Identification of an animal sucrose transporter - Journal of Cell Science

1984

Research Article

Identification of an animal sucrose transporter Heiko Meyer, Olga Vitavska and Helmut Wieczorek* Department of Biology and Chemistry, University of Osnabrück, 49069 Osnabrück, Germany *Author for correspondence ([email protected])

Accepted 7 February 2011 Journal of Cell Science 124, 1984-1991 © 2011. Published by The Company of Biologists Ltd doi:10.1242/jcs.082024

Journal of Cell Science

Summary According to a classic tenet, sugar transport across animal membranes is restricted to monosaccharides. Here, we present the first report of an animal sucrose transporter, SCRT, which we detected in Drosophila melanogaster at each developmental stage. We localized the protein in apical membranes of the late embryonic hindgut as well as in vesicular membranes of ovarian follicle cells. The fact that knockdown of SCRT expression results in significantly increased lethality demonstrates an essential function for the protein. Experiments with Saccharomyces cerevisiae as a heterologous expression system revealed that sucrose is a transported substrate. Because the knockout of SLC45A2, a highly similar protein belonging to the mammalian solute carrier family 45 (SLC45) causes oculocutaneous albinism and because the vesicular structures in which SCRT is located appear to contain melanin, we propose that these organelles are melanosome-like structures and that the transporter is necessary for balancing the osmotic equilibrium during the polymerization process of melanin by the import of a compatible osmolyte. In the hindgut epithelial cells, sucrose might also serve as a compatible osmolyte, but we cannot exclude the possibility that transport of this disaccharide also serves nutritional adequacy. Key words: Drosophila melanogaster, Sucrose transporter, Sugar transport

Introduction Transmembrane sugar transport in prokaryotes as well as in fungi and plants is attained by a variety of transport proteins specific for mono-, di- and oligosaccharides (Maiden et al., 1987; Sauer and Stolz, 1994; Szkutnicka et al., 1989). By contrast, sucrose transport has never been reported for animal cells. In lieu thereof, two major families of monosaccharide transporters that are essential for glucose uptake have been extensively studied, sodium-glucose linked transporters (SGLT) and glucose transporters (GLUT). The SGLT family represents Na+ symporters (Wright et al., 1994), whereas the members of the GLUT family are facilitative transporters (Thorens, 1996). One feature common to SGLT, GLUT and other animal sugar transporters is the fact that sucrose has never been reported to be a transported substrate. In previous studies of active K+ transport across caterpillar intestinal epithelium, we have learned that the disaccharide sucrose has to be present in the bathing medium to sustain ion transport activity (Meyer et al., 2004). Because this observation was inconsistent with the role of sucrose as a simple extracellular osmolyte, we hypothesized that sucrose has some function in transepithelial transport or is itself a transported substrate. Although proteins mediating the diffusion or active transport of sucrose are unknown in animal cells, the genomes of insects and vertebrates do include open reading frames that code for proteins similar to plant sucrose uptake transporters (SUTs). In addition, a disaccharide transporter with high similarity to these animal proteins and also to plant SUTs (supplementary material Fig. S1), was identified in the fission yeast Schizosaccharomyces pombe. Apparently, this protein (SUT1) mainly transports maltose and, with reduced affinity, also sucrose (Reinders and Ward, 2001). In contrast to other disaccharide transporters from fungi, e.g. the maltose permeases AGT1 or MAL61, the transporters with homology to plant SUTs contain a highly conserved sequence motif (RxGRR) between the second and the third transmembrane helix. Because this motif occurs in every known sucrose transporter at exactly the same

position, it is considered to be a sucrose transporter signature, presumably useful to identify new members belonging to this protein family (Lemoine, 2000). In addition to sharing the RxGRR motif (supplementary material Fig. S1, emphasized in blue), the homologous protein from Drosophila melanogaster analysed in our present study (SCRT; CG4484, FBgn0035968, http://flybase.org) appears to have 12 membrane-spanning domains (supplementary material Fig. S1, emphasized in black), an attribute that is also shared with sucrose transporters. Another structural aspect appears to be the long cytosolic, -helical N-terminal region (supplementary material Fig. S1, emphasized in green), which is present in SCRT, but also in all low-affinity plant sucrose transporters with apparent Km values equal to or higher than 10 mM (Schulze et al., 2000). In summary, the predicted SCRT structure is typical for a protein belonging to the major facilitator superfamily (MFS) (Marger and Saier, 1993) and reveals characteristic similarities to SUTs. In the present study we determined, on the basis of genomic information from D. melanogaster (FBgn0035968, http://flybase.org), the stage- and tissue-dependent occurrence of SCRT. In addition, we analysed the effects of RNAi-mediated scrt knock-down in transgenic flies and expressed the protein in Saccharomyces cerevisiae to glimpse at its function. We here present evidence that SCRT is a transporter of sucrose. It is the first sucrose transporter to be demonstrated in any animal. Results Expression of SCRT

The scrt gene (synonym CG4484) is located on the third chromosome at position 67A3. Sequence analysis predicts a single transcript of 2951 nucleotides resulting in a protein consisting of 599 amino acids (FBgn0035968, http://flybase.org). Initial experimental evidence for this transcript can be inferred from expressed sequence tags: among others, the clones LP06168, CK02046 and EK216041 (FBgn0035968, http://flybase.org)


