicted genes were searched for coding sequences containing Interpro IPR003663 sugar/inositol transporter signature using HMMER 2.3.2 (Eddy, 1998); http://hmmer.janelia.org). Gene predictions were verified by comparison with ESTs manually, and by assembly of revised gene sequences using Sequencher software (v. 4.2, Genecodes Software). Sequences for genes not predicted by automated methods were completed using genomic scaffold sequences, on the basis of EST sequences. Predicted protein sequences belonging to family Sugar_tr and others (Pfam family PF00083) were analysed individually for the presence of signatures from the Interpro database (Zdobnov & Apweiler, 2001). They were also compared individually with proteins predicted by the D. melanogaster genome database, and with the global protein database, using BlastP software (Altschul et al., 1990). This analysis identified two groups of proteins which were not sugar transporters: a group of proteins similar to synaptic vesicle proteins, containing Interpro signature IPR005988, and a group of proteins highly similar to organic cation transporters (Blast score <e-80). The remaining 54 proteins were divided into two groups on the basis of the presence or absence of the IPR003663 hexose transporter signature, and into sub-groups depending on whether they contained the MFS signature IPR005828 or the MFS signature IPR011701.

Phylogenetic analysis

Predicted protein sequences of sugar transporters were aligned using ClustalX (Thompson et al., 1997) and checked manually with BioEdit 7.0.9 (Hall, 1999). To generate the comparisons of sequences containing the IPR003663 signature, N and C termini regions were excluded from phylogenetic analysis because they could not be aligned with confidence. Bayesian inference (BI) and maximum likelihood (ML) analyses were conducted, following selection of the WAG model (Whelan & Goldman, 2001), using the optimal instantaneous rate matrix estimated in ProtTest 1.4 (Abascal et al., 2005).

A model was chosen for BI and ML through ProtTest 1.4 (Abascal et al., 2005). The BI analysis was run in MrBayes 3.1.2 (Ronquist & Huelsenbeck, 2003) with default values of Metropolis-coupled Markov chain Monte Carlo (MCMC). Each run was 10 million generations long, sampled every 1000th step and the first 25% were discarded as burn-in. The ML was performed on RaxML 7.0.3 (Stamatakis, 2006) with the starting tree selected from 100 consensus trees generated through RaxML. Bootstrap support was estimated from 1000 parametric replicates, with probability values added to the most likely consensus tree. Tree figures were generated using Tracer and FigTree software (Rambaut and Drummond, 2007).

Aphid culture

Parthenogenetic females of the A. pisum (Harris) clone LL01 were maintained on pre-flowering Vicia faba cv. “The Sutton” at 18 °C in a long-day regime of 16 h of light and 8 h of dark. For artificial diet feeding trials, 2-day-old parthenogenetic apterous aphids were transferred to sterile diet of formulation A (Prosser & Douglas, 1992) containing 150 mM amino acids and 500 mM sucrose. Aphids were maintained on an artificial diet for 24 h and day-3 aphids were transferred to test diets containing 250 mM, 500 mM and 1000 mM sucrose. At day 5, RNA was isolated from the aphids. Developmental expression of Ap_ST3 was determined by transferring 3-day-old diet-maintained aphids to 500 mM sucrose diets and total RNA was isolated from day 5, day 7 and day 9 aphids. Feeding chambers containing 10 aphids were maintained under the same environmental conditions as cultures on plants. Ap_ST3 transcript abundance was determined by real-time PCR.

RNA isolation and cDNA synthesis

Total RNA was isolated from whole aphids maintained on plants or chemically defined diets using RNeasy mini kit (Qiagen, Valencia, CA, USA). Aphid guts, heads embryos and bacteriocytes were dissected in nuclease-free 0.9% saline and tissues were immediately lysed in TRI Reagent (Sigma, St. Louis, MO, USA) and total RNA was purified according to manufacturer's recommendations. Total RNA from whole insects and dissected tissues was treated with DNaseI and purified using Qiagen RNeasy column. First-strand cDNA was synthesised from 0.5 µg total RNA in a 20 µl reaction using Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA) with 50 µM oligo(dT)20 primer, according to manufacturer's guidelines. First-strand cDNAs were diluted to 100 µl with nuclease free water.

