Rabphilin potentiates soluble N-ethylmaleimide sensitive factor attachment protein receptor function independently of rab3

Abstract

Rabphilin, a putative rab effector, interacts specifically with the GTP-bound form of the synaptic vesicle-associated protein rab3a. In this study, we define in vivo functions for rabphilin through the characterization of mutants that disrupt the Caenorhabditis elegans rabphilin homolog. The mutants do not display the general synaptic defects associated with rab3 lesions, as assayed at the pharmacological, physiological, and ultrastructural level. However, rabphilin mutants exhibit severe lethargy in the absence of mechanical stimulation. Furthermore, rabphilin mutations display strong synergistic interactions with hypomorphic lesions in the syntaxin, synaptosomal-associated protein of 25 kDa, and synaptobrevin soluble N-ethylmaleimide sensitive factor attachment protein receptor (SNARE) genes; double mutants were nonresponsive to mechanical stimulation. These synergistic interactions were independent of rab3 function and were not observed in rab3-SNARE double mutants. Our data reveal rab3-independent functions for rabphilin in the potentiation of SNARE function.

Fig. 1.

C. elegans rabphilin gene structure. A, A schematic diagram of the C. elegans and rat rabphilin proteins illustrating the conservation in domain structure. The percentage of amino acid identities among the two proteins in the zinc-finger (Zn²⁺) and calcium–phospholipid binding (C2) domains is indicated. B, Protein sequence alignments of the C. elegans, Drosophila, bovine, and rat rabphilin zinc-finger and C2 domains. Amino acids conserved among three of the four proteins in the alignment are boxed. The standard single-letter code was used. Introduced gaps are represented as dashes. Alignments were made using ClustalW (Thompson et al., 1994). The Drosophilasequence presented is a conceptual translation with no experimental confirmation. C, Northern blot of wild-type mixed-staged poly(A)-selected RNA probed with a rabphilin genomic clone. RNA size markers are labeled on the right (in kilobases).Arrows depict the two major rabphilin transcripts,C. briggsae and C. elegans rabphilin genes. Exons are depicted as patterned boxes. Exons encoding the zinc-finger domain (▨), the C2 domains (▩), and the alternatively spliced domains (▤) are uniquely labeled. Putative regulatory elements are depicted assmall black filled boxes. The position of thejs232 lesion and Tc1 transposon insertion ( ) from which thejs232 lesion was derived are depicted below the diagram. The genomic regions included in two plasmid clones used in the study are also diagrammed.

Fig. 2.

Rabphilin is localized to synaptic-rich regions of the C. elegans nervous system. Lateral views of the head region of whole adult animals fixed, permeabilized, coincubated with mouse and rabbit antibodies, and visualized with fluorescent tagged secondary antibodies are shown. RAB-3 (A) and rabphilin (B) colocalize in wild-type animals double-labeled with α-RAB-3 and α-RBF-1 antibodies. Synaptotagmin (C) and rabphilin (D) colocalize in wild-type animals double-labeled with α-SNT-1 and α-RBF-1 antibodies. RAB-3 (E) and rabphilin (F) both accumulate in neuronal cell bodies inunc-104(e1265) animals double-labeled with α-RAB-3 and α-RBF-1 antibodies. Synaptotagmin (G) and rabphilin (H) both remain localized to synapse-rich regions in rab-3(js49) animals double-labeled with α-SNT-1 and α-RBF-1 antibodies. Scale bar, 20 μm.

Fig. 3.

Rabphilin protein is absent inrbf-1(js232) mutant animals. Lateral views of the head region of whole adult animals fixed, permeabilized, incubated with mouse and rabbit antibodies, and visualized with fluorescent tagged secondary antibodies are shown. RAB-3 (A) and rabphilin (B, C) are localized in wild-type animals labeled with α-RAB-3 antibodies (A) as well as antibodies directed against the RBF-1 N terminus (B) and RBF-1 C terminus (C). However, rabphilin but notRAB-3 is absent in rabphilin mutants labeled with α-RAB-3 (D), α-RBF-1 N terminus (E), and α-RBF-1 C terminus (F) antibodies. Scale bar, 20 μm.

Fig. 4.

Rabphilin mutants appear normal in many assays of synaptic function. A, Rabphilin animals show wild-type responses to the acetylcholinesterase inhibitor aldicarb. Animals were placed on plates containing various concentrations of aldicarb, and their behavior was assayed 4.5 hr later. The percentage of animals paralyzed is plotted as a function of aldicarb concentration for the indicated genotypes in this representative experiment.B–D, Pharyngeal recordings from wild-type animals and mutants. Characteristic electropharyngeogram traces from the wild-type strain N2 (B), the rabphilin mutant (C), and the rab-3(js49) mutant (D) are shown. Recordings from rabphilin mutants were more variable than those from wild-type animals. The lower trace in C illustrates a typical abnormal electropharyngeogram observed in the rabphilin mutant. Alltraces are plotted as millivolts versus time.E–G, Ultrastructure of neuromuscular junctions from a wild-type animal (E) and two neighboring sections of the rabphilin mutant (F, G). Electron micrographs of ventral cord of adult animals are shown. A thick electron-dense specialization is visible at the opposition nerve and muscle (arrow) surrounded by a cluster of vesicles. The arrowhead identifies a representative vesicle. Scale bar, 500 nm.

Fig. 5.

Rabphilin mutants are lethargic.A, Superimposed images of animals of different genotypes foraging on an E. coli bacterial lawn before (left) and after (right) a stimulus; see Materials and Methods for details. Two images separated by 120 sec (left) or 5 sec (right) were pseudocolored in green and red and superimposed. Animals that have moved appear as two images, onegreen and one red, whereas animals that have remained stationary or moved a minimal distance appear as a singleyellow or partially yellow image. The genotypes are rbf-1(js232), unc-64(e246), and jsEx510[pRF6; pMR1]. B, Quantification of locomotor rates of animals of different genotypes foraging on an E. coli bacterial lawn before and after a stimulus; see Materials and Methods for details. The mean locomotor velocity of 20–25 animals is plotted at various time points before and after a stimulus applied at t = 0. Genotypes: as described above and rab-3(js49).

Fig. 6.

Rabphilin lesion effects on basal and stimulated locomotion. Mean basal velocities (open bars) and mean stimulated velocities (filled bars) of rabphilin and rab-3 single, double, and triple mutant combinations are shown. Error is expressed as SEM. A, Expression of the rabphilin lethargic phenotype requires rab-3 activity. The genotypes are rbf-1(js232) andrab-3(js49). B, Synaptic regulators show negligible interactions with rabphilin. The genotypes areunc-10(md1117), snt-1(md290),aex-3 (y255), rbf-1(js232), andrab-3(js49). rab-3(y251) was used for the double mutant with snt-1(md290).rab-3(y251) behaves like a null mutant (Nonet et al., 1997). C, Neuronal SNARE mutations show strong interactions with rabphilin. The genotypes arerab-3(js49), rbf-1(js232), and SNAREs as indicated on the figure.

Fig. 7.

Synergistic locomotor defects in rabphilin–SNARE double mutants. Superimposed images of animals of different genotypes foraging on an E. coli bacterial lawn after a stimulus are shown; see Materials and Methods for details. Two images separated by 5 sec were pseudocolored ingreen and red and superimposed. Animals that have moved appear as two images, one green and onered, whereas animals that have remained stationary or moved a minimal distance appear as a single yellow orpartially yellow image. The genotypes areric-4(md1088), rab-3(js49), and rbf-1(js232).