Figures
Abstract
Transcription factors govern many of the time- and tissue-specific gene expression events in living organisms. CEH-60, a homolog of the TALE transcription factor PBX in vertebrates, was recently characterized as a new regulator of intestinal lipid mobilization in Caenorhabditis elegans. Because CEH-60’s orthologs and paralogs exhibit several other functions, notably in neuron and muscle development, and because ceh-60 expression is not limited to the C. elegans intestine, we sought to identify additional functions of CEH-60 through DNA adenine methyltransferase identification (DamID). DamID identifies protein-genome interaction sites through GATC-specific methylation. We here report 872 putative CEH-60 gene targets in young adult animals, and 587 in L2 larvae, many of which are associated with neuron development or muscle structure. In light of this, we investigate morphology and function of ceh-60 expressing AWC neurons, and contraction of pharyngeal muscles. We find no clear functional consequences of loss of ceh-60 in these assays, suggesting that in AWC neurons and pharyngeal muscle, CEH-60 function is likely more subtle or redundant with other factors.
Citation: Van de Walle P, Muñoz-Jiménez C, Askjaer P, Schoofs L, Temmerman L (2020) DamID identifies targets of CEH-60/PBX that are associated with neuron development and muscle structure in Caenorhabditis elegans. PLoS ONE 15(12): e0242939. https://doi.org/10.1371/journal.pone.0242939
Editor: Denis Dupuy, INSERM U869, FRANCE
Received: September 3, 2020; Accepted: November 11, 2020; Published: December 11, 2020
Copyright: © 2020 Van de Walle et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Raw and processed DamID data are publicly available at ArrayExpress (accession E-MTAB-9539).
Funding: PVdW is an n SB PhD fellow of the FWO Flanders (1S00617N, https://www.fwo.be). The authors are grateful to the FWO Flanders (G095915) and KU Leuven (https://www.kuleuven.be, C16/19/003), the Spanish Ministry of Science and Innovation (PID2019-105069GB-I00 and MDM-2016-0687 to PA) and the Spanish Research Council (2019AEP142 to PA) for financial support. We are especially grateful to the Genie COST action (BM1408) for supporting this research.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
PBC-class transcription factors fulfill a wide range of developmental functions in many organisms (reviewed in [1]). In the nematode Caenorhabditis elegans, they are represented by CEH-20, -40 and -60. While CEH-20 and -40 have been extensively characterized as drivers of neuronal, muscle and general mesodermal development [2–8], CEH-60 remained poorly understood. Recently, it has been shown that CEH-60, like other PBC-class proteins, interacts with a conserved partner UNC-62 and that this interaction occurs in the adult intestine to control lipid mobilization through yolk protein production [9, 10]. It is also argued that the default mode of action of CEH-60 is as a repressor of transcription. This is supported by the fact that in absence of functional CEH-60, ~75% of differential transcripts are upregulated vs ~25% downregulated and that upregulated genes associate more closely with CEH-60 binding sites than downregulated genes [9].
The role of CEH-60 may not be limited to its adult-specific function in lipid mobilization, as CEH-60 activity is also observed in a specific pair of sensory neurons, identified as AWC ("Amphid Wing C"), and in the pharyngeal muscle cells PM6 [10, 11]. AWC neurons are a pair of ciliated neurons in the head region of the animal, their neuron bodies residing near the nerve ring that forms the nexus of the C. elegans nervous system [12]. The best-characterized function of AWC neurons is sensing volatile odors that may signal the presence of food in natural environments to regulate movement towards these odors, termed chemotaxis [13]. To a lesser degree, AWC neurons are involved in sensing temperature [14] and electric fields [15]. Pharyngeal muscles are involved in food intake, and can thus regulate metabolism [16].
Other PBX proteins in C. elegans, CEH-20 and CEH-40, as well as their vertebrate counterparts, have well-characterized roles in neuronal development [17–19] and muscle development [20–24]. Because of ceh-60's functionally orphan neuronal and pharyngeal expression pattern and involvement of its closest relatives in numerous other developmental processes, we decided to look for direct gene targets of CEH-60 using DNA adenine methyltransferase identification (DamID).