Ap_ST3 transcript relative abundance by real-time PCR

Real-time quantative PCR was performed on aphid cDNA and relative expression of Ap_ST3 was determined using a Step One Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) with comparative Ct methodology. PCR primers were designed using Primer express software for real-time PCR v3.0 (Applied Biosystems) using cDNA sequences for Ap_ST3 (LOC100163094, ACYPI004204, XM_001950662) and GAPDH (LOC100169122, ACYPI009769, XM_001943014 and LOC100168448, ACYPI009148, XM_001943017). Primers for Ap_ST3: Ap_ST3-sybr-fwd, 5′- CGCTGTTGTTTACGGCGTTT -3′ and Ap_ST3-sybr-rev, 5′- CGTGCAATATGTCCTGGATTTC -3′; primers for GAPDH: GAPDH-sybr-fwd, 5′- CAATGGAAACAAGATCAAGGTGTT -3′ and GAPDH-sybr-rev, 5′- ACCAGCAGATCCCCATTGG -3′. Reaction mixtures (20 µl) contained 1× Power SYBR Green PCR Master Mix (Applied Biosystems), 20 µM PCR primers and either 2 µl cDNA template or water (for no-template control reactions). Reactions were run in triplicate. Analysis of amplification profiles and melt-curves was performed using StepOne software v2.0.1 (Applied Biosystems), according to the manufacturer's guidelines. Expression of Ap_ST3 was normalized to GAPDH.

Localization of Ap_ST3 RNA by in situ hybridization

Aphid guts were dissected from plant-reared aphids in nuclease-free 0.9% saline and treated as described in Price et al. (2007a). Sense and antisense Ap_ST3 DIG-11-dUTP labelled RNA probes (700 bp probe, corresponding to 201–900 bp Ap_ST3) were synthesized from Ap_ST3 cDNA in pLitmus28i (NEB) using T7 RNA polymerase (Promega, Madision, WI, USA) and DIG RNA labeling mix (Roche, Indianapolis, IN, USA). Riboprobes were incubated with gut tissues and hybridized riboprobes were visualised with BM purple substrate (Roche). Endogenous phosphatase activity was blocked by adding 5 mM levamisole to the colour substrate.

Ap_ST3 cDNA cloning and constructs for expression in yeast

The entire coding sequence for A. pisum hexose transporter 3 (Ap_ST3, XM_001950662) was amplified from first-strand cDNA with Phusion polymerase (Finnzymes) with primers Ap_ST3-fwd (5′- ATGGAGTTGGAGACGTTGATGG -3′) and Ap_ST3-rev (5′-TCATGCTCTGGTCATTTTGTGG -3′). Amplification reactions were carried out with a Perkin Elmer (Boston, MA, USA) 2400 cycler using an initial hold at 98 °C 30 s, followed by 25 cycles of 98 °C 10 s, 66 °C 10 s, 72 °C 45 s; and after cycling 72 °C for 5 min. Ap_ST3 cDNA was cloned into pJET1.2 (Fermentas Life Sciences, Glen Burnie, MD, USA) and checked by sequencing both DNA strands. Ap_ST3 cDNA for expression in yeast was modified by reamplification using a forward primer Ap_ST3-pDR195-fwd, which has a yeast-optimized Kozak translation initiation sequence (Kozak, 1991) and a XhoI site (5′- ATCTCGAGATAATGGAGTTGGAGACGTTGATGG -3′); and a reverse primer Ap_ST3-pDR195-rev, which encodes a Ala-Ala-Ala linker region (corresponding to a NotI site), a c-myc epitope (EQKLISEEDL), stop codon and a BamHI site (5′- ATGGATCCTCACAGATCCTCTTCTGAGATGAGTTTTTGTTCAGCGGCCGCTGCTCTGGTCATTTTGTGGTTC -3′). Amplification was performed using conditions as described previously except the template DNA was Ap_ST3-pJET1.2 with 10 amplification cycles. The modified Ap_ST3 PCR product was digested with BamHI and XhoI and cloned into corresponding sites in yeast shuttle vector pDR195 (Rentsch et al., 1995) and several independent clones were checked by DNA sequencing.

