Calcium/Calmodulin-Dependent Protein Kinase II Regulates Caenorhabditis elegans Locomotion in Concert With a G_o/G_q Signaling Network

Corresponding author: James H. Thomas, University of Washington, Department of Genetics, Box 357360, Seattle, WA 98195., jht{at}genetics.washington.edu (E-mail)

Communicating editor: P. ANDERSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Caenorhabditis elegans locomotion is a complex behavior generated by a defined set of motor neurons and interneurons. Genetic analysis shows that UNC-43, the C. elegans Ca²⁺/calmodulin protein kinase II (CaMKII), controls locomotion rate. Elevated UNC-43 activity, from a gain-of-function mutation, causes severely lethargic locomotion, presumably by inappropriate phosphorylation of targets. In a genetic screen for suppressors of this phenotype, we identified multiple alleles of four genes in a G_o/G_q G-protein signaling network, which has been shown to regulate synaptic activity via diacylglycerol. Mutations in goa-1, dgk-1, eat-16, or eat-11 strongly or completely suppressed unc-43(gf) lethargy, but affected other mutants with reduced locomotion only weakly. We conclude that CaMKII and G_o/G_q pathways act in concert to regulate synaptic activity, perhaps through a direct interaction between CaMKII and G_o.

ORGANISMS respond to the environment by modulating their behavior. To understand how behavior is modulated, the cellular and molecular components that control particular behaviors must be defined. The model organism Caenorhabditis elegans is particularly well suited to such analysis since it has a relatively simple nervous system and is highly amenable to both genetic manipulation and behavioral analysis. Understanding how behavior is controlled at the cellular and molecular level in a relatively simple organism can provide insight into how behavior is controlled in more complex organisms such as mammals. In C. elegans, environmental influences have been shown to modulate several behaviors, including locomotion, feeding, egg laying, and defecation (AVERY and THOMAS 1997 ; BARGMANN and MORI 1997 ; DRISCOLL and KAPLAN 1997 ; JORGENSEN and RANKIN 1997 ). In particular, work from several investigators has described some of the neurons and molecules that control several aspects of locomotion, including the response to food, chemical stimuli, and mechanical stimuli (DRISCOLL and KAPLAN 1997 ; JORGENSEN and RANKIN 1997 ). However, despite the relative simplicity of C. elegans, many aspects of the neuronal and molecular control of locomotion have not yet been defined.

Neuronal connectivity maps, neuronal ablations, and the analysis of mutants that perturb locomotion have shown that at least three partially separable processes control C. elegans locomotion: the generation of coordinated sinusoidal body bends, selection of forward or backward movement, and determination of locomotion rate (DRISCOLL and KAPLAN 1997 ; JORGENSEN and RANKIN 1997 ). C. elegans moves by coordinated sinusoidal body bends that propagate smoothly along the length of the animal. These bends are generated by several classes of cholinergic motor neurons and two classes of GABA-ergic motor neurons that form neuromuscular junctions with body-wall muscles. A bend is generated by the local excitation of muscles on one side of the animal (via cholinergic motor neurons) with reciprocal inhibition of corresponding muscles on the opposite side of the animal (via GABA-ergic motor neurons; WHITE et al. 1986 ; MCINTIRE et al. 1993A , MCINTIRE et al. 1993B ). Selection of forward or backward movement is controlled by five interconnected command interneurons, which receive input from sensory neurons and other interneurons and send output to motor neurons (CHALFIE et al. 1985 ; WHITE et al. 1986 ). These command interneurons mediate forward and backward movement in response to specific sensory cues such as food and touch, and they set the duration of forward and backward movement in the absence of specific cues (CHALFIE et al. 1985 ; ZHENG et al. 1999 ).

Determination of locomotion rate has also begun to be elucidated. Modulation of locomotion rate in response to several stimuli, including food, has been described (e.g., SAWIN 1996 ). Animals exhibit hyperactivity in the absence of food and reduced locomotion when returned to food, presumably to optimize foraging. These responses appear to be mediated in part by serotonergic neurons in the pharynx and dopaminergic sensory neurons (SAWIN 1996 ). Consistent with these observations, exogenous serotonin and dopamine reduce locomotion, whereas mutants with reductions in these neurotransmitters move hyperactively (HORVITZ et al. 1982 ; SCHAFER and KENYON 1995 ; SEGALAT et al. 1995 ; NURRISH et al. 1999 ). A G_o/G_q heterotrimeric G-protein signaling network expressed throughout the nervous system has been shown to regulate locomotion rate, partly by affecting serotonergic signaling (MENDEL et al. 1995 ; SEGALAT et al. 1995 ; BRUNDAGE et al. 1996 ; HAJDU-CRONIN et al. 1999 ; LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). Specifically, loss-of-function (lf) mutations in goa-1 (G_o) cause hyperactivity, whereas lf mutations in egl-30 (G_q) cause severe lethargy. The egl-30 G_q regulates a phospholipase C signaling pathway that facilitates synaptic transmission by body-wall muscle motor neurons and perhaps other neuronal cell types (BRUNDAGE et al. 1996 ; LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). The goa-1 G_o appears to mediate serotonergic antagonism of the egl-30 pathway (HAJDU-CRONIN et al. 1999 ; LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). goa-1 and egl-30 also regulate several other behaviors in addition to locomotion rate, including egg laying (TRENT et al. 1983 ; PARK and HORVITZ 1986 ; MENDEL et al. 1995 ; REINER et al. 1995 ; SEGALAT et al. 1995 ; BRUNDAGE et al. 1996 ).

The C. elegans Ca²⁺-calmodulin-dependent serine/threonine protein kinase II (CaMKII) encoded by unc-43 is also widely expressed in neurons and regulates locomotion rate, as well as other behaviors (REINER et al. 1999 ; E. NEWTON and J. H. THOMAS, unpublished results). A gain-of-function (gf) mutation in unc-43 causes severe lethargy, as well as body-wall muscle hypercontraction, reduced egg laying, and reduced defecation (REINER et al. 1999 ). Several of these effects are genetically separable, indicating that unc-43 regulates defecation, body-wall muscle tone, and locomotion rate through different effectors (REINER et al. 1999 ). CaMKII has been shown to be an important regulator of synaptic strength by electrophysiological studies in the marine snails Aplysia and Hermissenda (NELSON and ALKON 1997 ), Drosophila (GRIFFITH et al. 1994 ), and rodents (MALINOW et al. 1989 ; SILVA et al. 1992A ; MAYFORD et al. 1995 ; ROTENBERG et al. 1996 ). These and other studies indicate that CaMKII plays a role in sensitization, learning and memory, fear and aggressive responses, and olfactory attenuation (SILVA et al. 1992B ; GRIFFITH et al. 1993 ; CHEN et al. 1994 ; BACH et al. 1995 ; WEI et al. 1998 ). CaMKII is unique among Ca²⁺-responsive proteins because it forms multimers with subunits that interact cooperatively to produce a nonlinear graded response to calcium (HANSON and SCHULMAN 1992 ; HANSON et al. 1994 ; DE KONINCK and SCHULMAN 1998 ). Adjacent subunits with bound Ca²⁺-calmodulin can phosphorylate each other, resulting in a large increase in Ca²⁺-calmodulin affinity and partial kinase activity even after Ca²⁺-calmodulin dissociates. This mechanism of activation enables CaMKII to integrate Ca²⁺ signaling events over time, which may be the basis for the role of CaMKII in regulating synaptic strength (HANSON et al. 1994 ; DE KONINCK and SCHULMAN 1998 ).

To understand how unc-43 controls locomotion rate in C. elegans, we performed a genetic suppressor screen with unc-43(gf) to identify genes that act with unc-43 to control locomotion rate. From our screen, we recovered multiple alleles of the genes goa-1, dgk-1, and eat-16, all involved in the goa-1/egl-30 G-protein network (HAJDU-CRONIN et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ), and alleles of a fourth gene, eat-11, that probably affects this same pathway. Our results indicate that UNC-43 may regulate this G-protein signaling network to control locomotion rate in C. elegans.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strain maintenance:
Worms were cultured using standard methods (BRENNER 1974 ). The standard N2 Bristol strain was used as wild type, and all strains were grown at 20° except those containing goa-1(n363) and goa-1(sa734). We found that these strains were healthier when grown at 15°. For behavioral assays, these strains were staged as L4 larvae and grown for 38 hr at 15°, which is equivalent to 24 hr of growth at 20° (J. H. THOMAS, unpublished observations). These strains were then placed at the assay temperature for 30 min prior to the start of each assay. Assays with wild type and unc-43(n498) animals grown under the same conditions were performed and found to be equivalent to assays performed with animals grown for 24 hr at 20°.

Mutations analyzed:
The following C. elegans mutations were analyzed in this work: eat-11(ad541, sa581, sa586, sa603, sa604, sa762, sa765, sa833) I, eat-16(sa609, sa735, sa768, sa839, sy438) I, egl-30(ad805) I, goa-1(n363, n1134, sa585, sa734, sa837, sa841) I, glr-1(n2461) III, unc-93(e1500) III, unc-103(n500) III, unc-43(n498, n1186) IV, dgk-1(sa605, sa748, sa760, sa766, sy428) X, unc-110(sa859) X, nIs51[egl-10(+)], syIs36[egl-30(+)], and syIs9[goa-1(gf)].