DamID is a technique to map the interactions between a protein of interest (POI) and the genome. It has been successfully performed in Drosophila melanogaster [25], mammalian cell lines [26] and C. elegans [27]. This technique was originally developed to characterize the interactions of chromatin proteins with DNA [25], but can also be used to find target genes of transcription factors [28–30]. In this paper, DamID is used to map the interactions between the transcription factor CEH-60 and the promoter regions of its gene targets. Compared to immunoprecipitation techniques such as ChIP-seq, DamID has the advantage of providing an accumulated signal of protein binding instead of a “snapshot”.
DamID relies on the fusion of a POI to a Dam domain, which will methylate the adenine of GATC sites in the genome that are spatially close to where the fusion protein interacts with the DNA [25]. Because GATC methylation does not occur naturally in eukaryotes, the methylation sites can be used as markers for sites of POI-DNA interaction.
In this study, we use DamID to find the genomic targets of CEH-60 and classify them. Genes linked to neuronal development, along with muscle structure genes, indeed emerge as some of the strongest target candidates. We subsequently investigate the morphology and sensory function of AWC neurons as well as pharyngeal pumping in ceh-60 mutants. In these assays we find no clear differences between mutants and controls, suggesting that the function of CEH-60 in neurons and muscle cells may be more subtle or redundant with other factors.
2. Materials and methods
2.1 Strains and culture methods
C. elegans were grown under standard conditions [31], fed Escherichia coli OP50, and raised at 20°C. For details on strain names and genotypes, see Table 1.
2.2 DamID: Plasmids and strains
To obtain tissue-specific control over the dam::ceh-60 transgene, we used a FLP-based gene expression toolkit [32]. In the experimental DamID strain LSC1724, an inducible heat-shock promoter is followed by a fusion of an mCherry gene with a fragment encoding the histone HIS-58, flanked by FRT sites, integrated on chromosome II. This FRT-excisable reporter cassette is followed by a fusion gene of ceh-60 and dam. Additionally, integrated on chromosome IV, this strain contains a ceh-60 promoter driving expression of the flippase FLP, followed by an mNeonGreen reporter gene. Another strain, LSC1722, in which the dam::ceh-60 fusion gene is replaced with gfp::dam, serves as the control for aspecific DamID signal (= noise) during analysis.
Only in tissues in which FLP is expressed, the FRT-flanked cassette on chromosome II is excised, inactivating the mCherry and his-58 reporters, and ceding heat-shock control to the dam::ceh-60 or gfp::dam fusion genes. In this way, Dam fusion proteins are only expressed in tissues in which the flippase, driven by a tissue-specific promoter (here, ceh-60p), is active. Heat shock promoters are used in non-heatshocked conditions to maintain a low and steady state of expression.
2.3 DamID: Sampling and library preparation
DamID sampling and library preparation were carried out essentially as described in [33]. Strains carrying either the gfp::dam (LSC1722) or dam::ceh-60 (LSC1724) transgene were grown under standard conditions. Standard hypochlorite treatment was used to harvest eggs, which were allowed to hatch overnight in M9 buffer. For growing animals for DamID library preparation, four 100 mm NGM plates seeded with Dam-negative E. coli GM119 were used. Dam-negative bacteria do not show GATC methylation, which occurs naturally in most other E. coli strains and would otherwise contaminate C. elegans DamID libraries with bacterial GATC methylation. Young adult or L2 animals were rinsed off plates and 30 μL aliquots were snap-frozen in liquid nitrogen. Genomic DNA was extracted and purified using the Qiagen DNeasy Blood and Tissue kit according to the manufacturer's instructions. 