Complementation of S. cerevisiae hexose transport mutant pDR195-Ap_ST3 and pDR195-empty vector was used to transform S. cerevisiae hxt-null mutant EBY.VW4000 (Wieczorke et al., 1999) according to Gietz & Woods (2002). Transformants were selected on synthetic complete (SC) agar (without uracil) containing 2% maltose instead of glucose. Transformants were screened by colony PCR Ap_ST3-fwd and Ap_ST3-rev, with PCR cycling as described previously. For functional complementation studies EBY.VW4000 transformed with Ap_ST3-pDR195 (EBY-Ap_ST3 cells) and control cells EBY.VW4000 transformed with pDR195 empty vector (EBY-vec cells) were grown overnight in SC (without uracil) containing 2% maltose and washed twice in water and cells were diluted to OD600= 1, 1/5, 1/25 and 1/125. Ten ml of each dilution of cells was plated onto SC agar (without uracil) with 60 mM sugars: D(+) galactose, D(+) glucose, D(+) mannose and β(−) fructose. Growth of cultures was assessed after incubation at 30 °C for 3 days.

Immunohistochemical techniques and protein gel blot analysis

Yeast cells were grown in SC without uracil with 2% maltose and until OD600= 1 and 5 ml of cells were fixed at room temperature for 40 min in 4% (v/v) paraformaldehyde. Cells were harvested by centrifugation and washed twice in sorbitol buffer (1.2 M sorbitol, 50 mM KPO4, pH 7) and cell walls were digested at 37 °C for 45 min by resuspending cells in 500 ml sorbitol buffer containing 0.1% β-mercaptoethanol, 0.02% glusulase (PerkinElmer) and 5 mg ml−1 zymolyase (Zymolase-100T from Arthrobacter luteus; Seikagaku Biobusiness Corporation, Tokyo, Japan) equivalent to 0.5 units ml−1. Spheroplasts were washed in sorbitol buffer, then PBS and attached to poly-L-lysine-coated coverslips. Samples were incubated with 1% BSA in PBS for 60 min at room temperature and then overnight at 4 °C with mouse monoclonal antibody (clone 9B11) against c-myc tag (Cell Signalling Technology, Beverley, MA, USA) at 1:1500 dilution. Afterwards, the samples were washed 3 times 10 min each with PBS and incubated for 60 min at room temperature with goat anti-mouse fluorescein isothiocyanate conjugated antibody (Jackson ImmunoResearch Laboratories, Inc, West Grove, PA, USA) at 1:200 dilution. Coverslips were treated with 0.02% (w/v) DAPI solution in PBS for 10 min in order to visualize nuclei, washed three times 10 min each in PBS and mounted in Vectashield (Vector Laboratories, Inc., Burlingame, CA, USA), mounting medium. Fluorescence images were acquired using Carl Zeiss 510 laser confocal microscope (Zeiss LSM 510; Carl Zeiss).

Protein extracts of membrane fractions from EBY.VW4000 were prepared as described by Sauer & Stolz (2000). Membrane proteins were resolved by SDS-PAGE on a 12.5% polyacrylamide gels containing 0.1% SDS in a discontinuous pH system (Laemmli, 1970), using ATTO AE-6450 equipment. Gels were transferred to nitrocellulose by electroblotting in Bjerrun and Schafer-Nielsen buffer with SDS (48 mM Tris, 39 mM glycine, 20% methanol, 1.3 mM SDS, pH 9.2). Recombinant proteins were probed with myc-tag (9B11) mouse mAb (Cell Signalling Technology) at 1:3000 dilution and specifically bound antibodies were visualized using peroxidase-coupled anti-mouse antibodies (Bio-Rad, Hercules, CA, USA) in conjunction with Immobilon Western HRP Substrate (Millipore, Billerica, MA, USA), and then exposed to film.

