| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroscience Research Institute, AIST, Tsukuba, Ibaraki 305-8566, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionExperimental proceduresReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
The small GTPase proteins including Rab are modified by protein prenylation, which adds C15 farnesyl or C20 geranylgeranyl lipids to the C-terminus. This step is catalyzed by protein prenyltransferases (Casey & Seabra 1996; Shen & Seabra 1996). In the case of Rab proteins, prenylation transferring C20 geranylgeranyl lipids is absolutely critical for the proper function of Rab proteins in cellular processes. Without prenylation, Rab proteins are not correctly localized in target donor membranes. For this prenylation step, Rab proteins form a stable trimeric protein complex with Rab escort protein (REP) and Rab geranylgeranyltransferase (RabGGTase). REP binds to all newly synthesized Rab proteins and mediates their prenylation by presenting them to RabGGTase. (Andres et al. 1993). RabGGTase is a heterodimer enzyme composed of 
- and β-subunit. RabGGTase neither recognizes short peptides containing the Rab C-terminal prenylation motif, nor does it recognize the Rab protein alone (Seabra et al. 1992b; Anant et al. 1998). Instead, it binds REP. Although detailed molecular mechanisms for Rab protein prenylation have not been elucidated yet, recent biochemical analyses have led to the alternative two possible pathways for Rab protein modification. In the classical pathway, an unprenylated Rab protein first associates with REP and then Rab-REP complex is recognized by RabGGTase, which adds two geranylgeranyl moieties to the C-terminus of Rab protein without prior dissociation of REP (Thoma et al. 2001b,c). In the alternative pathway, REP first forms a complex with RabGGTase under the conditions in which their concentrations are relatively high compared with the Rab protein. This complex could associate with Rab protein (Thoma et al. 2001a). In both cases, following to the dissociation of RabGGTase from the complex, REP delivers the prenylated Rab protein to its target membrane (Alexandrov et al. 1994).
Although Rab prenylation by REP and RabGGTase is essential, a paradox of genetic hierarchy exists between these protein functions in vivo. In humans, the lack of functional Rep1 but not RabGGTase, results in choroideremia (CHM), which is X-linked slow-onset retinal degeneration that affects photoreceptors, retinal pigment epithelium and the choroid (Seabra et al. 1992a; Seabra 1996). Rab27a also seems to play an important role in triggering the degenerative process in CHM, suggesting the target for Rep1 and RabGGTase (Seabra et al. 1995). Because patients with CHM only experience age-related blindness, Rep2, a homolog of Rep1, appears to effectively compensate for the loss of Rep1 in all tissues except the eye. Furthermore, a mutation in zebrafish Rep1 results in a 90% reduction in hair-cell number and partial retinal degeneration (Starr et al. 2004). However, other obvious defects are not observed until 5-days post-fertilization in spite of the lack of Rep2 ortholog in zebrafish. No clear explanation exists why such tissue-specific roles occur in each REP proteins, and there is a large gap between a proposed REP protein importance and mutational phenotypes. These diseases and disease models in the vertebrates also lead to the question of whether REP is really essential for several Rab functions in vivo.
To obtain molecular and genetic insights into how this Rab/REP/RabGGTase complex is organized and is functioning in vivo, as well as the genetic hierarchy among components, we used the free-living nematode C. elegans, which is useful for molecular analyses and has abundant known genetic data. The C. elegans genome encodes 29 predicted Rab proteins and one each of the RabGGTase - and β-subunits. However, no REP homologue has yet been reported. Among the Rab proteins, RAB-3 and RAB-27 are known to be exocytic Rabs. Both Rabs are widely expressed in the nervous system, are co-localized with synaptic sites, and regulate synaptic transmission (Nonet et al. 1997; Mahoney et al. 2006). Mutations in both genes cause decreased transmitter release and both mutants show resistance to aldicarb. Genetic analyses suggest that these two Rab proteins function in a parallel pathway for synaptic vesicle release. In addition, only rab-27 (referred to as aex-6) mutants have severe defects in defecation behaviors, suggesting other functions or sites of action of RAB-27 (Thomas 1990; Mahoney et al. 2006). In this study, we identified a C. elegans rep-1 gene and found that C. elegans has only one type of REP, which is more similar to mammalian Rep2 than Rep1. Our behavioral analyses using both the rep-1 mutant and rep-1 RNAi-treated animals suggest that C. elegans REP-1 is required in specific Rab pathways and in specific tissues. We also showed that several Rab proteins could be correctly localized to their target membranes without REP-1. These findings indicate a possible mechanism on the formation of the Rab/REP/RabGGTase complex and the mechanism of Rab prenylation and localization in vivo.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Exogenously applied melatonin directly modulates neuronal activity and locomotion behavior in C. elegans (Tanaka et al. 2007). When wild-type (N2) animals are placed on melatonin assay plates containing 100 µM melatonin, their locomotion rates decrease significantly to approximately one-half of the rates on control plates (Fig. 1). To understand the underlying molecular mechanism for this phenomenon, we carried out genetic screening to isolate mutants resistant to exogenous melatonin. From this screening we isolated the ta208 mutant, which has a strong resistance to exogenous melatonin: the mutant does not exhibit any decrease in locomotion rates on melatonin assay plates (Fig. 1). We suspected that the gene corresponding to ta208 might be involved in neuronal activity such as synaptic transmission. We therefore performed SNP-based genetic mapping to identify the corresponding gene. Based on the melatonin-resistant and small progeny number phenotypes observed in the ta208 mutant (see below), we successfully mapped ta208 within three cosmids (M01G5, Y67D2 and Y22D7AL) on chromosome III. We next performed direct sequencing of the predicted genes located on these cosmids and found a mis-sense mutation in the Y67D2.1 gene (Fig. 2). To confirm that Y67D2.1 really corresponds to ta208, the rescue ability using wild-type genomic DNAs was examined (see Fig. 2A for rescue constructs). In all of the defective phenotypes, melatonin resistance (Fig. 1), small progeny number (Fig. 4A), aldicarb resistance (Fig. 3C), defecation defects (data not shown), were partially or fully rescued to wild-type levels. We concluded that ta208 is an allele of Y67D2.1. The full-length cDNA of the Y67D2.1 gene was isolated from an yk clone (yk1365) and an RT-PCR product. Although four splicing variants are suggested in the Y67D2.1 gene (Wormbase, <http://www.wormbase.org>), only a single transcript corresponding to Y67D2.1a was isolated in our RT-PCR analyses. We also found that the splice leader 2 (SL2) is attached to the 5' end of the Y67D2.1a transcript. This suggests that Y67D2.1 gene may form a polycistronic gene complex with the former gene Y67D2.2.