Animal disaccharide transporter especially indicate transcription of the predicted isoform. Using specific primers, we could confirm this prediction. By cloning and sequencing the respective amplicon, we verified that the isolated cDNA corresponds to the predicted transcript. To further ascertain the issue, we analysed the stage dependence of transcript occurrence. Northern blots with total RNA isolated from embryonic, larval, pupal and adult Drosophila confirmed the presence of scrt mRNA in all developmental stages. Although embryonic and pupal stages exhibited moderate expression levels, strong expression was evident in adult flies. Third instar larvae showed a weaker expression than other stages tested (Fig. 1A). To ascertain whether these differences in transcript abundance are also valid regarding protein expression, we raised polyclonal antibodies against the central cytoplasmic loop of Drosophila SCRT (supplementary material Fig. S1, emphasized in red). The monospecificity of the antibodies was confirmed by western blots. In embryonic, larval and adult D. melanogaster protein preparations the antibodies detected a single band, which corresponded well with the predicted SCRT size of 66 kDa (Fig. 1B). The observation that no signal was visible in protein preparations isolated from RNAi-mediated scrt knockdown flies (actin-Gal4 ⫻ UAS-scrt-RNAi) provides strong evidence for the monospecificity of the antibodies and the validity of protein detection. In pupal protein preparations, the molecular mass of the detected band was slightly increased. However, because the pupal scrt transcript detected by northern blot (Fig. 1A) had apparently the same size as the transcripts from the remaining stages, which minimizes the possibility of alternative splicing, we consider the larger protein as a post-translationally modified version of SCRT. Indeed, sequence analysis revealed the presence of two N-glycosylation sites and a high number of putative phosphorylation sites in the protein (Motif Scan, http://myhits.isb-

Fig. 1. Stage-dependent expression of scrt transcript and SCRT protein. (A)Expression of scrt analysed by northern blot. scrt mRNA is present in embryonic, larval, pupal and adult Drosophila total RNA preparations as shown by hybridization of an antisense riboprobe (arrow). The bottom panel shows part of the radiant red stained gel with ribosomal RNA (rRNA) visible to demonstrate the loading of comparable RNA amounts (15g per lane). (B)Expression of SCRT analysed by western blot. SCRT is present in embryonic, larval, pupal and adult Drosophila protein preparations (arrow). The size of the detected protein corresponds to the predicted molecular mass of SCRT (66 kDa). The slightly increased molecular mass of the larval protein is presumably related to post-translational modifications. In adults expressing scrt-specific hpRNA (adult RNAi), the amount of SCRT is reduced below detection limits.

1985

sib.ch/cgi-bin/motif_scan), which renders glycosylation or phosphorylation to be reasonable modifications. To detect the protein in situ, we applied the antibodies against a number of tissues in different developmental stages. Of these tissues tested for SCRT expression, the most distinct signals were apparent in the late embryonic hindgut and in ovarian follicle cells (Figs 2, 3). Embryonic whole mount preparations showed a strong immunostaining covering the complete hindgut (Fig. 2B–D). Significantly, this staining is only observed in the final embryonic stage (stage 17) prior to hatching and first meal, indicating that Drosophila SCRT, like other proteins of metabolic relevance (Arbeitman et al., 2002), is already expressed in the late embryo to serve its purpose immediately in the first instar larval stage. The observation that the relative expression of SCRT in embryos is significantly higher compared to third instar larvae (Fig. 1) further supports the assumption that SCRT activity is required in very early larval stages already. To further localize Drosophila SCRT in the hindgut, we also investigated cross-sections of the complete embryo and found clear signals only at the very apical side of the hindgut epithelial cells, whereas the rest of the tissues remained unstained (Fig. 2E–J). Control experiments without primary antibodies did not yield any signal (Fig. 2A). The location of Drosophila SCRT at that exclusive site along the gastrointestinal tract suggests a role in substrate uptake across the apical membrane. In ovarian follicle cells of adult flies, small vesicle-like structures (1–2 m in diameter) were stained (Fig. 3A), whereas control sections without primary antibodies did not exhibit any pattern (not shown). In Drosophila and other insects, follicle cells produce the chorion (eggshell) including the dorsal appendages of developing embryos. During the formation of these structures, many cell constituents, including proteins, are co-secreted by follicle cells and can be found in the chorion after oviposition (Waring, 2000). To ascertain whether SCRT is also secreted, we immunostained the embryonic chorion and detected Drosophila SCRT in vesicular structures of the dorsal appendages, which looked very similar to the signal we had observed in the follicle cells (Fig. 3D,G). Interestingly, the same vesicular structures were labelled by a monoclonal antibody against melanin (Fig. 3H), suggesting that these structures represent melanin-containing organelles. However, in follicle cells the anti-melanin antibody did not detect considerable amounts of melanin (Fig. 3B), indicating that at this developmental stage melanin synthesis was not yet occurring. Although the anti-melanin signal was restricted to the dorsal appendages, an anti-SCRT signal was also observed as a cell-shaped staining covering the whole chorion. Because the same pattern was also detected by other antibodies used as controls (anti-aldolase, Fig. 3E; anti-tubulin, not shown), this signal might be unspecific. Stains without primary antibodies also resulted in a weak cell-shaped staining, corroborating an unspecific signal, whereas dorsal appendages remained completely unstained (Fig. 3F). SCRT expression is vital for Drosophila

To ascertain the in vivo relevance of SCRT expression, we generated RNAi lines utilizing the UAS–Gal4 system and analysed the effects of scrt downregulation. High knockdown efficiency was confirmed by northern blots (Fig. 4B). With both Gal4 driver lines being balanced over CyO, and the UAS line being homozygous for the scrt-RNAi construct, we could easily distinguish flies with expression of scrt-specific hairpin RNA (hpRNA, normal wings) from flies lacking the actin–Gal4 driver and thus also hpRNA expression (curly


1986

Journal of Cell Science 124 (12)

Fig. 2. Immunodetection of SCRT in the embryonic hindgut. In late Drosophila embryos (stage 17) anti-SCRT antibodies detect protein expression in the apical membranes of hindgut epithelial cells. Expression seems to be uniformly distributed across the whole tissue: (A) control, (B) brightfield, (C) fluorescence, (D) merge. Embryonic cross-sections show a clear signal at the apical side of the hindgut: (E,H) brightfield, (F,I) fluorescence, (G,J) merge.