Identification of suppressor mutations:
unc-43(n498) hermaphrodites were treated with ethane methylsulfonate as described (SULSTON and HODGKIN 1988 ) and F₂ self-progeny were screened for suppression of the unc-43(n498) locomotion defect under a dissecting microscope. Animals that moved well spontaneously or animals that moved well after the plate was tapped were picked as putatively suppressed animals. The broods of these animals were rescored for suppression. In total, 28,000 haploid genomes were screened. Extragenic suppressors were crossed out of the unc-43(n498) background based on their hyperactive and reduced egg-retention phenotypes and retested for their ability to suppress the unc-43(n498) locomotion defect by backcrossing to the unmutagenized unc-43(n498) parent strain. Suppressor mutations that were closely linked to unc-43 and caused unc-43(lf) locomotory phenotypes were considered likely to be intragenic revertants. All mutations analyzed in this study are independent. All mutations were backcrossed to N2 at least twice before analysis.

Mapping to specific chromosomes was performed using dpy-5(e61) I, rol-6(e187) II, unc-32(e189) III, unc-5(e53) IV, dpy-11(e224) V, dpy-3(e27) X, and lon-2(e678) X. We tested each of our suppressor alleles for complementation of goa-1(n1134), dgk-1(sy428), and eat-16(sy438). For eat-11 complementation tests, we first tested sa765 for complementation of eat-11(ad541), and then subsequently tested all other alleles for complementation of sa765. Since sa604 and sa833 exhibited decreased egg retention in comparison to other eat-11 alleles, we tested sa604 and sa833 for complementation of ad541 as well as sa765. We scored complementation tests in the unc-43(n498)/+; dpy-11(e224)/+ background for all genes. We ruled out second-site noncomplementation by scoring for wild-type and unc-43(n498) homozygous progeny in the self-progeny broods of noncomplementing mutant heterozygotes. All noncomplementing mutations were closely linked by this test.

Construction of double mutant strains:
For the egl-30(ad805); unc-43(n1186) double mutant, n1186/+ males were mated to ad805 hermaphrodites and weakly Egl F₁ progeny were picked to individual plates. Plates that segregated convulsive-Unc animals were used to pick convulsive-Unc animals that were weakly Egl to individual plates. Their progeny were examined to confirm that n1186 was homozygous, and many of these progeny were picked to homozygose ad805 in the next generation. Animals that produced all Egl progeny were kept. The resulting strains were then tested for homozygosity of both mutations by crossing with N2 males and observing both convulsive-Unc animals and Egl animals segregating from all heterozygotes.

For syIs9[goa-1(gf)]; unc-43(n1186), the linked double mutant dpy-20(e1282) unc-43(n1186) was first constructed. e1282-n1186/+ males were then crossed to dpy-20(e1282); syIs9[goa-1 (gf)] hermaphrodites. F₁ progeny were picked and plates that segregated Dpy, convulsive-Unc animals were used to pick many syIs9 animals (Unc, Egl, non-Dpy) to individual plates. From the broods of these parents, animals that were Unc, Egl, non-Dpy and had a slightly flaccid body posture, as exhibited by unc-43(n1186) animals (REINER et al. 1999 ), were picked. The resulting strains were then tested for homozygosity of n1186 and syIs9 by crossing with N2 males and observing both convulsive-Unc animals and Egl animals segregating from all heterozygotes.

Double mutants with unc-43(n498) and goa-1, dgk-1, eat-16, or eat-11 mutations were constructed by picking individual unc-43(n498) homozygous animals (severe Unc) from the broods of double mutant heterozygotes (weaker Unc). Double mutants were picked in the next generation from those broods that segregated one quarter suppressed animals (double mutants) and three quarters severe Unc animals.

Locomotion assays:
Radial locomotion assays were performed at 23° on 8.5-cm plates harboring a 1.5-day-old lawn of OP50 bacteria. Animals were picked as L4 larvae and assayed after 24 hr of growth at 20°. Five animals were placed in the center of a plate and radial distance traveled was measured at 5, 10, and 15 min after the start of the assay. Two alleles of each suppressor gene were assayed, with the exception of egl-10 and egl-30, for which only one transgene was available. The assay was performed twice per genotype.

Body-bend assays were performed at 20° on 5-cm plates with a reproducibly thin lawn of OP50 bacteria that had been applied 8 hr prior to the assay. Animals were picked as L4 larvae and assayed after 24 hr of growth at 20° (or the equivalent; see section on strain maintenance). One animal was transferred to the assay plate, left undisturbed for 5 min, and then assayed for 3 min. At least four animals per genotype were assayed. Body bends were counted by observing flexing in the middle of the animal, using the vulva as a reference point. A flex was counted as a body bend when the vulva reached the peak or trough of the sine wave. Strains containing goa-1(null) or dgk-1(null) often alternated rapidly between forward and backward movement. Frequently, only partial body bends were completed during this behavior. Since partial body bends were not counted in the assay, our data is an underestimate of the movement rate of strains containing goa-1(null) and dgk-1(null). For assays with unc-93(gf), unc-110(gf), and unc-103(gf) mutants, body bends were counted anterior to the vulva since body bends did not always propagate along the entire length of the animal.

Aldicarb assays:
Aldicarb assays were performed at 23°. Stock solutions were prepared by dissolving aldicarb (Chem Service, West Chester, PA) in 70% ethanol to a final concentration of 100 mM. Aldicarb plates were prepared by adding the aldicarb stock solution to NG agar to a final concentration of 1 mM. Plates were stored at 4° until used. Twelve hours before an assay, a single drop of OP50 bacterial solution was added to each plate and incubated at 23°. Parallel experiments were performed on plates from the same batch. Aldicarb response was assayed by picking 19–25 animals to a single assay plate and scoring paralysis at 10-min intervals. Animals were scored as paralyzed when no spontaneous movement was exhibited, no movement was elicited by tapping the plate, and no movement was elicited by harsh touch to the anterior or posterior. unc-43(n1186) animals that initially appeared paralyzed by the above criteria would occasionally resume movement after harsh touch to the anterior or posterior. Therefore, we scored these animals twice at each timepoint and counted an animal as paralyzed when the above criteria for paralysis were met both times. Some strains were scored at 10-min intervals for an entire 120-min period. Strains for which the data exhibited a clear trend at early timepoints were scored at 10-min intervals for the first 60 min and then scored again at the 120-min timepoint.

Egg-staging assays:
Suppression of the unc-43(n498) egg-laying defect was quantified by assaying the stages of eggs laid at 20° on plates harboring a 2-day-old lawn of OP50 bacteria. Animals were picked as L4 larvae and assayed after 24 hr of growth at 20° (or the equivalent; see section on strain maintenance). Two alleles of each suppressor gene were assayed. A total of 10–14 animals were placed on the assay plate and allowed to recover from the transfer for 30 min. Eggs laid during the recovery period were removed, and at 10-min intervals, eggs laid were examined under Nomarski optics to ascertain their developmental stages (SULSTON et al. 1983 ; SCHIERENBERG 1986 ). After examination, eggs were removed from the assay plate before beginning the next 10-min egg-laying period. Animals were assayed for 2–3 hr.

Sequencing of goa-1(sa734):
goa-1(sa734) was outcrossed three times before sequencing. Sequencing was performed on bulk PCR product generated directly from genomic DNA with Taq and Pfu polymerases in a ratio of 100:1. Sequencing reactions were performed with Taq Dye Terminator reagents (Applied Biosystems, Foster City, CA). Primers for amplification and sequencing were designed using program Primer 3.0 from goa-1 genomic sequence in the cosmid C26C6 (as reported in GenBank). The goa-1(sa734) allele is a C to T change at base pair 154 that results in a Q52stop mutation. This mutation was confirmed by sequencing both strands.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mutations in goa-1, dgk-1, eat-16, and eat-11 are recovered as suppressors of unc-43(gf):
The gf mutation unc-43(n498) causes pleiotropic defects including reduced egg laying, reduced defecation, and lethargy (REINER et al. 1999 ). n498 causes an E108K change in the kinase domain of UNC-43. This change is predicted to result in a partially Ca²⁺-calmodulin independent kinase by analogy to a rat CaMKII mutant in the same residue (REINER et al. 1999 ). Comparison with the CaMKI crystal structure (GOLDBERG et al. 1996 ) suggests that this mutated residue destabilizes the autoinhibitory loop of the kinase without affecting substrate specificity, since the residue is far from the substrate binding cleft. In strong support of this hypothesis, lf mutations in unc-43 confer phenotypes that are generally reciprocal to the unc-43(gf) phenotypes, including hyperactivation of the egg-laying muscles and increased frequency of the enteric muscle contractions required for defecation (REINER et al. 1999 ). The reciprocity of the locomotion rate phenotype is more difficult to assess since unc-43(lf) disrupts locomotory coordination in a manner that obscures the determination of locomotion rate. The coordination defects exhibited by unc-43(lf) animals include kinking when moving backward and spontaneous convulsions that involve the simultaneous contraction of dorsal and ventral body-wall muscles (REINER et al. 1999 ). Since these defects severely alter the coordinated pattern of locomotion, the locomotion rate of unc-43(lf) animals cannot be ascertained readily. However, unc-43(lf) animals variably exhibit short bursts of rapid forward movement, which may indicate underlying hyperactivity. The spontaneous convulsions exhibited by these animals may also reflect underlying hyperactivity (REINER et al. 1999 ).