|
|
|
|
The Y67D2.1a gene contains a 1533-nt open reading frame and encodes a 510-amino acids protein that is similar to a REP or a GDP/GTP dissociation inhibitor (GDI). REP proteins have been identified from many organisms including humans, mouse, zebrafish, and several other invertebrates (Alexandrov et al. 1994; Cremers et al. 1994; Starr et al. 2004). Because REP proteins are similar to GDI, the gene had not been identified as REP in the C. elegans genome, but had been described as a GDI. However, both the conserved 128th leucine (Fig. 2C, closed arrowhead) and 136th phenylalanine (Fig. 2C, arrow) in GDI proteins are not conserved in the predicted Y67D2.1a protein (Fig. 2C). In contrast, the Y67D2.1a protein has a phenylalanine at the 184th position (open arrowhead in Fig. 2C), which is highly conserved in the REP family and is absent from GDI proteins (Pylypenko et al. 2003). This suggests that the Y67D2.1 may encode a REP.
If the Y67D2.1 gene really encodes a C. elegans REP, the encoded protein should physically interact with Rab or RabGGTase, and the lipid modification of Rab proteins should be altered in the ta208 mutant animals. We examined protein–protein interaction by using the LexA yeast two-hybrid system and the lipid modification in Rab protein by western blotting analysis. Both the bait plasmid containing the full-length Y67D2.1 cDNA and each prey plasmid expressing RabGGTase subunit (M57.2, (Lackner et al. 2005)) or several Rab proteins were simultaneously introduced into yeast host cells, and the growth of transformed yeast on the plate lacking leucine was confirmed as positive interaction of two proteins. In six Rab proteins which we used in this study, the Y67D2.1 protein showed clear positive interaction with only RAB-5, RAB-7 and RAB-11 proteins (Fig. 3A). However, the Y67D2.1 protein did not show positive interaction with other Rab proteins such as RAB-3, RAB-27, RAB-10, and unexpectedly, with RabGGTase subunit. Interaction of Y67D2.1 product and these proteins may be quite weak in yeast cell, or somewhat toxic for yeast. This result, however, strongly suggests that the Y67D2.1 protein can physically interact with several Rabs in vivo.
Another characteristic property as a REP is that this protein helps Rab proteins to be modified by adding lipid moieties at their C-termini. Thus, lack of REP should increase the amount of unprenylated Rabs compared to prenylated Rabs. Using anti-RAB-27 antibodies, we examined the amount of prenylated and unprenylated RAB-27 protein in the worm lysates from the ta208 mutant animals and Y67D2.1 RNAi-treated animals (Fig. 3B). In the wild-type worm lysate, only prenylated RAB-27 band was detected in the Western blot. This prenylated RAB-27 was completely lost in the lysate from aex-6(sa24) mutant animals, a null mutant of rab-27 (Mahoney et al. 2006). Although only the prenylated RAB-27 was detected in the ta208 worm lysate, knockdown of Y67D2.1 gene by RNAi significantly increased the unprenylated RAB-27 level. Similar amount of unprenylated RAB-27 was also observed in the RabGGTase subunit-RNAi worms, suggesting that both proteins should be involved in the Rab modification pathway. We also found that much prenylated RAB-27 remained in both samples from the Y67D2.1 and RabGGTase RNAi-treated animals. This may result in incomplete knockdown of the gene products in RNAi-insensitive neuronal cells or easier breakdown of the unprenylated Rab protein in vivo.
The predicted C. elegans REP-1 protein shows approximately 30% identity (70% similarity) to both human and mouse Rep1 proteins in total amino acid length and contains all the conserved domains found in the REP protein family (Rasteiro & Pereira-Leal 2007). The ta208 mutation substitutes the 107th glutamate to lysine (E107K; Fig. 2B,D). This glutamate is well-conserved in Rep2 proteins in humans and mouse, but not in Rep1 of these species (Fig. 2D). Zebrafish Rep1 also has a conserved glutamate in this position. This ta208 mutation phenotypically causes several defects, but the amino acid change does not seem to affect largely on the biochemical properties of REP-1; physical interaction between the mutant REP-1(E107K) protein and several Rab proteins which interacted with wild-type REP-1 looks unchanged (Fig. 3A). Although the ta208 mutation may alter the interaction with RabGGTase or other Rab proteins, which was not detected in yeast two-hybrid analysis, the functional properties of this region are not well known. Our hypothesis is, however, that this glutamate should have an important role for REP function that may be specific to Rep2 in the vertebrates. Because we have not isolated another rep-1 allele, it is not clear whether this ta208 allele is a null mutation or not. However, both weaker resistance to aldicarb in the rep-1(ta208) mutant compared to rab-3 or aex-6 single mutants (see below) and less unprenylated RAB-27 compared to RNAi worms suggests that ta208 is probably a weak hypomorphic allele.
REP-1 regulates RAB-27 function in synaptic transmission, but not RAB-3 function
In C. elegans, RAB-3 and RAB-27 are neuronal Rab proteins that regulate synaptic transmission at neuromuscular junctions. Mutations in both genes cause a resistance to aldicarb, an inhibitor of acetylcholinesterase, suggesting decreased neurotransmitter release from presynaptic terminals (Nonet et al. 1997; Mahoney et al. 2006). Because REP protein regulates the localization and function of various Rab proteins through their geranylgeranylation, we examined aldicarb sensitivity in the ta208 mutant and performed genetic analyses to investigate whether REP-1 similarly regulates both RAB-3 and RAB-27 function. We first tested the extent to which steady-state acetylcholine release is affected in the rep-1(ta208) mutant animals and found that these mutant animals showed mild resistance to aldicarb (Fig. 4A). We confirmed that this resistant phenotype was retained by introducing a wild-type rep-1 genomic DNA (Fig. 4B), suggesting that steady-state acetylcholine release is decreased by the ta208 mutation. To examine if the mutant really has defects in presynaptic neurons, not in the post-synaptic muscles, we also tested for levamisole sensitivity in the ta208 mutant. Levamisole is a potent agonist of the C. elegans nicotinic Ach receptors. Because the ta208 mutant animals had normal or slightly enhanced sensitivity to levamisole (data not shown), we confirmed that the disruption of REP-1 function causes decreased neurotransmitter release from presynaptic terminals, similar to the rab-3 and aex-6 mutants.
