PHA-4/FoxA controls the function of pharyngeal and extrapharyngeal enteric neurons in C. elegans
Zion Walker, Wen Xi Cao, Eduardo Leyva-Díaz, Mayeesa Rahman, Surojit Sural, Michelle A. Attner, Oliver Hobert
TL;DR
This study shows that the PHA-4/FoxA transcription factor is essential for the function of gut-related neurons in C. elegans, both during development and in adulthood.
Contribution
The study reveals that PHA-4/FoxA is continuously required for the function of multiple types of enteric neurons, beyond its known role in gut development.
Findings
PHA-4/FoxA is required for terminal differentiation and lifelong function of postmitotic enteric neurons in C. elegans.
PHA-4/FoxA is expressed in extrapharyngeal neurons like AVL, DVB, RIS, and PVT, which control gut behaviors.
PHA-4/FoxA is the only known transcription factor required for the function of all types of enteric neurons in a nervous system.
Abstract
In this study, Walker et al. show that the FoxA transcription factor PHA-4 is required for terminal differentiation and function of postmitotic enteric neurons in C. elegans. PHA-4 function goes beyond its pioneer activity during gut development, with continuous expression being a universal requirement for the activity of distinct enteric neurons in controlling various gut behaviors. FoxA transcription factors pattern gut tissue across animal phylogeny. Beyond their early patterning function, little is known about whether they control the terminal differentiation and/or function of the fully mature enteric nervous system, the intrinsic nervous system of the gut. We show here that the expression and function of the sole Caenorhabditis elegans FoxA homolog, PHA-4, reach beyond its previously described pioneer factor roles in patterning the foregut. Through the engineering of…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8- —NIH 10.13039/100000002
- —Howard Hughes Medical Institute 10.13039/100000011
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGenetics, Aging, and Longevity in Model Organisms · Congenital gastrointestinal and neural anomalies · Developmental Biology and Gene Regulation
The animal gut, subdivided into foregut, midgut, and hindgut, is composed of a multitude of distinct cell types. These include enteric neurons that innervate gut tissue to control various aspects of gut function—most notably, the contraction of gut musculature. The enteric nervous system was initially dubbed the “second brain” of an animal to illustrate its physical separation from the CNS and its autonomous function (Gershon 1998). Similarities in the organization and function of the vertebrate enteric nervous system to the nervous system of comparatively more primitive animals such as cnidarians have resulted in the enteric nervous system also being dubbed the “first brain”; that is, an evolutionary antecedent of a more complex nervous system (Furness and Stebbing 2018). These evolutionary considerations, as well as the involvement of enteric nervous systems in many aspects of human health and disease (Rao and Gershon 2018), have spurred substantial interest in a better understanding of the developmental programs that specify enteric nervous systems. Notably, while early patterning mechanisms that result in the eventual specification of enteric neurons are relatively well understood in a number of distinct invertebrate and vertebrate animal systems (Hartenstein 1997; Laranjeira and Pachnis 2009; Mango 2009; Sasselli et al. 2012; Nagy and Goldstein 2017), the molecular mechanisms that operate in postmitotic enteric neurons to specify their terminal differentiation program and to maintain their functional features during the life of the animal remain incompletely understood.
FoxA-type forkhead domain transcription factors control the specification of gut tissue across animal phylogeny, ranging from protostomes to deuterostomes (Friedman and Kaestner 2006; Mango 2009; Adler et al. 2014; Annunziata et al. 2019). For example, in the nematode Caenorhabditis elegans, expression of the sole FoxA transcription factor ortholog, PHA-4, demarcates all cells in the developing embryo that generate foregut, midgut, and hindgut tissue (Azzaria et al. 1996; Horner et al. 1998; Kalb et al. 1998; Mango 2009; Niu et al. 2011; Ma et al. 2021). Previous genetic loss-of-function studies have firmly established functions of PHA-4/FoxA in all sections of the gut at different stages of embryonic development (Mango et al. 1994; Horner et al. 1998; Kalb et al. 1998; Gaudet and Mango 2002; Mango 2009; Fakhouri et al. 2010). In the foregut and hindgut, pha-4 functions as an organ selector to control the lineage specification and morphogenesis of these tissues during embryonic patterning. The organ selector role of PHA-4 critically depends on a phylogenetically conserved pioneer factor function of PHA-4, which involves nucleosome displacement and RNA Pol II recruitment (Hsu et al. 2015).
In the midgut (intestine), pha-4 may not have a developmental function (a notion that we probe further here) but has a well-documented role in intestinal homeostasis (Panowski et al. 2007; Wu et al. 2018; Torzone et al. 2023). Outside the gut, PHA-4 has only been reported to be expressed in a small number of postembryonically generated cells, which share with the alimentary tract the feature of also forming tubular structures—specifically, the somatic gonad (Kalb et al. 1998; Updike and Mango 2007; Chen and Riddle 2008). PHA-4 appears to function in the proper differentiation of these cells as well (Ao et al. 2004; Updike and Mango 2007; Chen and Riddle 2008).
We set out to investigate C. elegans PHA-4 function because of our overall interest in studying terminal differentiation programs of enteric neurons of the foregut and hindgut (Hobert et al. 1999; Gendrel et al. 2016; Vidal et al. 2022). Two enteric neurons, AVL (located in the ventral head ganglion) and DVB (located in the dorsorectal ganglion of the tail), which we refer to here as “hindgut enteric neurons” (HENs), innervate the hindgut of the worm to control enteric muscle contraction during defecation (Fig. 1A; McIntire et al. 1993; Avery and Thomas 1997). AVL and DVB are GABAergic neurons that are activated by the intestinally released neuropeptide NLP-40, which generates all-or-none action potentials in these neurons (Mahoney et al. 2008; Wang et al. 2013; Jiang et al. 2022). The rhythmic activation of AVL and DVB results in the GABA-dependent contraction of electrically coupled hindgut enteric muscles (stomatointestinal, anal depressor, and anal sphincter muscles), leading to the expulsion of gut contents (Thomas 1990; Jiang et al. 2022). The terminal identity features of the AVL and DVB are genetically specified by the LIM homeobox gene lim-6 and the orphan nuclear hormone receptor nhr-67 (Hobert et al. 1999; Gendrel et al. 2016).
In addition, 20 enteric neurons are located within the foregut of the worm, which we term here “pharyngeal enteric neurons” (PENs) (Fig. 1A). These 20 neurons can be classified into 14 distinct types based on position, neurite morphology, synaptic connectivity, and molecular profile (Albertson and Thomson 1976; Cook et al. 2020; Taylor et al. 2021). Like enteric neurons in all other animal nervous systems, PENs form a densely interconnected, autonomously acting network that controls foregut contraction during feeding behavior, known as pharyngeal pumping (Albertson and Thomson 1976; Cook et al. 2020).
All 20 PENs require the Six-type homeodomain transcription factor CEH-34 to acquire their unique terminal properties, with CEH-34 interacting with distinct homeodomain cofactors to specify the distinct identity of individual enteric neurons (Vidal et al. 2022). We had previously reported that of the many distinct functions of PHA-4 during organogenesis and specification of distinct foregut tissue, one is the proper induction of CEH-34 expression in foregut enteric neurons (Vidal et al. 2022). As we describe here, not only is PHA-4 expressed during embryonic induction of CEH-34 expression, but its expression is maintained in all enteric neurons throughout postdevelopmental stages into adulthood. Surprisingly, we also discovered previously unnoted and continuous PHA-4 expression in the extrapharyngeal HENs (AVL and DVB), as well as in two other extrapharyngeal neurons: RIS and PVT. RIS was previously shown to control foregut-associated behavior (pharyngeal pumping) (Steuer Costa et al. 2019), and we show here that PVT controls hindgut-associated behavior (defecation).
