![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 17, Issue 7, 3021-3030, July 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Program in Molecular, Cell, and Developmental Biology, Oklahoma Medical Research Foundation, Oklahoma City, OK 73104; and Department of Biological Sciences, University of Southern California, Los Angeles, CA 90089-2520
Submitted February 24, 2006;
Revised April 10, 2006;
Accepted April 14, 2006
Monitoring Editor: Sandra Schmid
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Recently, plasma membrane neurotransmitter transporters have been shown to play a significant role in maintaining appropriate levels of neurotransmitters in the synaptic cleft. Proteins in the sodium-dependent neurotransmitter symporter family (Pfam designation, SNF) transport neurotransmitters (and related molecules) from the synaptic cleft in a Na+- and Cl–-dependent manner, thereby terminating synaptic transmission and providing a means of recycling neurotransmitters. SNFs can rapidly redistribute between plasma membrane and vesicular membranes; this is regulated by cycles of phosphorylation and dephosphorylation (reviewed in Deken et al., 2001). SNFs are also the focus of much interest as drug targets; drugs blocking specific SNFs are currently used to treat a number of pathological conditions, including depression and epilepsy.
There are at least 18 members of the SNF family in humans, and 14 each in Drosophila melanogaster and Caenorhabditis elegans. Sequence comparisons suggest that humans, fruit flies, and nematodes share a core set of SNF proteins, which include the dopamine, serotonin, and GABA transporters. In addition, each species has a unique set of SNF proteins. We chose the nematode C. elegans as a model for studying SNF function because it is well-suited for genetic analysis, has a simple nervous system, and the complete sequence of its genome has been determined. When this study was initiated, there were already two characterized SNF members in C. elegans: dat-1 (T23G5.5) encodes the dopamine transporter (Nass et al., 2001) and mod-5 (Y54E10BR.7) encodes the serotonin transporter (Ranganathan et al., 2001). By our analysis, there are 12 additional family members in the C. elegans genome. We have assigned them the C. elegans gene designation "snf," consistent with the Pfam designation.
The GABAergic system in C. elegans mediates several distinct behaviors, and mutations affecting GABAergic function result in several easily scorable mutant phenotypes (reviewed in Schuske et al., 2004). The 26 C. elegans GABAergic cells modulate locomotion (DD and VD motor neurons), foraging (RMEs), and defecation (AVL and DVB) behaviors (McIntire et al., 1993b; Schuske et al., 2004). Mutations in the GABA biosynthetic enzyme glutamic acid decarboxylase (GAD) (unc-25; Jin et al., 1999), vesicular transporter (unc-47; McIntire et al., 1997), and excitatory and inhibitory receptors (exp-1; Beg and Jorgensen, 2003; unc-49; Bamber et al., 1999), have been described in detail. In addition, earlier studies had demonstrated uptake of exogenous GABA into a subset of C. elegans neurons (McIntire et al., 1993b).
Two C. elegans genes, snf-3 and snf-11, encode proteins with significant similarity to mammalian GABA transporters. We show here that snf-11 encodes a functional GABA transporter with properties similar to the high-affinity mammalian transporter GAT1, we describe the expression pattern for the gene, and we characterize the phenotypes of snf-11 null mutants.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The deletion mutations snf-11(ok156) and snf-3(ok293) were provided by the C. elegans Gene Knockout Consortium, and snf-11(tm625) was provided by Shohei Mitani (Tokyo Women’s Medical College, Tokyo, Japan). A strain containing a Pmyo-3::GFP transgene (ceEx64) was provided by Ken Miller (Oklahoma Medical Research Foundation). Additional strains were provided by the Caenorhabditis Genetics Center (St. Paul, MN).
Isolation of snf-11 cDNAs
We used a PCR-based screening approach to identify and enrich for snf-11 cDNAs in the RB1 cDNA library. Nested primers were used to amplify a fragment of the snf-11 cDNA from the library. The "titer" of the cDNA in the library was then determined by dilution and PCR amplification. The library was then diluted accordingly and used to infect host bacteria in a 96-well microtiter tray. Phage were then harvested, pooled, and analyzed by a second round of PCR. Repeated rounds of PCR amplification, dilution of phage, and phage growth achieved significant enrichment of the target cDNA. Sequencing of the ends of the cDNA indicated that the insert was nearly full length, missing only six base pairs of sequence at the 5' end (determined by 5' rapid amplification of cDNA ends). A primer was synthesized with the missing 5' sequences and used with a reverse primer in the 3'-untranslated region to amplify the full-length cDNA (using the Expand High Fidelity PCR kit; Roche Diagnostics, Indianapolis, IN).
snf-3 cDNA
We obtained two putative snf-3 cDNA clones from Yuji Kohara (National Institute of Genetics, Mishima, Japan): yk802e9 contains a presumed missense mutation (A1979G) that converts Q660R, and yk39f1 is a partial cDNA that starts 13-base pairs into exon 2. A 1-kb BamHI fragment from yk802e9 was inserted into the BamHI site of yk39f1 to yield a full-length snf-3 cDNA.