200 ng of each genomic DNA sample was digested by incubation with 10 units of DpnI in 10 μL for 6 hours at 37°C, followed by inactivation at 80°C for 20 minutes. DpnI-digested DNA was incubated overnight at 16°C with 2 μL ligation buffer, 1 μL Roche T4 DNA ligase and 0.8 μL of double-stranded adaptors (AdRt: CTAATACGACTCACTATAGGGCAGCGTGGTCGCGGCCGAGGA, 50 μM, AdRb: TCCTCGGCCG, 50 μM, mixed separately, heated to 95°C and cooled to room-temperature before addition to ligation mix) in a total volume of 20 μL. Ligase was inactivated by incubation for 10 minutes at 65°C. DpnI-digested adaptor-ligated DNA fragments were purified with AgenCourt AMPure XP® beads and a magnetic particle concentrator, after which the sample was digested with DpnII (NEB) by incubation for 1 hour at 37°C, followed by inactivation at 80°C for 20 minutes and another purification step with AgenCourt AMPure XP® beads. DpnII-digested DNA was then amplified using Taq polymerase and 1.25 μL of 50 μM Adr primer (NNNNGGTCGCGGCCGAGGATC). Amplified adaptor-ligated DNA was analyzed by agarose gel electrophoresis and size-selected for amplicons between 400–1200 bp, the expected size of naturally-occurring GATC-flanked reads, using AgenCourt AMPure XP® beads and a magnetic particle concentrator according to the manufacturer's instructions. Using a 0.3 bead-to-sample ratio and selecting the supernatant, amplicons larger than 1200 bp are discarded. Next, using a 0.8 beads-to-sample ratio and selecting the bead-bound DNA, amplicons smaller than 400 bp are discarded. The DNA pool is separated by size through gel electrophoresis to confirm the presence of DNA only in the expected 400–1200 bp range. Adaptor ligated DNA is enriched by PCR and prepared for sequencing using the NEBNext Singleplex oligos for Illumina® by adding 25 μL of NEBNext Q5 Hot Start HiFi PCR Master Mix to 5 μL Index Primer and 5 μL Universal PCR primer to 15 μL of Adaptor-ligated DNA fragments, followed by 8 cycles of PCR amplification as per the manufacturer's instructions. Finally, the sequencing libraries are purified with AgenCourt AMPure XP® beads and size of the amplicons is confirmed by agarose gel electrophoresis and ExperionTM automated electrophoresis system (Bio-Rad) before sending out the libraries for Illumina sequencing at EMBL GeneCore (Heidelberg, Germany).
2.4 DamID: Data analysis
Next-generation sequencing data generated from DamID DNA samples was processed using a variant of the DamIDSeq R pipeline optimized for mapping DamID reads to gene regions termed GeneDamIDseq [34], freely available as R module. The GeneDamIDSeq module takes as input a text file describing the nature and path of (compressed) fastq files which contain the raw sequencing data for all conditions. Reads that do not contain the adapter sequence followed by the GATC motif are discarded, while the adapter sequence is trimmed from the remaining reads. Trimmed reads are mapped to the C. elegans reference genome BSgenome.Celegans.UCSC.ce11, obtained from WormBase, using Bowtie. Read counts for each binned genomic region are summed per sample and normalized for total read number per sample. In GeneDamIDSeq, the binned regions correspond to genes on the C. elegans genome.
The resulting output contains read numbers for each bin (i.e. each C. elegans gene) for each of three replicates for experimental (i.e. dam::ceh-60) and control (gfp::dam) conditions. First, genes were selected where the dam::ceh-60 normalized read number in each single lane was higher than the read number in each single gfp::dam lane and fold-change of average read number of dam::ceh-60 over gfp::dam was higher than 1.7. Correlation analysis on mapped GATC reads was performed by calculating Spearman’s correlation coefficient for all comparisons within young adult or L2 datasets, after removing GATC sites with 0 or 1 read, leaving ~19,000 reads out of ~27,000 per sample.
Corresponding gene names for all targets were obtained using the BioMart Data Mining Tool on WormBase ParaSite. Gene Ontology analysis was carried out using PANTHER 15's statistical overrepresentation test using default settings for biological processes. Gene target lists of L2 and young adult animals were compared for significant overlap using a chi-square test. Single-cell RNA-seq data from [35] was mined for expression profiles of CEH-60 DamID gene targets in order to obtain expression values in each tissue for genes present in both L2 and young adult datasets, L2 only, and young adult only. For each list, the total transcripts per million value was calculated per tissue. DamID and ChIP-seq peaks were tested for overlap using ChIP-seq NarrowPeaks lists for CEH-60 or UNC-62 [9], and 3kb promoter regions upstream of CEH-60 targets identified by GeneDamIDseq. CEH-60 DamID vs CEH-60 ChIP-seq targets, and CEH-60 DamID vs UNC-62 ChIP-seq were compared for significant overlap using a chi-square test. Raw and processed DamID data was deposited to ArrayExpress (accession E-MTAB-9539).
2.5 Microscopy
For characterization of AWC neuron morphology, synchronized young adult animals carrying the odr-1::rfp transgene were anesthetized with 1 mM tetramisole and visualized using a confocal FluoView1000 microscope (Olympus, Japan). Z-stack images of the head region of each animal were converted into maximum intensity projections using Fiji [36].
