Paralog-specific miRNA regulation uncouples ASER gene repression from the canonical die-1 / cog-1 switch

Abstract

The Pristionchus pacificus ASER-specific Ppa-gcy-22 .3 and Ppa-gcy-22 .5 paralogs differ in their regulatory dependencies despite sharing an upstream terminal selector machinery. P. pacificus gcy-22.3 expression is sensitive to Ppa-die-1 and Ppa-cog-1 perturbation, while Ppa-gcy-22 .5 is largely independent of these canonical ASE laterality regulators. Both paralogs require the miRNA, Ppa-miR-8345 , for repression in ASEL; loss of Ppa-miR-8345 produces a complete ASEL-to-ASER conversion, in contrast to the hybrid ASEL/ASER states caused by Ppa-die-1 or Ppa-cog-1 mutation alone. These findings indicate Ppa-miR-8345 acts through additional regulatory outputs beyond the die-1 / cog-1 feedback loop, revealing gene-by-gene rewiring of terminal differentiation programs.

Figure 1. Transgenic promoter reporters and HCR-FISH show representative expression of P. pacificus ASER-specific Ppa-gcy genes .

The paralogs Ppa-gcy-22 .3 and Ppa-gcy-22 .5 mark ASER fate (green). The Ppa-gcy-8.1 expression (magenta) in the AFD neurons represents the HCR-FISH staining control, while the ASEL-specific fate is marked by Ppa-gcy-7.2 expression (red). (A) Ppa-gcy-22 .3p::GFP expression in the wild type is found only in the ASER neuron. (B-B') Reduction of Ppa- DIE-1 function results in 2xASER* expression of Ppa-gcy-22 .3p::GFP . (C-C') Ppa-miR-8345 mutants only show the 2xASER Ppa-gcy-22 .3p::GFP misexpression. (D) A proposed model for the paralog-specific negative regulatory loop controlling Ppa-gcy-22 paralog expression. (E) HCR-FISH shows Ppa-gcy-22 .3 expression in the wild type is found only in the ASER neuron. (F) Ppa-die-1 mutants have the 2xASER* Ppa-gcy-22 .3 ectopic expression with a hybrid ASEL neuron. (G-G') Ppa-miR-8345 mutants show 2xASER misexpression of Ppa-gcy-22 .3 . (H-H') The loss of negative regulatory sites in the Ppa-cog-1 3' UTR (gain-of-function allele) results in the misexpression of Ppa-gcy-22 .3 . (I) Ppa-gcy-22 .5p::GFP expression in the wild type is found only in the ASER neuron. (J) Ppa-die-1 mutants show wildtype-like Ppa-gcy-22 .5p::GFP expression. (K) Ppa-miR-8345 mutants result in the 2xASER Ppa-gcy-22 .5p::GFP misexpression. (L) The loss of negative regulatory sites in the Ppa-cog-1 3' UTR results in the 2xASER Ppa-gcy-22 .5p::GFP expression. (M) Ppa-gcy-22 .5 expression in the wild type is found only in the ASER neuron. (N-N') A reduction-of-function mutation in Ppa-die-1 does not significantly alter Ppa-gcy-22 .5 expression. (O-O') Ppa-miR-8345 mutants show only 2xASER Ppa-gcy-22 .5 misexpression. (P-P') The loss of negative regulatory sites in the Ppa-cog-1 3' UTR does not affect Ppa-gcy-22 .5 expression. Arrowheads indicate misexpression of the ASER marker in the ASEL neuron. The percentage refers to the animals with specific ASER expression pattern with the sample number indicated in the parentheses. Results in B, F-H, and J were previously reported (Castro et al , 2026) ¹ . Scale bars: 5 μm.

Description

The establishment and maintenance of neuronal identity require activation and sustained expression of terminally differentiated genes such as neurotransmitters and taste receptors. For the specification of the ASE chemosensory neurons in C. elegans , the CHE-1/Glass-type zinc-finger protein is a paradigmatic terminal selector known to regulate a suite of 4 left-specific and 5 right-specific receptor-type guanylate cyclases ( gcys ) in the ASE neurons through direct binding of their regulatory regions (Yu et al. 1997; Johnston et al. 2005; Ortiz et al. 2006). In addition, three regulators- DIE-1 (zinc-finger), lsy-6 (miRNA), COG-1 (homeodomain) - form a negative regulatory feedback loop to produce mutually exclusive expression of either cog-1 or die-1 in the ASEL and ASER neurons, respectively (Palmer et al. 2002; Chang et al. 2003; Johnston and Hobert 2003; Johnston et al. 2005). Thus, ASEL-specific ( gcy-6 , gcy-7 ) as well as ASER-specific ( gcy-5 , gcy-22 ) gcy genes are controlled by the same genetic regulators (Johnston et al 2005). However, it remains unclear if such miRNA regulatory network abides by the same conserved principle to establish stable terminally differentiated receptor gene expression in other nematodes.