Postmitotic, postdevelopment expression of PHA-4 in these various enteric function-controlling neuron classes indicates that PHA-4 may have late functions after its earlier functions in foregut organogenesis. Through the engineering of neuron-specific cis-regulatory alleles, Cre-mediated cell-specific knockout, and degron-mediated temporally controlled PHA-4 removal, we show here that PHA-4 function extends beyond an early pioneer function to later postdevelopmental roles within pharyngeal enteric neurons as well as in the extrapharyngeal neurons that control enteric function. Our analysis reveals that PHA-4 represents the only currently known transcription factor whose expression and function cover the entire set of enteric neurons in an animal nervous system.
Results
Expression pattern of gfp-tagged pha-4
The expression pattern of pha-4 throughout the entire animal has been previously analyzed using multicopy reporter transgenes and PHA-4 antibody staining (Horner et al. 1998; Kalb et al. 1998; McKay et al. 2003; Großhans et al. 2005; Panowski et al. 2007; Chen and Riddle 2008; Riddle et al. 2016). Four-dimensional live imaging analysis of a PHA-4 fosmid-based reporter further refined the pattern of PHA-4 expression onset during embryogenesis (Fig. 1B,C; Ma et al. 2021). One notable aspect of these previous analyses was the observation of PHA-4 expression in a previously unidentified head and tail neuron in adult animals (Fig. 1C; Kalb et al. 1998; Panowski et al. 2007; Chen and Riddle 2008). To confirm all these previous patterns using an independent reagent and to identify the nature of the PHA-4-expressing head and tail neurons, we tagged the endogenous pha-4 locus with gfp using CRISPR/Cas9 genome engineering. The locus was tagged at the 3′ end to capture all three isoforms of the pha-4 locus (Fig. 2A). Resulting pha-4(ot946[pha-4::gfp]) animals are viable and appear superficially wild type, indicating that tagging does not obviously affect gene function.
Expression pattern of the gfp-tagged pha-4 locus. (A) pha-4 locus schematic on chromosome V, showing the pha-4 CRISPR/Cas9-engineered reporter allele pha-4(ot946[pha-4::gfp::loxP::3xflag]), indicating the location of the GFP, loxP, and 3xFLAG cassette insertion. The existence of different isoforms is inferred from the identification of different SL1 trans-spliced transcripts (Azzaria et al. 1996). (B) Expression of the pha-4(ot946) gfp reporter allele over the course of development, showing expression in the enteric system throughout different embryonic and postembryonic stages, including adulthood. Expression in the intestine strongly decreases in adulthood. Red asterisks indicate the arcade cells, shown at the left. Red stars indicate the pharyngeal–intestinal valve cells, shown at the right. Overlays with a panneuronal reporter are shown in Supplemental Figure S1A. (C) pha-4 is expressed in GABAergic neurons AVL, RIS, and DVB. Images of an adult worm show colocalization of pha-4::gfp expression with a unc-47prom::mCherry reporter transgene (otIs348) in two neurons situated on the ventral ganglion in the head (AVL and RIS) and one neuron on the dorsal side in the tail (DVB). (D) pha-4::gfp is expressed in the preanal ganglion neuron PVT. Magnified images of the preanal ganglion of an adult worm show colocalization of pha-4(ot946) expression with PVT as identified by NeuroPAL (otIs696). NeuroPAL cell IDs of neighboring neurons are also annotated.
Expression of this pha-4 reporter allele precisely recapitulated the previously reported sites of expression of PHA-4 throughout the entire alimentary system, from the arcade cells that attach the foregut to the mouth to all cells of the foregut and midgut (intestine) and cells associated with the hindgut (Fig. 2B). Expression initiates at the end of gastrulation and is evident throughout the pharyngeal primordium and the midgut (Fig. 2B). While expression in the midgut becomes significantly weaker over the course of embryonic and postembryonic development, pha-4 expression in cells of the foregut and hindgut remains strong into adulthood (Fig. 2B; Supplemental Fig. S1A), a possible consequence of transcriptional autoregulation of the pha-4 locus (Kaltenbach et al. 2005).
While previous studies noted expression only in an unidentified “head neuron” and a “rectal neuron” (Kalb et al. 1998; Panowski et al. 2007; Chen and Riddle 2008), we noted a total of four extrapharyngeal neurons that express pha-4(ot946) in the adult. Based on position, and with the use of molecular landmarks, we identified these as the GABAergic RIS, AVL, and DVB neurons and the peptidergic PVT tail neuron (Fig. 2B–D). Onset of PHA-4::GFP expression in RIS, AVL, and rectal cells, including PVT, was observed in the comma-stage embryos shortly after these neurons had been generated, while expression in DVB became apparent at the late L1 stage, when DVB was born (Fig. 2B). All four neurons are known to require a terminal selector transcription factor, the LIM homeobox gene lim-6, for their proper differentiation (Hobert et al. 1999; Aurelio et al. 2003; Gendrel et al. 2016); we found that in lim-6-null mutant adult animals, pha-4 reporter allele expression is still present in these neurons but is notably weaker, indicating that lim-6 is required to sustain proper pha-4 expression (Supplemental Fig. S1B).
pha-4-expressing extrapharyngeal neurons control enteric behaviors
Three of the four PHA-4-expressing, extrapharyngeal neurons have been previously linked to the control of gut function. The head neuron RIS, whose single axon extends around the isthmus of the pharynx (Fig. 3A), controls foregut function, as inferred from optogenetic activation of RIS affecting the serotonin-induced pharyngeal pumping rate (Steuer Costa et al. 2019). We sought to corroborate the involvement of RIS in controlling pharyngeal pumping by silencing RIS using a transgene of the flp-11 promoter driving expression of a histamine-gated chloride channel (HisCl) (Grubbs et al. 2020). We found that histamine-induced silencing of RIS in freely moving animals decreases pharyngal pumping behavior while not having any impact on hindgut-controlled defecation behavior (Fig. 3D).