Expression and Reporter Constructs
Modular C. elegans expression vectors (Fire et al., 1990) and GFP-containing reporter vectors (Miller et al., 1999) were obtained from Andy Fire (Stanford University School of Medicine, Stanford, CA). Transformation experiments involved injecting either plasmids (with "RM#" designations) or PCR fusion products (with "FRM" designations). PCR fusion products (Horton et al., 1989; Hobert, 2002) were generated using PCR products amplified from either green fluorescent protein (GFP)- or cDNA-containing plasmids, fused to PCR products derived from the snf-11 promoter region of cosmid T03F7.
The promoter for Punc-25::GFP and Punc-25::YFP plasmids was derived from pSC325 (Jin et al., 1999). Transgenic animals carrying either Punc-25::GFP or Punc-25::YFP robustly express GFP/yellow fluorescent protein (YFP) in the RMEs and ventral nerve cord GABAergic cells, but not detectably in the AVL, DVB, and RIS cells. In this respect, our constructs resemble pSC380 and similar constructs (Jin et al., 1999). Jin et al. (1999) note that extensive RNA processing may be necessary for expression of unc-25 reporters in AVL, DVB, and RIS.
Transgenic Methods
Transgenic nematodes were obtained by microinjection of DNA, essentially as described by Mello et al. (1991). Transformation markers included the pBX plasmid (Heinke and Ralf Schnabel, Max-Plank-Institute fur Biochemie, Martinsried, Germany), which rescues the temperature-sensitive lethality of pha-1(e2123) mutants, and the RM#611p plasmid, which contains the unc-25 promoter driving expression of GFP. In some experiments, GFP fusions were used as transformation markers.
Xenopus Oocyte Preparation and Injection
Xenopus oocytes were prepared as described previously (Quick et al., 1992). Briefly, oocytes were defolliculated in collagenase A (Roche Diagnostics) and maintained at 18°C in ND96 (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES, and 1.8 mM CaCl2) supplemented with 2.5 mM Na pyruvate, 50 µg/ml gentamicin, and 5% horse serum. The medium was changed daily. Increased expression of SNF-11 protein was achieved by inserting the alfalfa mosaic virus 5'-untranslated region upstream of the start codon, and a poly(50)-A tail downstream of the stop codon. cRNA synthesis was performed using mMessage mMachine (Ambion, Austin, TX). Individual oocytes were injected with 50 ng of cRNA twice: 1 and 4 d after the oocyte preparation. Expression was found to be maximal 4–9 d after the second injection.
[3H]GABA Uptake Assays
Individual oocytes were placed in multiwell plates in ND96. At timed intervals, GABA was added to a final concentration of 1 mM unless otherwise indicated; the concentration of [3H]GABA was 60 nM. The reaction was terminated by rapid removal of the oocyte followed by six washes in ice-cold ND96. Subsequently, the oocyte was solubilized in 10% SDS at 60°C for 2 h, and [3H]GABA uptake was determined by liquid scintillation spectrometry. For all uptake experiments, specific uptake at each concentration was determined by subtracting total uptake from nonspecific uptake measured in sister oocytes from the same oocyte batch.
Behavioral Assays
Swimming (thrashing), pharyngeal pumping, and defecation were all measured on young adult hermaphrodites raised at 20°C as described previously (Miller et al., 1996). Acute aldicarb assays were performed as described previously (Lackner et al., 1999; Nurrish et al., 1999) using 2 mM aldicarb. Paralysis was defined as the complete absence of spontaneous and provoked (prodded with a worm pick) locomotion. GABA transport-dependent behavioral assays were performed as described by McIntire et al. (1993b), except that the final concentration of GABA was 70 mM.
Generation and Purification of Polyclonal Antibodies
-SNF-11 antibodies were raised against a synthetic peptide (NH2-CRPDIEIKGAMENAEN-COOH) corresponding to a predicted antigenic region in the carboxyl tail of SNF-11. The peptide was coupled to goat serum albumin with glutaraldehyde and injected into two goats (G379 and G380). Polyclonal antibodies were affinity purified against SNF-11 peptide (coupled to rabbit serum albumin) immobilized on nitrocellulose membranes as described previously (Duerr et al., 1999).
Immunofluorescence Staining and Microscopy of Nematodes
Nematodes were stained using a freeze-fracture procedure adapted from Duerr et al. (1999), with major modifications by Kiely Grundahl (Oklahoma Medical Research Foundation). Nematodes were freeze fractured (with no subbing agent) and then washed from slides in ice-cold methanol. Nematodes were pelleted in 50-ml conical tubes, transferred to microfuge tubes, and then pelleted again. Nematodes were then fixed in acetone for 2 min, washed in phosphate-buffered saline (PBS), and then blocked with 10% donkey serum in antibody buffer (0.5% Triton X-100, 1 mM EDTA, and 0.1% bovine serum albumin in PBS with 0.05% sodium azide). Primary antibody incubations (1:50–1:100) were done overnight at 4°C. After thorough washing with antibody buffer, nematodes were incubated in secondary antibodies overnight at 4°C. Secondary antibodies were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA). After washing, nematodes were mounted in anti-fade medium.