To identify genes that act with unc-43 to control locomotion rate, we used the lethargic phenotype of unc-43(gf) as the basis for a genetic suppressor screen. unc-43(gf) animals rarely move if undisturbed (Fig 1B). We reasoned that screening for increased locomotion of unc-43(gf) animals might identify genes that act with unc-43 to control locomotion rate. After chemical mutagenesis of unc-43(gf) animals, we screened F₂ progeny for increased locomotion. We recovered 43 independent revertants from a screen of 28,000 haploid genomes. Twenty-four of these were closely linked to the unc-43 locus and exhibited unc-43(lf) locomotory characteristics. Therefore, these revertants are likely lf alleles of unc-43. The remaining 19 suppressor mutations were genetically unlinked to unc-43 and exhibited recessive inheritance, consistent with lf mutations. In addition to increasing unc-43(gf) locomotion (Fig 1D and Fig F), these 19 mutations also increased unc-43(gf) egg laying. We obtained the suppressors as single mutants in an unc-43(+) background and found that all 19 exhibited hyperactive locomotion compared to wild type, and most also exhibited decreased retention of eggs. These phenotypes suggested that the suppressors could be allelic to genes in the goa-1/egl-30 heterotrimeric G-protein signaling network. Complementation tests and genetic mapping showed that we had indeed isolated multiple alleles of goa-1, dgk-1, eat-16, and eat-11 (Table 1; Fig 1).

View larger version (111K): In this window In a new window Download PPT slide   Figure 1. Suppressors of the unc-43(gf) lethargy. Representative photographs compare (A) wild-type C. elegans with (B) unc-43(gf), (C) goa-1(sa734), (D) goa-1(sa734); unc-43(gf), (E) eat-16(sa609), and (F) eat-16(sa609); unc-43(gf) animals. The unc-43(gf) allele is n498. One animal was placed in the center of a bacterial lawn and photographed 30 min later. All animals were staged as L4 larvae and assayed 24 hr after growth at 20°. All photographs were taken at the same magnification. Note the increase in the number of tracks made in the bacterial lawn by suppressed animals in comparison to unc-43(gf).

goa-1, dgk-1, and eat-16 are expressed throughout the nervous system and are components of the goa-1/egl-30 network that has been shown to affect locomotion rate and egg-laying activity (Fig 2; MENDEL et al. 1995 ; SEGALAT et al. 1995 ; HAJDU-CRONIN et al. 1999 ; LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). goa-1, dgk-1, and eat-16 encode proteins that antagonize the EGL-30 (G_q) signaling pathway, which regulates production of the second messenger diacylglycerol (DAG). goa-1 encodes a G_o that may inhibit EGL-30 directly, act on a regulator of the EGL-30 pathway, or function in parallel (MENDEL et al. 1995 ; SEGALAT et al. 1995 ; HAJDU-CRONIN et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). eat-16 encodes a regulator of G-protein signaling (RGS) that appears to regulate EGL-30 activity directly (HAJDU-CRONIN et al. 1999 ), and dgk-1 encodes a diacylglycerol kinase that reduces levels of DAG, a product of phospholipase C activity (NURRISH et al. 1999 ). The effects of goa-1, dgk-1, and eat-16 mutations on locomotion rate and egg laying are opposite to those of egl-30 mutations. goa-1(lf), dgk-1(lf), and eat-16(lf) mutants exhibit hyperactivity and decreased retention of eggs, whereas egl-30(lf) mutants exhibit severe lethargy and increased retention of eggs (MENDEL et al. 1995 ; SEGALAT et al. 1995 ; BRUNDAGE et al. 1996 ; HAJDU-CRONIN et al. 1999 ; NURRISH et al. 1999 ). The goa-1/egl-30 network has been shown to regulate cholinergic neurotransmission between excitatory motor neurons and body-wall muscle (LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). Although its molecular identity is not known, eat-11 has been shown to interact genetically with egl-30; therefore, eat-11 is likely another gene that negatively regulates the EGL-30 signaling pathway (AVERY 1993 ; BRUNDAGE et al. 1996 ; LACKNER et al. 1999 ).

View larger version (17K): In this window In a new window Download PPT slide   Figure 2. A model for regulation of locomotion rate and egg-laying activity by the GOA-1/EGL-30 network. This model is compiled from the work of several groups (MENDEL et al. 1995 ; SEGALAT et al. 1995 ; BRUNDAGE et al. 1996 ; HAJDU-CRONIN et al. 1999 ; LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). GOA-1(Go) and EGL-30(Gq) couple to receptors in the plasma membrane and are each regulated by RGS proteins. EGL-30(Gq) activates phospholipase Cß (PLCß), encoded by the egl-8 gene. PLCß cleaves phosphatidylinositol 4,5-bisphosphate into DAG and inositol-1,4,5-triphosphate (BERRIDGE 1984 ). GOA-1, DGK-1, and EAT-16 antagonize signaling by the EGL-30 pathway. GOA-1 may inhibit EGL-30 directly, act on a regulator of the EGL-30 pathway such as EAT-16 or DGK-1, or function in parallel to EGL-30.

Since our screen recovered multiple alleles of genes in the goa-1/egl-30 network, we infer that unc-43 and the goa-1/egl-30 network act together to control locomotion rate. In addition, since the unc-43(gf) egg-laying defect was also suppressed by mutations in the goa-1/egl-30 network, we infer that unc-43 and this G-protein network also act together in the egg-laying system.

Analysis of locomotory behavior indicates that UNC-43 may regulate the GOA-1/EGL-30 network:
If mutations in goa-1, dgk-1, eat-16, and eat-11 suppress unc-43(gf) because one or more of the gene products is a direct target of the UNC-43 kinase, then null alleles of these genes may completely suppress the unc-43(gf) lethargy. To test this, we first measured the unc-43(gf) suppression using a radial locomotion assay. In this assay, animals are allowed to disperse from the origin of a circular plate. In addition to several of our suppressor alleles, we also assayed goa-1(n1134), a lf allele identified in other work (SEGALAT et al. 1995 ). We compared unc-43(gf) animals, suppressor (sup); unc-43(gf) double mutant animals, and sup single mutant animals. As expected, we found that unc-43(gf) animals have a severely reduced dispersal distance compared to wild type (Fig 3A). We found that alleles of goa-1, dgk-1, eat-16, and eat-11 suppress this locomotion defect significantly in sup; unc-43(gf) double mutant animals (Fig 3, B–E). However, the suppression of unc-43(gf) by the alleles of goa-1, dgk-1, and eat-16 that we examined in this assay is incomplete since the sup; unc-43(gf) double mutants do not disperse as far as the corresponding sup single mutants (Fig 3, B–D). In contrast, we found that the eat-11; unc-43(gf) double mutant animals disperse as well as the eat-11 single mutant animals at early timepoints (Fig 3E). However, even for eat-11, the dispersal of the single mutant exceeds that of the double mutant at later timepoints (data not shown).

View larger version (35K): In this window In a new window Download PPT slide   Figure 3. Radial locomotion of unc-43(gf) suppressors. (A) Wild-type animals disperse rapidly and evenly across the assay plate. unc-43(gf) animals are strikingly defective in dispersal compared with wild-type animals (P < 0.0001). (B–G) sup; unc-43(gf) strains disperse farther than unc-43(gf) [P = 0.02 for unc-43(gf) vs. goa-1(n1134); unc-43(gf), P = 0.003 for unc-43(gf) vs. dgk-1(sa748); unc-43(gf), and P < 0.0001 for all others]. The unc-43(gf) allele is n498. Each data point represents the combined measurements from 10 animals. The same unc-43(gf) curve is included in each panel for comparison. Error bars represent standard error of the mean. Differences between genotypes were analyzed for significance at the 15-min timepoint using Student's t-test. Other alleles of each suppressor gene were analyzed and the same trend as shown was observed. n1134 was identified in other work as a lf allele of goa-1 (SEGALAT et al. 1995 ). The curves resulting from the goa-1, dgk-1, and eat-16 single mutant data extend above the curve for wild-type because these single mutant animals tended to remain at the edge of the assay plate, rather than continuing to disperse evenly over the plate like wild-type animals.