wings). If the ubiquitous expression of scrt-specific hpRNA has no effect on the viability of the animals, a ratio between normal and curly wings of 1:1 would be anticipated. However, as depicted in Fig. 4A, flies with curly wings occurred approximately twice as often as flies with normal wings (67:33% on average), indicating a vital role for SCRT in D. melanogaster. Half of the knockdown flies died before reaching the adult stage. This result was valid for two different Gal4 driver lines (BL4414, BL25374) and was irrespective

of the crossing type (male UAS–scrt-RNAi crossed to virgin female actin–Gal4 and vice versa). The fact that, despite strong downregulation of scrt, some animals reached the adult stage whereas others did not is presumably based on the individual differences in RNAi penetration. Although almost negligible, the residual amount of scrt mRNA in the surviving RNAi flies (Fig. 4B) was apparently sufficient for the animals to reach adult stage. Except for the increased lethality however, no additional phenotype was apparent.

Fig. 3. Immunodetection of SCRT and melanin in follicle cells and in the embryonic chorion. (A)In ovary preparations, the follicle cells show a strong SCRT staining based on vesicular structures of about 1–2m in diameter. The nuclei of follicle cells are visible as black circles due to the absence of staining (yellow arrows). (B)The application of an anti-melanin antibody results only in a weak background staining. (C)Merge of A and B. (D)In the embryonic chorion, an unspecific cell-shaped staining (white arrow,) as well as a very specific staining of the dorsal appendages (green arrow) is apparent. This staining is unique to SCRT antibodies, whereas the cell-shaped staining is also detected by control antibodies (anti-aldolase, E). (F)Controls without primary antibodies show a weak cell-shaped staining but no staining of the dorsal appendages. (G)Staining in the appendages is based on vesicular structures of 1–2m in diameter (green arrows). (H)The same structures are labelled by an antibody against melanin (red arrows). (I)Merge of G and H.


Animal disaccharide transporter

Fig. 4. Effect of scrt RNAi on viability and gene expression. Downregulation of scrt mRNA was realized by crossing scrt-RNAi flies (stock 105439, Vienna Drosophila RNAi Center) to two different actin–Gal4 driver lines (BL4414 or BL25374, Bloomington Drosophila Stock Center). 1, 乆 BL4414 ⫻ 么 105439; 2, 乆 105439 ⫻ 么 BL4414; 3, 乆 BL25374 ⫻ 么 105439 and 4, 乆 105439 ⫻ 么 BL25374. The surviving G1 flies were collected, divided into the control group (CyO, curly wings) and scrt-RNAi-expressing group (Act5cGal4, normal wings) and counted. (A)Survival rate of flies expressing scrt-specific hpRNA (Act5cGal4) compared with control flies lacking the Gal4 driver element (CyO). Flies expressing scrt-specific hpRNA show significantly increased lethality compared with control flies. In all crossings, flies with curly wings occur about twice as often as flies with normal wings. In individual experiments, total numbers of 650 (1), 786 (2), 696 (3) and 715 (4) flies were counted. Bars show means ± s.d. (B)Subsequently to counting, total RNA was extracted from flies of each group, subjected to northern blot and probed with a scrt antisense riboprobe. Strong downregulation of scrt mRNA is apparent in the RNAi groups (Act5cGal4) compared to the control groups (CyO). The bottom panel shows part of the radiant red stained gel with ribosomal RNA (rRNA) visible to demonstrate the loading of comparable RNA amounts (15g/lane).

SCRT is a functional sucrose transporter

To assess a functional role of Drosophila SCRT, we expressed the protein in yeast cells, which are a commonly used system for the characterization of plant sucrose transporters (Riesmeier et al., 1992; Sauer and Stolz, 1994). Complementation assays showed that the expression of Drosophila SCRT in the yeast strain SuSy7 (Riesmeier et al., 1992) enabled the transformed cells to grow on minimal media containing 2% sucrose as the sole carbon source at a pH of 4, but not at neutral or slightly alkaline pH (Fig. 5). One key feature of the SuSy7 strain is its inability to grow on sucrose due to a deletion of its invertase gene. In yeast, invertase is a secreted enzyme that mediates the extracellular hydrolysis of sucrose to glucose and fructose. Thus, the lack of invertase prevents the extracellular hydrolysis of sucrose and thereby the subsequent uptake of the respective monosaccharides. A second key feature of SuSy7 is the insertion of the sucrose synthase gene from potato, enabling the strain to metabolize intracellular sucrose. Due to these genetic modifications, SuSy7 is eminently suited to identify sucrose transporters by complementation assays. The fact that expression of Drosophila SCRT enabled this strain to grow normally on a sucrose-containing minimal medium provides strong evidence that SCRT is able to transport sucrose. As a positive control, StSUT1 which is a well-characterized sucrose transporter from Solanum tuberosum (Riesmeier et al., 1993), was analogously expressed and led to growth characteristics comparable to those observed with Drosophila SCRT. By contrast, the transformation of empty expression

1987

Fig. 5. Complementation of sucrose uptake by SCRT expression. The functionality of SCRT is shown by complementation of a yeast strain incapable of sucrose uptake (SuSy7). Under acidic conditions (pH 4), the expression of SCRT under the control of the PMA1 promoter enables this strain to grow normally compared with the very slow growth rate of cells transformed with the empty expression vector as a control. At neutral pH values (pH 7), the growth rate of the cells expressing SCRT is largely reduced, and diminishes completely under alkaline conditions (pH 8.5). As a positive control, StSUT1 was expressed in the same strain and showed growth characteristics similar to SCRT. Transformed cells were grown for 5 days at 29°C on minimal media.

