A neuronal acetylcholine receptor regulates the balance of muscle excitation and inhibition in Caenorhabditis elegans

Abstract

In the nematode Caenorhabditis elegans, cholinergic motor neurons stimulate muscle contraction as well as activate GABAergic motor neurons that inhibit contraction of the contralateral muscles. Here, we describe the composition of an ionotropic acetylcholine receptor that is required to maintain excitation of the cholinergic motor neurons. We identified a gain-of-function mutation that leads to spontaneous muscle convulsions. The mutation is in the pore domain of the ACR-2 acetylcholine receptor subunit and is identical to a hyperactivating mutation in the muscle receptor of patients with myasthenia gravis. Screens for suppressors of the convulsion phenotype led to the identification of other receptor subunits. Cell-specific rescue experiments indicate that these subunits function in the cholinergic motor neurons. Expression of these subunits in Xenopus oocytes demonstrates that the functional receptor is comprised of three alpha-subunits, UNC-38, UNC-63 and ACR-12, and two non-alpha-subunits, ACR-2 and ACR-3. Although this receptor exhibits a partially overlapping subunit composition with the C. elegans muscle acetylcholine receptor, it shows distinct pharmacology. Recordings from intact animals demonstrate that loss-of-function mutations in acr-2 reduce the excitability of the cholinergic motor neurons. By contrast, the acr-2(gf) mutation leads to a hyperactivation of cholinergic motor neurons and an inactivation of downstream GABAergic motor neurons in a calcium dependent manner. Presumably, this imbalance between excitatory and inhibitory input into muscles leads to convulsions. These data indicate that the ACR-2 receptor is important for the coordinated excitation and inhibition of body muscles underlying sinusoidal movement.

Figure 1. acr-2 locus.

(A) Genetic position of acr-2 on the left arm of the X chromosome. (B) Representative cosmids and subclones used in germline transformation rescue of acr-2(n2420gf). Rescue of spontaneous convulsion is indicated by a plus sign (+); and no rescue by a minus sign (−). Canonical cosmids F38B6 and K11G12 are shown in parentheses. (C) ACR-2 is a member of the UNC-29 class non–α-subunits of the acetylcholine receptor family . (D) The n2420 mutation causes a valine-to-methionine change in the 13′ position of the pore-forming second transmembrane domain (TM2, residues are colored in teal). The amino acid at the 13′ position is oriented toward the pore; orange mark residues lining the pore .

Figure 2. Mutant phenotypes and pharmacological analysis of acr-2(n2420gf) mutants.

(A) Suppression of the convulsions of acr-2(n2420gf) mutants by exogenous mecamylamine (100 µM). The effects of mecamylamine are reversible. (B) acr-2(n2420gf) mutants are hypersensitive, and acr-2(lf) mutants are moderately resistant to aldicarb. The graph shows the time course of the response of adult hermaphrodites to 0.5 mM aldicarb, from three trials with ten animals per genotype per trial. *p<0.001 between wild type and mutants (two-way ANOVA). (C) acr-2(n2420gf) mutants are hypersensitive to levamisole. The graph shows the time course of the response of 1-d-old hermaphrodites to 1 mM levamisole, from three trials with ten animals per genotype per trial. *p<0.001 between wild type and acr-2(n2420gf) (two-way ANOVA). (D) Quantification of convulsion rates of 1-d-old hermaphrodites of various genotypes. (E) Developmental onset of the convulsion phenotype of acr-2(n2420gf) animals occurs during the L3 stage. OA, 2-d-old adult; YA, 1-d-old adult.

Figure 3. acr-2 is expressed and functions in the cholinergic ventral cord motor neurons.