In the left/right asymmetric ASE gustatory neuron pair of the predatory, entomophilic nematode Pristionchus pacificus , the Ppa- CHE-1-dependent ASER-specific Ppa-gcy-22 subfamily ( gcy-22.1, gcy-22.2, gcy-22.3, gcy-22.4, gcy-22.5) provides a comparative context to examine the role of the regulatory miRNA in establishing neuronal asymmetry. These five Ppa-gcy-22 paralogs are distributed across 4 loci over 3 chromosomes (I, IV, X), with Ppa-gcy-22 .3 and Ppa-gcy-22 .5 located on Chromosomes IV and X, respectively. Notably, Ppa-die-1 r eduction-of-function mutants show a fully penetrant misexpression of the ASER marker Ppa-gcy-22 .3 in the ASEL, while Ppa-gcy-22 .5 expression remains wildtype-like in Ppa-die-1 (csu225) , indicating that detectable repression of Ppa-gcy-22 .5 in ASEL does not require normal DIE-1 function under these assay conditions. (Castro et al 2026). Here we show that two ASER-expressed Ppa-gcy-22 paralogs differ in their dependence on canonical ASE laterality regulators: Ppa-gcy-22 .3 is sensitive to Ppa-die-1 and Ppa-cog-1 perturbation, whereas Ppa-gcy-22 .5 is largely independent of those transcriptional regulators but remains dependent on Ppa-miR-8345 for repression in ASEL. Understanding the roles of miRNA-mediated regulation can expose the veneer of genetic conservation to show paralog-specific branches of terminal differentiation programs.

To determine the extent of paralog-specific regulation in the ASE neurons, we used both transgenic reporters and HCR-FISH (Hybridization Chain Reaction In situ Fluorescent Hybridization) to characterize Ppa-gcy-22 .3 and Ppa-gcy-22 .5 expression in mutants comprising this miRNA regulatory loop. While the reporters allow us to assess promoter activity through likely conserved Ppa- CHE-1-dependent ASE-motifs in the cis -regulatory regions of the gcy genes (Etchberger et al 2007), the recent utilization of HCR-FISH permits the simultaneous monitoring of multiple mRNA transcripts to determine if the ASEL neuron expresses a stable hybrid state (a Ppa-gcy-22 paralog and Ppa-gcy-7.2 ) or if the ASEL neuron is transformed completely into the ASER fate (expresses only the gcy-22 paralogs). We use “2xASER” to denote animals in which both ASE neurons express ASER markers and lack detectable ASEL-specific Ppa-gcy-7.2 expression. “2xASER*” is used to denote ectopic ASER-marker expression in ASEL neurons while ASEL-specific Ppa-gcy-7.2 expression persists, indicating a hybrid ASEL/ASER state.

As reported previously, Ppa-gcy-22 .3p::GFP and Ppa-gcy-22 .3 steady-state transcript expression patterns show mis-expression in the ASEL in reduction-of-function Ppa-die-1 (csu225) mutants (Castro et al 2026), whereas Ppa-gcy-22 .5p::GFP and Ppa-gcy-22 .5 transcript expression in Ppa-die-1 (csu225) remain wildtype-like and restricted to the ASER ( Fig. 1 B, F, J, N). With the exception of a single animal misexpressing Ppa-gcy-22 .5 in both ASE neurons (2% 2xASER*, n=50), we found predominantly wild-type Ppa-gcy-22 .5 expression patterns in Ppa-die-1 mutants ( Fig. 1N ). By contrast, GFP reporters and FISH detected misexpression of both Ppa-gcy-22 .3 (86% 2xASER) and Ppa-gcy-22 .5 (89% 2xASER, n=35) in both ASE neurons in Ppa-miR-8345(csu259) mutants, without any ASEL-specific Ppa-gcy-7.2 expression ( Fig. 1 C, G, K, O). Despite functional similarity, Ppa-miR-8345 does not appear to be an ortholog of C. elegans lsy-6 , because its precursor sequence, predicted hairpin structure, and genomic location differ substantially (Castro et al, 2026). Finally, while Ppa-gcy-22 .3 transcript expression in the Ppa-cog-1 gain-of-function allele shows the 2xASER* phenotype with hybrid ASEL (62%)( Fig. 1H- H'), Ppa-gcy-22 .5 expression was wildtype-like and detectable only in the ASER (100%, n=66)( Fig. 1P- P'). Interestingly, in the Ppa-cog-1 (csu255) gain-of-function allele, Ppa-gcy-22 .5p::GFP expression occasionally exhibited the 2xASER phenotype (17%, n=100)( Fig. 1L ), although this effect was not detected at the endogenous transcript level. Taken together, these results indicate that loss of Ppa-miR-8345 produces a more complete molecular conversion of ASEL toward ASER identity than either Ppa-die-1 reduction of function or Ppa-cog-1 derepression alone. In Ppa-die-1 and Ppa-cog-1 mutant animals, Ppa-gcy-22 .3 can be ectopically expressed in ASEL while ASEL-specific Ppa-gcy-7.2 expression persists, consistent with a hybrid ASEL/ASER state. By contrast, Ppa-miR-8345 mutants ectopically express both Ppa-gcy-22 .3 and Ppa-gcy-22 .5 in both ASE neurons and lack detectable ASEL-specific Ppa-gcy-7.2 expression. These observations suggest that Ppa-miR-8345 controls additional regulatory output(s), beyond the canonical die-1/cog-1 bistable loop that are required to suppress ASER-specific gcy paralogs in ASEL. The differential response of Ppa-gcy-22 .3 and Ppa-gcy-22 .5 further suggests that activation of ASEL-specific Ppa-gcy-7.2 and repression of ASER-specific Ppa-gcy-22 paralogs are genetically separable outputs, rather than inseparable consequences of a single binary fate switch.