The incomplete suppression of unc-43(gf) could be the result of non-null sup alleles. To control for this, we obtained dgk-1(sy428) and eat-16(sy438), which were identified as putative null alleles by genetic criteria in other work (HAJDU-CRONIN et al. 1999 ). We also sequenced the entire goa-1 coding region in goa-1(sa734) because this allele behaves similarly to goa-1(n363), a deletion allele that removes a region containing the goa-1 gene and perhaps other genes (SEGALAT et al. 1995 ). We found that sa734 contains an early stop mutation and is thus an excellent candidate for a molecular null. Assaying these alleles, we found that eat-16(sy438) shows strong but incomplete suppression in the radial assay, similar to the eat-16 alleles isolated in our screen (data not shown). In addition, our sa609 allele of eat-16 was also shown to behave genetically as a null (HAJDU-CRONIN et al. 1999 ). Interestingly, we found that null alleles of goa-1 and dgk-1 severely reduce dispersal of both sup single mutant and sup; unc-43(gf) double mutant animals. These alleles cause animals to move in a rapid but ineffective manner that results in little dispersal from a point of origin (data not shown). This ineffective mode of locomotion appears to result from an increase in wave amplitude, which has been previously reported for goa-1 (MENDEL et al. 1995 ; HAJDU-CRONIN et al. 1999 ), and an increase in the frequency of direction reversals. Therefore, we found that radial locomotion rate is a poor measure of the suppression conferred by null alleles of goa-1 and dgk-1.

As an alternative means of measuring the suppression of the unc-43(gf) lethargy, we measured the body-bend rate of single and double mutant animals. Table 2 shows that unc-43(gf) animals have markedly fewer body bends per minute than wild-type animals and that sup; unc-43(gf) double mutant animals exhibit a significantly higher body-bend rate than unc-43(gf) animals. We found that goa-1(sa734) suppresses unc-43(gf) to the level of the goa-1(sa734) single mutant in this assay, suggesting that UNC-43 may regulate GOA-1 activity. In contrast, the suppression by dgk-1(sy428) is incomplete. For eat-11, we analyzed sa833 since this allele confers greater hyperactivity and egg-laying activity than other eat-11 alleles (data not shown). We found that the suppression by eat-11(sa833) is also incomplete. Although the strength of the sa833 phenotype suggests that this may be a strong loss-of-function allele, rigorous determination of the eat-11 null phenotype awaits the cloning of the eat-11 gene. eat-16(sy438) conferred variable suppression, with some individual sup; unc-43(gf) animals showing complete suppression and others exhibiting weaker suppression. Because of this variability, we also assayed eat-16(sa609) and obtained similar results (data not shown). Since neither sy438 nor sa609 is a clear molecular null by sequence analysis (HAJDU-CRONIN et al. 1999 ), the variability of the eat-16 suppression could be due to residual EAT-16 activity. However, because sy438 and sa609 were shown to behave genetically as null alleles, we suggest that the body-bend assay may be a less sensitive measure of the eat-16 suppression than the radial assay, in which both sy438 and sa609 clearly conferred incomplete suppression. Differences between the radial assay and the body-bend assay are not surprising since these assays measure locomotion rate differently. Although other explanations are possible, a simple interpretation of our analysis of locomotion rate is that UNC-43 may regulate GOA-1 activity.

If the suppression of unc-43(gf) reflects a direct biochemical interaction between UNC-43 and the GOA-1/EGL-30 network, mutations in this network should suppress unc-43(gf) specifically and not strongly affect gf mutations that reduce locomotion rate by other mechanisms. To test the specificity of the unc-43(gf) suppression by mutations in the goa-1/egl-30 network, we examined the effect of goa-1(null) mutations on gf mutations in unc-93, unc-103, and unc-110 (Table 3). Like unc-43(gf), these gf mutants exhibit few body bends/minute. We found that goa-1(null) mutations increase the body-bend rate of unc-93(gf), unc-103(gf), and unc-110(gf) animals only slightly in comparison to their effect on unc-43(gf) animals. The weak effect of goa-1(null) on unc-110(gf) and unc-93(gf) mutants must be indirect and nonspecific since unc-93 and unc-110 function in body-wall muscle (LEVIN and HORVITZ 1992 ; D. JOHNSTONE and J. H. THOMAS, unpublished results), whereas goa-1 acts neuronally to control locomotion (MENDEL et al. 1995 ; SEGALAT et al. 1995 ; NURRISH et al. 1999 ). In contrast, unc-103 may function in some of the same neurons as the goa-1/egl-30 network since unc-103 probably acts in excitatory motor neurons (D. REINER and J. H. THOMAS, unpublished results). However, since goa-1(null) has only a weak effect on unc-103(gf), we conclude that this interaction is also indirect and nonspecific. Thus, gf mutations in unc-43, unc-93, unc103, and unc-110 reduce locomotion rate to a similar degree, but goa-1 mutations suppress unc-43(gf) significantly better. These results suggest that unc-43 and the goa-1/egl-30 pathway regulate locomotion rate via the same mechanism, supporting a model in which UNC-43 regulates GOA-1 activity.

Since the goa-1/egl-30 network has been shown to regulate synaptic transmission at cholinergic synapses (LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ), we examined the interaction between unc-43 and the goa-1/egl-30 network by assaying the response of animals to the acetylcholinesterase inhibitor aldicarb. Loss-of-function mutations in goa-1 or dgk-1 confer hypersensitivity to the paralytic effects of aldicarb (NURRISH et al. 1999 ), whereas lf mutations in egl-30 or egl-8 confer resistance to aldicarb-induced paralysis (LACKNER et al. 1999 ; MILLER et al. 1999 ). If unc-43 regulates the goa-1/egl-30 pathway, mutations in unc-43 should confer an altered response to aldicarb. To test this, we subjected wild-type, unc-43(gf), and unc-43(null) animals to 1 mM aldicarb and scored paralysis over time. We found that the unc-43 mutants show strikingly altered responses to aldicarb-induced paralysis in comparison to wild type: unc-43(gf) confers resistance to aldicarb-induced paralysis, whereas the putative null allele unc-43(n1186) (REINER et al. 1999 ) confers hypersensitivity (Fig 4A). In agreement with our measurements of locomotion rate, goa-1(null) suppresses unc-43(gf) completely: goa-1 (null); unc-43(gf) animals are as hypersensitive to aldicarb-induced paralysis as the goa-1(null) single mutant (Fig 4B). In contrast, the other suppressors confer incomplete suppression (Fig 4, C–E). These results indicate that unc-43 and the goa-1/egl-30 pathway regulate cholinergic synaptic transmission similarly and further support a model in which UNC-43 regulates GOA-1 activity.

View larger version (34K): In this window In a new window Download PPT slide   Figure 4. Effect of aldicarb on unc-43 mutants and unc-43(gf) suppressed strains. (A) Comparison between the response of wild-type, unc-43(gf), and unc-43(null) animals to 1 mM aldicarb exposure over 2 hr. unc-43(gf) animals show a decreased rate of paralysis compared with wild type (P = 0.002 for data at the 120-min timepoint), and unc-43(null) animals show an increased rate of paralysis in comparison with wild type (P < 0.0001 for data at the 60-min timepoint). (B) The goa-1(sa734); unc-43(gf) double mutant shows the same rapid rate of paralysis as the goa-1(sa734) single mutant (P = 0.25 for data at the 20-min timepoint). goa-1(sa734) is a putative null allele (see text). (C–E) sup; unc-43(gf) strains with putative null alleles of dgk-1 and eat-16 and with eat-11(sa833) exhibit paralysis rates that are intermediate. eat-16(sy438) gave similar results to eat-16(sa609). The unc-43(gf) allele is n498. unc-43(null) is the putative null allele n1186 (REINER et al. 1999 ). Each data point represents data combined from at least two independent experiments. Strains were analyzed in parallel for each experiment. Error bars represent standard error of the mean. The same wild-type and unc-43(gf) curves were included in each panel for comparison. Differences between unc-43 mutants and wild type were analyzed using Student's t-test. goa-1(sa734) vs. goa-1(sa734); unc-43(gf) was analyzed using Fisher's exact test applied to raw data since the goa-1(sa734) data at the 20-min timepoint had a standard deviation of zero. Data were not collected at the 70- to 110-min timepoints for some of the strains because their response trend became apparent at earlier timepoints.

The interaction between unc-43 and the goa-1/egl-30 network also occurs in the egg-laying system:
Although we isolated mutations in the goa-1/egl-30 network as suppressors of the unc-43(gf) lethargic phenotype, we found that these same mutations also suppressed the unc-43(gf) egg-laying defect. However, since the targets of unc-43 may vary in different tissues, the interaction between unc-43 and the goa-1/egl-30 pathway in the egg-laying system need not be the same as the interaction in the locomotory system. To test whether or not the gene products of the goa-1/egl-30 network might be targets of the UNC-43 kinase in the egg-laying system, we compared the egg-laying behavior of unc-43(gf), sup; unc-43(gf), and sup single mutant animals. To assess the activity of the egg-laying muscles, we scored the developmental stages of eggs laid (Table 4). Wild-type animals lay eggs at about the gastrulation stage of embryogenesis and do not accumulate excess eggs in their gonad. unc-43(gf) animals lay later-staged eggs and become bloated with retained eggs, indicating reduced activity of the egg-laying muscles. For goa-1, dgk-1, eat-16, and eat-11, the sup; unc-43(gf) strains lay eggs at significantly earlier stages than unc-43(gf) and retain fewer eggs in their gonad than unc-43(gf). The suppression by goa-1(sa734) is the strongest: goa-1(sa734); unc-43(gf) animals lay eggs as early as goa-1(sa734) single mutant animals, indicating complete suppression. In contrast, the dgk-1, eat-16, and eat-11 sup; unc-43(gf) strains show strong but incomplete suppression. In support of the specificity of the interaction between unc-43 and the goa-1/egl-30 network in the egg-laying system, unc-93(gf) and unc-103(gf), which cause animals to become bloated with retained eggs in addition to their effect on locomotion (GREENWALD and HORVITZ 1980 ; PARK and HORVITZ 1986 ), were not noticeably suppressed for egg laying by goa-1(sa734) (data not shown). These results are similar to the results we obtained for the suppression of the unc-43(gf) lethargy and indicate that UNC-43 may regulate GOA-1 activity in the egg-laying system.