Primary Cell Culture
Embryonic cells were isolated and cultured in a procedure modified from Christensen et al. (2002). Briefly, worms were grown in 2xYT liquid media with Escherichia coli (OP50 strain) and harvested by gentle centrifugation. Eggs were released by agitating worms in a bleach/NaOH solution, chitinased, and passed through a 5-µm filter in an osmotically balanced egg buffer. Dissociated cells were plated onto peanut lectin-coated coverslips at a density of 
1 x 106 cells/ml and allowed to differentiate and grow at 20°C in L-15–10 media for 7–10 d. In GABA uptake experiments, the growth media were replaced with fresh media containing 840 µM GABA, and cells were incubated for 1 h at 23°C. For immunofluorescence, coverslips were washed in egg buffer, and fixed immediately with 1% formaldehyde, 0.20% glutaraldehyde in 0.1 M PIPES for 10 min at 23°C. After washing in PBS, cells were incubated 60–90 min at 23°C in a PBS-based blocking buffer containing 0.5% porcine gelatin, 0.01 M glycine, 0.5% Tween 20, 0.05% sodium azide, 1 mM EDTA, and 10% donkey serum. Primary antibody staining was at 23°C for 45–60 min with rabbit -GABA (Sigma-Aldrich, St. Louis, MO), and mouse and chicken -GFP antibodies (Chemicon International, Temecula, CA) diluted 1:1000 in the blocking buffer without donkey serum. After three PBS washes, cells were incubated at 23°C for 45–60 min in a secondary antibody solution consisting of Cy3-conjugated donkey -rabbit F(ab')2 and Cy2-conjugated donkey -mouse and -chicken F(ab')2 (The Jackson Laboratory, Bar Harbor, ME), each diluted 1:1000 in the above-mentioned buffer. Coverslips were then mounted on slides with ProLong anti-fade medium (Invitrogen, Carlsbad, CA).
Microscopy and Imaging
Images were collected on a Leica TCS NT confocal microscope. Nematode images were collected at 10 or 40x magnification, and 512 x 512 or 1024 x 1024 pixels with 0.5-µm Z-steps. Cell culture images were collected at 100x magnification with 2x zoom and 512 x 512 pixels with 0.2-µm Z-steps. Images were cropped to size, assembled, and annotated using Adobe Photoshop 5.5 (Adobe System, Mountain View, CA). All images within a given experiment were collected in the same session using identical settings and were processed identically for publication.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). However, reporter expression studies indicated that transcription of srh-45 requires only the small (383 base pairs) region just upstream of srh-45 and is not affected by any sequences within or upstream of the snf-11 coding sequence; thus, snf-11 does not seem to be part of an operon.
|
|
-alanine, betaine, or taurine, indicating that these compounds are not substrates.
|
-SNF-11 antibody strongly stained the four RME neurons (RMED, RMEV, RMEL, and RMER; Figure 4). Faint staining of three additional GABAergic neurons (AVL, DVB, and RIS; Table 1) was sometimes observed. All seven of these neurons have been shown previously to possess GABA uptake activity (McIntire et al., 1993b). Several non-GABAergic neurons, including RID, also seemed to stain. We observed staining of both the processes and the soma of each neuron. In RMED and RMEV, a punctate staining pattern was observed in the posteriorly directed processes, possibly corresponding to synapses (Figure 4). The ventral nerve cord DD and VD inhibitory motor neurons did not stain. Faint staining of the body wall, anal, and uterine muscles with the -SNF-11 antibody was observed in some animals. In snf-11(ok156) mutants, -SNF-11 staining was completely absent, confirming the specificity of the staining (Figure 4).
|
|
-SNF-11 polyclonal antibodies. We also noted expression in two additional neurons near the pharynx as well as two neurons in the retrovesicular ganglion. There were no apparent differences in expression between the two reporters, suggesting that the 1.9-kb region is sufficient to drive expression in all snf-11–positive cells. In contrast to observations reported previously (Jiang et al., 2005), we did not observe snf-11 expression in the ventral cord inhibitory (DD and VD) motor neurons (Figure 5). However, we did observe robust expression in the body wall, anal, and uterine muscles that was not noted previously (McIntire et al., 1993b; Jiang et al., 2005). In young animals, expression of the Psnf-11::GFP reporter in muscle cells is the most prominent aspect of the expression pattern (Figure 5A).
|
). To our surprise, we found that a snf-11 null mutation had no dramatic effects on any of these GABAergic behaviors. Thrashing rates in liquid were not significantly different from those of wild-type (N2) animals [snf-11(ok156) = 163 ± 2.4, N2 = 160 ± 3.6 (thrashes per minute)]. Likewise, the enteric muscle contraction step of the defecation motor program was normal [snf-11(ok156) = 99 ± 1%, N2 = 92 ± 3%], although there was a small but statistically significant difference between mutant and wild-type animals. The foraging behavior seemed to be completely normal in snf-11 mutants.