Whereas the asymmetrically expressed ASE gcy genes undergo a hybrid precursor state in late embryogenesis and early larval stages in C. elegans , transcripts of the ASEL- and ASER-specific gcy genes have not been detected to co-localize in the same ASE neuron in P. pacificus ( Johnston et al 2005, Castro et al 2026). P. pacificus gcy-22.3 and Ppa-gcy-7.2 however, do exhibit precursor hybrid states in the AFD thermosensory neurons during late embryogenesis that resolve into a Ppa-gcy-8.1 -only state by the J1 early larval stage. Thus, hybrid fates in the ASE neurons do not represent delayed paedomorphic phenotypes but rather bona fide changes in post-mitotic terminal cell fates. It is unclear which other target of Ppa-miR-8345 mediates the suppression of Ppa-gcy-22 .5 in the ASEL neuron. Our finding shows that regulation of terminal selector outputs and asymmetric effector genes may be rewired on a gene-by-gene basis, rather than acquired as a single coordinated regulatory module.

Methods

P. pacificus and other nematode strains were maintained at ~20°C on NGM plates seeded with E. coli OP50 . For reporter analyses, we crossed Ppa-gcy-22 .3p::gfp and Ppa-gcy-22 .5p::gfp males with Ppa-miR-8345(csu259) and Ppa-cog-1 (csu255) mutant hermaphrodites and identified transgenic homozygous mutant progeny by sequencing PCR products. For FISH, we performed third-generation HCR v3.0 with split-initiator probe pairs using three differently conjugated fluorophores (B2, B4, B5)(Molecular Instruments, Los Angeles, CA) as previously described (Castro et al. 2026). Although the transcript levels for Ppa-gcy-22 .5 (PPA03763) on www.pristionchus.org appear only in some but not all expression datasets (compare stage-specific expression in Baskaran et al, 2015 with Han and Lo, 2022), we were able to consistently observe HCR-FISH signals for Ppa-gcy-22 .5, albeit at distinctively lower level than Ppa-gcy-22 .3 (PPA04464). We ordered the probe sets as DNA oligo pools at 50 pmol (IDT, San Diego, CA) and used a 10x higher concentration of probes than previously published (20 pmol) (Ramadan and Hobert 2024). Because no obvious stage-dependent differences were observed among post-embryonic stages, animals from J2 through adult stages were pooled for phenotype scoring (J2, J3, J4, adult). For Ppa-gcy-22 .5 , a new round of HCR-FISH was performed a year later (wild type, Ppa-die-1 , Ppa-cog-1 ) and combined with previously published results.

strain name	genotype	source
PS312	Wildtype
RLH334	csuEx90 [ Ppa-gcy-22 .3p::gfp; Ppa-egl-20p::rfp]	Castro et al, 2026
RLH333	csuEx90 [ Ppa-gcy-22 .3p::gfp; Ppa-egl-20p::rfp]; Ppa-die-1 (csu225)	Castro et al, 2026
RLH405	csuEx90 [ Ppa-gcy-22 .3p::gfp; Ppa-egl-20p::rfp]; Ppa-miR-8345 (csu259)	this study
RLH378	csuEx105 [ Ppa-gcy-22 .5p::gfp; Ppa-egl-20p::rfp]	Castro et al, 2026
RLH346	csuEx105 [ Ppa-gcy-22 .5p::gfp; Ppa-egl-20p::rfp]; Ppa-die-1 (csu225)	Castro et al, 2026
RLH397	csuEx105 [ Ppa-gcy-22 .5p::gfp; Ppa-egl-20p::rfp]; Ppa-miR-8345 (csu259)	this study
RLH387	csuEx105 [ Ppa-gcy-22 .5p::gfp; Ppa-egl-20p::rfp]; Ppa-cog-1 (csu255)	this study

Funding Statement

null