Effect of RIS and PVT inhibition on enteric behaviors. (A,B) Schematic showing the anatomical location and morphology of the unilateral RIS neuron (A) and the PVT neuron (B), taken from WormAtlas (https://www.wormatlas.org; Hall and Altun 2007). (C) Establishing Cre-driver lines with PVT specificity. Representative images of animals assayed in day 1 adult stage. Expression of a cnc-11 promoter fusion (otEx8346) (top) and sre-22 promoter fusion (otEx8333) (bottom) in the head, midbody, and tail. Expression in PVT and a few other cell types is annotated. Asterisks indicate the expression of the coinjection marker unc-122p::gfp in coelomocytes. Expression was confirmed in 14 array-positive otEx8333 animals and 25 array-positive otEx8345 animals. (D) Histamine-mediated silencing of RIS impacts adult pumping but not defecation behavior. Pharyngeal pumping in RISp::HisCl-expressing worms (qnEx643) was observed in the absence of histamine (0 µM) and at 8 and 10 µM concentrations of histamine. The frequency of pBocs, the frequency of expulsion, and the expulsion:pBoc ratio in RISp::HisCl-expressing adult worms treated with histamine were assayed and are individually plotted. (E) Histamine-mediated silencing of PVT impacts defecation but not pharyngeal pumping behavior. Three PVTp::HisCl array-expressing lines (otEx8336, otEx8337, and otEx8342; lines 1–3) were assayed. (Left) Pharyngeal behavior was observed at 0 and 10 µM concentrations of histamine. (Right) The frequency of pBocs, the frequency of expulsion, and the expulsion:pBoc ratio in adult worms of two lines expressing PVTp::HisCl (otEx8336 and otEx8337; lines 2 and 3) treated with histamine were assayed and are individually plotted. Dark worms are non-array-carrying siblings from the PVTp::HisCl lines (i.e., controls). (F) Enteric behaviors are unaffected in homt-1(zw94)-null mutant animals. (G) Cre-mediated removal of egl-3 from PVT impacts adult defecation. PVTp::Cre; egl-3loxP lines (otEx8339 and otEx8340). (H) nlp-40 locus schematic on chromosome I, showing the nlp-40 CRISPR/Cas9-engineered reporter allele syb3208, with the location of the reporter cassette insertion indicated. Previously reported nlp-40 reporter transgene expression in the intestine appears too dim to be visible with this reporter allele. (D–G) Data points show individual worms assayed. The horizontal line represents median value of biological replicates. () P < 0.05, () P < 0.01, () P < 0.001, (**) P < 0.0001, (ns) not significant in Mann–Whitney U-test (F), Dunn's multiple comparisons test after Kruskal–Wallis test (G), and Šídák's multiple comparisons test after two-way ANOVA (D,E).
AVL and DVB are known hindgut-innervating enteric neurons (HENs), which control enteric muscle contraction during defecation (Fig. 1A; McIntire et al. 1993). The defecation motor program occurs in three steps: contraction of posterior body muscles (pBoc), contraction of anterior body muscles (aBoc), and expulsion (Exp) (Thomas 1990). A pacemaker in the intestinal epithelial cells initiates the pBoc step, which acts upstream of AVL and DVB (Dal Santo et al. 1999). The Exp step requires activation of AVL and DVB to constrict hindgut enteric muscles to release gut contents through the anus. The defecation cycle is highly regular both within and across individuals over time, with well-fed young adult animals never missing a pBoc, aBoc, or Exp step (Thomas 1990).
PVT is a peptidergic interneuron in the preanal ganglion whose single axon extends from the tail to the isthmus of the pharynx (Fig. 3B). PVT displays an intriguing lineage relationship with some cellular constituents of the gut (Fig. 1B), but no function in relation to the enteric system has been previously investigated. To manipulate PVT activity, we first used scRNA data (Taylor et al. 2021) to identify potential drivers with PVT-specific expression, which we validated with reporter gene constructs (Fig. 3C). We used the promoter of sre-22 to drive the histamine-gated chloride channel (HisCl) in PVT and found that, after histamine treatment, the frequency of pBoc events in such transgenic animals was unaffected, but the rate of expulsions was significantly reduced (Fig. 3E). Compared with untreated PVTp::HisCl worms, histamine-treated worms expressing PVTp::HisCl exhibited significantly fewer expulsion events, with an average of every third pBoc event not followed by an expulsion event (Fig. 3E). PVT silencing does not affect pharyngeal pumping (Fig. 3E).
PVT has also recently been shown to control locomotory quiescence during sleeplike behavior via releasing melatonin (Niu et al. 2020). We found that animals lacking the melatonin-synthesizing homt-1 gene display no effect in defecation behavior (Fig. 3F), indicating that PVT may act via other signaling pathways to control defecation behavior. Because PVT expresses a host of neuropeptides (Taylor et al. 2021), we tested whether removal of the neuropeptide-processing enzyme egl-3 in PVT would affect defecation behavior. To this end, we used a floxed egl-3 locus (Marquina-Solis et al. 2024) and drove Cre expression with the sre-22 driver. We found that in worms with egl-3 removed from PVT, the frequency of pBoc events was still unaffected, but not all pBoc events were followed by expulsions (Fig. 3G). Unlike control animals (floxed egl-3 animals with no Cre expression), worms with egl-3 removed from PVT displayed both missed expulsions and failed expulsions in which a weak contraction of the intestinal muscle was observed, but the concerted contraction of the hindgut enteric muscles and expulsion of gut contents were absent (Supplemental Videos S1, S2). While the timing of the weak muscle contraction occurs ∼3 sec after the pBoc in the defecation cycle, which is typical when AVL and DVB are active, the weak muscle response is not sufficient to complete the expulsion of gut contents from the body (Supplemental Video S2). The aberrant expulsion phenotype observed in worms with egl-3 removed from PVT points toward a potential role of PVT in the neuropeptidergic activation of AVL and DVB during the Exp step of defecation.
The neuropeptide NLP-40 regulates calcium transients in AVL and DVB in order to orchestrate the release of GABA onto the hindgut enteric muscles and is thought to be secreted from the intestine in response to intestinal calcium waves of the pBoc and aBoc (Wang et al. 2013). Intriguingly, as per scRNA data (Taylor et al. 2021), the neuropeptide most strongly expressed in PVT is NLP-40. We validated this observation by engineering a T2A:gfp::h2b cassette into the nlp-40 locus, confirming strong expression in PVT (Fig. 3H). We also found clear expression of the NLP-40 receptor AEX-2 in PVT using an SL2::gfp::h2b cassette (Ripoll-Sánchez et al. 2023), suggesting that NLP-40 release from PVT may cooperate to amplify intestinally released NLP-40 to control the defecation behavior that is disrupted in nlp-40 mutants (Wang et al. 2013).
Cis-regulatory elements driving neuronal PHA-4 expression
An analysis of pha-4 function in PENs and HENs, particularly in regard to the control of enteric behaviors, is complicated by (1) the early lethality of pha-4-null mutants and (2) its expression in multiple distinct cells of the alimentary tract. As we exemplified in a previous study of an essential gene (Reilly et al. 2022), the dissection of the cis-regulatory architecture of a genetic locus provides a potential avenue to generate *cis-*regulatory mutant alleles in which expression of a gene is lost selectively in only specific subsets of cells that normally express this gene. Previous work had shown that a genomic 7 kb fragment that contains ∼2 kb of sequences of the longest isoform of the pha-4 locus (pha-4a), as well as several exons and introns, drives lacZ reporter expression throughout the entire alimentary tract, while a 300 bp fragment upstream of the first isoform drives expression exclusively in the midgut (Fig. 4A; Azzaria et al. 1996; Kalb et al. 1998). We recapitulated the intestine-specific expression with a *gfp-*based transgene that captures the region upstream of the longest pha-4 isoform (pha-4prom1) (Fig. 4A,B). Interestingly, 2242 bp of sequences upstream of the second isoform (i.e., the first intron of the large isoform, pha-4prom2), pha-4b (Fig. 4A), drove exclusive expression of a gfp reporter in all PENs and HENs (AVL and DVB) as well as the RIS and PVT neurons (Fig. 4C). We confirmed this notion by crossing these transgenic animals with red fluorophore transgenes that marks either the entire nervous system or, selectively, the GABAergic identity of the HENs (Supplemental Fig. S1C,D). The onset of expression of the pha-4prom2::gfp transgene in HENs and PENs is comparable with the onset of expression of the endogenously tagged pha-4 locus (Figs. 2B, 4C).