Because the snf-11 null mutation did not result in a dramatic GABAergic phenotype, we used a pharmacological approach to look at more subtle defects. C. elegans mutants deficient in GABA are hypersensitive to the effects of the acetylcholinesterase inhibitor aldicarb, presumably because GABA normally antagonizes the action of acetylcholine (ACh) on the body wall muscles. A priori, we expected that snf-11 mutants would be at least slightly resistant to aldicarb because the elevated levels of extracellular GABA would offset the effects of aldicarb to some degree. Therefore, we used an aldicarb acute response assay (Lackner et al., 1999; Nurrish et al., 1999) to monitor the response of snf-11 mutants to 2 mM aldicarb. We found that snf-11 mutants are resistant to aldicarb (Figure 6). This resistance was dependent on endogenous GABA synthesis, however, because unc-25; snf-11 double mutants were as hypersensitive as unc-25 single mutants.
|
to create a contrived "GABA transport dependent" phenotype. As noted above, unc-25(null) mutants are defective in GABA-mediated behaviors, including defecation and foraging. However, application of exogenous GABA rescues the defecation and foraging defects in unc-25 mutants. Behavioral rescue is not only dependent on GABA but also on GABA transport—treatment with the GAT inhibitor nipecotic acid blocks behavioral rescue (McIntire et al., 1993b). Similarly, we found that the snf-11(ok156) mutation blocks GABA-mediated rescue of the unc-25 defecation defect (Figure 7). The snf-11 mutation also blocked rescue of the RME-mediated foraging defect. A null mutation in the second potential GABA transporter gene, snf-3, had no effect on GABA transport in this assay (Figure 7), suggesting that snf-11 encodes the sole GABA transporter in the defecation and foraging neurons.
|
GABA Uptake in C. elegans Embryonic Cell Culture Is Blocked by a snf-11 Null Mutation
Earlier studies on GABA uptake in C. elegans relied on whole-worm treatment and staining techniques; worms were exposed to high levels of exogenous GABA, followed by fixation and staining with -GABA antibodies (McIntire et al., 1993b). This technique has provided valuable insights into GABA uptake in C. elegans, but it has several limitations that reduce its usefulness. The C. elegans cuticle is relatively impermeable; consequently, an extremely high level of GABA is required to detect uptake. The cuticle also makes staining with antibodies technically difficult. To overcome these limitations, we took advantage of the newly developed C. elegans embryonic cell culture techniques to study GABA uptake. We used this approach to demonstrate GABA uptake in a variety of cell types in cell culture using physiological concentrations of GABA. In combination with C. elegans mutations affecting the GABA biosynthesis and uptake, this approach provides a valuable methodology for the study of GABA uptake.
Cells derived from unc-25(e156) (GAD) mutants expressing an integrated Punc-25::GFP transcriptional fusion were prepared using the methods described by Christensen et al. (2002). unc-25(e156) mutants do not synthesize detectable levels of GABA; consequently, it is possible to examine the uptake of GABA without the confounding presence of endogenous GABA. These cells were exposed to 840 µM GABA for 1 h, rinsed briefly, and then stained with an -GABA antibody. As noted in Materials and Methods, our Punc-25::GFP fusion is robustly expressed in the RMEs and ventral cord GABAergic cells but not detectably in the AVL, DVB, and RIS cells. In addition, the DVB and VD cells arise postembryonically and should not be present in our preparation. Strong -GABA staining was noted in several cell types (Figure 8), which fell into three categories: 1) GFP-positive neurons, 2) GFP-negative neurons, and 3) presumptive muscle cells. Although we did not make precise cell counts, -GABA staining was absent from approximately one-half of the GFP-positive neurons. We believe that the GABA-positive GFP-positive neurons are RMEs, based on the expression patterns described above for the Psnf-11 and Punc-25::GFP reporters in vivo. The GABA-positive GFP-negative neurons are presumably AVL, RIS, and non-GABAergic neurons (i.e., RID) that express snf-11. The GFP-positive GABA-negative cells are presumably ventral cord DD GABAergic cells, based on the absence of snf-11 expression in these cells in vivo.
|
-GABA antibodies. These cells also carry a Pmyo-3::GFP reporter (as an extrachromosomal array; ceEx64) that is expressed in all body wall muscle cells, and -GFP antibodies were included as staining controls. We found that the snf-11 mutation blocks all significant GABA uptake activity in the cultured embryonic cells, including body wall muscle cells (Figure 8). Therefore, the snf-11 gene seems to encode the sole high-affinity GABA transporter in C. elegans. Furthermore, our observations that muscles cells possess GABA uptake activity and that a snf-11 mutation blocks this uptake, confirm our observation of snf-11 expression in these cells. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). These GATs are expressed differentially in the rat nervous system and have distinct physiological and pharmacological properties. We found that SNF-11 is inhibited by the GAT1-specific antagonist SKF89976A and is not inhibited by GAT2/3-specific antagonist -alanine, indicating that SNF-11 belongs to the GAT1-subtype of GABA transporters. We also present evidence that snf-11 encodes the only high-affinity GABA transporter in C. elegans. There are two genes in C. elegans, snf-3 and snf-11, that encode proteins with significant similarity to mammalian plasma membrane GABA transporters. SNF-3 is 38% identical to GAT1, whereas SNF-11 is 46% identical to GAT1. In this study, we demonstrate that SNF-11 is expressed in all cells that possess significant GABA uptake activity. In contrast, the SNF-3 protein is not expressed in cells possessing GABA uptake activity; SNF-3 is primarily expressed in sheath and socket cells in the head, and epithelial cells surrounding the vulva (our unpublished data). Furthermore, a snf-11 null mutation seems to block all significant GABA uptake in a C. elegans primary cell culture paradigm. Finally, expression of the snf-3 cDNA under control of the snf-11 promoter does not rescue the snf-11 mutant phenotypes.