However, such a weak phenotype in the rep-1 mutant caused us to question whether REP-1 really regulates both RAB-3 and RAB-27 (mutants is referring as aex-6) functions for synaptic transmission. The rab-3; aex-6 double mutants show a severe resistant phenotype to aldicarb compared with the single mutants, suggesting that two parallel RAB pathways function for Ach release at the C. elegans neuromuscular junctions (Mahoney et al. 2006). To test whether REP-1 regulates one or both pathways, we generated double mutants of rep-1 and rab-3 or aex-6 mutants and examined their aldicarb sensitivity. Surprisingly, only the rab-3(js49); rep-1(ta208) double mutant animals showed increased resistance compared with single rab-3 mutant animals, whereas no significant difference was observed between the aex-6; rep-1 double mutant and aex-6 single mutant (Fig. 4A). These results suggest that aex-6 (rab-27) and rep-1 function in the same genetic pathway for synaptic transmission, whereas rab-3 and rep-1 may function in parallel pathways in the context of synaptic transmission. Although REP-1 did not interact with both RAB-3 and RAB-27 in yeast cells, these results suggest that REP-1 specifically regulates RAB-27 function via lipid modification, but that RAB-3 may not require REP-1 for its function and localization in the C. elegans synaptic transmission.
The rep-1 mutant has defects in gonad development and progeny number
In addition to aldicarb resistance, we found that rep-1(ta208) mutant animals produce many lethal embryos and have a reduced number of progeny compared to wild-type animals. At 20 °C, a single wild-type animal produces approximately 300 progeny on average, whereas a rep-1(ta208) mutant had only less than 100 progeny, which is a significantly lower reproductive rate (Fig. 5A). Because this low reproduction has not been found in rab-3, aex-6, or their double mutants, rep-1 is involved in another Rab protein pathway concerning reproduction. Interestingly, only the rep-1 mutant and rep-1(RNAi) animals have decreased, but a certain number of progeny compared with the complete sterility in RabGGTase RNAi-treated animals. This may also suggest a lower requirement of REP-1 for Rab protein modification by geranylgeranylation (see below). In addition, the rep-1(ta208) mutant occasionally shows abnormal germ-line development and gonad morphology (Fig. 5C,D). In some severely affected worms, gonads did not develop at all or small traces of gonads were located around the vulva. This may be a cause of low reproduction in rep-1 mutants.
|
RAB-27, which was originally isolated as a defecation defective mutant aex-6 (Thomas 1990), also regulates the defecation behavior of C. elegans. Defecation behavior in C. elegans is achieved by a cyclical stereotype motor program that consists of posterior body-wall muscle contraction (pBoc), followed by anterior body-wall muscle contraction (aBoc) and enteric muscle contraction to expel the gut contents (Exp). The aex-6 mutant animals almost completely lack the aBoc and Exp steps (Fig. 6A), compared to nearly 100% of those steps in the wild-type and rab-3 mutants animals. Normal defecation behaviors in rab-3 mutants suggests that RAB-27 only functions for vesicle transport in the defecation process, distinct from a parallel pathway of both Rabs for synaptic transmission (Mahoney et al. 2006). Because we found that the RAB-27 function in synaptic transmission is regulated by REP-1 (Fig. 4), we expected that the ta208 mutant would show defects in defecation behavior. However, rep-1(ta208) mutant animals showed only mild defecation defects (Fig. 6A). The Exp : pBoc ratio of the rep-1 mutant was slightly decreased compared with that of the wild type (wild type; 0.98 ± 0.012, rep-1; 0.86 ± 0.038; P < 0.05). We also performed genetic analyses of the defecation by using the aex-6; rep-1 double mutant, but genetic interactions between aex-6 and rep-1 were not detectable because the aex-6 single mutant alone shows sufficiently severe defecation defects (Fig. 6A).
|
Considering these data, we wondered whether RAB-27 can function without any lipid modification by RabGGTase or whether RAB-27 is transferred the lipid by a sole RabGGTase, that is, not forming a trimeric complex with REP-1. For this, we observed defecation behavior in animals whose RabGGTase activity was disrupted by RNAi treatment. Both M57.2 ( subunit) and B0280.1 (β subunit) RNAi-treated worms were severely affected in their defecation behavior, similar to the rab-27(RNAi) worms (Fig. 6B and data not shown). These results strongly suggest that RAB-27 may be modified solely by RabGGTase protein in defecation behavior, and that REP-1 is not necessary for the RAB-27 function in this pathway. This conclusion is not consistent with the REP-1 dependent-RAB-27 function for synaptic transmission in presynaptic neurons. We wondered whether RAB-27 expressing in other tissues does not require REP-1 for its function. By tissue-specifically expressing RAB-27 in aex-6 mutant background and observing their rescue abilities for defecation defects, we examined where RAB-27 does function for defecation behavior. Both the intestine-specific (by elt-2 promoter) and neuron-specific expression of rab-27 cDNA (by unc-119 promoter) showed the similar rescue abilities for the defects, suggesting that rab-27 probably functions in both the intestine and some neurons for defecation. As two GABAergic neurons, AVL and DVB, are known to regulate the aBoc and Exp muscle contractions (McIntire et al. 1993a,b; Branicky & Hekimi 2006), we are supposing that the rescue of neuronal expression of RAB-27 probably result from the RAB-27 function in these two GABAergic neurons. These results suggest that RAB-27 for defecation, which presumably functions in the intestine and AVL and DVB GABAergic neurons, could be modified by RabGGTase without interaction with REP-1. On the other hands, RAB-27 in several neurons, such as motor neurons at neuromuscular junctions, requires REP-1 for normal synaptic transmission responsible for aldicarb sensitivity.