An intronic cis-regulatory element drives PHA-4 expression in PENs and HENs. (A) pha-4 locus schematic with the reporter gene fusions indicated. lacZ fusions previously described by Azzaria et al. (1996) and Kalb et al. (1998) are schematized along with two newly generated promoter fragments fused to GFP that were used for expression analyses. (B) Expression pattern of a pha-4prom1::gfp reporter transgene (otEx8356). An ∼2 kb fragment upstream of the start codon of the pha-4a isoform drives expression specifically in the intestinal cells. (C) Expression pattern of a pha-4prom2::gfp reporter transgene (otEx8085). An ∼2.4 kb fragment upstream of the start codon of the second isoform, pha-4b, equivalent to the first intron of pha-4a, drives expression specifically in all enteric neurons of the pharynx and hindgut as well as RIS and PVT. The onset of GFP expression is detected in the embryo, and expression is maintained throughout adulthood.
Postmitotic enteric neuron expression of PHA-4 is essential for feeding and survival
Based on the cis-regulatory analysis described above, we used CRISPR/Cas9 to eliminate the promoter of the pha-4b isoform from the endogenous pha-4 locus that we had tagged with gfp (Fig. 5A). We found that in these animals, gfp expression is selectively lost from PENs, HENs, RIS, and PVT (Fig. 5B). Intriguingly, expression in these neurons appears intact initially but is lost by the 1.5-fold stage and remains absent in postembryonic animals (Fig. 5B), raising the possibility that the pha-4prom2 region represents an autoregulatory element. To test this notion, we crossed two independent, chromosomally integrated pha-4prom2 reporter lines into a pha-4-null mutant background. We found that expression is mostly lost from PENs but is unaffected in two extrapharyngeal neurons, likely AVL and RIS (Fig. 5C; Supplemental Fig. S1E). We conclude that pha-4 indeed autoregulates in many cell types, albeit apparently not in extrapharyngeal neurons.
Postmitotic enteric neuron expression of PHA-4 is essential for survival. (A) pha-4 locus schematic on chromosome V, showing different alleles used in this study. pha-4(ot1078 ot946) animals contain the C-terminal reporter cassette (ot946, including one loxP site), as well as an added second loxP site (ot1078). The gfp-tagged allele was also used to delete either the first intron only, resulting in pha-4(ot1505 ot946) animals, or the entire locus, including the reporter cassette, resulting in pha-4(ot1506)-null mutants. (B) Expression pattern of the pha-4 cis-regulatory mutant allele (ot1505 ot946), showing the lack of expression in RIS, AVL, and PVT. All other PHA-4::GFP cells in the head and tail regions of these images are nonneuronal PHA-4(+) cells. The L1 animals are shown from a dorsoventral perspective. Data points representing each individual worm assayed are plotted, and the horizontal line in the middle of the data points represents the median value of biological replicates. () P < 0.0001 in a Mann–Whitney U-test. (C) Expression pattern of the pha-4 reporter transgene pha-4prom2::mNeonGreen::PH (otIs946) in wild-type or pha-4(ot1506)-null mutant animals. Data points representing each individual worm assayed are plotted, and the horizontal line in the middle of data points represents the median value of biological replicates. () P < 0.001 in a Mann–Whitney U-test. (D) Bright-field image of L1 larva expressing a gfp-tagged pha-4(ot946) reporter allele with wild-type morphology (left) and L1 pha-4(ot1505 ot946) mutant lethal larva (right). The graph shows the survival rate of the progeny from worms heterozygous for pha-4(ot1505 ot946). One-quarter of their progeny is expected to be homozygous for pha-4(ot1505 ot946), and we observed about one-quarter of L1-arrested animals. Data points represent the percentage of viable progeny of a given parent worm. Ten parent worms were assayed per condition (see the Materials and Methods for more details). (E, left) Neuron-specific depletion of the floxed, gfp-tagged pha-4 locus (ot1078 ot946) using the Cre::egl-3(syb5804) and Cre::ceh-48(syb5859) alleles was confirmed by loss of PHA-4::GFP signals in neurons of L1-arrested animals. All remaining PHA-4::GFP cells in the head region are nonneuronal PHA-4(+) cells. (Right) Bright-field images of L1-arrested larva resulting from pha-4 removal from PENs using the Cre::egl-3(syb5804) or Cre::ceh-48(syb5859) alleles. Also shown is the survival rate of the progeny from worms heterozygous for pha-4(ot1078 ot946) in the Cre::egl-3(syb5804) or Cre::ceh-48(syb5859) allele backgrounds. One-quarter of their progeny is expected to carry a homozygously, panneuronally deleted pha-4 locus, and we indeed observed about one-quarter of L1-arrested animals. Data points represent the percentage of viable progeny of a given parent worm. Ten parent worms were assayed per condition (see the Materials and Methods for more details).*
Animals carrying this *cis-*regulatory allele of pha-4 display no obvious pharyngeal morphology defects but arrest development at the first larval stage as scrawny, starved-looking animals (Fig. 5D). This is exactly the phenotype expected from loss of function of PENs—specifically the M4 motor neuron, whose microsurgical removal results in exactly the same L1 arrest phenotype (Avery and Horvitz 1989).
Even though the temporal dynamics of expression of the pha-4 cis-regulatory allele (loss only in postmitotic neurons) and its resulting larval arrest phenotype already strongly argue for postmitotic neuronal function of pha-4, we sought to independently validate this notion using an orthogonal approach. To this end, we first engineered two loxP sites into the pha-4 locus (Fig. 5A). We independently engineered two strains in which we inserted a FLAG::NLS::Cre::SL2 expression cassette at the 5′ end of two different genes: egl-3 (neuropeptide-processing enzyme) and ceh-48 (homeobox gene; referred to here as Cre::egl-3 and Cre::ceh-48, respectively) (Supplemental Fig. S2). We had previously shown through gfp tagging of these loci that both loci are expressed exclusively in all postmitotic neurons of the central, peripheral, and enteric nervous systems (Leyva-Díaz and Hobert 2022). We crossed Cre::egl-3 and Cre::ceh-48 independently with the floxed pha-4 locus and confirmed the depletion of pha-4::gfp expression in normally *pha-4-*expressing neurons (Fig. 5E). In both resulting strains, we observed the same phenotype that we observed upon removal of the pha-4prom2 elements: In each strain, neuronal expression of PHA-4::GFP is lost by the twofold stage, and these animals arrest as scrawny, starved-looking L1 animals that fail to show proper pharyngeal pumping but display no obvious pharyngeal morphology defects (Fig. 5E). Taken together, the phenotypes of the cis-regulatory allele as well as the neuron-specific removal of floxed pha-4 let us conclude that pha-4 function extends beyond embryonic patterning to a function in postmitotic neurons.
Effect of pha-4 removal on neuron differentiation
We used both the cis-regulatory allele of pha-4 and the postmitotic Cre-mediated excision of the floxed pha-4 locus to ask whether postmitotic removal of pha-4 expression affected PEN differentiation. Analyzing unc-17/VAChT expression in the *cis-*regulatory allele as well as eat-4/VGluT and cat-1/VMAT expression in the floxed pha-4 allele showed gene expression defects in cholinergic, glutamatergic, and serotonergic PENs (Fig. 6A). Panneuronal identity, marked by an unc-75 reporter (Leyva-Díaz et al. 2025), is unaffected after postmitotic pha-4 removal (Fig. 6A), leading us to conclude that pha-4 participates in regulating the proper identity of PENs, but not their survival or overall panneuronal identity.