14C-glucose and 14C-fructose uptake studies

EBY-Ap_ST3 and control EBY-vec yeast cells were grown until mid-logarithmic phase (OD600 1–2), harvested by centrifugation (3000 g, 5 min at 4 °C), washed twice in transport buffer (50 mM potassium phosphate pH 7.0) and resuspended in transport buffer to 30 OD600 ml−1. Uptake experiments were performed at 30 °C and were initiated by mixing 75.7 ml of prepared cells with 24.3 ml labelled hexose solution containing 1.5 kBq of D-[U-14C]Glucose (3 mCi/mmol, GE Healthcare, Amersham, Buckinghamshire, UK) or 1.5 kBq D-[U-14C]Fructose (295 mCi/mmol, GE Healthcare) and unlabelled glucose or fructose at the concentrations used in the experiments. Uptake experiments were stopped by rapid vacuum-filtration of a 100 ml aliquot of cells on 0.45 mM nitrocellulose membranes (Sartorius AG, Goettingen, Germany) and cells were washed three times with ice-cold transport buffer before liquid scintillation counting. Competition for glucose or fructose uptake was performed by adding a 2-fold molar excess (100 mM) of unlabelled sugar to 50 mM glucose or fructose containing the appropriate labelled hexose. pH dependence was analysed by washing cells in 100 mM potassium phosphate adjusted to different pH values. The influence of pH gradient was determined by preincubating cells in 10 mM 2,4-dinitrophenol (DNP) or 10 mM carbonyl cyanide m-chlorophenylhydrazone (CCCP). Kinetic parameters Km and Vmax were estimated by Michaelis–Menten model (Prism vs. 5 software, GraphPad, San Diego). All experiments were performed at least in triplicate, and statistical analyses (pairwise t-tests to compare different treatments) were carried out using Prism software.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