Other genes in the goa-1/egl-30 network can suppress unc-43(gf):
Since we found that several genes in the goa-1/egl-30 network suppress unc-43(gf), we expected that other genes in this network would show a similar interaction. The suppressor alleles that we had isolated were lf mutations in genes that normally antagonize EGL-30 signaling. Therefore, we predicted that gf mutations in genes that positively regulate EGL-30 signaling would also suppress unc-43(gf). Such mutations were probably not isolated in our screen because gf mutations are rare. To test our prediction, we combined unc-43(gf) with transgenes that overexpress either EGL-30 or EGL-10, an RGS protein that is thought to inhibit goa-1 (Fig 2; KOELLE and HORVITZ 1996 ). The integrated transgenes syIs36[egl-30(+)] and nIs51[egl-10(+)] overexpress wild-type protein and confer phenotypes that are opposite to the lf phenotypes of these genes: animals carrying either transgene alone exhibit hyperactive locomotion and decreased retention of eggs (BRUNDAGE et al. 1996 ; KOELLE and HORVITZ 1996 ). Both transgenes strongly suppress the unc-43(gf) lethargy and egg-laying defects (Fig 3F and Fig G; Table 4). These results are consistent with UNC-43 regulation of the GOA-1/EGl-30 network in the locomotory and egg-laying systems. However, since the strength of this suppression is presumably dependent upon the amount of EGL-10 or EGL-30 that is expressed from the transgenes, we focused our analysis and conclusions on the lf mutations that suppress unc-43(gf).

UNC-43 may act through GOA-1 or EGL-30:
Since unc-43(gf) is likely to encode a kinase with Ca²⁺-independent activity, UNC-43(gf) may be largely independent of upstream regulators. Therefore, we expected that our screen would preferentially recover genes that act downstream or in parallel to unc-43. To further test whether the goa-1/egl-30 network acts downstream of unc-43, we made double mutants with unc-43(null) and either egl-30(lf) or syIs9[goa-1(gf)], an integrated transgene that overexpresses an activated form of GOA-1 (MENDEL et al. 1995 ). syIs9[goa-1(gf)] and egl-30(lf) animals lay eggs of later stages, become bloated with retained eggs, and are lethargic (MENDEL et al. 1995 ; BRUNDAGE et al. 1996 ; HAJDU-CRONIN et al. 1999 ). unc-43(null) animals lay eggs of earlier stages and exhibit complex locomotion phenotypes including spontaneous convulsions (REINER et al. 1999 ). If goa-1 and egl-30 act downstream of unc-43, we would expect the double mutants to exhibit the phenotypes of syIs9[goa-1(gf)] and egl-30(lf) single mutants. To compare the mutants we measured the egg-laying phenotype rather than locomotion rate since unc-43(null) severely disrupts coordinated movement. We found that the double mutants lay late-staged eggs like the respective syIs9[goa-1(gf)] or egl-30(lf) single mutants (Table 5). Furthermore, the double mutants become as bloated with retained eggs as the syIs9[goa-1(gf)] and egl-30(lf) single mutants (data not shown). Though we did not measure it, the locomotion phenotype of the double mutants appears similar to the respective syIs9[goa-1(gf)] or egl-30(lf) single mutants, rather than like unc-43(null): the double mutants are about as lethargic as the syIs9[goa-1(gf)] and egl-30(lf) single mutants, and we never observed spontaneous convulsions or bursts of rapid forward movement. Occasionally we observed double mutant animals that exhibited a very slight increase in locomotory activity over the respective syIs9[goa-1(gf)] or egl-30(lf) single mutant. The slight increase in activity may be due to residual EGL-30 activity from the ad805 allele and from transgene mosaicism in syIs9 [goa-1(gf)]; unc-43(null) animals. The results of this epistasis analysis are consistent with goa-1 and egl-30 acting downstream of unc-43 to control egg-laying activity and locomotion rate. This genetic relationship indicates that UNC-43 may regulate the GOA-1/EGL-30 network in the locomotory and egg-laying systems via the regulation of GOA-1 or EGL-30 activity.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The genetic interaction we have described between unc-43 and the goa-1/egl-30 network indicates that UNC-43 may directly regulate this network. Such a regulator is expected to act upstream of the goa-1/egl-30 network and, when activated, should be strongly suppressed by mutations in this network. Our screen with unc-43(gf) and our double mutant analysis with unc-43(null) are consistent with unc-43 acting upstream of goa-1 and egl-30. Quantitative analysis of the unc-43(gf) suppression demonstrates that mutations in the goa-1/egl-30 network suppress unc-43(gf) strongly and specifically. Since previous genetic analysis indicates that goa-1 may act upstream of egl-30 (HAJDU-CRONIN et al. 1999 ; MILLER et al. 1999 ), GOA-1 is a logical candidate for regulation by UNC-43. Our quantitative analysis of the unc-43(gf) suppression is consistent with this model since our data show that goa-1(null) suppresses the lethargy, aldicarb resistance, and egg-laying defects of unc-43(gf) to the level of the goa-1(null) single mutant. This is the expected result from a single target of UNC-43 regulation; therefore a simple model is that the effect of unc-43(gf) on locomotion rate and egg-laying activity is caused by inappropriate activation of GOA-1.

In addition to indicating that UNC-43 may regulate GOA-1 activity, our data indicate that GOA-1, in turn, may regulate EGL-30 activity rather than DGK-1 activity (see Fig 2). Since the suppression of unc-43(gf) by a putative null allele of dgk-1 is significantly weaker than the suppression by goa-1(null) in both the locomotory and egg-laying systems, DGK-1 may act partly or fully in parallel to GOA-1 rather than as an effector of GOA-1. This model for GOA-1 activity has been proposed by others (HAJDU-CRONIN et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). It has also been proposed that GOA-1 regulates EGL-30 activity by modulating EAT-16 (HAJDU-CRONIN et al. 1999 ). This model predicts that null alleles of eat-16 should suppress unc-43(gf) as strongly as goa-1(null). However, we found that the suppression by putative null alleles of eat-16 was clearly incomplete in the radial locomotion assay, the aldicarb response assay, and the egg-laying assay. These results suggest that eat-16 regulates egl-30 in parallel to goa-1, supporting the model in which GOA-1 regulates EGL-30 directly (HAJDU-CRONIN et al. 1999 ; MILLER et al. 1999 ). However, since the results of genetic analysis depend on the nature of the mutations analyzed, confirmation of these models awaits the identification of clear molecular null alleles of dgk-1 and eat-16, as well as biochemical analysis.

The genetic interaction between unc-43 and goa-1 suggests that UNC-43 could directly activate GOA-1 by phosphorylation or could indirectly activate GOA-1 by interacting with a GOA-1 regulator. EGL-10, the RGS protein that is thought to regulate GOA-1 activity, is an obvious candidate for such an interaction. Since RGS proteins decrease G activity by increasing their rate of GTP hydrolysis (HUNT et al. 1996 ), inhibition of EGL-10 (by UNC-43 phosphorylation) would increase GOA-1 activity. The amino acid sequences of both GOA-1 and EGL-10 contain CaMKII consensus phosphorylation sites (RXXS/T). Although such sites are not necessarily required or predictive of CaMKII phosphorylation (KENNELLY and KREBS 1991 ), one consensus site in the N terminus of EGL-10 has been perfectly conserved in the N terminus of human RGS7. Human RGS7 is a proposed functional homolog of EGL-10, sharing 75% amino acid identity in the N terminus and 53% overall amino acid identity (KOELLE and HORVITZ 1996 ). However, since egl-10(null) mutants are not as severely lethargic as unc-43(gf) (data not shown), a second UNC-43 target that also regulates GOA-1 activity is required to explain the entire unc-43(gf) effect on locomotion rate. Since Ca-MKII is a multifunctional CaM kinase able to phosphorylate many substrates in vitro (reviewed in HANSON and SCHULMAN 1992 ), models in which UNC-43 regulates more than one component of the GOA-1/EGL-30 network are plausible. However, we favor the simpler model in which UNC-43 regulates GOA-1 directly. Additional biochemical analysis will be required to examine these possibilities.