Effect of neuron-specific removal of pha-4 on PEN and HEN differentiation. (A) Loss of pha-4 affects the expression of neurotransmitter identity genes but not panneuronal identity genes. Images of L1 wild-type and L1-arrested larva show the reporter transgenes eat-4fosmid::mCherry (otIs518), mScarlet-I3::unc-75 (ot1539), and cat-1fosmid::mCherry (otIs625) in an Cre::egl-3(syb5804) pha-4loxP(ot1078 ot946) background. Images of an L1-arrested larva showing reporter allele unc-17::mKate2::3xFLAG (ot907) in the pha-4(ot1505) background. L1-arrested animals most likely to be homozygous pha-4(ot1505) were picked from a heterozygous plate based on scrawny appearance and developmental arrest; thus, some heterozygous animals may have been scored, dampening the mutant phenotype quantified. Statistical analysis was performed using Fisher's exact test. N is indicated within each bar and represents the number of animals scored. (B) Loss of pha-4 does not affect cell-specific identity genes in AVL, RIS, and DVB. Images of L1-arrested larva show the reporter transgene unc-47::mCherry (otIs348) in the pha-4(q490)-null background; lim-6::gfp (wgIs387 reporter transgene), flr-2::gfp (syb4861 reporter allele), flp-11b::gfp (ynIs40 reporter transgene), and unc-25::gfp (ot1372 reporter allele) in the pha-4(ot1506)-null mutant background; and unc-47::gfp (oxIs12 reporter transgene) and flr-2::gfp (syb4861 reporter allele) in the unc-47p::Cre (arTi479); pha-4loxP(ot1078 ot946) background. Statistical analysis was performed using Fisher's exact test. N is indicated within each bar and represents the number of animals scored. (C) Loss of pha-4 affects cell-specific identity genes in PVT and rectal K cells. Images of L1-arrested larva show the reporter transgenes pdf-1::gfp (syb3330 reporter allele), sre-22::gfp (otEx8333 reporter transgene), lim-6::gfp (wgIs387 reporter transgene), nlp-40::gfp (syb3208 reporter allele), and egl-38::gfp (wgIs171 reporter transgene) in the pha-4(ot1506) full locus deletion background. Statistical analysis was performed using Fisher's exact test. Sample size is indicated within each bar and represents the number of animals scored. (D) Loss of pha-4 does not affect intestine-specific identity genes. Images of L1-arrested larva show the reporter allele aex-5::gfp (ot1532) and ges-1p::gfp reporter transgene otIs904 (Sural et al. 2025) in the pha-4(ot1506)-null mutant background. Statistical analysis was performed using Fisher's exact test. N is indicated within each bar and represents the number of animals scored.
To assess the impact of pha-4 on the differentiation of the extrapharyngeal neurons that we newly identified to express pha-4 (RIS, AVL, PVT, and DVB), we did not rely on neuron-specific removal of pha-4 but rather first examined a complete pha-4-null allele to assess the strongest possible phenotype. In spite of the overall disorganization of the head region in pha-4-null mutants, molecular markers for RIS, AVL, and PVT are selectively expressed enough to allow for proper assessment of their expression in pha-4-null mutants. For the RIS and AVL neurons, we used terminal function genes previously shown to be critical for the function of these neurons: the vesicular transporter unc-47/VGAT (AVL and RIS), the GABA biosynthetic enzyme unc-25/GAD, and the neuropeptides flp-11 (RIS) and flr-2 (AVL). Expression of these markers is not affected in L1-arrested pha-4(ot1506)-null mutant animals (Fig. 6B). Notably, flr-2 is normally also expressed in PENs (Vidal et al. 2022), and we found PEN expression to be lost in pha-4-null mutants (Fig. 6B).
To assess the effect of neuron-specific pha-4 removal on DVB, which is born after the L1 stage (i.e., after pha-4-null mutant animals die), we removed pha-4 cell-specifically from DVB by driving Cre recombinase under control of the GABA neuron-specific promoter of the unc-47/VGAT locus. An unc-47prom::Cre miniMOS line indeed removed endogenously *gfp-*tagged pha-4 from the RIS, AVL, and DVB neurons but not the PVT neurons, as expected (Fig. 7A). Postembryonic loss of pha-4 from DVB does not affect the expression of molecular markers flr-2 and unc-47/VGAT (Fig. 6B). However, we cannot exclude the possibility that the Cre-mediated gene removal is too late to affect differentiation of the neurons. Nevertheless, considering the result with the null allele, it appears that in contrast to strong differentiation defects observed in the PENs, the RIS, AVL, and DVB neurons are generated and appear to adopt their proper fate in the absence of pha-4 function.
pha-4 acts in HENs to control defecation behavior. (A) Cell-specific removal of pha-4 from the gfp-tagged, floxed pha-4 locus (ot1078 ot946) generated via the CRISPR/Cas9 system via unc-47p::Cre (arTi479). Elimination of PHA-4::GFP expression was confirmed at the day 1 adult stage. (B) pha-4 removal from AVL, RIS, and DVB affects defecation behavior. The frequency of expulsion and the expulsion:pBoc ratio in adult worms with cell-specific removal of the pha-4 locus (floxed reporter allele ot1078 ot946) via unc-47p::Cre (arTi479). (C, left) Neuron-specific removal of PHA-4::GFP (floxed reporter allele ot1078 ot946) via the sre-22 promoter PVTp::Cre (otEx343) was confirmed in adult animals. All remaining PHA-4::GFP cells in the tail region are nonneuronal PHA-4(+) cells. (Right) pha-4 removal from PVT affects defecation behavior. The frequency of expulsion in adult worms with PVT-specific removal of pha-4 (floxed reporter allele ot1078 ot946) via PVTp::Cre (otEx343). Data points representing each individual worm assayed are plotted, and the horizontal line represents the median value of biological replicates in B and C. () P < 0.05, (**) P < 0.01, (***) P < 0.0001, (ns) not significant in Dunn's multiple comparisons test after Kruskal–Wallis test (B) and Mann–Whitney U-test (C).
In the PVT neurons, pha-4 appears to play an earlier patterning role. PVT is the sister of rectal cells that Mango et al. (1994) had noted to be absent in pha-4 mutants. Consistent with a lineage defect, we found that in pha-4-null mutants, expression of a terminal marker for PVT identity, sre-22, is frequently lost, as is expression of a reporter allele of the pdf-1 neuropeptide (Fig. 6C). The nlp-40 neuropeptide reporter allele shows reduced expression in a tail cell outside of the preanal ganglion region, where PVT normally resides (Fig. 6C). We further corroborated a potential lack of differentiated PVT by examining the expression of a regulator of PVT identity: the LIM homeobox gene, lim-6 (Hobert et al. 1999; Aurelio et al. 2003). We found no observable lim-6 signal in the tail region (Fig. 6C), consistent with a loss of the PVT neuron.