(A) Expression pattern from a 3.5-kb acr-2 promoter-driven Pacr-2-GFP transgene (juIs14) in an L1 (upper image) and an L4 larva. Middle image is a side view, and the bottom image is a ventral view of an L4 larva. The number and positions of the cells and the sides of the commissures indicate that they are embryonically born DA and DB and postembryonically born VA and VB neurons. Expression is also seen in DVC in the tail (arrow, middle panel). Scale bars indicate 10 µm. (B). The top image shows nonoverlapping expression of Pacr-2-GFP (juIs14) (green) and Pttr-39-mcherry (juIs223) (red) marking the DD and VD GABAergic neurons. The bottom images show nonoverlapping expression of the 1.8-kb promoter-driven Pacr-2-mcherry transgene (juEx2045) (red) and those of Pglr-1-GFP (nuIs1) (green) or Pnmr-1-GFP (akIs3) (green), which label the command interneurons in the head (left panels) and tail (right panels) ganglia. Transcriptional activity of the acr-2 1.8-kb promoter is not seen in any head neurons. Scale bars indicate 10 µm. (C) Quantification of cell-type–specific transgenic rescue of the convulsion defects in acr-2(n2420gf) animals. Pacr-2::acr-2 contains the full-length genomic coding sequence driven by the 3.5-kb-long promoter. Pacr-2::acr-2(mini) contains cDNA that replaced genomic DNA from exon 2, driven by the 1.8-kb promoter. Punc-25::acr-2 contains the full-length genomic coding sequences driven by the 1.4-kb unc-25 promoter. n = 10, 10, 5, 10, and 8 for N2, acr-2(n2420gf), acr-2(n2420gf);Pacr-2-acr-2(oxEx707), acr-2(n2420gf); Pacr-2-acr-2(mini) (juEx2336), and acr-2(n2420gf);Punc-25-acr-2(juEx32), respectively.

Figure 4. Acetylcholine neurotransmission is reduced in acr-2 null mutants.

(A) acr-2 gene structure with the intragenic mutations identified from the acr-2(n2420gf) suppressor screen. acr-2(n2595 n2420) contains a nonsense mutation at the codon for tryptophan 175 in addition to the original gain-of-function mutation and is used as the null mutant in this figure. ok1887 removes the first three exons and inserts 420 bp of unrelated DNA. Details of the nucleotide and amino acid changes are shown in Table S1 and Figure S1. Boxes indicate exons; lines, introns; M, transmembrane domain; an asterisk (*) indicates mutations at the splice junctions; X indicates stop codon mutations. (B) Average speed of wild-type (3.9 mm/min±0.3 SEM, n = 16), acr-2(n2595 n2420) (2.8 mm/min±0.1 SEM, n = 16), and acr-2(n2595 n2420) worms carrying a Pacr-2::acr-2 construct (4.0 mm/min±0.4 SEM, n = 16). An ANOVA test followed by a Dunn's post test was used to analyze the data; *p<0.05. (C) Acetylcholine mini frequencies recorded in 2 mM external CaCl₂ from the wild type (21.9 events/s±3.2 SEM, n = 5) and acr-2(n2595 n2420) (12.9 events/s±1.6 SEM, n = 9) are significantly different (*p = 0.015). GABA mini frequencies recorded in 2 mM external CaCl₂ from the wild type (14.9 events/s±5.4 SEM, n = 5) and acr-2(n2595 n2420) (10.4 events/s±2.2 SEM, n = 9) are not significantly different (p = 0.38). Data were analyzed using a two-tailed unpaired t-test. (D) Representative traces of minis recorded at two holding potentials, −60 and −10 mV, on body muscle cells from wild-type and acr-2(n2595 n2420) worms in 2 mM external CaCl₂.

Figure 5. acr-2(n2420gf) mutants exhibit reduced GABA neurotransmission.

(A) Representative traces (upper panel) and frequencies (lower panel) of endogenous postsynaptic currents recorded at two holding potentials, −60 and −10 mV, from wild-type and acr-2(n2420gf) worms in 0.5 mM external CaCl₂. The frequency of miniature postsynaptic currents from cholinergic neurons is increased in the gain-of-function mutant (wild type: 12.1 events/s±1.1 SEM, n = 8; acr-2(n2420gf): 18.3 events/s±2.2 SEM, n = 9; *p = 0.0307). The frequency of miniature currents induced by GABAergic motor neurons was only slightly reduced in the gain-of-function mutant (wild type: 9.2 events/s±2.3 SEM, n = 8; acr-2(n2420gf): 7.2 events/s±2.1 SEM, n = 9; p = 0.5342). Data were analyzed using a two-tailed unpaired t-test. (B) Representative traces (upper panel) and frequencies (lower panel) of endogenous postsynaptic currents recorded at two holding potentials, −60 and −10 mV, from wild-type and acr-2(n2420gf) worms in 2 mM external CaCl₂. Acetylcholine currents frequencies recorded from the wild type (18.9 events/s±2.3 SEM, n = 10) and acr-2(n2420gf) (18.7 events/s±2.6 SEM, n = 14) are not significantly different (p = 0.9555). GABA currents recorded from the wild type (9.9 events/s±1.7 SEM, n = 10) and acr-2(n2420gf) (1.2 events/s±0.6 SEM, n = 14) are significantly different (*p = 0.0007). Data were analyzed using a two-tailed unpaired t-test with Welch's correction. (C) Representative traces (upper panel) and frequencies (lower panel) of endogenous postsynaptic currents recorded at two holding potentials, −60 and −10 mV, from wild-type and acr-2(n2420gf) worms in 5 mM external CaCl₂. Acetylcholine currents frequencies recorded from the wild type (24.9 events/s±3.4 SEM, n = 8) and acr-2(n2420gf) (7.6 events/s±2.7 SEM, n = 7) are significantly different (*p = 0.0017). GABA currents recorded from the wild type (13.4 events/s±2.0 SEM, n = 8) and acr-2(n2420gf) (2.3 events/s±1.0 SEM, n = 7) are significantly different (**p = 0.0004). Data were analyzed using a two-tailed unpaired t-test.