Previous identification of putative CaMKII phosphorylation targets has relied almost exclusively on candidate gene approaches and in vitro phosphorylation assays (HANSON and SCHULMAN 1992 ). Such approaches have surely missed some targets and implicated other, nonphysiological targets. G subunits have not been previously implicated as CaMKII phosphorylation targets, though G_i and G_t (transducin) have been shown to be phosphorylated in vitro by protein kinase C, and G_q and G_s subunits have been shown to undergo tyrosine phosphorylation (KATADA et al. 1985 ; ZICK et al. 1986 ; MOYERS et al. 1995 ; UMEMORI et al. 1997 ). RGS proteins belong to a relatively new protein family and their phosphorylation status has not been reported. Interestingly, a member of a different family of GTPase-activating proteins that regulates the small G protein Ras has been shown to be phosphorylated in a CaMKII-dependent manner in rat (CHEN et al. 1998 ).

Our genetic analysis does not exclude the possibility that UNC-43 acts in parallel to the GOA-1/EGL-30 network. CaMKII has been shown to regulate neuronal activity by several different mechanisms, including interactions with adenylyl cyclase, -amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA)-type glutamate receptors, and Eag-related K⁺ channels (MCGLADE-MCCULLOH et al. 1993 ; GRIFFITH et al. 1994 ; BARRIA et al. 1997 ; WEI et al. 1998 ; REINER et al. 1999 ). However, if the unc-43(gf) lethargy were due to inappropriate regulation of these genes, we would have expected to recover them in our screen. To examine the possibility that some genes were missed in our screen, we tested whether lf mutations in glr-1, a conserved AMPA-type glutamate receptor (HART et al. 1995 ; MARICQ et al. 1995 ), would suppress unc-43(gf). We found that glr-1(lf) suppressed none of the unc-43(gf) phenotypes (data not shown). Therefore, since we recovered multiple alleles of goa-1, dgk-1, eat-16, and eat-11, and the suppression by these alleles is striking, we propose that CaMKII also regulates neuronal activity by controlling G_o/G_q pathways.

unc-43 and the goa-1/egl-30 network are widely expressed throughout the nervous system (MENDEL et al. 1995 ; SEGALAT et al. 1995 ; KOELLE and HORVITZ 1996 ; HAJDU-CRONIN et al. 1999 ; LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ; REINER et al. 1999 ; E. NEWTON and J. H. THOMAS, unpublished results). This coexpression makes a direct interaction between UNC-43 and the GOA-1/EGL-30 network plausible. Work by other groups indicates that the goa-1/egl-30 network regulates synaptic transmission in body-wall muscle motor neurons and perhaps other cell types (LACKNER et al. 1999 ; MILLER et al. 1999 ; NURRISH et al. 1999 ). Specifically, goa-1 and egl-30 are thought to mediate presynaptic modulation of motor neuron synaptic transmission, in part, by effecting changes in localization of UNC-13, a DAG-binding protein predicted to regulate synaptic vesicles (MARUYAMA and BRENNER 1991 ; BETZ et al. 1997 ). goa-1 is thought to mediate the effect of humorally acting serotonin and perhaps other neuromodulators (NURRISH et al. 1999 ). An interaction between unc-43 and the goa-1/egl-30 network in motor neurons could explain the effect of these genes on locomotion rate; however, other neurons are also implicated in controlling locomotion rate. For example, disruption of the mechanosensory neurons that mediate the response to gentle body touch also results in lethargy (reviewed in DRISCOLL and KAPLAN 1997 ). Since the neuronal circuitry controlling locomotion rate has not been fully defined, and unc-43 and the goa-1/egl-30 network have broad neuronal expression, experiments with mosaic animals will be required to determine where these genes are acting to control locomotion rate.

unc-43 and members of the goa-1/egl-30 network are also coexpressed in the egg-laying system. goa-1, egl-10, and eat-16 have been shown to be expressed in the hermaphrodite-specific neuron (HSN) motor neurons that control egg laying, and goa-1 and eat-16 have also been shown to be expressed in the egg-laying muscles (MENDEL et al. 1995 ; SEGALAT et al. 1995 ; KOELLE and HORVITZ 1996 ; HAJDU-CRONIN et al. 1999 ). Since UNC-43 is also present in both the HSN motor neurons and the egg-laying muscles (E. NEWTON and J. H. THOMAS, unpublished results), an interaction between unc-43 and the goa-1/egl-30 network could occur in either cell type. The mechanism by which the goa-1/egl-30 network controls egg-laying behavior has not been well defined. Heterotrimeric G proteins are activated by ligand-bound seven-pass transmembrane receptors (SIMON et al. 1991 ). goa-1 does not appear to be an effector of serotonin in the egg-laying system since exogenous serotonin stimulates egg laying (HORVITZ et al. 1982 ), whereas goa-1 activity inhibits egg laying (MENDEL et al. 1995 ; SEGALAT et al. 1995 ). This observation has led to the suggestion that goa-1 couples to a different neurotransmitter in the egg-laying system (NURRISH et al. 1999 ). The control of egg laying by unc-43 and the goa-1/egl-30 network is probably complex since these genes may function both presynaptically (in the HSN neurons) and postsynaptically (in the egg-laying muscles). However, despite this complexity, our genetic analysis indicates that unc-43 and the goa-1/egl-30 network function similarly in the egg-laying system.

An interaction between CaMKII, G_o, and G_q pathways could be relevant to mammalian behavior since there is a high degree of conservation between these C. elegans proteins and their mammalian counterparts. In particular, GOA-1, EGL-30, and UNC-43 share 70–80% overall amino acid identity with mammalian G_o, G_q, and CaMKII, respectively (LOCHRIE et al. 1991 ; BRUNDAGE et al. 1996 ; REINER et al. 1999 ). G_o subunits, several RGS proteins, and CaMKII are highly expressed in the mammalian brain (STERNWEIS and ROBISHAW 1984 ; ERONDU and KENNEDY 1985 ; KOELLE and HORVITZ 1996 ; GOLD et al. 1997 ), indicating that an interaction between the mammalian proteins is plausible. Strikingly, mice lacking either G_o or CaMKII exhibit increased locomotory activity (SILVA et al. 1992A , SILVA et al. 1992B ; JIANG et al. 1998 ). The striking similarity at the behavioral level of perturbation of these genes in mice and C. elegans indicates that the gene interactions we have described for C. elegans may be relevant to mammalian behavior.

	ACKNOWLEDGMENTS

We thank Michael Ailion, Takao Inoue, Duncan Johnstone, Elizabeth Newton, David Reiner, and the rest of the Thomas lab for their comments on this manuscript and for helpful discussions. We thank Jennifer Knapp and Dave Reiner for isolating sa581, sa585, sa586, sa603, sa604, sa605, and sa609. We also thank Dave Reiner for constructing egl-30(ad805); unc-43(n1186) and the initial goa-1; unc-43(n498) double mutant and for providing unpublished results for unc-103. We thank Elizabeth Newton for providing unpublished results for the unc-43 expression pattern and Duncan Johnstone for identifying unc-110(sa859). We thank Paul Sternberg and Yvonne Hajdu-Cronin for providing dgk-1(sy428) and eat-16(sy438) prior to their publication. Some strains used in this work were provided by the Caenorhabditis Genetics Research Center, which is funded by the National Institutes of Health (NIH) National Center for Research Resources. NIH grant N530187 to J.H.T. supported this work.

Manuscript received May 10, 2000; Accepted for publication June 27, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AVERY, L., 1993  The genetics of feeding in Caenorhabditis elegans.. Genetics 133:897-917[Abstract].

AVERY, L., and J. H. THOMAS, 1997 Feeding and defecation, pp. 679–716 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BACH, M. E., R. D. HAWKINS, M. OSMAN, E. R. KANDEL, and M. MAYFORD, 1995  Impairment of spatial but not contextual memory in CaMKII mutant mice with a selective loss of hippocampal LTP in the range of the theta frequency. Cell 81:905-915[Medline].

BARGMANN, C. I., and I. MORI, 1997 Chemotaxis and thermotaxis, pp. 717–737 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BARRIA, A., D. MULLER, V. DERKACH, L. C. GRIFFITH, and T. R. SODERLING, 1997  Regulatory phosphorylation of AMPA-type glutamate receptors by CaM-KII during long-term potentiation. Science 276:2042-2045[Abstract/Free Full Text].

BERRIDGE, M. J., 1984  Inositol trisphosphate and diacylglycerol as second messengers. Biochem. J. 220:345-360[Medline].

BETZ, A., M. OKAMOTO, F. BENSELER, and N. BROSE, 1997  Direct interaction of the rat unc-13 homologue Munc13-1 with the N terminus of syntaxin. J. Biol. Chem. 272:2520-2526[Abstract/Free Full Text].

BRENNER, S., 1974  The genetics of Caenorhabditis elegans.. Genetics 77:71-94[Abstract/Free Full Text].

BRUNDAGE, L., L. AVERY, A. KATZ, U. J. KIM, and J. E. MENDEL et al., 1996  Mutations in a C. elegans Gq gene disrupt movement, egg laying, and viability. Neuron 16:999-1009[Medline].