We also examined the impact of pha-4 on the rectal epithelial K cell, which is lineally related to PVT (Fig. 1B) and is the precursor of the postembryonically born DVB hindgut innervation neurons. We found that expression of the K-cell marker egl-38, a paired domain transcription factor (Chamberlin et al. 1997; Vidal et al. 2015), is affected in pha-4-null mutants (Fig. 6C), again consistent with pha-4 having early developmental defects in generating components of the hindgut. This is further supported by our analysis of the expression of aex-5, which codes for an endopeptidase involved in defecation behavior (Mahoney et al. 2008). An aex-5 reporter allele that we generated by CRISPR/Cas9 genome engineering (Sural et al. 2025) is expressed in the rectal gland cells (one of which, rectD, is also lineally related to PVT) (Fig. 1B) and in intestinal–rectal valve cells (virL/R), and this expression is lost in pha-4-null mutants (Fig. 6D).
lim-6 is also a terminal selector of AVL and RIS neuron identity, and, again consistent with the lack of differentiation defects of AVL and RIS (Hobert et al. 1999; Gendrel et al. 2016), we found lim-6 expression in these neurons to be apparently unaffected in pha-4-null mutants (Fig. 6B). We also noted ectopic lim-6 expression in the head, perhaps indicating that pharyngeal cells that fail to differentiate properly in pha-4 mutants undergo a cell fate switch (Fig. 6B).
pha-4 acts in GABAergic HENs and in PVT to control defecation behavior
We found that in pha-4(ot1506)-null mutants, the expression of several markers of terminal differentiation of intestinal cells (ges-1 and aex-5) is unaffected (Fig. 6D). Previous work has shown that pha-4 is required for proper function and homeostasis of mature intestinal cells (Panowski et al. 2007; Wu et al. 2018; Torzone et al. 2023), indicating that in specific cellular contexts, pha-4 may not have a role in the proper differentiation of a cell type but rather in its proper functioning. We set out to test whether a similar scenario applies to the extrapharyngeal neurons (HENs, RIS, and PVT), whose overall differentiation appears unaffected in pha-4-null mutants. To avoid the larval lethality associated with either pha-4-null mutants or panneuronal removal of pha-4, we set out to remove pha-4 selectively from HENs (AVL and DVB). To this end, we expressed Cre recombinase specifically in GABAergic neurons using the unc-47/VGAT promoter. An unc-47prom::Cre miniMOS line indeed removed endogenously *gfp-*tagged pha-4 from the RIS, AVL, and DVB neurons but not from PVT or other rectal cells, conforming with the specificity of the unc-47prom driver (Fig. 7A). Because AVL and DVB neurons (both of which are motor neurons that innervate enteric muscle) were previously shown to control defecation behavior (McIntire et al. 1993; Choi et al. 2021), we analyzed this behavior in animals lacking pha-4 in GABAergic neurons and indeed observed defecation defects. While the rate of pBoc events in these worms was only mildly affected, not all pBoc events were followed by expulsion events (Fig. 7B). In some worms, expulsion events were entirely absent for the six observed defecation cycles (Fig. 7B), which is consistent with the phenotype of worms in which AVL and DVB are ablated (McIntire et al. 1993; Choi et al. 2021). Removing pha-4 from GABAergic neurons does not affect pharyngeal pumping behavior (Fig. 7B; Supplemental Fig. S3).
To assess whether PVT requires pha-4 to regulate defecation behavior, we employed the above-described sre-22prom PVT driver to express Cre in the PVT neuron of animals that contain a floxed copy of the pha-4 locus. These animals also exhibit significant defects in the expulsion step of defecation, with only about half of pBocs being followed by an expulsion step (Fig. 7C). These defects were similar to those observed upon selective removal of egl-3 from PVT (Fig. 3). We conclude that pha-4 is required for features of PVT that are critical for its function in controlling defecation behavior.
pha-4 is continuously required to sustain function of HENs and PENs
While Cre-mediated removal of pha-4 establishes a function for pha-4 in postmitotic PENs and HENs, it does not address whether pha-4 may also function postembryonically to maintain function of these neurons through to adulthood. Previous work using a temperature-sensitive allele of the pha-4 locus indicated that pha-4 removal at the first larval stage resulted in animal lethality (Gaudet and Mango 2002), but it was not clear in which tissue type pha-4 was required or whether this requirement persisted into later stages of animal life. To address this issue, we used the auxin-inducible degron system (Zhang et al. 2015). We tagged the endogenous pha-4 locus with an auxin-inducible degron (AID) sequence (Fig. 8A) and generated a transgenic line in which the TIR1(F79G) ubiquitin ligase (Hills-Muckey et al. 2022) is driven by the pha-4prom2 driver described above, which is expressed in the PENs, HENs (AVL and DVB), and RIS but not in PVT. After addition of the auxin derivative 5-Ph-IAA, we observed that GFP::loxP::3xFLAG::AID2-tagged PHA-4 was indeed depleted from the PENs, AVL, and RIS but not PVT or DVB (Fig. 8B). Using this system, we observed feeding and defecation behavior of adult animals that have been treated with 5-Ph-IAA since either the embryonic, L1 larval, or L4 larval stage (Fig. 8C). We conclude that PHA-4 is required continuously after embryonic development for maintenance of enteric functions.
pha-4 is continuously required in enteric neurons to maintain enteric function. (A) pha-4 locus schematic showing the location of an additional AID tag inserted via CRISPR/Cas9 engineering into the already gfp-tagged pha-4 reporter allele ot1078 ot946, resulting in pha-4(ot1078 ot946 syb5755), which are labeled as pha::gfp::AID in the ensuing panels. (B) Temporally controlled removal of PHA-4 in pha-4(ot1078 ot946 syb5755) animals from enteric neurons via pha-4prom2::TIR1(F79G)(otIs908) in the presence of 5-Ph-IAA. Animals were treated with either ethanol only (0 µM 5-Ph-IAA) or 100 or 200 µM 5-Ph-IAA starting at various stages, and GFP expression in pha-4(+) neurons was characterized. White arrows point to TIR1(F79G) and pha-4(ot1078 ot946 syb5755) expression in pharyngeal neurons. Statistical analysis was performed using Fisher's exact test. N is indicated within each bar and represents the number of neurons scored. (C) Temporally controlled removal of pha-4 from enteric neurons results in defects in enteric behaviors. Pharyngeal pumping, the frequency of expulsion, and the expulsion:pBoc ratio in adult worms treated with auxin starting at different developmental stages are shown. Animals were treated with either ethanol only (0 µM 5-Ph-IAA) or the indicated concentrations of 5-Ph-IAA. Data points representing each individual worm assayed are plotted, and the horizontal line represents the median value of the biological replicates. () P < 0.05, (**) P < 0.01, (***) P < 0.0001, (ns) not significant in Šídák's multiple comparisons test after two-way ANOVA.
Discussion
The role of FoxA genes in gut development is evident across animal phylogeny. Previous studies on the C. elegans FoxA ortholog PHA-4 have firmly established its role as an organ selector gene for foregut development (Mango et al. 1994; Mango 2009). Target genes for PHA-4 have been defined during this embryonic patterning function in different parts of the foregut (Gaudet and Mango 2002; Gaudet et al. 2004; Fakhouri et al. 2010), and, on a mechanistic level, PHA-4 has been shown to act as a pioneer factor to enable gene activation (Hsu et al. 2015). We have extended here the description of PHA-4 function by revealing postmitotic functions of PHA-4 during terminal differentiation of enteric neurons of not only the foregut but also the hindgut, as well as other neurons that control the function of the foregut or hindgut. These postmitotic functions extend beyond into larval and adult stages. Because pioneer factors—of which the vertebrate PHA-4 ortholog FoxA is the paradigmatic example—are generally thought to provide the initial trigger for opening chromatin (Zaret and Carroll 2011), a continuous requirement for PHA-4 indicates functions of PHA-4 that go beyond such a pioneer function. We propose that within the foregut enteric nervous system, continuously expressed PHA-4 cooperates with a previously identified panenteric homeodomain transcription factor, the Six-type homeodomain protein CEH-34, that is also required to initiate and maintain enteric neuron differentiation (Vidal et al. 2022). A direct role of PHA-4 in maintaining, together with CEH-34, the fully differentiated state of pharyngeal enteric neurons is supported by the presence of ChIP-seq peaks for PHA-4 (identified by the modENCODE project [Gerstein et al. 2010]) in the vicinity of terminal marker genes including, for example, the cat-1/VMAT locus, whose expression we found to be affected by postdevelopmental removal of PHA-4.