vectors as a negative control resulted in negligible background growth of the cells. To confirm the results obtained by the complementation tests, we expressed SCRT in the yeast strain RS453, which is a commonly used strain for the expression and characterization of monosaccharide (Sauer and Stadler, 1993) and disaccharide transporters (Sauer and Stolz, 1994). We made use of the fact that RS453 cells grown on glucose-containing media strongly repress the uptake of sucrose as well as extracellular hydrolysis of sucrose by invertase (Sauer and Stolz, 1994). Accordingly, glucose-grown yeast cells are naturally well suited for the heterologous expression and characterization of sucrose transporters. Northern blots ensured that the SCRT expression construct was transcribed by the transformed yeast cells. As expected, an antisense probe detected a clear band of anticipated size, whereas the application of a sense probe did not elicit any signal (not shown). The transformed strain was subsequently used for uptake measurements with [14C]sucrose as a substrate. As shown in Fig. 6, the expression of Drosophila SCRT under the control of the strong PMA1 promoter (Sauer and Stolz, 1994) enabled the yeast cells to accumulate sucrose in a time-dependent manner, being linear up to at least 2 minutes. Control experiments showed that the kinetics of sucrose uptake were considerably blunted in yeast cells transformed with the empty expression vector. Clearly, SCRT is a transporter of sucrose. To ascertain whether the pH dependence observed in the complementation tests reflected pH-dependent sucrose transport, we performed uptake experiments at different pH values. Sucrose

1988



Fig. 6. Time dependence of sucrose uptake by yeast cells (RS453) expressing SCRT. Compared to yeast cells transformed with empty expression vectors (open squares), cells expressing SCRT (closed squares) show a significant increase in time-dependent uptake of [14C]sucrose. Uptake was measured in the presence of 100 mM sucrose at pH 4. Data points correspond to the means ± s.d. of seven independent preparations. FW, fresh weight.

transport was most efficient under acidic conditions, whereas neutral or slightly alkaline pH abolished transport completely (Fig. 7). For reasons of comparison, we did the same experiments with yeast cells expressing the plant sucrose transporter StSUT1 and observed an almost identical stimulation of sucrose uptake with decreasing pH [not shown here, but already reported for, e.g. AtSUC1 and AtSUC2 (Sauer and Stolz, 1994)]. Experiments regarding the concentration dependence of SCRT under conditions of initial velocity revealed that half maximal [14C]sucrose uptake requires concentrations in the range of 15 mM (Fig. 8). Thus, the affinity of Drosophila SCRT for sucrose was in line with that of low affinity sucrose transporters from plants, e.g. proteins belonging

Fig. 8. Concentration dependence of sucrose uptake by yeast cells (RS453) expressing SCRT. Sucrose uptake rate measured after 1 minute of transport (initial velocity) increases with elevated sucrose concentrations. Values are presented with background transport (empty vector) subtracted. The hyperbolic curve was fitted using Origin 7 (Originlab) and the apparent Km was 16±5 mM. Uptake was measured at pH 4 and the data points correspond to the means ± s.d. of three independent preparations. FW, fresh weight.

to the SUT4 or the SUT2/SUC3 subfamilies, which show similar apparent Km values (Schulze et al., 2000; Weise et al., 2000). To obtain initial data on the substrate specificity of SCRT, sucrose uptake was competed with other mono- and disaccharides that might be potential substrates for the transporter. The addition of non-labelled sucrose itself led, due to the lower specific radioactivity, to a 70% decrease of detected radioactivity (Table 1). Among the other sugars, the most efficient competitor of sucrose uptake was trehalose. It is noteworthy that the addition of glucose augmented sucrose transport instead of diminishing it. The latter result was also reported for different sucrose transporters from plants, e.g. SUC1 and SUC2 from Arabidopsis thaliana (Sauer and Stolz, 1994). As a possible reason for this effect, the authors proposed that the increased transport activity is based on a more efficient yeast plasma membrane energization caused by the increased metabolism of the optimal carbon source glucose. An almost complete inhibition of transport was caused by the addition of the protonophore carbonyl cyanide m-chlorophenylhydrazone (CCCP), which is consistent with a proton-coupled sucrose uptake mechanism (Table 1).

Table 1. Transport properties of SCRT in yeast cells Sugar added None Sucrose Trehalose Maltose Lactose Galactose Glucose CCCP Fig. 7. pH dependence of sucrose uptake by yeast cells (RS453) expressing SCRT. Sucrose uptake after 10 minutes of transport was measured at different pH values. Transport activity is high under acidic conditions whereas neutral or slightly alkaline pH abolishes the uptake of radiolabelled substrate. Data are presented with background transport (empty vector) subtracted and correspond to the means + s.d. of three independent assays. FW, fresh weight.

Activity (%) 100 31±16 56±16 75±11 86±10 73±4 124±1 7±3

Putative substrates were added at fivefold excess (final concentration 300 mM) and CCCP was added to a final concentration of 50 M, 2 minutes before the assay was started by the addition of radiolabelled sucrose. Values are presented with background transport (empty vector) subtracted and represent relative uptake rates after 10 minutes of transport at pH 4 compared with [14C]sucrose uptake without additional sugars or inhibitors. Values correspond to the means ± s.d. of at least three independent assays.


Animal disaccharide transporter

Fig. 9. Rooted ML tree (lnL–15818.5) of human proteins belonging to the SLC45 family, SCRT from D. melanogaster, and similar proteins from fungi and plants as outgroups. The tree shows the relationship of the proteins to each other. Bootstrap values are given at the branches. The scale bar indicates substitutions per position. SpSUT1, Schizosaccharomyces pombe glucoside transporter (NP_594387.1); NfSUT, Neosartorya fischeri putative sucrose transporter (XP_001266016.1); AfSUT, Aspergillus fumigatus sucrose transporter (XP_754012.1); PmSUT, Penicillium marneffei putative sucrose transporter (XP_002147999.1); StSUT1, Solanum tuberosum sucrose transporter (CAA48915.1); VvSUT2, Vitis vinifera sucrose transporter (AAL32020.1); AtSUC3, Arabidopsis thaliana sucrose transporter (NP_178389.1); LeSUT4, Lycopersicon esculentum sucrose transporter (AAG09270.1). NCBI sequence accession numbers (http://www.ncbi.nlm.nih.gov/sites/entrez) are given in brackets.