REP-1 is expressed in the nervous system
To further understand the REP-1 function in vivo, we examined the expression pattern of the rep-1 gene using GFP fusion constructs. Because rep-1 mRNA is transcribed by the SL2 sequence, we generated GFP fusion constructs that contained the upstream promoter region of the former gene Y67D2.2, and examined GFP expression patterns (see Fig. 2B for plasmid constructs). GFP expression was observed in several neurons including head neurons, motor neurons located in the ventral nerve cord, HSN and CAN neurons, and tail neurons (Fig. 7). However, rep-1 does not seem to be expressed in all the neurons although rab-3 and rab-27 are expressed in almost all or all neurons (Nonet et al. 1997; Mahoney et al. 2006). Not only was GFP involved in the neuronal expression, it was also expressed in various muscles such as body-wall, pharyngeal, intestinal and anal sphincter, in addition to the seam cells, hypodermis and the intestine (Fig. 7). These broad expression patterns of rep-1 may explain various defects caused by the ta208 mutation.
|
Our drug experiments suggest that REP-1 regulates RAB-27 function but not RAB-3 function for neuronal synaptic transmission. If REP-1 in neurons specifically regulates the lipid modification of RAB-27, target membrane-localization of RAB-27 could be disrupted. We thus examined whether the mutation in rep-1 alters the localization patterns of specific RAB proteins. In wild-type animals, both GFP::RAB-3 and GFP::RAB-27 fusion proteins were enriched at synapse-rich regions of the nervous system such as the nerve ring around the head (Fig. 8A,C). In the rep-1(ta208) mutant background, a similar localization pattern for GFP::RAB-3 was obtained (Fig. 8C,E). However, for RAB-27, broader GFP expression was observed: the fusion protein was localized not only at the synapse-rich nerve ring but also around the cell body region of the head neurons (shown by a blanket in Fig. 8D,F). These results are consistent with the aldicarb analyses that suggest the specific requirement of REP-1 for RAB-27 function.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Identification of Caenorhabditis elegans REP-1
The ta208 mutant was isolated as a defective response to melatonin, which suggests a lower neuronal activity such as decreased synaptic transmission in the mutant. We concluded that this ta208 is an allele of the gene Y67D2.1 for following reasons. First, the mutant was clearly mapped on three cosmids and a mis-sense mutation at a well conserved glutamate in Rep2 proteins was found in the Y67D2.1 ORF. Second, all of the mutant phenotypes including melatonin resistance, aldicarb resistance, decreased number of progeny, defecation defects, were rescued by introducing either PCR fragments or a plasmid covering Y67D2.1 genomic region. Third, an Y67D2.1 RNAi-treatment caused quite similar phenotypes with ta208 mutant phenotypes. Forth, some phenotypes observed in the ta208 mutant were also observed in rab mutants and localizations of several Rab proteins were affected in the ta208 mutant background. Although we have not succeeded to isolate other alleles of the gene Y67D2.1, all of the above evidences support our conclusion.
Does the gene Y67D2.1 really encode a C. elegans REP-1? The ta208 mutation substitutes the 107th glutamate to lysine in the predicted Y67D2.1 protein (E107K; Fig. 2D) and this mutation causes several behavioral defects concerned in the Rab protein pathways. The mutation also causes altered localization patterns of several GFP::RAB fusion proteins, suggesting that Y67D2.1 protein functions in the Rab signaling pathway. Y67D2.1 encodes a 510 amino acid protein that is similar to both GDI and REP. The C. elegans GDI is predicted to be encoded by the gene Y57G11C.10 (Wormbase, <http://www.wormbase.org>). Both proteins have similar amino acid length and no clear domain differences are found between them. However, several phenylalanines are strictly conserved in each protein family. The 297th phenylalanine in human Rep1 protein is required for the Rab/RabGGT/REP complex formation (Pylypenko et al. 2003). The Y67D2.1 protein also has phenylalanine in this region and does not have GDI-family conserved phenylalanine. Biochemical analyses also support the idea that Y67D2.1 encodes a REP. The Y67D2.1 protein physically interacts with several Rab proteins in yeast. Furthermore, Y67D2.1 RNAi-treated animals increased the amount of unprenylated RAB-27 protein similar to RabGGTase RNAi-treated animals, suggesting that both genes regulate lipid modification of Rab proteins. Although positive interaction between REP-1 and RabGGTase could not be observed in our yeast assay, this does not mean that the Y67D2.1 product does not interact with RabGGTase in C. elegans. We are suspecting that the expression of full-length RabGGTase may be toxic for yeast cells, because we also failed to detect positive interaction between and β subunit of the C. elegans RabGGTase in the yeast (data not shown). In conclusion, all of these data strongly indicate that Y67D2.1 encodes a C. elegans REP protein, not a GDI, and that ta208 is an allele of the C. elegans REP gene.
The C. elegans REP protein is highly homologous to both the vertebrate Rep1 and Rep2 proteins. Interestingly, amino acid alignments with several REP proteins indicate that the mutated 107th glutamate in the ta208 mutant is well conserved in mammalian Rep2 proteins (Fig. 2D). Initially, we wondered whether this region containing the 107th glutamate might function for the binding site of Rab or RabGGTase. REP proteins should have at least two binding sites that bind Rab protein and RabGGTase to assist Rab protein geranylgeranylation. However, Pylypenko et al. (2003) showed that the domain containing the well-conserved 184th phenylalanine in worm REP-1 (279th in human Rep1, Fig. 2C) is one of the essential domains for binding to RabGGTase (Pylypenko et al. 2003). Furthermore, more of the middle region (around the 300th amino acid in C. elegans REP-1) is important to form a complex with Rab proteins (Alory & Balch 2003). Therefore, the region containing the ta208 mutation probably does not function directly for either the binding site to Rab or RabGGTase. Consistent with this, in yeast two-hybrid assay, the binding property of mutant REP-1(E107K) protein with several Rab proteins does not seem to be altered compared to the wild-type REP-1. Considering the large difference from negatively charged glutamate to positively charged lysine in the amino acid sequence, we think that the ta208 mutation could act largely through its effect on protein structure or protein folding. The unfolded mutant REP-1 protein may be degradated earlier than normal protein in vivo, and could cause weak hypomorphic phenotypes in several observations.
Interestingly, zebrafish has a single Rep1 gene and has glutamate at this conserved position (Starr et al. 2004). Thus, the REP protein containing glutamate in this position is probably ancestral form, and REP proteins having histidine like mammalian Rep1 might have been derived from Rep2 form, by gene duplication in mammalian taxa. Amino acid change from the glutamate in Rep2 to histidine in Rep1 may provide a characteristic feature for mammalian Rep1, such as higher affinity for RabGGTase in the Rep1-Rab27 complex (Larijani et al. 2003). Unfortunately, there have been few studies of the Rep2 protein group, so functional property of this glutamate has not been understood yet. Further analyses using C. elegans will provide a new insight on functional consequences between Rep1 form and Rep2 form.