As noted above, there are four subtypes of GABA transporters in rats; these subtypes are each encoded by separate genes. In contrast, both C. elegans and Drosophila melanogaster possess a single GABA transporter gene. In C. elegans, the GABAergic nervous system is very simple, consisting of 26 cells capable of synthesizing GABA (Schuske et al., 2004), and mediates a simple repertoire of behaviors. Presumably, the existence of multiple GAT subtypes in mammals reflects the complexity of the GABAergic nervous system in these organisms.
Comparison of the SNF-11 sequence to mammalian GAT1 sequences (Supplemental Figure 1) reveals similarities and differences that may be relevant to transporter function and regulation. The N-terminal cytoplasmic domain of GAT1 is known to regulate transporter function through interaction with syntaxin 1A (Beckman et al., 1998; Deken et al., 2000), and by interaction with the GAT1 fourth intracellular loop (Hansra et al., 2004). The SNF-11 N-terminal cytoplasmic domain is much shorter than that of GAT1 (18 versus 52 amino acids), yet it shares some features with the region proximal to the GAT1 first transmembrane domain (TM I). This region of GAT1 binds to syntaxin, and a cluster of three aspartic acid residues mediates this binding (Deken et al., 2000). Alignment of SNF-11 and GAT1 indicates that SNF-11 does not have the three aspartic acids, but rather two glutamic acid residues at comparable positions (Supplemental Figure 1), which may be sufficient to mediate syntaxin binding.
The fourth intracellular loop of GAT1 (between TM VIII and TM IX) is involved in intramolecular interactions and regulation of transport kinetics (Hansra et al., 2004). In this region of the protein, SNF-11 is 17/29 identical (23/29 similar) to GAT1. In particular, there is a cluster of basic amino acids, just proximal to the ninth transmembrane domain, that is required for the interaction with the GAT1 N-terminal domain (Hansra et al., 2004); in this region, there are four arginine residues in GAT1, versus three arginines and two lysines in SNF-11. Therefore, it is likely that regulatory interactions ascribed to the GAT1 fourth intracellular loop are conserved in SNF-11.
The C-terminal cytoplasmic tail of rat GAT1 contains two important localization motifs: a hydrophobic VMI tripeptide (VMV in the human protein), and the C-terminal AYI, which is a type II PDZ domain interaction motif (Farhan et al., 2004). We note that SNF-11 contains neither of these motifs.
snf-11–deficient Phenotypes
The loss of snf-11 function in C. elegans does not result in a dramatic mutant phenotype. This observation is consistent with previous observations made on worms treated with the GAT inhibitor nipecotic acid (McIntire et al., 1993b), which likewise exhibited no obvious behavioral abnormalities. However, we were able to demonstrate phenotypic differences between wild-type and snf-11 mutant animals using two pharmacobehavioral assays: 1) a GABA-dependent defecation assay and 2) an acute aldicarb assay.
The GABA-dependent defecation assay provides a means of detecting GABA uptake activity in the defecation neurons, AVL and DVB. Foraging behavior, which is modulated by the RMEs, can be assayed simultaneously, but is less amenable to quantification. We found that GABA uptake in both defecation and foraging neurons is completely dependent on snf-11 activity. The defecation assay, in particular, could provide a convenient means of identifying additional snf-11 mutations.
Aldicarb resistance is usually interpreted as a measure of ACh release (presynaptic) or response (postsynaptic); however, because GABA antagonizes the action of ACh, GABA levels may contribute to resistant or sensitive phenotypes. This is clearly seen in the case of unc-25 (GAD) mutants, which are hypersensitive to aldicarb. GAT1-deficient mice have elevated levels of extracellular GABA, resulting in increased activation of post-synaptic GABAA receptors (Jensen et al., 2003). Based on this observation, we expected that snf-11 mutants also would have elevated levels of extracellular GABA. We found that snf-11 mutants are resistant to aldicarb in an acute assay (Figure 6). This resistance requires GABA synthesis and release, however, because unc-25; snf-11 double mutants are as hypersensitive as unc-25 single mutants (Figure 6).
The loss of GAT1 function in mice results in significant neurological defects, including tremors, ataxia, and anxiety (Chiu et al., 2005). The reduced GABA uptake activity results in elevated extracellular GABA levels, which in turn result in an increased GABAA receptor-mediated tonic conductance in cerebellar granule and Purkinje cells (Chiu et al., 2005; also Jensen et al., 2003). Jensen et al. (2003) also report that GAT1-deficient mice lack a presynaptic GABAB-mediated conductance. Clearly, the loss of GAT1 activity in mice has significant effects on GABAergic signaling and, consequently, on GABAergic behaviors.