2.6 Chemotaxis assay.
Butanone chemotaxis assays were performed essentially as described in [37] for naive sensing conditions. Synchronized animals were grown at 20°C on E. coli OP50 until the young adult stage, after which they were washed off the plates with M9 buffer and collected in 15 mL conical tubes. Worms were allowed to settle without centrifugation to minimize sampling stress and they were washed twice with M9 buffer to remove residual bacteria. Assay plates were prepared by spotting 1μL of 1M NaN3 on both sides of the plate and additionally adding 1μL of 95% EtOH to one spot and 1μL of 10% butanone in a previously prepared 95% EtOH solution to the other spot. Approximately 100 animals per replicate were spotted at the origin. Chemotaxis assay plates were incubated for 1 hour at room temperature, after which the number or worms at the origin, EtOH-spot, butanone-spot and the total number of worms on the plate were counted. The chemotaxis index (CI) was calculated as Conditions were compared using a one-way ANOVA with Dunnett's multiple comparison post-hoc test.
2.7 Pharyngeal pumping assays
Pharyngeal pumping assays were performed essentially as described in [38]. Non-starved day 1 adult animals were filmed on NGM plates with E. coli OP50 for three times 15 seconds using a Leica M165 FC microscope outfitted with an MU035 AmScope microscope eyepiece camera. Recordings were played back at one fourth of normal speed to quantify pumping rate. Six animals were imaged per condition for three periods of 15 seconds each. Pumping rates were averaged per animal. Pumping rates for all conditions were compared using a one-way ANOVA with Dunnett’s multiple comparison post-hoc test.
Isthmus peristalsis rate coinciding with movement of food from the corpus of the pharynx to the terminal bulb was measured using dsRed-expressing OP50 bacteria as a food source, as described in Van Sinay et al. [39] with some adaptations. Well-fed young-adult animals were transferred to a 2% agarose pad containing a 2 μL drop of OP50-dsRed E. coli bacteria, allowed to air dry and covered with a coverslip. Animals were imaged on a Zeiss Axio Observer Z1 equipped with a hEGFP/HcRed filter using only the 555 mm light source. MetaMorph® software was used for image acquisition. Image analysis was performed in Fiji [36]. Isthmus peristalsis rates were quantified during 100 seconds for at least six animals per condition. Frames during which the animal was out of frame or out of focus were censored. Rates were compared using a one-way ANOVA with Dunnett’s multiple comparison post-hoc test.
3. Results and discussion
3.1 DamID reveals putative gene targets of CEH-60
To identify the gene targets of CEH-60, we performed DamID on L2 larvae and young adult animals carrying the dam::ceh-60 transgene, using non-specific gfp::dam animals as controls. Highly reproducible results were obtained from three biological replicas as shown by Spearman’s correlation analysis (S1 Fig). High correlation values are observed within dam::ceh-60 samples (≥0.67 in YA, ≥0.65 in L2). Correlation between gfp::dam and dam::ceh-60 is also high (≥0.52 in YA, ≥0.51 in L2). This is not surprising, as open chromatin is more accessible than dense chromatin to both proteins: transcription factor fusions, such as Dam::CEH-60, and diffusible GFP::Dam [40, 41]. We assigned loci as putative CEH-60 targets when the normalized dam::ceh-60 read number in each single replica was higher than the normalized read number in each single gfp::dam lane, and the fold-change of average read number of dam::ceh-60 over gfp::dam was higher than 1.7, resulting in 872 candidate gene targets in young adult samples (S1 Table).
Gene Ontology (GO) analysis of these 872 candidates reveals several statistically overrepresented biological processes, most prominent among which are categories related to muscle, embryo and neuron development (Fig 1 and S3 Table). Among the GO terms, most contain 20–40 gene targets, representing only a small set of the 872 hits, indicating that CEH-60 likely does not target a single pathway or process, but instead regulates many processes or pathways, possibly in subtle ways.
Non-exhaustive list of biological processes overrepresented in CEH-60 DamID targets. False discovery rate corrected p values are shown for the specified biological process. GO terms were grouped thematically (neuron/embryo/muscle development) by color. Gene ontology analysis was carried out with PANTHER 15 using statistical overrepresentation test for biological processes (complete). A complete list of overrepresented GO terms can be found in S3 Table.