In vivo functions of Caenorhabditis elegans REP
REP proteins recruit various newly synthesized Rab proteins to their specific target membranes depending on their geranylgeranylation by the stable REP/Rab/RabGGTase complex (Andres et al. 1993). Without proper geranylgeranylation of Rab proteins, they cannot be delivered to their target membranes and fail to have proper function. Thus REP proteins control Rab-related endocytic and exocytic pathways through the regulation of distribution patterns of each Rab protein. In this study, we have shown that the rep-1(ta208) mutant animals had mild resistance to aldicarb but normal response to levamisole, suggesting a weak defect in neurotransmitter release from presynaptic neurons. Previous studies have shown the involvement of two Rab proteins, that is, RAB-3 and RAB-27, in the neuronal exocytic pathways in C. elegans, although both Rabs has not been demonstrated to function in the same neurons (Nonet et al. 1997; Mahoney et al. 2006). Surprisingly, our genetic analyses strongly suggest that REP-1 functions only in the RAB-27 pathway, and not in the RAB-3 pathway for synaptic transmission (Fig. 4). Possible hypotheses from these results are that REP-1 is expressed only in neurons in which RAB-27 is required for transmitter release, or C. elegans REP-1 can only be required for RAB-27 function. In former case, other REP proteins such as a Rep1-like protein should be expressed in RAB-3-positive neurons, including cholinergic motor neurons, and must function for RAB-3 post-translational modification in cooperation with RabGGTase. This possibility may be supported by rep-1 expression analysis because the rep-1 promoter activity seems to be restricted to a subset of neurons, and not expressed in all neuronal cells (Fig. 7). However, no candidate rep gene has been found in the C. elegans genome, except for rep-1. Thus, the existence of another REP protein is unlikely in C. elegans. In latter case, a kind of Rab proteins such as RAB-3 may not require geranylgeranylation by the REP/RabGGTase complex or RAB-3 could be modified by RabGGTase alone. We prefer the latter possibility: in C. elegans, some Rab could bind to RabGGTase and could be attached to lipid moieties for their function, without forming a trimeric complex containing REP-1. Several neuronal cells possibly do not express any REP proteins (Fig. 7), and several Rabs including RAB-3 does not bind to REP-1 protein in yeast cells (Fig. 3). This may mean that rep-1 is expressed in certain cells in which REP-1 function is highly required for Rab functioning in these cells.
This variability of REP-1 requirement for Rab protein modification seems to be dependent in the cells that each Rab is expressed. As for defecation, for example, both aex-6 (rab-27) mutant animals and rab-27 RNAi-treated animals showed severe defecation defects. If aex-6 and rep-1 function through the same signaling pathways and RAB-27 requires REP-1 for its function, rep-1 knockdown mutants and aex-6 knockdown mutants would exhibit the similar phenotype. However, both the knockdown of rep-1 activity by RNAi and ta208 mutation only caused mild defecation defects (Fig. 5). These results suggest that the requirement of REP-1 for RAB-27 function in defecation is quite weak. These controversial results between the aldicarb experiments and defecation experiments mean that the requirement of REP-1 for the same Rab protein depends on its site of action. Although defecation defects in aex-6 mutants can be partially rescued by both an intestine-specific expression or a pan-neuronal expression of RAB-27, depending on our knowledge about C. elegans defecation behavior, we strongly believe that the neuronal cells in which RAB-27 functions for defecation is not the same neurons corresponding to the aldicarb sensitivity, but the two GABAergic AVL and DVB neurons (Branicky & Hekimi 2006). Thus, the RAB-27 expressed in the intestine and/or GABAergic neurons probably can interact with RabGGTase without REP-1. The precise mechanism for cell-type specific REP requirement is unknown, but it might be possible that cytoplasmic environments, such as the intracellular pH in cells or the existence of small molecules attached to Rab proteins, can affect the folded structure of RAB-27 protein and facilitate the protein-protein interaction between RAB-27 and RabGGTase. We suppose that the failure of interaction in yeast cells could be caused by yeast-specific cellular environment. Alternatively, a novel regulator to help their interaction might not be expressed in yeast cells and the C. elegans intestinal cells. By applying genetic approach, for example, isolating genetic enhancers in the rep-1(–) background, such regulators for Rab and RabGGTase interaction might be isolated.
Molecular basis of REP specificity to Rab proteins
Alexandrov et al. (1994) reported that the process of Rab prenylation starts with the association of an unprenylated Rab protein with REP. The complex is then recognized by RabGGTase, which attaches two geranylgeranyl moieties to the C-terminus of the Rab protein. Finally, REP is thought to escort the prenylated Rab protein to its target membrane (Alexandrov et al. 1994). It is thought that the formation of a hetero-trimeric complex of Rab/REP/RabGGTase is necessary to correctly localize Rab protein and that a lack of REP or RabGGTase would cause Rab-deficient abnormalities via the disturbance of Rab localization. Consistent with this, the disruption of RabGGTase function caused quite severe defects in both the defecation (Fig. 5) and GFP::Rab localizations (Fig. 9). In contrast, the disruption of rep-1 had little effect in the defecation presumably through the RAB-27 and only mild effects in RAB-10 and RAB-11 localization. Therefore, we hypothesize a new model in which the C. elegans RabGGTase can interact with a kind of Rab proteins without REP-1 and can add lipid moieties for its correct localization. In contrast of the requirement of escort protein for Rab prenylation, post-translational prenylation in other small GTPases are proceeded directly by the prenylating enzymes such as farnesyl transferase or geranylgeranyl transferase type-I (Casey & Seabra 1996). Like these type-I prenyltransferase enzymes, the type-II C. elegans RabGGTase may have a weak binding affinity directly to several Rab proteins. The phenotypes observed in other REP-deficient organisms also support our hypothesis in which some Rabs require REP-1 but some does not. In zebrafish, Rep1 null mutant animals develop up to 5-day post-fertilization and show quite similar appearance with wild-type fish. Mutant fish do not respond to an acoustic stimulus because of the lack of hair-cell in the ear, but show normal avoidance behavior to a touch stimulus (Starr et al. 2004). Although the mutant larvae began to die at the beginning on the sixth day, these phenotypes suggest that some Rab proteins in some tissues must be functional despite the complete loss of REP protein in zebrafish.