DRGP, EF, AMRG and JAG thank the Biochemical and Biotechnological Sciences Research Council (Crop Science Initiative; Grant BB/E006280/1) for funding. CWR and AED thank the Sarkaria Foundation for financial support.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Abascal, F., Zardoya, R. and Posada, D. (2005) Prottest: selection of best-fit models of protein evolution. Bioinformatics 21: 21042105.
  • Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Basic local alignment search tool. J Mol Biol 215: 403410.
  • Arbuckle, M.I., Kane, S., Porter, L.M., Seatter, M.J. and Gould, G.W. (1996) Structure-function analysis of liver-type (GLUT2) and brain-type (GLUT3) glucose transporters: Expression of chimeric transporters in Xenopus oocytes suggests an important role for putative transmembrane helix 7 in determining substrate selectivity. Biochemistry 35: 1651916527.
  • Ashford, D.A., Smith, W.A. and Douglas, A.E. (2000) Living on a high sugar diet: the fate of sucrose ingested by a phloem-feeding insect, the pea aphid Acyrthosiphon pisum. J Insect Physiol 46: 335341.
  • Chintapalli, V.R., Wang, J. and Dow, J.A.T. (2007) Using FlyAtlas to identify better Drosophila models of human disease. Nat Genet 39: 715720.
  • Eddy, S.R. (1998) Profile hidden Markov models. Bioinformatics 14: 755763.
  • Garcia, J.C., Strube, M., Leingang, K., Keller, K., and Mueckler, M.M. (1992) Amino acid substitutions at Tryptophan-388 and Tryptophan-412 of the Hepg2 (GLUT1) glucose transporter inhibit transport activity and targeting to the plasma membrane in Xenopus oocytes. J Biol Chem 267: 77707776.
  • Gietz, R.D. and Woods, R.A. (2002) Transformation of yeast by lithium acetate/single-stranded carrier DNA/polyethylene glycol method. Methods Enzymol 350: 8796.
  • Hall, T.A. (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/Nt. Nucleic Acids Symp Ser 41: 9598.
  • Hashiramoto, M., Kadowaki, T., Clark, A.E., Muraoka, A., Momomura, K., Sakura, H., Tobe, K., Akanuma, Y., Yazaki, Y., Holman, G.D. and Kasuga, M. (1992) Site-directed mutagenesis of Glut1 in helix 7 residue 282 results in perturbation of exofacial ligand binding. J Biol Chem 267: 1750217507.
  • Karley, A.J., Ashford, D.A., Minto, L.M., Pritchard, J. and Douglas, A.E. (2005) The significance of gut sucrase activity for osmoregulation in the pea aphid, Acyrthosiphon pisum. J Insect Physiol 51: 13131319.
  • Kikawada, T., Saito, A., Kanamori, Y., Nakahara, Y., Iwata, K.I., Tanaka, D. et al. (2007) Trehalose transporter 1, a facilitated and high-capacity trehalose transporter, allows exogenous trehalose uptake into cells. Proc Natl Acad Sci USA 104: 1158511590.
  • Kozak, M. (1991) Structural features in eukaryotic messenger-RNAs that modulate the initiation of translation. J Biol Chem 266: 1986719870.
  • Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680685.
  • Manolescu, A.R., Witkowska, K., Kinnaird, A., Cessford, T. and Cheeseman, C. (2007) Facilitated hexose transporters: new perspectives on form and function. Physiology 22: 234240.
  • Moriwaki, N., Matsushita, K., Nishina, M. and Kono, Y. (2003) High concentrations of trehalose in aphid hemolymph. Appl Entomol Zool 38: 241248.
  • Mueckler, M., Weng, W.F. and Kruse, M. (1994) Glutamine-161 of GLUT1 glucose transporter is critical for transport activity and exofacial ligand binding. J Biol Chem 269: 2053320538.
  • Mueckler, M. and Makepeace, C. (2004) Analysis of transmembrane segment 8 of the GLUT1 glucose transporter by cysteine-scanning mutagenesis and substituted cysteine accessibility. J Biol Chem 279: 1049410499.
  • Mueckler, M. and Makepeace, C. (2006) Transmembrane segment 12 of the Glut1 glucose transporter is an outer helix and is not directly involved in the transport mechanism. J Biol Chem 281: 3699336998.
  • Pao, S.S., Paulsen, I.T. and Saier, M.H. Jr (1998) Major facilitator superfamily. Microbiol Mol Biol Rev 62: 134.
  • Pascual, J.M., Wang, D., Yang, R., Shi, L., Yang, H. and De Vivo, D.C. (2008) Structural signatures and membrane helix 4 in GLUT1 – inferences from human blood-brain glucose transport mutants. J Biol Chem 283: 1673216742.
  • Price, D.R.G., Wilkinson, H.S. and Gatehouse, J.A. (2007a) Functional expression and characterisation of a gut facilitative glucose transporter, NIHT1, from the phloem-feeding insect Nilaparvata lugens (rice brown planthopper). Insect Biochem Mol Biol 37: 11381148.
  • Price, D.R.G., Karley, A.J., Ashford, D.A., Isaacs, H.V., Pownall, M.E., Wilkinson, H.S. et al. (2007b) Molecular characterization of a candidate gut sucrase in the pea aphid, Acyrthosiphon pisum. Insect Biochem Mol Biol 37: 307317.
  • Prosser, W.A. and Douglas, A.E. (1992) A test of the hypotheses that nitrogen is upgraded and recycled in an aphid (Acyrthosiphon pisum) symbiosis. J Insect Physiol 38: 9399.
  • Rambaut, A. and Drummond, A.J. (2007) Tracer v1.4. http://beast.bio.ed.ac.uk/Tracer .
  • Rentsch, D., Laloi, M., Rouhara, I., Schmelzer, E., Delrot, S. and Frommer, W.B. (1995) Ntr1 encodes a high-affinity oligopeptide transporter in Arabidopsis. FEBS Lett 370: 264268.
  • Rhodes, J.D., Croghan, P.C. and Dixon, A.F.G. (1997) Dietary sucrose and oligosaccharide synthesis in relation to osmoregulation in the pea aphid, Acyrthosiphon pisum. Physiol Entomol 22: 373379.
  • Ronquist, F. and Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 15721574.
  • Sauer, N. and Stolz, J. (2000) Expression of foreign transport proteins in yeast. In Membrane Transport (Baldwin, S., ed.), pp. 79105. Oxford University Press, Oxford.
  • Seatter, M.J., De la Rue S.A., Porter L.M. and Gould G.W. (1998). QLS motif in transmembrane helix VII of the glucose transporter family interacts with the C-1 position of D-glucose and is involved in substrate selection at the exofacial binding site. Biochem. 37:13221326.
  • Stamatakis, A. (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22: 26882690.
  • Tamori, Y., Hashiramoto, M., Clark, A.E., Mori, H., Muraoka, A., Kadowaki, T., Holman, G.D. and Kasuga, M. (1994) Substitution at Pro(385) of GLUT1 perturbs the glucose transport function by reducing conformational flexibility. J Biol Chem 269: 29822986.
  • Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. and Higgins, D.G. (1997) The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25: 48764882.
  • Uldry, M. and Thorens, B. (2004) The SLC2 family of facilitated hexose and polyol transporters. Pflugers Archiv 447: 480489.
  • Ward, A., Sanderson, N.M., O'Reilly, J., Rutherford, N.G., Poolman, B. and Henderson, P.J.F. (2000) The amplified expression, identification, purification, assay and properties of hexahistidine-tagged bacterial membrane transport proteins. In Membrane Transport (Baldwin, S., ed.), pp. 141166. Oxford University Press, Oxford.
  • Whelan, S. and Goldman, N. (2001) A general empirical model of protein evolution derived from multiple protein families using a maximum-likelihood approach. Mol Biol Evol 18: 691699.
  • Wieczorke, R., Krampe, S., Weierstall, T., Freidel, K., Hollenberg, C.P. and Boles, E. (1999) Concurrent knock-out of at least 20 transporter genes is required to block uptake of hexoses in Saccharomyces cerevisiae. Febs Lett 464: 123128.
  • Wright, E.M. and Loo, D.D. (2000) Coupling between Na+, sugar, and water transport across the intestine. Ann N Y Acad Sci 915: 5466.
  • Zdobnov, E.M. and Apweiler, R. (2001) InterProScan – an integration platform for the signature-recognition methods in InterPro. Bioinformatics 17: 847848.
  • Zhou, J.-J., Kan, Y., Antoniw, J., Pickett, J.A. and Field, L.M. (2006) Genome and EST analyses and expression of a gene family with putative functions in insect chemoreception. Chem Senses 31: 453465.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1. Sequence similarity tree (Maximum Likelihood method) of sugar transporters containing the IPR003663 signature in Acyrthosiphon pisum (Blue), Drosophila melanogaster (Red), Apis mellifera (Brown) and other insects (Purple). For each branch, the posterior probability is shown. The scale bar indicates an evolutionary distance of 0.4 amino acid substitutions per position in the sequence.