Looking at GO terms in more detail, CEH-60 target loci identified by DamID contain multiple genes regulating neuron projection development specifically, or neuronal development more generally. These include the FEZ1 ortholog unc-76, important for axon fasciculation and extension [42], zag-1, a Zn-finger homeobox transcription factor gene regulating axon development and neuronal differentiation [43], unc-130, coding for a Forkhead transcription factor involved in axon extension and guidance [44], the kinase gene sad-1, which regulates presynaptic vesicle clustering and axon termination [45], the cell-adhesion molecule gene sax-7, involved in neuronal development [46], the serine/threonine kinase gene sax-1, involved in the development of axons of sensory neurons [47], the contactin gene rig-6, which mediates neuronal cell migration [48], the synaptic guidepost protein coding gene syg-2, important for synaptic specificity [49], the heterochronic regulator of cell fate decision during larval development lin-14 [50], the homeobox transcription factor gene ceh-14, which is important for development of thermosensory neurons [51] and the kinesin motor protein encoding gene vab-8, which functions in posteriorly directed neuron migration [52]. The presence of this large group of drivers of neuronal development, most of them involved in axon extension, suggests a role for CEH-60 in the development of ceh-60 expressing AWC neurons and their related sensory function.
The other large group of candidate genes identified by gene ontology analysis is involved in muscle development and contraction. These include the troponin T1 orthologs mup-2, tnt-2 and T22E5.6. These three genes form part of the troponin complex which enables Ca2+-dependent muscle contraction and in C. elegans is also important for muscle development and axon fasciculation [47, 53], the myosin light chain-encoding genes mlc-1, mlc-2, mlc-3 and mlc-4 [54], the myosin heavy chain genes myo-3 [55] and unc-54 [56], the myosin light-chain kinase-like unc-22 gene [57], the Obscurin-coding gene unc-89, a kinase important for muscle cell architecture [58, 59], the tropomyosin ortholog lev-11, and the paramyosin-coding gene unc-15, a major structural component of invertebrate muscles [60]. Many of the targets in this category code for large structural proteins or kinases in muscles, which suggests a possible role for CEH-60 in muscle structure. In support of this, the ceh-60 paralogs ceh-20 and ceh-40, and CEH-60’s in vivo interaction partner UNC-62 have all been implicated in muscle patterning, vulval muscle development and post-embryonic muscle cell differentiation in C. elegans [2, 4–6]. Vertebrate Pbx genes, orthologs of ceh-20, -40 and -60, are also known to be involved in muscle development [20–24].
While ceh-60 is not expressed in the body wall muscle of C. elegans, which comprises the largest muscle system in the animal, ceh-60 is distinctly expressed in the smaller pharyngeal muscle system [10], indicating a possible role for ceh-60 in this tissue. As “regulation of muscle contraction” is one of the most overrepresented biological processes among CEH-60 DamID targets, we speculate that CEH-60 could be involved in contraction of the pharynx, which is indispensable for feeding [61]. Indeed, for 9 out of the 23 genes classified under “muscle structure development”, the largest muscle-related biological process in the GO analysis (Fig 1), there is evidence for expression in pharyngeal muscle: unc-89, unc-22, unc-15, emb-9, dyb-1, alp-1, unc-96, mlc-1 and unc-27 [54, 62–69]. These observations indicate that ceh-60’s function in the muscle system could be related to the pharynx. Alternatively, CEH-60’s action could involve the body wall muscles, but expression in these tissues may be low or time-specific.
Apart from muscle structure and neuronal development, embryonic development is also overrepresented among gene targets. While we did not study the function of CEH-60 in embryonic development in detail, a recent system-level screen of early embryonic development showed that ceh-60 knockdown causes disorganized cell division during two early cleavage points of embryonic development (P3 and AB8) and sporadic misarrangement prior to division [70]. Curiously, ceh-20 and unc-62 knockdown were also tested in the high-throughput study by Guan et al. [70], but without apparent defects. This could indicate that the role of ceh-60 in embryonic development may be mechanistically distinct from its postembryonic function, although no concrete evidence is present for this hypothesis.