Discussion We have provided the first experimental evidence for a sucrose transporter in animal cells. However, as depicted in Table 1, the major insect blood sugar trehalose (Wyatt and Kale, 1957) has a certain potential to compete with sucrose for transport, indicating that this disaccharide might also be a substrate for SCRT, yet with a reduced potency. The Drosophila SCRT resembles sucrose transporters in plants, including a crucial region that is considered to be a sucrose transporter signature. Its localization in the apical membrane of hindgut epithelial cells suggests a role in transmembrane and/or transepithelial sucrose transport. An intestinal sucrose transporter would appear particularly advantageous for animals such as Drosophila that rely primarily on a fruit diet, which contains sucrose in concentrations of up to 0.2 mol/l (Beruter, 2004; Etxeberria and Gonzalez, 2005). This high abundance is a strong argument for the assumption that at least in the hindgut tissue sucrose represents the main substrate of SCRT. Furthermore, if the blood sugar trehalose was the transported disaccharide in the hindgut, we would assume the transporter to be located in the basolateral instead of the apical membrane of epithelial cells. The pH-dependent growth of transformed yeast cells at pH 4 but not at neutral or slightly alkaline pH (Fig. 5), together with the result that uptake of radiolabelled sucrose requires acidic conditions (Fig. 7), suggests that Drosophila SCRT might be, like the plant sucrose transporters, an H+/sucrose symporter. Additional evidence for proton-coupled transport arises from our finding that the uncoupler CCCP abolishes sucrose transport almost

1989

completely (Table 1). Indeed, the pH in the insect hindgut is slightly acidic (Harrison, 2001), thus supporting our speculation. The transport of sucrose or other disaccharides such as the insect blood sugar trehalose into melanin-containing organelles might also be an H+ co-transport because in mammals, mature melanin-synthesizing melanosomes are, in contrast to premature melanosomes, not thought to be acidic (Ancans et al., 2001; Tabata et al., 2008). Nevertheless, the function of Drosophila SCRT in the melanosomal membrane is less clear. We suggest that it plays a role in volume control during the polymerization of tyrosine at the beginning of melanin synthesis. Indeed, the putative sucrose transporter membrane-associated transporter protein (MATP), which was identified in the Japanese killifish and exhibits high similarities to SCRT and also to plant sucrose transporters (supplementary material Fig. S1), has been proposed to provide osmotic balance in melanosomes during melanin biosynthesis (Fukamachi et al., 2001). Similar postulations were made for human MATP (SLC45A2), in which distinct point mutations, presumably inactivating the protein, result in a novel form of oculocutaneous albinism, OCA4 (Newton et al., 2001). Furthermore, MATP-mutant mice exhibit small, crenated melanosomes compared with the ovoid or round melanosomes in the wild type, which was considered an indication for an osmoregulatory dysfunction (Lehman et al., 2000). Because in vivo substrates of the vertebrate melanosomal transporters are entirely unknown, our data on the highly similar SCRT might be particularly advantageous in understanding the function of sucrose transporters in melanin synthesis. Although there is so far no experimental proof of the necessity of disaccharides in melanosomal volume regulation, the potential to act not only as compatible osmolytes but also as stabilizers of protein structures and enzymatic activities has been shown extensively for sucrose (Cioni et al., 2005; Meyer et al., 2004; Ou et al., 2001; Wang et al., 1995) and for trehalose (Felix et al., 1999; Kaushik and Bhat, 2003; Zancan and Sola-Penna, 2005). Thus, in addition to the proposed osmoregulatory functions, sucrose or trehalose might be crucial for protein functionality inside the melanosomal microcompartment. Besides SLC45A2, which displays the highest degree of amino acid identity with SCRT (supplementary material Fig. S1), the remaining three members of the human SLC45 family also show remarkable similarities to Drosophila SCRT. In a phylogenetic analysis, SCRT is sister to a clade comprising SLC45A1, SLC45A2 and SLC45A4. SLC45A3 is sister to this clade (Fig. 9). In contrast to SLC45A1 and SLC45A4 that have not yet been assigned to any physiological process, SLC45A3 has been reported to be involved in prostate cancer (Helgeson et al., 2008; Tomlins et al., 2007). Because up to now no physiological substrate for this putative transporter has been identified, its involvement in the disease is rather elusive. Our observation that, except for increased lethality, scrt knockdown flies do not show any apparent phenotype is presumably due to the residual expression of scrt mRNA in the surviving RNAi flies (Fig. 4B). Apparently, the remaining expression of SCRT in these flies is not only sufficient for the individuals to develop to adults, but might also be adequate to conceal additional phenotypes like albinism, which could be anticipated to occur according to the observations reported on similar vertebrate proteins (MATPs). In future experiments, the use of tissue- or stage-specific driver lines might be beneficial in identifying additional RNAimediated phenotypes. However, the fact that ubiquitous downregulation of SCRT induces lethality is quite an interesting

1990


result on its own because it demonstrates that the physiological functions fulfilled by SCRT are absolutely essential. The existence of a sucrose transporter in D. melanogaster extends our current understanding of sugar transport in animals. Protein expression in the Drosophila hindgut emphasizes this in particular by indicating a direct involvement of SCRT in sucrose uptake in the intestinal tract. Future research will clarify how extensively the common conception that sugar transport across animal membranes is restricted to monosaccharides is violated. Materials and Methods