In contrast, the loss of GAT activity in worms has no major behavioral consequences, suggesting that these animals are not dramatically hyper-GABAergic. Perhaps because of the structure and geometry of GABAergic synapses in C. elegans, diffusion seems to be adequate to reduce extracellular GABA to tolerable levels. In addition, GABA is the most abundant inhibitory neurotransmitter in the mouse central nervous system and modulates a wide range of behaviors. In contrast, GABA is primarily used at neuromuscular synapses in C. elegans, and its action is counteracted by excitatory cholinergic input. These differences may also account for the relative subtlety of the snf-11 mutant phenotypes compared with those of the mouse GAT1 knockout.
Expression and Localization of SNF-11 in C. elegans
Our results indicate that all GABAergic synapses are associated with either presynaptic or postsynaptic (or both) expression of SNF-11. The SNF-11 protein was detected in both neuronal cell bodies and processes and was concentrated at presumptive synapses in some cells (RMED and RMEV). In mammals, the four GAT subtypes are differentially expressed and localized, although there is more overlap in expression than previously realized. Until recently, expression studies suggested that GAT1 is exclusively expressed in neurons. However, recent studies show that GAT1 is expressed in both neurons and glial cells. There is strong evidence for expression of GAT3, GAT1, and GAT2 (in order of abundance) in glial cells in the mature neocortex, and GAT1 and GAT2 in neurons. However, only GAT1 seems to have a synaptic localization pattern (Conti et al., 1998; Minelli et al., 2003; Chiu et al., 2002). The apparent localization of SNF-11 to synapses in a subset of neurons is consistent with the localization of the GAT1 subtype of GABA transporter.
The absence of snf-11 expression in the ventral cord inhibitory (DD and VD) motor neurons is puzzling. The ventral cord inhibitory motor neurons are required for coordinated movement, and they function in opposition to the cholinergic excitatory motor neurons. SNF-11 is expressed in the postsynaptic muscle cells, and this muscle expression presumably provides the GABA clearance function at these neuromuscular synapses. We suggest that the expression of plasma membrane GABA transporters in the ventral cord inhibitory motor neurons either may confer no energetic advantage, or it may be incompatible with their function. Recent studies on the dopamine transporter in C. elegans suggest that the channel characteristics of neurotransmitter transporters may contribute significantly to the electrical properties of cells (Carvelli et al., 2004). Possibly, the channel properties of SNF-11 would compromise the conductive properties of the ventral cord inhibitory motor neurons.
Discrepancies with a Previous Study on snf-11
We noted several discrepancies between our results and those reported recently by Jiang et al. (2005). We found that snf-11 is expressed in a subset of GABAergic neurons as well as the body wall, anal, and uterine muscles. In contrast to Jiang et al. (2005), we did not observe expression in the ventral nerve cord inhibitory motor neurons. However, we note that the reporter construct described by Jiang et al. (2005) is missing the 752-base pair proximal-most portion of the regulatory region described in this study, which may account for the observed differences in expression. Several additional lines of evidence also support our observations. We and other investigators (McIntire et al., 1993b) have been unable to demonstrate uptake of exogenous GABA into the ventral cord inhibitory (DD and VD) motor neurons. The homologous neurons in Ascaris (DI and VI) also lack GABA uptake activity (Guastella et al., 1991). Although previous studies (McIntire et al., 1993b) found no evidence of GABA uptake into muscles in C. elegans, our observations are consistent with the results of GABA uptake experiments in Ascaris (Guastella et al., 1991).
We also report phenotypic difference between our snf-11 null mutants and the partial RNAi knockdown reported by Jiang et al. (2005). The snf-11 null mutants have no obvious phenotype, and the absence of GABA uptake activity seems to have little effect on GABAergic behaviors. In contrast, Jiang et al. (2005) reported that RNA interference (RNAi) knockdown leads to movement and defecation defects. Although it is possible that snf-11 mutants compensate during their development for the absence of snf-11 function, it should be noted that exposure to GAT inhibitors does not result in behavioral phenotypes such as those observed by Jiang et al. (2005) but rather results in phenotypes identical to those of the snf-11 null mutants. Furthermore, the RNAi phenotype reported by Jiang et al. (2005) is precisely the opposite of what one would expect from an excess of GABAergic signaling. Because eliminating GABAergic function results in constipation (McIntire et al., 1993a,b), it is surprising that Jiang et al. (2005) observed constipation in snf-11 knockdown animals, which ought to have excess GABAergic signaling. Likewise, Jiang et al. (2005) report that snf-11 RNAi leads to aldicarb hypersensitivity, which is a phenotype associated with reduced GABAergic signaling. In contrast, we observed that snf-11 null mutants are aldicarb resistant, which is consistent with increased GABA function. Clearly, the snf-11 knockouts provide a better means of assessing the phenotypic effects of eliminating GABA uptake activity.