Curiously, unc-62 itself is also found among the candidate gene targets of CEH-60. UNC-62 is an intestinal interaction partner of CEH-60 important for lipid transportation [9, 10]. The possibility of genetic cross-regulation by direct binding of CEH-60/PBX to the UNC-62/MEIS-encoding DNA is interesting, and suggestive of a feedback mechanism for this conserved regulatory complex. One group of genes that is notably absent among the DamID gene targets however, encompasses those coding for vitellogenins, precursors to yolk proteins, discussed below.
3.2 Developmental expression shift of CEH-60 targets
We performed DamID on both young adult animals and L2 larvae in order to get a more dynamic view of how CEH-60 gene targets may change when its expression pattern shifts with respect to development, i.e. from limited to the neurons and PM6 during early larval stages to also including intestinal cells in L4 and adult stages [10]. Applying the same approach as for young adult animals, we found 587 candidate CEH-60 targets in L2 DNA (S2 Table). Of these, 43% (252/587 targets, p < 2.2E-16) overlap with those found in young adults. The overlapping genes include muscle proteins mlc-3, myo-3, tnt-2, unc-15, unc-54 and T22E5.10 and the neuronal development regulators lin-14, unc-130 and syg-2, in addition to the vitellogenesis regulators unc-62 and lrp-2. The observation that many genes are also “lost” from the L2 to the young adult stage suggests that early stage-specific marks placed by Dam::CEH-60 may be masked during development by an accumulation of methylation in open chromatin by the continuous activity of GFP::Dam in non-dividing cells. Indeed, we found no significant difference in signal intensity between overlapping and non-overlapping (<log2FCoverlap> = 1.36, <log2FCnon-overlap> = 1.51, p = 0.29) genes, indicating that non-overlapping genes are not simply noise in the L2 dataset that was not filtered out.
We compared the expression sites of L2 and young adult DamID gene targets to gain insight into possible shifts in binding preferences of CEH-60 upon reaching adulthood. For example, an increase in primarily intestinal genes in young adult animals (compared to larval animals) could reflect the activation of ceh-60’s intestinal expression from the L4 stage onward [9, 10]. To this end, we queried three lists of CEH-60 gene targets against publicly available single-cell RNA-seq data from L2 larvae [35]: (1) those overlapping between L2 and young adult, (2) those occurring only in L2 and (3) those occurring only in young adults. Because no peer-reviewed, whole-animal single-cell RNA-seq data are available that would allow a similar analysis based on expression patterns of young adults, we were limited to assigning genes—also the young adult ones—to larval expression patterns.
In presumed CEH-60 target genes that overlap between L2 and young adult datasets, body wall muscle expression is most enriched: body wall muscles represent 61% of all CEH-60 target gene expression, while all other tissues represent only 5–9% each. Of all body wall muscle transcripts measured in the Cao et al. dataset [35], are represented by CEH-60 DamID targets found in both L2 and young adults, while this value ranges from 1–2.5% for other tissues.
In genes identified as possible CEH-60 targets in the L2, but not in the young adult stage, no clear tissue preference can be observed: as opposed to what is seen in the shared gene group, here, body wall muscles account for a share of only 19% of target gene expression, which is similar to the 12–17% weight of nearly all other tissues: gonad, hypodermis, neurons, pharynx and glia. Intestinal expression is least abundant, accounting for only 9%.
In genes present as putative CEH-60 targets only in young adults, but not in L2 larvae, body wall muscles again take the lead in tissue weight, claiming 39% of target gene expression, while neurons and pharynx represent 15% and 12% respectively. Based on ceh-60’s known temporal change in expression pattern [9, 10], the limited weight of the intestine in this young-adult-only list may initially seem counter-intuitive. We cannot emphasize explicitly enough, however, that assignment of tissue-resolved expression is based on RNA-seq data from L2 larvae [35]. CEH-60 gene targets may be expressed in different tissues throughout life, just like ceh-60 itself, and targets may for example be expressed in the intestine in young adults, but not in L2 larvae. Thus, our expression analysis should be used as an indication of which categories of genes may be targeted by CEH-60, until future single-cell resolved sequencing data of adult animals will become available that will allow to fully account for life-stage-specific changes. In all three lists, body wall muscles are the most highly represented, corresponding to the previously described GO-proposed roles of CEH-60 in muscle function. In the overlapping gene list, body wall muscle expression is highest (61 vs 19% and 39%), which could be indicative of a constitutive ‘core function’ of CEH-60 in muscles throughout postembryonic life. Neurons represent less expression in overlapping genes than in L2 or young adult-specific genes, which may indicate stage-specific neuronal roles for CEH-60. The intestine represents only 6–10% of total expression in all three datasets, although expression is highest for young-adult only genes, which may reflect the reported increase in intestinal CEH-60 from the L4 stage onwards (cf above, [9, 10]).