Strains and growth conditions

D. melanogaster Oregon R strain was grown on a medium containing 1.35% agar, 3.75% baker’s yeast, 6.4% sucrose and 16.5% maize grit. Compounds were mixed, boiled briefly and transferred to breeding containers. Breeding was conducted at 22°C. To knock down scrt mRNA expression in the complete animal, flies from the scrt-specific RNAi stock 105439 (Vienna Drosophila RNAi Center, VDRC, http://www.vdrc.at/) were either crossed to the driver line BL4414 (Act5C-Gal4/CyO) or to the line BL25374 (Act5C-Gal4/CyO, Bloomington Drosophila Stock Center, http://flystocks.bio.indiana.edu/) and reared at 29°C, 12:12 photoperiod and 70% relative humidity. In all cases, 5–8 virgin females were crossed to 10–15 males, with the mated flies being transferred to new breeding containers every second day. The genotype of the resulting G1 progeny was identified utilizing the CyO marker for chromosome 2. Flies showing the curly wing phenotype were considered wild type, whereas flies with normal wings expressed scrt-specific hpRNA. High knockdown efficiency was confirmed by northern blots. For cloning in Escherichia coli, strain DH5 (Hanahan, 1983) was used. In the case of transport tests with [14C]sucrose as a substrate, SCRT was expressed in the S. cerevisiae strain RS453 (Sauer and Stadler, 1993). For complementation experiments, the yeast strain SuSy7 (Riesmeier et al., 1992), a kind gift from Norbert Sauer, Friedrich-Alexander University, Erlangen-Nürnberg, Germany, was used as an expression system. Molecular cloning of full length scrt and the central loop

Full length scrt was obtained by RT-PCR using D. melanogaster embryonic cDNA as a template and the primers 5⬘-ATGGTTGGAGTTACTGCCGAC-3⬘ (forward, fw) and 5⬘-CTAGACATATAACACAAACAT-3⬘ (reverse, rv), respectively. The cDNA was reverse transcribed from embryonic total RNA (1 g per reaction) using the AMV First Strand cDNA Synthesis Kit for RT-PCR (Roche, Basel, Switzerland). In order to ligate the resulting fragment into the NEV-E yeast expression vector (Sauer and Stolz, 1994), EcoRI restriction sites were introduced at both ends into the cDNA by appropriate primer design. To amplify the sequence coding for the central cytoplasmic loop, the primers 5⬘-CGTGAAATTCCATTGCCCTTG-3⬘ (fw) and 5⬘GCGCATCGAGTAGGGCATGAT-3⬘ (rv) were used. Subsequently, the PCR-product was ligated into the pET16b expression vector (Novagen, EMD Biosciences, Madison, WI) employing NdeI and BamHI restriction sites, respectively. Expression in E. coli Rosetta (DE3) cells (Novagen, Darmstadt, Germany) was done essentially as described (Meyer et al., 2009) and the resulting soluble protein (fused to a Nterminal 10⫻His-tag) was purified via Ni-NTA Agarose (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Templates for riboprobe synthesis were generated with primers 5⬘-ATGGTTGGAGTTACTGCCGAC-3⬘ (fw) and 5⬘AAGATGTGAAATTCAAGGCTG-3⬘ (rv). All constructs were validated by sequencing. Phylogenetic analysis

We used a maximum likelihood (ML) analysis to reconstruct the phylogenetic relationship of proteins belonging to the human SLC45 family and SCRT from D. melanogaster (CG4484, http://flybase.org), and complemented the data with similar proteins from fungi and plants as outgroups. The amino acid data set was aligned with the aid of MUSCLE (Edgar, 2004). The most appropriate substitution model for the ML analysis was LG+I+G+F as determined on the basis of the Akaike information criterion using ProtTest (Abascal et al., 2005). ML analyses were conducted with RAxML 2.7.6 (Stamatakis, 2006) using 100 replicate searches starting from randomized maximum parsimony trees. Confidence values for the edges of the best ML tree were determined on the basis of 100 bootstrap replicates. Expression in S. cerevisiae and transport assays

All plasmids were transformed into yeast strains via electroporation (1 pulse, 1.5 kV, 25 F, 200 ) with 1 g plasmid DNA per transformation approach. The resulting strains were grown on minimal media containing 0.67% yeast nitrogen base (BD, Franklin Lakes, NJ), 1% casaminoacids (BD), 2% D-glucose (Serva, Heidelberg, Germany) and 25 mM sodium phosphate (pH 6.0). After reaching an optical density of 1.5 at 600 nm, 20 ml of cells were harvested by centrifugation (5 minutes at 3.500 g), washed first with 20 ml H2O followed by 20 ml of incubation buffer (minimal media as described above, with substitution of glucose by variable concentrations of sucrose at different pH-values adjusted with 25 mM sodium phosphate), centrifuged

again and resuspended in 1 ml of incubation buffer. Cells were kept on ice until uptake assays were started by the addition of radiolabelled substrate solution ([14C]sucrose 0.2 Ci/l; GE Healthcare, Munich, Germany). Assays with 1 mM, 5 mM and 25 mM sucrose in the incubation buffer employed 0.4 Ci, whereas assays with 60 mM sucrose contained 1 Ci and assays with 100 mM sucrose contained 2.5 Ci. In all cases, aliquots (100 l) were taken and transferred into 1 ml of incubation buffer to stop the uptake process. After filtration (cellulose acetate filters, 0.45 m, Schleicher and Schuell, Germany), the filters containing yeast cells were transferred into scintillation vials and the amount of accumulated [14C]sucrose was measured by scintillation counting (Beckman LS 6500). Complementation tests

For complementation, transformed SuSy7 strains were grown at 29°C on SD minimal media containing 0.67% Yeast Nitrogen Base (BD), 0.002% tryptophan (Sigma), 2% sucrose (analytical grade; Serva, Heidelberg, Germany) and 25 mM sodium phosphate (variable pH values). Immunohistochemistry