CHALFIE, M., J. E. SULSTON, J. G. WHITE, E. SOUTHGATE, and J. N. THOMSON et al., 1985  The neural circuit for touch sensitivity in Caenorhabditis elegans.. J. Neurosci. 5:956-964[Abstract].

CHEN, C., D. G. RAINNIE, R. W. GREENE, and S. TONEGAWA, 1994  Abnormal fear response and aggressive behavior in mutant mice deficient for -calcium-calmodulin kinase II. Science 266:291-294[Abstract/Free Full Text].

CHEN, H. J., M. ROJAS-SOTO, A. OGUNI, and M. B. KENNEDY, 1998  A synaptic Ras-GTPase activating protein (p135 SynGAP) inhibited by CaM kinase II. Neuron 20:895-904[Medline].

DE KONINCK, P. and H. SCHULMAN, 1998  Sensitivity of CaM kinase II to the frequency of Ca2+ oscillations. Science 279:227-230[Abstract/Free Full Text].

DRISCOLL, M., and J. KAPLAN, 1997 Mechanotransduction, pp. 645–677 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

ERONDU, N. E. and M. B. KENNEDY, 1985  Regional distribution of type II Ca2+/calmodulin-dependent protein kinase in rat brain. J. Neurosci. 5:3270-3277[Abstract].

GOLD, S. J., Y. G. NI, H. G. DOHLMAN, and E. J. NESTLER, 1997  Regulators of G-protein signaling (RGS) proteins: region-specific expression of nine subtypes in rat brain. J. Neurosci. 17:8024-8037[Abstract/Free Full Text].

GOLDBERG, J., A. C. NAIRN, and J. KURIYAN, 1996  Structural basis for the autoinhibition of calcium/calmodulin-dependent protein kinase I. Cell 84:875-887[Medline].

GREENWALD, I. S. and H. R. HORVITZ, 1980  unc-93(e1500): a behavioral mutant of Caenorhabditis elegans that defines a gene with a wild-type null phenotype. Genetics 96:147-164[Abstract/Free Full Text].

GRIFFITH, L. C., L. M. VERSELIS, K. M. AITKEN, C. P. KURIACOU, and W. DANHO et al., 1993  Inhibition of calcium/calmodulin-dependent protein kinase in Drosophila disrupts behavioral plasticity. Neuron 10:501-509[Medline].

GRIFFITH, L. C., J. WANG, Y. ZHONG, C. F. WU, and R. J. GREENSPAN, 1994  Calcium/calmodulin-dependent protein kinase II and potassium channel subunit eag similarly affect plasticity in Drosophila. Proc. Natl. Acad. Sci. USA 91:10044-10048[Abstract/Free Full Text].

HAJDU-CRONIN, Y. M., W. J. CHEN, G. PATIKOGLOU, M. R. KOELLE, and P. W. STERNBERG, 1999  Antagonism between G(o) and G(q) in Caenorhabditis elegans: the RGS protein EAT-16 is necessary for G(o) signaling and regulates G(q) activity. Genes Dev. 13:1780-1793[Abstract/Free Full Text].

HANSON, P. I. and H. SCHULMAN, 1992  Neuronal Ca2+/calmodulin-dependent protein kinases. Annu. Rev. Biochem. 61:559-601[Medline].

HANSON, P. I., T. MEYER, L. STRYER, and H. SCHULMAN, 1994  Dual role of calmodulin in autophosphorylation of multifunctional CaM kinase may underlie decoding of calcium signals. Neuron 12:943-956[Medline].

HART, A. C., S. SIMS, and J. M. KAPLAN, 1995  Synaptic code for sensory modalities revealed by C. elegans GLR-1 glutamate receptor. Nature 378:82-85[Medline].

HORVITZ, H. R., M. CHALFIE, C. TRENT, J. E. SULSTON, and P. D. EVANS, 1982  Serotonin and octopamine in the nematode Caenorhabditis elegans.. Science 216:1012-1014[Abstract/Free Full Text].

HUNT, T. W., T. A. FIELDS, P. J. CASEY, and E. G. PERALTA, 1996  RGS10 is a selective activator of G i GTPase activity. Nature 383:175-177[Medline].

JIANG, M., M. S. GOLD, G. BOULAY, K. SPICHER, and M. PEYTON et al., 1998  Multiple neurological abnormalities in mice deficient in the G protein Go. Proc. Natl. Acad. Sci. USA 95:3269-3274[Abstract/Free Full Text].

JORGENSEN, E. M., and C. RANKIN, 1997 Neural plasticity, pp. 769–790 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

KATADA, T., A. G. GILMAN, Y. WATANABE, S. BAUER, and K. H. JAKOBS, 1985  Protein kinase C phosphorylates the inhibitory guanine-nucleotide-binding regulatory component and apparently suppresses its function in hormonal inhibition of adenylate cyclase. Eur. J. Biochem. 151:431-437[Medline].

KENNELLY, P. J. and E. G. KREBS, 1991  Consensus sequences as substrate specificity determinants for protein kinases and protein phosphatases. J. Biol. Chem. 266:15555-15558[Free Full Text].

KOELLE, M. R. and H. R. HORVITZ, 1996  EGL-10 regulates G protein signaling in the C. elegans nervous system and shares a conserved domain with many mammalian proteins. Cell 84:115-125[Medline].

LACKNER, M. R., S. J. NURRISH, and J. M. KAPLAN, 1999  Facilitation of synaptic transmission by EGL-30 Gq and EGL-8 PLCbeta: DAG binding to UNC-13 is required to stimulate acetylcholine release. Neuron 24:335-346[Medline].

LEVIN, J. Z. and H. R. HORVITZ, 1992  The Caenorhabditis elegans unc-93 gene encodes a putative transmembrane protein that regulates muscle contraction. J. Cell Biol. 117:143-155[Abstract/Free Full Text].

LOCHRIE, M. A., J. E. MENDEL, P. W. STERNBERG, and M. I. SIMON, 1991  Homologous and unique G protein alpha subunits in the nematode Caenorhabditis elegans.. Cell. Regul. 2:135-154[Medline].

MALINOW, R., H. SCHULMAN, and R. W. TSIEN, 1989  Inhibition of postsynaptic PKC or CaMKII blocks induction but not expression of LTP. Science 245:862-866[Abstract/Free Full Text].

MARICQ, A. V., E. PECKOL, M. DRISCOLL, and C. I. BARGMANN, 1995  Mechanosensory signalling in C. elegans mediated by the GLR-1 glutamate receptor. Nature 378:78-81[Medline].

MARUYAMA, I. N. and S. BRENNER, 1991  A phorbol ester/diacylglycerol-binding protein encoded by the unc-13 gene of Caenorhabditis elegans.. Proc. Natl. Acad. Sci. USA 88:5729-5733[Abstract/Free Full Text].

MAYFORD, M., J. WANG, E. R. KANDEL, and T. J. O'DELL, 1995  CaMKII regulates the frequency-response function of hippocampal synapses for the production of both LTD and LTP. Cell 81:891-904[Medline].

MCGLADE-MCCULLOH, E., H. YAMAMOTO, S. E. TAN, D. A. BRICKEY, and T. R. SODERLING, 1993  Phosphorylation and regulation of glutamate receptors by calcium/calmodulin-dependent protein kinase II. Nature 362:640-642[Medline].

MCINTIRE, S. L., E. JORGENSEN, and H. R. HORVITZ, 1993a  Genes required for GABA function in Caenorhabditis elegans.. Nature 364:334-337[Medline].

MCINTIRE, S. L., E. JORGENSEN, J. KAPLAN, and H. R. HORVITZ, 1993b  The GABAergic nervous system of Caenorhabditis elegans.. Nature 364:337-341[Medline].

MENDEL, J. E., H. C. KORSWAGEN, K. S. LIU, Y. M. HAJDU-CRONIN, and M. I. SIMON et al., 1995  Participation of the protein Go in multiple aspects of behavior in C. elegans.. Science 267:1652-1655[Abstract/Free Full Text].

MILLER, K. G., M. D. EMERSON, and J. B. RAND, 1999  Go and diacylglycerol kinase negatively regulate the Gq pathway in C. elegans.. Neuron 24:323-333[Medline].

MOYERS, J. S., M. E. LINDER, J. D. SHANNON, and S. J. PARSONS, 1995  Identification of the in vitro phosphorylation sites on Gs alpha mediated by pp60c-src. Biochem. J. 305:411-417.

NELSON, T. J. and D. L. ALKON, 1997  Biochemistry of molluscan learning and memory. Bioessays 19:1045-1053.

NURRISH, S. J., L. SÉGALAT, and J. M. KAPLAN, 1999  Serotonin inhibition of synaptic transmission: G_o decreases the abundance of UNC-13 at release sites. Neuron 24:231-242[Medline].

PARK, E. C. and H. R. HORVITZ, 1986  Mutations with dominant effects on the behavior and morphology of the nematode Caenorhabditis elegans.. Genetics 113:821-852[Abstract/Free Full Text].