3.3 DamID and ChIP-seq are complementary tools for identifying gene targets and revealing new transcription factor functions
When comparing our DamID results with those based on an available ChIP-seq dataset [9], we find that 301 of 872 DamID-based CEH-60 target genes have one or more corresponding ChIP peaks (35%, p < 2.2E-16). Our result of 35% corresponds well with percentages of 32–49% reported for similar comparisons in literature [30, 71]. While this overlap is convincing, it does show that many DamID targets are not found by ChIP-seq and vice versa. This indicates that DamID and ChIP-seq approaches could be complementary when searching for gene targets of transcription factors. Possibly, sterical considerations limit the interactions revealed by either technique: a POI::Dam fusion protein may have different access to the DNA than a POI::GFP protein pulled down with immunoprecipitation. Additionally, the relative accessibility of different tissues might differ between the two techniques.
Among the overlapping targets, many belong to the broad functional categories delineated above, including muscle proteins (lev-11, mlc-1, -2 and -4, mup-2, T22E5.10, tnt-2, unc-54, unc-89) and regulators of neuronal development (lin-14, rig-6, sad-1, sdn-1, unc-130, unc-73, vab-8). On the other hand, one notable set of targets lacking in our DamID-based data are a group of intestinally-expressed genes, the vit genes.
vit genes code for vitellogenin proteins, and are lacking from our gene target list in both young adult and L2 animals, even though they are obvious candidates for direct regulation by intestinal CEH-60 in young adults [9–11]. A recent study of CEH-60 gene targets employing a different approach, did find that CEH-60 directly interacts with the promoters of vit genes in adult animals [9]. In line with this, CEH-60’s known interaction partner UNC-62 is also known to bind to the VPE1 element in the vit gene promoters [72, 73]. One possibility for the peculiar absence of vit genes in our dataset, is that our DamID pipeline may have been more stringent than standard ChIP-seq analysis, as indicated by the 872 gene targets identified through DamID by us, and the 9010 targets identified through ChIP-seq [9]. Alternatively, our DamID sampling may not have been optimal for capturing binding of vit genes in the C. elegans intestine, as this only occurs from the L4 stage on. While we sampled during the young adult stage, methylation of the GATC sites in vit promoters may have required more time to become significantly enriched. This could be a second reason why we are not observing an increase in intestinal targets in young adult animals compared to L2 larvae (Table 2), in addition to the L2 origin of the tissue-specific RNA-seq data (cf above).
When comparing ChIP-seq data of CEH-60’s known interaction partner UNC-62 [9] with DamID targets of CEH-60, we find that 82 out of 872 CEH-60 DamID targets correspond to one or more of the 2053 ChIP-seq peaks for UNC-62 (9.4%, p = 8.21E-13). This shows that there is a significant overlap in gene targets of CEH-60 and UNC-62, as would be expected from their conserved in vivo interaction. Of these 82 shared genes between CEH-60 DamID and UNC-62 ChIP-seq targets, 69 also occur in the 301 overlapping targets of CEH-60 DamID and CEH-60 ChIP-seq datasets (84.1%, p < 2.2E-16). Together, these data show that gene targets generated using DamID share a significant number of hits with gene targets generated using other methods for the same transcription factor or transcription factors that are—at least in some proven cases—part of the same complex, such as UNC-62.
Comparing our results to other DamID studies reveals that the number of putative gene targets identified by us, 872 for YA and 587 for L2 animals, is within the expected range [27, 29, 74, 75]. Still, false positives are definitely possible and gene targets identified by DamID alone should be regarded as putative. We here relied on DamID as a tool to build hypotheses on new functions of the transcription factor CEH-60, which can be subsequently studied using functional assays.