SCRT antisera were raised in guinea pigs (Charles River, Kisslegg, Germany) against the expressed and Ni-NTA-purified central cytoplasmic loop of the protein (amino acid positions 286–390; supplementary material Fig. S1). Antibody specificity was confirmed by antigen blocking (50 l of sera were incubated with 40 g of purified antigen in 1 ml PBS for 1 hour at room temperature) and the resulting sera were controlled by western blot and tissue staining. For whole mount preparations, embryos were dechorionized, devitellinized and fixed (Patel, 1994). If the embryonic chorion was stained, incubation in Klorix:water (1:1) was omitted. Fixation was performed in 4% formaldehyde in PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4) for 25 minutes. Antibody application was carried out in 2% bovine serum albumin, 0.1% SDS in PBS. Blocking was done in the same buffer for 1 hour at room temperature. Primary antibodies (anti-SCRT, raised in this study, and antimelanin antibodies, a kind gift from Joshua Nosanchuk, Albert Einstein College of Medicine, New York) and secondary antibodies (anti-guinea-pig Cy3-conjugate and anti-mouse FITC-conjugate, both from Sigma) were applied overnight (4°C) at dilutions of 1:1.000 and 1:200, respectively. Washing was done in PBS with 0.05% Tween 20. Prior to VectaShield (Vector Laboratories, Burlingame, CA) coating, whole mount stained embryos were incubated in glycerin for 48 hours. For cryo-sections (5 m), stained specimen were subjected to an increasing sucrose gradient (10, 20 and 30% in PBS) prior to embedding them in Tissue Freezing Medium (Jung, Germany). Freezing was conducted in isopentane cooled in liquid nitrogen. Subsequent to sectioning (Leica CM1900 cryomicrotome), frozen tissues were transferred to superfrost glass slides (Microm International, Walldorf, Germany), fixed by short-term warming (10 seconds, 50°C), washed three times (PBS, 5 minutes each) and embedded in VectaShield. Fluorescence was visualized using either a reverse microscope (Olympus IX70, 100⫻, NA 1.35) or a confocal laser scanning microscope (LSM 5 Pascal, Zeiss, 40⫻, NA 1.3). Western blot and northern blot analysis

SDS-PAGE and western blots were carried out essentially as described (Wieczorek et al., 1991) with the variations that proteins (15 g per lane) were transferred to nitrocellulose membranes instead of PVDF, and milk powder (5%) was used as blocking reagent instead of gelatine. Purified SCRT antiserum was applied at a dilution of 1:5,000 and visualized by anti-guinea pig alkaline phosphatase conjugated antibody (1:10,000; Sigma). Northern blots were generally conducted as described (Meyer et al., 2010) with total RNA (15 g per lane) and a hybridization temperature of 65°C.

This work was supported in part by the DFG (Research training group 612) and the Hans Mühlenhoff foundation. We thank Klaus W. Beyenbach (Cornell University) for valuable comments on the manuscript and Torsten Struck for giving support regarding the phylogenetic analysis. Supplementary material available online at http://jcs.biologists.org/cgi/content/full/124/12/1984/DC1

References Abascal, F., Zardoya, R. and Posada, D. (2005). ProtTest: selection of best-fit models of protein evolution. Bioinformatics 21, 2104-2105. Ancans, J., Tobin, D. J., Hoogduijn, M. J., Smit, N. P., Wakamatsu, K. and Thody, A. J. (2001). Melanosomal pH controls rate of melanogenesis, eumelanin/phaeomelanin ratio and melanosome maturation in melanocytes and melanoma cells. Exp. Cell Res. 268, 26-35. Arbeitman, M. N., Furlong, E. E., Imam, F., Johnson, E., Null, B. H., Baker, B. S., Krasnow, M. A., Scott, M. P., Davis, R. W. and White, K. P. (2002). Gene expression during the life cycle of Drosophila melanogaster. Science 297, 2270-2275. Beruter, J. (2004). Carbohydrate metabolism in two apple genotypes that differ in malate accumulation. J. Plant Physiol. 161, 1011-1029.


Animal disaccharide transporter Cioni, P., Bramanti, E. and Strambini, G. B. (2005). Effects of sucrose on the internal dynamics of azurin. Biophys. J. 88, 4213-4222. Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792-1797. Etxeberria, E. and Gonzalez, P. (2005). Accumulation of sucrose and other soluble solids in citrus juice cells. Proc. Fla. State Hort. Soc. 118, 43-48. Felix, C. F., Moreira, C. C., Oliveira, M. S., Sola-Penna, M., Meyer-Fernandes, J. R., Scofano, H. M. and Ferreira-Pereira, A. (1999). Protection against thermal denaturation by trehalose on the plasma membrane H+-ATPase from yeast. Synergetic effect between trehalose and phospholipid environment. Eur. J. Biochem. 266, 660-664. Fukamachi, S., Shimada, A. and Shima, A. (2001). Mutations in the gene encoding B, a novel transporter protein, reduce melanin content in medaka. Nat. Genet. 28, 381-385. Hanahan, D. (1983). Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557-580. Harrison, J. F. (2001). Insect acid-base physiology. Annu. Rev. Entomol. 46, 221-250. Helgeson, B. E., Tomlins, S. A., Shah, N., Laxman, B., Cao, Q., Prensner, J. R., Cao, X., Singla, N., Montie, J. E., Varambally, S. et al. (2008). Characterization of TMPRSS2:ETV5 and SLC45A3:ETV5 gene fusions in prostate cancer. Cancer Res. 68, 73-80. Kaushik, J. K. and Bhat, R. (2003). Why is trehalose an exceptional protein stabilizer? An analysis of the thermal stability of proteins in the presence of the compatible osmolyte trehalose. J. Biol. Chem. 278, 26458-26465. Lehman, A. L., Silvers, W. K., Puri, N., Wakamatsu, K., Ito, S. and Brilliant, M. H. (2000). The underwhite (uw) locus acts autonomously and reduces the production of melanin. J. Invest. Dermatol. 115, 601-606. Lemoine, R. (2000). Sucrose transporters in plants: update on function and structure. Biochim. Biophys. Acta 1465, 246-262. Maiden, M. C., Davis, E. O., Baldwin, S. A., Moore, D. C. and Henderson, P. J. (1987). Mammalian and bacterial sugar transport proteins are homologous. Nature 325, 641643. Marger, M. D. and Saier, M. H., Jr (1993). A major superfamily of transmembrane facilitators that catalyse uniport, symport and antiport. Trends Biochem. Sci. 18, 13-20. Meyer, H., Wieczorek, H. and Zeiske, W. (2004). K+ transport in the caterpillar intestine epithelium: role of osmolytes for the K+-secretory capacity of the tobacco hornworm midgut. J. Comp. Physiol. B 174, 527-539. Meyer, H., Panz, M., Zmojdzian, M., Jagla, K. and Paululat, A. (2009). Neprilysin 4, a novel endopeptidase from Drosophila melanogaster, displays distinct substrate specificities and exceptional solubility states. J. Exp. Biol. 212, 3673-3683. Meyer, H., Von Ohlen, T., Panz, M. and Paululat, A. (2010). The disintegrin and metalloprotease Meltrin from Drosophila forms oligomers via its protein binding domain and is regulated by the homeobox protein VND during embryonic development. Insect Biochem. Mol. Biol. 40, 814-823. Newton, J. M., Cohen-Barak, O., Hagiwara, N., Gardner, J. M., Davisson, M. T., King, R. A. and Brilliant, M. H. (2001). Mutations in the human orthologue of the mouse underwhite gene (uw) underlie a new form of oculocutaneous albinism, OCA4. Am. J. Hum. Genet. 69, 981-988. Ou, W. B., Park, Y. D. and Zhou, H. M. (2001). Molecular mechanism for osmolyte protection of creatine kinase against guanidine denaturation. Eur. J. Biochem. 268, 5901-5911. Patel, N. H. (1994). Imaging neuronal subsets and other cell types in whole-mount Drosophila embryos and larvae using antibody probes. Methods Cell Biol. 44, 445-487.