What is the molecular basis of specificity for the requirement of REP-1 in each Rab function? It seems unlikely that the four amino acids including prenylated cysteines at the C-terminus of each Rabs is important for binding specificity because no characteristic feature can not be found in Rab proteins used in this analyses (the last four amino acids in each Rab proteins are follows: RAB-3; QCNC, RAB-27; CANC, RAB-5; SCCK, RAB-7; GCNC, RAB-10; GGCC, RAB-11; CCIP). Our preliminary chimeric experiments also supported that the variable C-terminus regions of Rab proteins are not the intrinsic region for interaction with REP-1. Alternatively, this result suggests another possible mechanism for the specificity of REP-1-requirement in each Rab: the overall structure of each Rab may be highly variable, and this variation may largely affect the interaction with REP-1 and also interaction with RabGGTase, not forming trimeric complex. For examples, REP-1 did not show positive interaction with RAB-27 and RAB-10 in yeast cells, but it should interact with both Rab proteins in the C. elegans cells because the localization patterns of GFP fusion protein at synapses for RAB-27 and in the intestine for RAB-10 were mildly but clearly affected in either the rep-1(ta208) mutant or RNAi-treated animals (Figs 8 and 9). Thus, REP-1 probably can interact with most Rab proteins, but the binding affinity between REP-1 and each Rab may be easily affected by their environments, such as in yeast cells or in worm cells, also in neuronal cells or in the intestinal cells. Such subtle balance between Rab protein conformation and affinity to REP-1 could produce the controversial results of yeast two-hybrid assay in which might be sensitive to protein conformation than C. elegans’ cells. Protein-binding assay was partially consistent with the other results: both RAB-5 and RAB-7 are strongly dependent on REP-1 for their localization but both RAB-3 and RAB-10 are independent on REP-1. This suggests that REP-1-depndency in each Rab may correspond to the binding affinity of Rab proteins with REP-1. High binding affinity may precede the trimeric complex of Rab/REP/RabGGTase and REP-1-dependent Rab prenylation by RabGGTase. On the other hands, low binding affinity with REP-1 may skip the formation of Rab/REP complex and precede REP-1-independent Rab modification by sole RabGGTase.
In conclusion, we isolated a C. elegans rep-1 mutant and identified the rep-1 gene. From our behavioral and genetic analyses, we conclude that REP-1 functions in a tissue specific manner and that the effect of REP-1 differs from that in each Rab species. Using C. elegans, we will be able to elucidate the cellular and genetic mechanisms through which REP regulates various functions in vivo. Our experiments may also help to establish a remedy for diseases induced by REP-1 mutations, including chroideremia. Our study will provide some details with which to elucidate the function of REP, which may lead to the determination of the localization and signaling pathways of Rab proteins in the future.
|
Experimental procedures |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Nematodes were cultured at 20 °C on standard NGM agar plates seeded with the OP50 bacterial strain as a food source (Brenner 1974). The wild-type strain (Bristol N2) and several mutant strains were obtained from the Caenorhabditis Genetic Center (CGC). The jsIs682 (an integrated line of GFP::RAB-3) and jsEx740 (a transgenic line of GFP::RAB-27) strains were kindly provided by Michael Nonet. Double-mutant strains were constructed using standard genetic methods, and both mutations were confirmed by restriction-enzyme digestion or direct sequencing. The strains used in this study were aex-6(sa24), rab-3(js49), rep-1(ta208), jsIs682, jsEx740, pwIs72, pwIs170, pwIs214 and pwIs69.
Melatonin assay
The melatonin assay was performed as described previously (Tanaka et al. 2007). Briefly, synchronized young adult hermaphrodites were transferred to 100 µM melatonin assay plates. Fifteen minutes after transfer, the number of body bends was counted for 30 s. A body bend was defined as the center of the body passing through the midline of the amplitude.
Isolation and cloning of the rep-1 mutant
The ta208 mutant was originally isolated from a screening for melatonin-resistant mutants, as described previously (Tanaka et al. 2007). After backcrossing to the N2 wild type, we mapped ta208 based on the melatonin-resistant and small progeny number phenotypes. Using the standard SNP-based mapping method (Wicks et al. 2001; Davis et al. 2005), the ta208 mutant was mapped within three cosmids (M01G5, Y67D2 and Y22D7AL) on Chromosome III. By direct sequencing of the mutant genome DNA, we searched for a mutation in the predicted genes contained in these cosmids. A mis-sense mutation was found in gene Y67D2.1. For the rescue experiments, two 11-kb fragments covering from the entire Y67D2.1 gene to the Y67D2.2 genomic region (Fig. 2) were amplified using a long PCR and were co-injected into rep-1 mutants. Because subsequent transgenic animals showed increased numbers of progeny, we confirmed that the Y67D2.1 gene probably corresponds to ta208 (see Results). The full-length rep-1 cDNA was isolated by RT-PCR using SL2- and oligo-dT primers. We also obtained the entire rep-1 cDNA from Y. Kohara's yk clone (yk1365e09 or yk1154b02). Both the RT–PCR products and corresponding yk clones were fully sequenced and their sequence consistency was confirmed.
Behavioral assays
Drug assays using aldicarb and levamisole were performed as described previously (Doi & Iwasaki 2002). Sensitivity to aldicarb was examined by placing approximately 30 young hermaphrodites on plates containing 1 mM aldicarb, and the time course of paralysis was recorded every 10 min. Animals were considered as paralyzed when they failed to respond to several stimuli given with a platinum wire. Each assay was performed three times. Sensitivity to 100 µM levamisole was similarly examined. The total number of progeny per worm was counted by placing a single L4 hermaphrodite onto an NGM plate. The worm was transferred to a new plate every 2 days, and progeny from all of the plates were summed. Defecation assays were performed as previously described (Doi & Iwasaki 2002).
Plasmid construction
All of the plasmids were constructed using standard molecular methods. For rep-1 expression analysis, a 3.8 kb upstream sequence from the Y67D2.1 start codon was amplified using primers with restriction enzyme sites; this fragment was inserted between the SalI and BamHI sites of the pPD95.77 GFP vector generating the plasmid pDK222. The plasmid pDK224, containing a 3.2-kb upstream sequence of the Y67D2.2 gene to 23 bp downstream of the Y67D2.1 start codon, was similarly generated by inserting a 14.6-kb PCR-amplified fragment between the SalI and KpnI sites of pPD95.77. The rescuing plasmid pDK225 was generated as follows. A 1.7-kb fragment containing a GFP sequence and unc-54 3'-UTR was amplified from pPD95.77 and inserted between the KpnI and NaeI sites of the pBluescript(SK+) vector. A 14.3-kb fragment and a 6.6 kb fragment, covering the full region of the Y67D2.2 promoter region and the Y67D2.1 coding region, were then sequentially subcloned into the GFP/pBluescript vector to fuse with GFP at its C-terminus. The tissue-specific aex-6-rescuing plasmids were generated as follows. The full-length aex-6 (rab-27) cDNA was amplified from yk1575a08 clone and inserted between the BamHI and EcoRI sites of the pPD95.65. A 4.5 kb elt-2 promoter region or a 2.2 kb unc-119 promoter region were inserted between PstI and BamHI sites to generate the intestine-specific (pDK323) or the neuron-specific rescue plasmid (pDK356), respectively. All of the primers used in this study can be available upon request.