Figure 6. Mutations in acetylcholine receptor subunits suppress convulsions of acr-2(n2420gf) mutants.

(A) acr-12 gene structure with the loss-of-function mutations identified as extragenic suppressors of acr-2(n2420gf). ok367 has a 1,368-bp deletion and is a null allele of acr-12. Boxes indicate exons; lines, introns; TM, transmembrane. (B) The convulsions of acr-2(n2420gf) mutants are suppressed by loss-of-function mutations in unc-63, unc-38, unc-50, or unc-74, but not in unc-29, lev-1, and several other acr genes tested. The alleles used were unc-63(e384), unc-38(e284), unc-50(n2623), unc-74(n2614), unc-29(x29), lev-1(e211), acr-16(ok789), acr-5(ok180), lev-8(ok1519), ric-3(hm9), and lev-10(x17), all of which are null mutations. Red columns represent ancillary proteins, and green columns represent acetylcholine receptor subunits. A minimum of eight to ten animals per genotype were scored. (C) Cell-type–specific transgenic expression of acr-12 and unc-63 shows that both are required in the ventral cord cholinergic motor neurons to suppress acr-2(n2420gf). The alleles used were unc-63(e384) and acr-12(ok367). Pmyo-3 promoter drives expression in all body muscles, and Prab-3 drives expression in all neurons . The Pacr-2 promoter contains 1.8 kb of acr-2 upstream sequences driving expression in A and B neurons (Figure 3A), and the Punc-25 promoter drives expression in the four RME and 19 D neurons . The colors of the columns represent grouping of genotypes. A minimum of eight to ten animals per genotype were scored.

Figure 7. Composition of the ACR-2R receptor reconstituted in Xenopus oocytes.

(A) The ACR-2R receptor shows high sensitivity to acetylcholine and is insensitive to levamisole or choline. Mecamylamine completely blocks the current. The ACR-2R receptor includes all five subunits. (B) The ACR-2(V13'M)R receptor shows high sensitivity to acetylcholine and an increased sensitivity to DMPP, and weakly responds to levamisole and choline. The ACR-2(V13'M)R receptor includes all five subunits. (C) Quantification of the relative efficacies of cholinergic agonists compared to acetylcholine. Numbers above the bars represent the numbers of oocytes recorded for each condition. Example of traces are shown in (A and B). Replacement of ACR-2(+) by ACR-2(n2420) increased agonist efficacy for DMPP, choline, and levamisole, whereas nicotine efficacy was unchanged. ACh: 100 µM acetylcholine; Nic: 100 µM nicotine; DMPP: 100 µM DMPP; Cho: 1 mM choline; and Lev: 100 µM levamisole. (D) Inclusion of ACR-3 greatly increases the current of the ACR-2R. cRNAs for acr-12, ric-3, unc-38, unc-50, unc-63, and unc-74 were coinjected for each condition. Average peak current for acr-2 + acr-3 coinjection was 111±68 nA (n = 34). (E) Dose–response curves for acetylcholine action on the ACR-2 and ACR-2(V13'M) receptors show that EC50 is comparable. All recordings were made with 1 mM external CaCl₂. Error bars are standard errors of the mean. Asterisks indicate that data are significantly different at p<0.05 (*) or p<0.01 (**). ACh: acetylcholine; Nic: nicotine; DMPP: 1,1-dimethyl-4-phenylpiperazinium; Cho: choline; Lev: levamisole; Mec: mecamylamine.