1991

Reinders, A. and Ward, J. M. (2001). Functional characterization of the alpha-glucoside transporter Sut1p from Schizosaccharomyces pombe, the first fungal homologue of plant sucrose transporters. Mol. Microbiol. 39, 445-454. Riesmeier, J. W., Willmitzer, L. and Frommer, W. B. (1992). Isolation and characterization of a sucrose carrier cDNA from spinach by functional expression in yeast. EMBO J. 11, 4705-4713. Riesmeier, J. W., Hirner, B. and Frommer, W. B. (1993). Potato sucrose transporter expression in minor veins indicates a role in phloem loading. Plant Cell 5, 1591-1598. Sauer, N. and Stadler, R. (1993). A sink-specific H+/monosaccharide co-transporter from Nicotiana tabacum: cloning and heterologous expression in baker’s yeast. Plant J. 4, 601-610. Sauer, N. and Stolz, J. (1994). SUC1 and SUC2: two sucrose transporters from Arabidopsis thaliana; expression and characterization in baker’s yeast and identification of the histidine-tagged protein. Plant J. 6, 67-77. Schulze, W., Weise, A., Frommer, W. B. and Ward, J. M. (2000). Function of the cytosolic N-terminus of sucrose transporter AtSUT2 in substrate affinity. FEBS Lett. 485, 189-194. Stamatakis, A. (2006). RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688-2690. Szkutnicka, K., Tschopp, J. F., Andrews, L. and Cirillo, V. P. (1989). Sequence and structure of the yeast galactose transporter. J. Bacteriol. 171, 4486-4493. Tabata, H., Kawamura, N., Sun-Wada, G. H. and Wada, Y. (2008). Vacuolar-type H(+)-ATPase with the a3 isoform is the proton pump on premature melanosomes. Cell Tissue Res. 332, 447-460. Thorens, B. (1996). Glucose transporters in the regulation of intestinal, renal, and liver glucose fluxes. Am. J. Physiol. 270, G541-G553. Tomlins, S. A., Laxman, B., Dhanasekaran, S. M., Helgeson, B. E., Cao, X., Morris, D. S., Menon, A., Jing, X., Cao, Q., Han, B. et al. (2007). Distinct classes of chromosomal rearrangements create oncogenic ETS gene fusions in prostate cancer. Nature 448, 595-599. Wang, A., Robertson, A. D. and Bolen, D. W. (1995). Effects of a naturally occurring compatible osmolyte on the internal dynamics of ribonuclease A. Biochemistry 34, 15096-15104. Waring, G. L. (2000). Morphogenesis of the eggshell in Drosophila. Int. Rev. Cytol. 198, 67-108. Weise, A., Barker, L., Kuhn, C., Lalonde, S., Buschmann, H., Frommer, W. B. and Ward, J. M. (2000). A new subfamily of sucrose transporters, SUT4, with low affinity/high capacity localized in enucleate sieve elements of plants. Plant Cell 12, 1345-1355. Wieczorek, H., Putzenlechner, M., Zeiske, W. and Klein, U. (1991). A vacuolar-type proton pump energizes K+/H+ antiport in an animal plasma membrane. J. Biol. Chem. 266, 15340-15347. Wright, E. M., Loo, D. D., Panayotova-Heiermann, M., Lostao, M. P., Hirayama, B. H., Mackenzie, B., Boorer, K. and Zampighi, G. (1994). ‘Active’ sugar transport in eukaryotes. J. Exp. Biol. 196, 197-212. Wyatt, G. R. and Kale, G. F. (1957). The chemistry of insect hemolymph. II. Trehalose and other carbohydrates. J. Gen. Physiol. 40, 833-847. Zancan, P. and Sola-Penna, M. (2005). Trehalose and glycerol stabilize and renature yeast inorganic pyrophosphatase inactivated by very high temperatures. Arch. Biochem. Biophys. 444, 52-60.