REINER, D. J., D. WEINSHENKER, and J. H. THOMAS, 1995  Analysis of dominant mutations affecting muscle excitation in Caenorhabditis elegans.. Genetics 141:961-976[Abstract].

REINER, D. J., E. M. NEWTON, H. TIAN, and J. H. THOMAS, 1999  Diverse behavioural defects caused by mutations in Caenorhabditis elegans unc-43 CaM kinase II. Nature 402:199-203[Medline].

ROTENBERG, A., M. MAYFORD, R. D. HAWKINS, E. R. KANDEL, and R. U. MULLER, 1996  Mice expressing activated CaMKII lack low frequency LTP and do not form stable place cells in the CA1 region of the hippocampus. Cell 87:1351-1361[Medline].

SAWIN, E. R., 1996 Genetic and cellular analysis of modulated behaviors in Caenorhabditis elegans. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA.

SCHAFER, W. R. and C. J. KENYON, 1995  A calcium-channel homologue required for adaptation to dopamine and serotonin in Caenorhabditis elegans.. Nature 375:73-78[Medline].

SCHIERENBERG, E., 1986  Developmental strategies during early embryogenesis of Caenorhabditis elegans.. J. Embryol. Exp. Morphol. 97:31-44.

SÉGALAT, L., D. A. ELKES, and J. M. KAPLAN, 1995  Modulation of serotonin-controlled behaviors by Go in Caenorhabditis elegans.. Science 267:1648-1651[Abstract/Free Full Text].

SILVA, A. J., C. F. STEVENS, S. TONEGAWA, and Y. WANG, 1992a  Deficient hippocampal long-term potentiation in -calcium-calmodulin kinase II mutant mice. Science 257:201-206[Abstract/Free Full Text].

SILVA, A. J., R. PAYLOR, J. M. WEHNER, and S. TONEGAWA, 1992b  Impaired spatial learning in -calcium-calmodulin kinase II mutant mice. Science 257:206-211[Abstract/Free Full Text].

SIMON, M. I., M. P. STRATHMANN, and N. GAUTAM, 1991  Diversity of G proteins in signal transduction. Science 252:802-808[Abstract/Free Full Text].

STERNWEIS, P. C. and J. D. ROBISHAW, 1984  Isolation of two proteins with high affinity for guanine nucleotides from membranes of bovine brain. J. Biol. Chem. 259:13806-13813[Abstract/Free Full Text].

SULSTON, J. E., and J. HODGKIN, 1988 Methods, pp. 587–606 in The Nematode Caenorhabditis elegans, edited by W. B. WOOD. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SULSTON, J. E., E. SCHIERENBERG, J. G. WHITE, and J. N. THOMSON, 1983  The embryonic cell lineage of the nematode Caenorhabditis elegans.. Dev. Biol. 100:64-119[Medline].

TRENT, C., N. TSUNG, and H. R. HORVITZ, 1983  Egg-laying defective mutants of the nematode C. elegans.. Genetics 104:619-647[Abstract/Free Full Text].

UMEMORI, H., T. INOUE, S. KUME, N. SEKIYAMA, and M. NAGAO et al., 1997  Activation of the G protein Gq/11 through tyrosine phosphorylation of the alpha subunit. Science 276:1878-1881[Abstract/Free Full Text].

WEI, J., A. Z. ZHAO, G. C. CHAN, L. P. BAKER, and S. IMPEY et al., 1998  Phosphorylation and inhibition of olfactory adenylyl cyclase by CaM kinase II in neurons: a mechanism for attenuation of olfactory signals. Neuron 21:495-504[Medline].

WHITE, J. G., E. SOUTHGATE, J. N. THOMSON, and S. BRENNER, 1986  The structure of the nervous system of the nematode Caenorhabditis elegans.. Philos. Trans. R. Soc. Lond. B Biol. Sci. 314:1-340.

ZHENG, Y., P. J. BROCKIE, J. E. MELLEM, D. M. MADSEN, and A. V. MARICQ, 1999  Neuronal control of locomotion in C. elegans is modified by a dominant mutation in the GLR-1 ionotropic glutamate receptor. Neuron 24:347-361[Medline].

ZICK, Y., R. SAGI-EISENBERG, M. PINES, P. GIERSCHIK, and A. M. SPIEGEL, 1986  Multisite phosphorylation of the alpha subunit of transducin by the insulin receptor kinase and protein kinase C. Proc. Natl. Acad. Sci. USA 83:9294-9297[Abstract/Free Full Text].

This article has been cited by other articles:

				C. J. Locke, B. B. Kautu, K. P. Berry, S. K. Lee, K. A. Caldwell, and G. A. Caldwell Pharmacogenetic Analysis Reveals a Post-Developmental Role for Rac GTPases in Caenorhabditis elegans GABAergic Neurotransmission Genetics, December 1, 2009; 183(4): 1357 - 1372. [Abstract] [Full Text] [PDF]

				J. S. Dittman and J. M. Kaplan Behavioral Impact of Neurotransmitter-Activated G-Protein-Coupled Receptors: Muscarinic and GABAB Receptors Regulate Caenorhabditis elegans Locomotion J. Neurosci., July 9, 2008; 28(28): 7104 - 7112. [Abstract] [Full Text] [PDF]

				S. L. Williams, S. Lutz, N. K. Charlie, C. Vettel, M. Ailion, C. Coco, J. J.G. Tesmer, E. M. Jorgensen, T. Wieland, and K. G. Miller Trio's Rho-specific GEF domain is the missing G{alpha}q effector in C. elegans Genes & Dev., November 1, 2007; 21(21): 2731 - 2746. [Abstract] [Full Text] [PDF]

				Q. Liu, B. Chen, Q. Ge, and Z.-W. Wang Presynaptic Ca2+/Calmodulin-Dependent Protein Kinase II Modulates Neurotransmitter Release by Activating BK Channels at Caenorhabditis elegans Neuromuscular Junction J. Neurosci., September 26, 2007; 27(39): 10404 - 10413. [Abstract] [Full Text] [PDF]

				J. M. Wheeler and J. H. Thomas Identification of a Novel Gene Family Involved in Osmotic Stress Response in Caenorhabditis elegans Genetics, November 1, 2006; 174(3): 1327 - 1336. [Abstract] [Full Text] [PDF]

				S. Suo, Y. Kimura, and H. H. M. Van Tol Starvation Induces cAMP Response Element-Binding Protein-Dependent Gene Expression through Octopamine-Gq Signaling in Caenorhabditis elegans J. Neurosci., October 4, 2006; 26(40): 10082 - 10090. [Abstract] [Full Text] [PDF]

				N. K. Charlie, M. A. Schade, A. M. Thomure, and K. G. Miller Presynaptic UNC-31 (CAPS) Is Required to Activate the G{alpha}s Pathway of the Caenorhabditis elegans Synaptic Signaling Network Genetics, February 1, 2006; 172(2): 943 - 961. [Abstract] [Full Text] [PDF]

				K. Afshar, F. S. Willard, K. Colombo, D. P. Siderovski, and P. Gonczy Cortical localization of the G{alpha} protein GPA-16 requires RIC-8 function during C. elegans asymmetric cell division Development, October 15, 2005; 132(20): 4449 - 4459. [Abstract] [Full Text] [PDF]

				M. A. Schade, N. K. Reynolds, C. M. Dollins, and K. G. Miller Mutations That Rescue the Paralysis of Caenorhabditis elegans ric-8 (Synembryn) Mutants Activate the G{alpha}s Pathway and Define a Third Major Branch of the Synaptic Signaling Network Genetics, February 1, 2005; 169(2): 631 - 649. [Abstract] [Full Text] [PDF]

				N. K. Reynolds, M. A. Schade, and K. G. Miller Convergent, RIC-8-Dependent G{alpha} Signaling Pathways in the Caenorhabditis elegans Synaptic Signaling Network Genetics, February 1, 2005; 169(2): 651 - 670. [Abstract] [Full Text] [PDF]

				A. H. Hawasli, O. Saifee, C. Liu, M. L. Nonet, and C. M. Crowder Resistance to Volatile Anesthetics by Mutations Enhancing Excitatory Neurotransmitter Release in Caenorhabditis elegans Genetics, October 1, 2004; 168(2): 831 - 843. [Abstract] [Full Text] [PDF]

				J. J. Moresco and M. R. Koelle Activation of EGL-47, a G{alpha}o-Coupled Receptor, Inhibits Function of Hermaphrodite-Specific Motor Neurons to Regulate Caenorhabditis elegans Egg-Laying Behavior J. Neurosci., September 29, 2004; 24(39): 8522 - 8530. [Abstract] [Full Text] [PDF]

				J. Bandyopadhyay, J. Lee, J. Lee, J. I. Lee, J.-R. Yu, C. Jee, J.-H. Cho, S. Jung, M. H. Lee, S. Zannoni, et al. Calcineurin, a Calcium/Calmodulin-dependent Protein Phosphatase, Is Involved in Movement, Fertility, Egg Laying, and Growth in Caenorhabditis elegans Mol. Biol. Cell, September 1, 2002; 13(9): 3281 - 3293. [Abstract] [Full Text] [PDF]