Perhaps most remarkably, our discovery of expression of PHA-4 in extrapharyngeal neurons pointed to a fundamental relatedness of HENs and PENs and also identified other neurons that control the enteric system. PHA-4-expressing AVL and DVB neurons have long been recognized as hindgut-innervating neurons that control defecation behavior. The extrapharyngeal RIS interneuron, which we discovered here to also express PHA-4, is a known regulator of pharyngeal pumping (Steuer Costa et al. 2019). Even though its cell body is located outside the foregut in the ventral ganglion, the extension of its unipolar process along the nerve ring that wraps around the isthmus of the pharynx puts RIS in a suitable position to control pharyngeal pumping apparently via the release of neuropeptides (FLP-11) perceived by neuropeptide receptors expressed in pharyngeal enteric neurons. The expression of PHA-4 in the PVT neuron guided us to explore a potential function of PVT in enteric behavior as well. We discovered that PVT plays a role in controlling the expulsion step of defecation, apparently mediated via neuropeptide release. Removing pha-4 from PVT using a cell-specific Cre driver recapitulated the expulsion defects seen in worms where PVT was silenced, suggesting that pha-4 is required for the enteric function of PVT.
Previous work has suggested that the neuropeptide NLP-40 is secreted from the intestine to stimulate proper levels of GABA release in AVL and DVB (Wang et al. 2013). Because we found nlp-40 to also be very strongly expressed in PVT and because the NLP-40 receptor AEX-2 is expressed not only in AVL and DVB but also in PVT (Ripoll-Sánchez et al. 2023), it is conceivable that PVT may operate as an amplifier. It may detect NLP-40 released from the intestine via the AEX-2 receptor to promote further NLP-40 release from PVT to stimulate AEX-2 in AVL and DVB during the expulsion step.
With its extension of a long process into the nerve ring from the tail of the animal, PVT represents a potential route of communication from the brain to the hindgut, akin to the vagus nerve. The lineage relationship of PVT with other parts of the hindgut (Fig. 1B) could have already predicted a function of PVT in controlling hindgut function. However, lineage relationships of functionally related neurons are certainly not a universal feature within the gut (e.g., the hindgut-innervating neuron AVL has no lineage relationship to any part of the gut) or outside the gut (e.g., amphid sensory neurons have very limited lineal relationships). Last, we note that the function of the peptidergic RIS and PVT neurons in controlling enteric function can be viewed as conceptually similar to the control of vertebrate enteric nervous system function (e.g., gut motility) by neuropeptides produced by the CNS (Furness 2012).
While PHA-4 may act together with (and upstream of) a terminal selector, CEH-34, to control multiple aspects of PENs in a master regulatory manner, PHA-4 function in extrapharyngeal neurons (AVL, DVB, RIS, and PVT) may be restricted to controlling select aspects of the functional properties of these neurons. Notably, all these four neurons utilize a shared LIM homeodomain transcription factor, lim-6 (Hobert et al. 1999; Aurelio et al. 2003; Gendrel et al. 2016), for their proper differentiation. In each of these neurons, PHA-4 may assist LIM-6 in controlling some critical aspect of their function. A similarly selective mode of action of PHA-4 in controlling cell function rather than cell differentiation is observed in the midgut endoderm, which also forms properly in pha-4 mutants but displays a number of functional defects, such as in lipid homeostasis (Wu et al. 2018; Torzone et al. 2023). Hence, our findings demonstrate that PHA-4 acts in a highly contextualized environment. In some cells, it acts to control cell differentiation in a “master regulatory manner” in combination with other transcription factors, while in other cell types, its function is committed to controlling a highly selective subset of as yet undiscovered genes to control select aspects of their functional properties.
We conclude that PHA-4 is currently the only gene whose expression is tightly restricted to every single neuron of an enteric nervous system, where it fulfills a number of diverse functions from a pioneer function role in early patterning to the initiation of terminal differentiation programs to the maintenance of enteric neuron functionality throughout the life of the animal.
Materials and methods
Strains
All C. elegans strains were cultured at 20°C on nematode growth medium (NGM) plates seeded with Escherichia coli (OP50 strain) bacteria as a food source unless stated otherwise. The wild-type strain used was Bristol N2. All experiments were performed on hermaphrodites. A comprehensive list of the strains utilized in this study is provided in Supplemental Table S1. Details for the generation of several transgenic and CRISPR/Cas9-engineered strains are provided in the Supplemental Material.
Reagents for PVT neuron visualization and manipulation
To identify drivers for PVT neuron expression, pZW28 (cnc-11p::gfp::h2b::tbb-2 3′UTR) was generated by cloning a 757 bp cnc-11 promoter fragment (−757 to −1 position with respect to the start codon of cnc-11). pZW30 (sre-22p::gfp::h2b::tbb-2 3′UTR) and pZW32 (sre-22p::HisCl::sl2::gfp::tbb-2 3′UTR) were generated by cloning a 1144 bp sre-22 promoter fragment (−1144 to −1 position with respect to the start codon of sre-22). The tbb-2 3′ UTR sequence is a 156 bp sequence immediately downstream from the stop codon of the tbb-2 gene. The sre-22 driver was preferred for further usage (HisCl and Cre expression) over the cnc-11 driver because of its greater specificity, with the only non-PVT site of expression being dim expression in the PVQ neurons. While we cannot exclude a role for PVQ in interpreting the results of the sre-22prom-driven HisCl silencing, the sre-22prom-driven excision of pha-4 from PVT, with ensuing defects matching those of sre-22prom::HisCl, argues against a contribution from PVQ, because pha-4 is not expressed in PVQ.
All plasmids used in this study were generated via NEBuilder HiFi DNA assembly using the manufacturer's protocol. All plasmid sequences were validated using Sanger (Azenta) and/or Oxford Nanopore (Plasmidsaurus) sequencing. Prior to microinjection, all plasmids were linearized using a restriction site in the plasmid backbone. The linearized plasmids were then injected into both gonadal arms of young adult animals. F1 progeny were selected based on the expression of the coinjection marker, and those that transmitted the array to F2 progeny were used to establish transgenic lines. Three independent lines (derived from independently injected P0 animals) for each injection type were utilized in subsequent experiments.
nlp-40 and pdf-1 expression in PVT, initially suggested by the CeNGEN Atlas (Taylor et al. 2021), was examined by inserting a T2A::3xNLS::GFP cassette at the 3′ end of the nlp-40 and pdf-1 coding sequences (SunyBiotech). This reporter cassette was subsequently shown to produce lower reporter expression than other reporter cassettes (Sun et al. 2023), which explains why the nlp-40 reporter allele shows expression in PVT but not the midgut, as reported previously (Wang et al. 2013).
Assays for larval arrest of pha-4 mutants
To assess the consequences of neuronal removal of pha-4, we analyzed the progeny of animals in which either the neuronal enhancer of the pha-4 locus was deleted (ot1505 ot946 allele) or floxed pha-4 was removed with a panneuronal Cre driver. For each parental genotype, 10 age-synchronized L4 stage hermaphrodites were transferred from uncrowded, nonstarved plates to 10 individual fresh NGM plates seeded with E. coli OP50 bacteria. After 24 h at 20°C, parent worms were transferred to new NGM plates seeded with OP50 bacteria. The number of live larvae and unhatched embryos was counted for each plate after an additional 24 h at 20°C. Parent worms were iteratively transferred and progeny counted in 24 h intervals two additional times, and the total hatch rate was pooled for each individual parent.