Neurotransmitter transporters potentially function in both the clearance and recycling of neurotransmitters from synaptic clefts. The results described by Jiang et al. (2005) strongly imply that the recycling function is more critical for GABAergic neurotransmission in C. elegans because snf-11 RNAi results in a GABAergic "loss-of-function" phenotype. Alternatively, Jiang et al. (2005) suggest that GABA receptors have been down-regulated; however, other studies fail to support this view (Gally and Bessereau, 2003). In contrast, our results clearly indicate that the GABA transporter in C. elegans functions chiefly in the clearance of GABA. Not only do snf-11 null mutants lack a discernible GABA loss-of-function phenotype, but our results also demonstrate that these mutants are, indeed, hyper-GABAergic.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org). 
Address correspondence to: James B. Rand ( james-rand{at}omrf.ouhsc.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Beckman, M. L., Bernstein, E. M., Quick, M. W. (1998). Protein kinase C regulates the interaction between a GABA transporter and syntaxin 1A. J. Neurosci 18, 6103–6112.
Beg, A. A. and Jorgensen, E. M. (2003). EXP-1 is an excitatory GABA-gated cation channel. Nat. Neurosci 6, 1145–1152.[CrossRef][Medline]
Borden, L. A. (1996). GABA transporter heterogeneity: pharmacology and cellular localization. Neurochem. Int 29, 335–356.[CrossRef][Medline]
Brenner, S. (1974). The genetics of Caenorhabditis elegans. Genetics 77, 171–94.
Carvelli, L., McDonald, P. W., Blakely, R. D., Defelice, L. J. (2004). Dopamine transporters depolarize neurons by a channel mechanism. Proc. Natl. Acad. Sci. USA 101, 16046–16051.
Chiu, C. S., Brickley, S., Jensen, K., Southwell, A., Mckinney, S., Cull-Candy, S., Mody, I., Lester, H. A. (2005). GABA transporter deficiency causes tremor, ataxia, nervousness, and increased GABA-induced tonic conductance in cerebellum. J. Neurosci 25, 3234–3245.
Chiu, C. S., et al. (2002). Number, density, and surface/cytoplasmic distribution of GABA transporters at presynaptic structures of knock-in mice carrying GABA transporter subtype 1-green fluorescent protein fusions. J. Neurosci 22, 10251–10266.
Christensen, M., Estevez, A., Yin, X., Fox, R., Morrison, R., McDonnell, M., Gleason, C., Miller, D. M., Strange, K. (2002). A primary culture system for functional analysis of C. elegans neurons and muscle cells. Neuron 33, 503–514.[CrossRef][Medline]
Conti, F., Melone, M., De Biasi, S., Minelli, A., Brecha, N. C., Ducati, A. (1998). Neuronal and glial localization of GAT-1, a high-affinity gamma-aminobutyric acid plasma membrane transporter, in human cerebral cortex: with a note on its distribution in monkey cortex. J. Comp. Neurol 396, 51–63.[CrossRef][Medline]
Deken, S. L., Beckman, M. L., Boos, L., Quick, M. W. (2000). Transport rates of GABA transporters: regulation by the N-terminal domain and syntaxin 1A. Nat. Neurosci 3, 998–1003.[CrossRef][Medline]
Deken, S. L., Beckman, M. L., Quick, M. W. (2001). PICKing on transporters. Trends Neurosci 24, 623–625.[CrossRef][Medline]
Duerr, J. S., Frisby, D. L., Gaskin, J., Duke, A., Asermely, K., Huddleston, D., Eiden, L. E., Rand, J. B. (1999). The cat-1 gene of Caenorhabditis elegans encodes a vesicular monoamine transporter required for specific monoamine-dependent behaviors. J. Neurosci 19, 72–84.
Farhan, H., Korkhov, V. M., Paulitschke, V., Dorostkar, M. M., Scholze, P., Kudlacek, O., Freissmuth, M., Sitte, H. H. (2004). Two discontinuous segments in the carboxyl terminus are required for membrane targeting of the rat gamma-aminobutyric acid transporter-1 (GAT1). J. Biol. Chem 279, 28553–28563.
Fire, A., Harrison, S. W., Dixon, D. (1990). A modular set of lacZ fusion vectors for studying gene expression in Caenorhabditis elegans. Gene 93, 189–198.[CrossRef][Medline]
Gally, C. and Bessereau, J. L. (2003). GABA is dispensable for the formation of junctional GABA receptor clusters in Caenorhabditis elegans. J. Neurosci 23, 2591–2599.
Guastella, J., Johnson, C. D., Stretton, A.O.W. (1991). GABA-immunoreactive neurons in the nematode Ascaris. J. Comp. Neurol 307, 584–597.[CrossRef][Medline]
Guastella, J. and Stretton, A.O.W. (1991). Distribution of 3H-GABA uptake sites in the nematode Ascaris. J. Comp. Neurol 307, 598–608.[CrossRef][Medline]
Hansra, N., Arya, S., Quick, M. W. (2004). Intracellular domains of a rat brain GABA transporter that govern transport. J. Neurosci 24, 4082–4087.
Hobert, O. (2002). PCR fusion-based approach to create reporter gene constructs for expression analysis in transgenic C. elegans. Biotechniques 32, 728–730.[Medline]
Horton, R. M., Hunt, H. D., Ho, S. N., Pullen, J. K., Pease, L. R. (1989). Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77, 61–68.[CrossRef][Medline]
Jensen, K., Chiu, C. S., Sokolova, I., Lester, H. A., Mody, I. (2003). GABA transporter-1 (GAT1)-deficient mice: differential tonic activation of GABAA versus GABAB receptors in the hippocampus. J. Neurophysiol 90, 2690–2701.