Transgenic animals
Transgenic strains bearing extrachromosomal arrays were generated using a previously described method (Mello et al. 1991). For the expression analyses, pDK222 or pDK224 (30 ng/µL) was co-injected into lin-15(n765ts) mutants with the injection marker lin-15(+) plasmid (50 ng/µL). For the rescue experiments, pDK225 (30 ng/µL) or PCR fragments (25 ng/µL each) were co-injected with the lin-15(+) plasmid (50 ng/µL) into lin-15(n765ts); rep-1(ta208) mutants. At least, five independent stable transgenic lines were obtained and used for analyses.
RNAi interference
RNA interference (RNAi) was performed based on the standard feeding protocol (Kamath et al. 2001). The rep-1 and rab-27 cDNA was prepared from the corresponding yk clones, yk1365e09 and 1575a08, respectively. The full-length RabGGTase (M57.2) cDNA was amplified by RT-PCR. Each cDNA was subcloned into the feeding RNAi vector L4440, and subsequent plasmids were transformed into the HT115 (DE3) bacterial cell. Two wild-type L4 worms were placed on the feeding RNAi plates seeded with each transformed bacterial cell, and the phenotypes of the F1 progeny were examined.
Yeast two-hybrid assay
LexA yeast two-hybrid assay was performed according to the manufactures’ protocol (Clonetech). The full-length REP-1 cDNA or mutated REP-1(E107K) cDNA was inserted between the BamHI and NcoI sites of the pLexA vector as "Bait" plasmids. The RabGGTase subunit cDNA and each Rab cDNA were inserted between the EcoRI and XhoI sites of the pB42AD vector to construct "Prey" plasmids. These plasmids were introduced into reporter strain EGY48. Positive interaction was examined by checking the growth of transformed yeast on the plate that lack tryptophan, histidine, uracil and leucine, containing 2% galactose and 1% Raffinose.
Western analysis
Worms were collected from a few NGM plates or feeding RNAi plates, and washed three times by M9. The 80 µL of Laemmli sample buffer was added to the packed worms (about 20 µL volume), and boiled for 5 min. After a brief centrifugation of the sample, each 15 µL sample was loaded onto a 5–20% gradient SDS-PAGE and transferred to nitrocellulose membrane (Bio-Rad). Primary anti-RAB-27 antibodies (rabbit, gift from M. Fukuda) was used at a concentration of 2 µg/mL. Secondary HRP-conjugated anti-Rabbit IgG antibodies (goat, Bio-Rad) was used at a 15 000 dilution, and blots were detected using a HRP chemiluminescent kit (Bio-Rad).
GFP:RAB quantification
For the RAB-3 and RAB-27 localization analyses, jsIs682 (GFP::RAB-3) and jsEx740 (GFP::RAB-27) strains were crossed to rep-1(ta208) mutants. GFP expression in wild-type or rep-1-mutant homozygous animals was observed using a Zeiss Pascal 5 confocal microscope system, using the same laser and detector settings. Z-series images were stacked to a single image. For endocytotic Rab localization analyses, we examined the localization patterns of RAB-5 (pwIs72), RAB-7 (pwIs170), RAB-10 (pwIs206) and RAB-11 (pwIs69). All the strains express GFP-fused Rab proteins under the control of the vha-6 intestine-specific promoter. Strains carrying each GFP::RAB transgene were treated by feeding RNAi, and GFP localization patterns in the anterior intestinal cells of the subsequent F1 animals were examined. All of the images were captured using a Zeiss Pascal 5 confocal microscope equipped with a 63x objective lens. All of the observations were performed using the same laser and detector settings. The images were converted into 8-bit tiff images, and the number of GFP puncta in each image was calculated using the ImageJ program.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
aPresent address: Evaluation and Licensing Division, Pharmaceutical and Food Safety Bureau, Ministry of Health, Labor and Welfare, Tokyo 100-8916, Japan.
* Correspondence: doi-m{at}aist.go.jp |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Alory, C. & Balch, W.E. (2003) Molecular evolution of the Rab-escort-protein/guanine-nucleotide-dissociation-inhibitor superfamily. Mol. Biol. Cell 14, 3857–3867.
Anant, J.S., Desnoyers, L., Machius, M., Demeler, B., Hansen, J.C., Westover, K.D., Deisenhofer, J. & Seabra, M.C. (1998) Mechanism of Rab geranylgeranylation: formation of the catalytic ternary complex. Biochemistry 37, 12559–12568.[CrossRef][Medline]
Andres, D.A., Seabra, M.C., Brown, M.S., Armstrong, S.A., Smeland, T.E., Cremers, F.P. & Goldstein, J.L. (1993) cDNA cloning of component A of Rab geranylgeranyl transferase and demonstration of its role as a Rab escort protein. Cell 73, 1091–1099.[CrossRef][Medline]
Branicky, R. & Hekimi, S. (2006) What keeps C. elegans regular: the genetics of defecation. Trends Genet. 22, 571–579.[CrossRef][Medline]
Brenner, S. (1974) The genetics of Caenorhabditis elegans. Genetics 77, 71–94.
Casey, P.J. & Seabra, M.C. (1996) Protein prenyltransferases. J. Biol. Chem. 271, 5289–5292.
Chen, C.C., Schweinsberg, P.J., Vashist, S., Mareiniss, D.P., Lambie, E.J. & Grant, B.D. (2006) RAB-10 is required for endocytic recycling in the Caenorhabditis elegans intestine. Mol. Biol. Cell 17, 1286–1297.
Cremers, F.P., Armstrong, S.A., Seabra, M.C., Brown, M.S. & Goldstein, J.L. (1994) REP-2, a Rab escort protein encoded by the choroideremia-like gene. J. Biol. Chem. 269, 2111–2117.