Table S1. Survey of genes encoding proteins belonging to clan MFS (Pfam CL0015) in the genome of Acyrthosiphon pisum, and assignment to different families within the clan. ‘Predicted genes’ = total number of different potential genes identified; ‘Reference sequences’ = predicted genes for which consistent evidence in the form of predictions from different methods and complete open reading frames available. Most ‘reference sequences’ also have corresponding expressed sequence tags. Sequences identified by BLAST analysis, cut off = 0.01. (Numbers of predicted genes differ from detailed analysis due to manual correction of duplication, incorrect assignments, etc.; note that many Sugar_tr genes are included in MFS_1).

Table S2. Summary of genes predicted to encode proteins belonging to Pfam Sugar_tr family (PF00083) in Acyrthosiphon pisum. Additional genes encoding synaptic vesicle proteins and genes in family MFS_1 which are similar to organic cation transporters are included. ‘ST_ID’ (Ap_ST) numbers are temporary designations for protein products, which will be replaced by approved gene symbols. Interpro ID indicates the presence of signatures diagnostic for sugar transporters from the InterPro database. ‘Most similar Drosophila protein’ indicates best orthologue in the Drosophila melanogaster genome.

Table S3. Numbers of predicted genes encoding putative sugar transporters in different species. Predicted proteins from genome sequence data were scanned for matches using the Interpro IPR003663 sugar/inositol transporter signature, and the Pfam PF00083 Sugar_tr family motif. The scan was carried out using the HMMER program, with cut-off for matching set to e-value < 1.0e-10.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
IMB_918_sm_SupFig1.tif6210KSupporting info item
IMB_918_sm_table_s1-3.pdf472KSupporting info item

Please note: Neither the Editors nor Wiley Blackwell are responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.