Measurement of enteric behaviors
All pharyngeal pumping assays were performed on freely moving animals. Age-synchronized adult worms were transferred from uncrowded, nonstarved plates to fresh NGM plates seeded with a uniform thin layer of E. coli OP50 bacteria. After a 5 min acclimatization period, worms were placed under a Nikon Eclipse E400 upright microscope equipped with differential interference contrast (DIC) optics. Before recording behavior, worms were allowed to acclimate to the light source intensity under the microscope for an additional 5 min. The movement of the grinder of the pharynx was then recorded with a handheld tally counter using a 20× air objective lens. The number of grinder movements in 20 sec was recorded and multiplied by three to obtain pharyngeal pumps per minute. The pharyngeal pumping rate was recorded from at least eight animals each day, and data were pooled from at least two independent days.
All defecation assays were performed on freely moving animals. Age-synchronized adult worms were transferred from uncrowded, nonstarved plates to fresh NGM plates seeded with a uniform thin layer of E. coli OP50 bacteria. After a 5 min acclimatization period, worms were placed under a Nikon Eclipse E400 upright microscope equipped with DIC optics. Before recording behavior, worms were allowed to acclimate to the light source intensity under the microscope for an additional 5 min. Defecation behavior was observed under a 20× air objective lens. Each animal was observed for a 6 min period, during which the timings of contraction of posterior body muscles (pBoc) and expulsion (Exp) muscle contraction that releases food contents from the anal opening were recorded. Defecation behavior was recorded from at least six animals on each day, and data were pooled from at least two independent days.
Histamine-mediated chemogenetic silencing of neurons
The inducible silencing of RIS and PVT neurons was conducted using the histamine-gated chloride channel HisCl1 from Drosophila melanogaster and by the addition of exogenous histamine as described previously (Pokala et al. 2014). Histamine dihydrochloride (Sigma-Aldrich H7250) was dissolved in water to prepare 1 M stock solutions. NGM agar containing 0 µM (control), 8 µM, and 10 µM histamine dihydrochloride was added to 60 mm plates. Age-synchronized adults were transferred to NGM agar control and histamine plates immediately prior to experiments. After a recovery period, worms were observed for pharyngeal pumping and defecation behavior under a Nikon Eclipse E400 upright microscope equipped with DIC optics.
Auxin-mediated protein degradation
Auxin-inducible degron (AID)-tagged proteins were conditionally degraded when exposed to 5-Ph-IAA in the presence of TIR1(F79G). All AID2 experiments were carried out as described previously with some modifications (Sural et al. 2024). To generate the experimental strain, the conditional allele pha-4(ot1078 ot946 syb5755) was crossed with otIs908, which expresses TIR1(F79G) in all neurons of the pharyngeal nervous system and AVL, RIS, and DVB. The synthetic auxin analog 5-Ph-IAA was acquired from BioAcademia (30-003-10) and dissolved in ethanol (EtOH) to prepare 100 mM stock solutions. NGM agar plates seeded with a uniform lawn of E. coli OP50 bacteria were coated with 100 µL of freshly prepared 5-Ph-IAA stock solution and allowed to dry and diffuse overnight at room temperature to final concentrations of 100 and 200 µM. Age-synchronized embryos (obtained via the alkaline bleaching method), L1s, or L4s were transferred to the plates 1 day after adding 5-Ph-IAA, and the animals were grown for an additional 24 h at 25°C until they reached the desired developmental stage for the experiment. In control conditions, worms were transferred onto EtOH-coated plates. All plates were shielded from light for the duration of the experiment to limit degradation of 5-Ph-IAA.
Microscopy
Worms were anesthetized using 100 mM sodium azide (NaN_3_) and mounted on 5% agarose pads on glass slides. Animals were imaged on an inverted Zeiss LSM 980 laser scanning confocal microscope using a 40× water immersion objective lens or a Zeiss Imager Z2 upright microscope equipped with Colibri 7 LEDs (Zeiss) using a 40× oil immersion objective lens. For each image, Z-stacks of uniform 0.31–1 µm thickness (depending on the age of the worm) were obtained for the entire sample.
Reporter gene expression in different neurons was visualized in wild-type and mutant animals and usually was assigned to one of the following categories by the observer: “on” (fluorescence levels comparable with wild-type animals), “dim” (fluorescence still detectable but much dimmer than wild-type animals), or “off” (fluorescence not detectable).
Statistical analysis
All images were processed and analyzed using ImageJ FIJI (Schindelin et al. 2012). All plotting and statistical tests were performed on GraphPad Prism 10. Sample sizes for pharyngeal pumping and defecation assays were a minimum of 32 and 12 animals, respectively, based on standards established in prior studies using these behavioral readouts. These sample sizes have been shown to reliably detect physiologically relevant differences. For microscopy images, a minimum of eight animals were imaged per condition, and a representative image is shown in the figures. The figure legends include the statistical tests applied in each figure along with the corresponding *P-*values. P < 0.05 was considered to be statistically significant.
Supplemental Material
Supplement 1
Supplement 2
Supplement 3
Supplement 4
Supplement 5
Supplement 6
Supplement 7
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Adler CE, Seidel CW, Mc Kinney SA, Sánchez Alvarado A. 2014. Selective amputation of the pharynx identifies a Fox A-dependent regeneration program in planaria. e Life 3: e 02238. 10.7554/e Life.0223824737865 PMC 3985184 · doi ↗ · pubmed ↗
- 2Albertson DG, Thomson JN. 1976. The pharynx of Caenorhabditis elegans. Philos Trans R Soc Lond B Biol Sci 275: 299–325. 10.1098/rstb.1976.00858805 · doi ↗ · pubmed ↗
- 3Annunziata R, Andrikou C, Perillo M, Cuomo C, Arnone MI. 2019. Development and evolution of gut structures: from molecules to function. Cell Tissue Res 377: 445–458. 10.1007/s 00441-019-03093-931446445 · doi ↗ · pubmed ↗
- 4Ao W, Gaudet J, Kent WJ, Muttumu S, Mango SE. 2004. Environmentally induced foregut remodeling by PHA-4/Fox A and DAF-12/NHR. Science 305: 1743–1746. 10.1126/science.110221615375261 · doi ↗ · pubmed ↗
- 5Aurelio O, Boulin T, Hobert O. 2003. Identification of spatial and temporal cues that regulate postembryonic expression of axon maintenance factors in the C. elegans ventral nerve cord. Development 130: 599–610. 10.1242/dev.0027712490565 · doi ↗ · pubmed ↗
- 6Avery L, Horvitz HR. 1989. Pharyngeal pumping continues after laser killing of the pharyngeal nervous system of C. elegans. Neuron 3: 473–485. 10.1016/0896-6273(89)90206-72642006 · doi ↗ · pubmed ↗
- 7Avery L, Thomas JH. 1997. Feeding and defecation. In C. elegans II (ed. Riddle DL et al.), pp. 679–716. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.21413237 · pubmed ↗
- 8Azzaria M, Goszczynski B, Chung MA, Kalb JM, Mc Ghee JD. 1996. A fork head/HNF-3 homolog expressed in the pharynx and intestine of the Caenorhabditis elegans embryo. Dev Biol 178: 289–303. 10.1006/dbio.1996.02198812130 · doi ↗ · pubmed ↗