Davis, M.W., Hammarlund, M., Harrach, T., Hullett, P., Olsen, S. & Jorgensen, E.M. (2005) Rapid single nucleotide polymorphism mapping in C. elegans. BMC Genomics 6, 118.[CrossRef][Medline]
Doi, M. & Iwasaki, K. (2002) Regulation of retrograde signaling at neuromuscular junctions by the novel C2 domain protein AEX-1. Neuron 33, 249–259.[CrossRef][Medline]
Kamath, R.S., Martinez-Campos, M., Zipperlen, P., Fraser, A.G. & Ahringer, J. (2001) Effectiveness of specific RNA-mediated interference through ingested double-stranded RNA in Caenorhabditis elegans. Genome Biol. 2, RESEARCH0002.[Medline]
Lackner, M.R., Kindt, R.M., Carroll, P.M., et al. (2005) Chemical genetics identifies Rab geranylgeranyl transferase as an apoptotic target of farnesyl transferase inhibitors. Cancer Cell 7, 325–336.[CrossRef][Medline]
Larijani, B., Hume, A.N., Tarafder, A.K. & Seabra, M.C. (2003) Multiple factors contribute to inefficient prenylation of Rab27a in Rab prenylation diseases. J. Biol. Chem. 278, 46798–46804.
Mahoney, T.R., Liu, Q., Itoh, T., Luo, S., Hadwiger, G., Vincent, R., Wang, Z.W., Fukuda, M. & Nonet, M.L. (2006) Regulation of synaptic transmission by RAB-3 and RAB-27 in Caenorhabditis elegans. Mol. Biol. Cell 17, 2617–2625.
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]
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]
Nonet, M.L., Staunton, J.E., Kilgard, M.P., Fergestad, T., Hartwieg, E., Horvitz, H.R., Jorgensen, E.M. & Meyer, B.J. (1997) Caenorhabditis elegans rab-3 mutant synapses exhibit impaired function and are partially depleted of vesicles. J. Neurosci. 17, 8061–8073.
Novick, P. & Zerial, M. (1997) The diversity of Rab proteins in vesicle transport. Curr. Opin. Cell Biol. 9, 496–504.[CrossRef][Medline]
Pereira-Leal, J.B. & Seabra, M.C. (2001) Evolution of the Rab family of small GTP-binding proteins. J. Mol. Biol. 313, 889–901.[CrossRef][Medline]
Pfeffer, S.R. (1994) Rab GTPases: master regulators of membrane trafficking. Curr. Opin. Cell Biol. 6, 522–526.[CrossRef][Medline]
Pylypenko, O., Rak, A., Reents, R., Niculae, A., Sidorovitch, V., Cioaca, M.D., Bessolitsyna, E., Thoma, N.H., Waldmann, H., Schlichting, I., Goody, R.S. & Alexandrov, K. (2003) Structure of Rab escort protein-1 in complex with Rab geranylgeranyltransferase. Mol. Cell 11, 483–494.[CrossRef][Medline]
Rasteiro, R. & Pereira-Leal, J.B. (2007) Multiple domain insertions and losses in the evolution of the Rab prenylation complex. BMC Evol. Biol. 7, 140.[CrossRef][Medline]
Seabra, M.C. (1996) New insights into the pathogenesis of choroideremia: a tale of two REPs. Ophthalmic Genet. 17, 43–46.[Medline]
Seabra, M.C., Brown, M.S., Slaughter, C.A., Sudhof, T.C. & Goldstein, J.L. (1992a) Purification of component A of Rab geranylgeranyl transferase: possible identity with the choroideremia gene product. Cell 70, 1049–1057.[CrossRef][Medline]
Seabra, M.C., Goldstein, J.L., Sudhof, T.C. & Brown, M.S. (1992b) Rab geranylgeranyl transferase. A multisubunit enzyme that prenylates GTP-binding proteins terminating in Cys-X-Cys or Cys-Cys. J. Biol. Chem. 267, 14497–14503.
Seabra, M.C., Ho, Y.K. & Anant, J.S. (1995) Deficient geranylgeranylation of Ram/Rab27 in choroideremia. J. Biol. Chem. 270, 24420–24427.
Shen, F. & Seabra, M.C. (1996) Mechanism of digeranylgeranylation of Rab proteins. Formation of a complex between monogeranylgeranyl-Rab and Rab escort protein. J. Biol. Chem. 271, 3692–3698.
Starr, C.J., Kappler, J.A., Chan, D.K., Kollmar, R. & Hudspeth, A.J. (2004) Mutation of the zebrafish choroideremia gene encoding Rab escort protein 1 devastates hair cells. Proc. Natl. Acad. Sci. USA 101, 2572–2577.
Sudhof, T.C. (2004) The synaptic vesicle cycle. Annu. Rev. Neurosci. 27, 509–547.[CrossRef][Medline]
Tanaka, D., Furusawa, K., Kameyama, K., Okamoto, H. & Doi, M. (2007) Melatonin signaling regulates locomotion behavior and homeostatic states through distinct receptor pathways in Caenorhabditis elegans. Neuropharmacology 53, 157–168.[CrossRef][Medline]
Thoma, N.H., Iakovenko, A., Goody, R.S. & Alexandrov, K. (2001a) Phosphoisoprenoids modulate association of Rab geranylgeranyltransferase with REP-1. J. Biol. Chem. 276, 48637–48643.
Thoma, N.H., Iakovenko, A., Kalinin, A., Waldmann, H., Goody, R.S. & Alexandrov, K. (2001b) Allosteric regulation of substrate binding and product release in geranylgeranyltransferase type II. Biochemistry 40, 268–274.[CrossRef][Medline]
Thoma, N.H., Niculae, A., Goody, R.S. & Alexandrov, K. (2001c) Double prenylation by RabGGTase can proceed without dissociation of the mono-prenylated intermediate. J. Biol. Chem. 276, 48631–48636.
Thomas, J.H. (1990) Genetic analysis of defecation in Caenorhabditis elegans. Genetics 124, 855–872.[Abstract]
Wicks, S.R., Yeh, R.T., Gish, W.R., Waterston, R.H. & Plasterk, R.H. (2001) Rapid gene mapping in Caenorhabditis elegans using a high density polymorphism map. Nat. Genet. 28, 160–164.[CrossRef][Medline]
Zerial, M. & McBride, H. (2001) Rab proteins as membrane organizers. Nat. Rev. Mol. Cell Biol. 2, 107–117.[CrossRef][Medline]
Accepted: 12 August 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