Jiang, G., Zhuang, L., Miyauchi, S., Miyake, K., Fei, Y. J., Ganapathy, V. (2005). A Na+/Cl–-coupled GABA transporter, GAT-1, from Caenorhabditis elegans: structural and functional features, specific expression in GABAergic neurons, and involvement in muscle function. J. Biol. Chem 280, 2065–2077.
Jin, Y., Jorgensen, E., Hartwieg, E., Horvitz, H. R. (1999). The Caenorhabditis elegans gene unc-25 encodes glutamic acid decarboxylase and is required for synaptic transmission but not synaptic development. J. Neurosci 19, 539–548.
Lackner, M. R., Nurrish, S. J., Kaplan, J. M. (1999). Facilitation of synaptic transmission by EGL-30 Gq and EGL-8 PLC: DAG binding to UNC-13 is required to stimulate ACh release. Neuron 24, 335–346.[CrossRef][Medline]
McIntire, S. L., Jorgensen, E., Horvitz, H. R. (1993a). Genes required for GABA function in Caenorhabditis elegans. Nature 364, 334–337.[CrossRef][Medline]
McIntire, S. L., Jorgensen, E., Kaplan, J., Horvitz, H. R. (1993b). The GABAergic nervous system of Caenorhabditis elegans. Nature 364, 337–341.[CrossRef][Medline]
McIntire, S. L., Reimer, R. J., Schuske, K., Edwards, R. H., Jorgensen, E. M. (1997). Identification and characterization of the vesicular GABA transporter. Nature 389, 870–876.[CrossRef][Medline]
Mello, C. C., Kramer, J. M., Stinchcomb, D., Ambros, V. (1991). Efficient gene transfer in C. elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J 10, 3959–3970.[Medline]
Miller, D. M., Desai, N. S., Hardin, D. C., Piston, D. W., Patterson, G. H., Fleenor, J., Xu, S., Fire, A. (1999). Two-color GFP expression system for C. elegans. Biotechniques 26, 914–918 920–921.[Medline]
Miller, K. G., Alfonso, A., Nguyen, M., Crowell, J. A., Johnson, C. D., Rand, J. B. (1996). A genetic selection for Caenorhabditis elegans synaptic transmission mutants. Proc. Natl. Acad. Sci. USA 93, 12593–12598.
Minelli, A., Barbaresi, P., Conti, F. (2003). Postnatal development of high-affinity plasma membrane GABA transporters GAT-2 and GAT-3 in the rat cerebral cortex [erratum in Brain Res. Dev. Brain Res. (2003) 145, 167–168]. Brain Res. Dev. Brain Res 142, 7–18.[Medline]
Nass, R., Miller, D. M., Blakely, R. D. (2001). C. elegans: a novel pharmacogenetic model to study Parkinson’s disease. Parkinsonism Relat. Disord 7, 185–191.[CrossRef][Medline]
Nurrish, S., Segalat, L., Kaplan, J. M. (1999). Serotonin inhibition of synaptic transmission: Go decreases the abundance of UNC-13 at release sites. Neuron 24, 231–242.[CrossRef][Medline]
Quick, M. W., Naeve, J., Davidson, N., Lester, H. A. (1992). Incubation with horse serum increases viability and decreases background neurotransmitter uptake in Xenopus oocytes. Biotechniques 13, 357–361.[Medline]
Ranganathan, R., Sawin, E. R., Trent, C., Horvitz, H. R. (2001). Mutations in the Caenorhabditis elegans serotonin reuptake transporter MOD-5 reveal serotonin-dependent and -independent activities of fluoxetine. J. Neurosci 21, 5871–5884.
Saitou, N. and Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol 4, 406–425.[Abstract]
Schuske, K., Beg, A. A., Jorgensen, E. M. (2004). The GABA nervous system in C. elegans. Trends Neurosci 27, 407–414.[CrossRef][Medline]
Spieth, J., Brooke, G., Kuersten, S., Lea, K., Blumenthal, T. (1993). Operons in C. elegans: polycistronic mRNA precursors are processed by trans-splicing of SL2 to downstream coding regions. Cell 73, 521–532.[CrossRef][Medline]
Südhof, T. C. (2004). The synaptic vesicle cycle. Annu. Rev. Neurosci 27, 509–547.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  G. P. Mullen, E. A. Mathews, M. H. Vu, J. W. Hunter, D. L. Frisby, A. Duke, K. Grundahl, J. D. Osborne, J. A. Crowell, and J. B. Rand Choline Transport and de novo Choline Synthesis Support Acetylcholine Biosynthesis in Caenorhabditis elegans Cholinergic Neurons Genetics, September 1, 2007; 177(1): 195 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Thimgan, J. S. Berg, and A. E. Stuart Comparative sequence analysis and tissue localization of members of the SLC6 family of transporters in adult Drosophila melanogaster J. Exp. Biol., September 1, 2006; 209(17): 3383 - 3404. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||