3.4 CEH-60 is not essential for normal AWC morphology or odor-sensing function
The transcription factor CEH-60 targets several genes important for neuronal development (Fig 1), and ceh-60 is expressed throughout life in the AWC neurons [10]. To assess whether the morphology of these neurons depends on CEH-60’s function, we crossed ceh-60(lst466) mutants with the AWC-specific odr-1p::rfp marker and observed the morphology of AWC neurons in both conditions (Fig 2). RFP signal can clearly be observed in the cytoplasm of both wild-type and ceh-60 mutant AWC neurons, with sensory cilia being present and reaching the tip of the nose of the animal. Also, the dendrite connecting the left and right AWC neuron by following a semicircular path along the nerve ring, appears similar in both conditions.
All animals carry the AWC-specific odr-1p::rfp fluorescent marker. We found no clear difference in morphology of the AWC neurons between mutants and controls. In both cases, the neuron bodies (dashed circles) are clearly formed and extend their projections towards the sensory tip of the nose (dashed arrows) and the nerve ring (solid arrows). Outline of animals is shown as a dotted white line. Scale bar = 20 μm. Both animals are L4 larvae.
We additionally tested whether despite normal overall morphological appearance, AWC function might be affected in ceh-60 mutants. The AWC neurons are best-characterized as sensors of volatile odors such as butanone [13]. Under normal circumstances, C. elegans is attracted to butanone. We found that the chemotaxis index (CI, Fig 3A), which here quantifies a population's attraction to butanone, does not differ between wild-type animals and either of the tested ceh-60 mutant alleles (Fig 3B), despite a proper response of the positive control (che-3 mutants, deficient in a known regulator of volatile odorant sensing [13]). This indicates that the odorant-sensing apparatus, which includes the AWC neurons, is functional in absence of CEH-60, and taken together with the morphological data (Fig 2), likely intact.
(A) The butanone sensing assay experimental setup. Animals are collected, washed to remove residual bacteria and released at the origin of a chemotaxis plate, to which 10% butanone (orange) and control (blue) spots have been applied (details: methods). After 1 hour, the chemotaxis index is calculated as CI = ([(nButanone)-(nEtOH)]/[(Total-nOrigin)], with n the number of animals at these positions. (B) Wild-type animals are able to sense butanone, as evidenced by their positive CI. che-3(e1124) animals have a documented defect in sensing butanone, and their (significantly lower) CI is used as a positive control for the assay. Neither ceh-60 mutant strain shows a significant difference in CI. **** p < 0.0001. NS, not significant. Each dot represents an assay of a biologically independent population, N = 6.
As tested by these assays, CEH-60 does not appear to affect AWC morphology or function. However, besides the here presented gene target studies and ceh-60’s expression in AWC, there are other reasons to believe that CEH-60 has a role to play in olfactory neurons. Previously, CEH-60 has been characterized as a modulator of fat mobilization through activation of vitellogenesis in the intestine [9, 10]. Olfactory sensing has often been linked to fat storage and metabolism, with a high-fat diet in mice causing decreased sense of smell and olfactory neuron activity [76], and human anorexia nervosa patients often experiencing increased olfactory sensing [77]. Recently, butanone sensing through AWC neurons in C. elegans was also found to influence fat storage and mobilization, likely signaling through the neuropeptide FLP-1 and its receptor NPR-4, the glucocorticoid-inducible kinase GSK-1 and DAF-16 in peripheral tissues [78]. Because CEH-60 does not appear to influence butanone sensing, we maintain that the method through which CEH-60 alters fat mobilization and storage, is through its documented regulation of vitellogenesis.
Future work may yet unveil CEH-60’s cryptic role or function in olfactory AWC neurons, further motivated by (1) the recent discovery that PBX1, coding for ceh-60’s homolog in vertebrates, acts as a terminal selector for olfactory bulb neuron differentiation in mice [79], and (2) the knowledge that in C. elegans, olfactory learning depends on the odor-sensing (ceh-60-expressing) AWC neurons but also involves signaling from DBL-1/TGF-β [80], which is here identified as a putative gene target for CEH-60 (S1 Table). DBL-1 could indeed be the signaling molecule linking several CEH-60-related phenotypes: DBL-1 regulates vitellogenesis [73], is needed for building of an impermeable epicuticle [81], functions as an olfactory learning cue in the neurons [73, 80, 81], and is a gene target of CEH-60 as identified by DamID (this study). So far however, the exact role of CEH-60 in AWC neurons remains undiscovered. Because simple morphological and functional studies here performed showed no distinct phenotype, it is possible that the function of CEH-60 in AWC neurons is subtle, redundant with other (transcription) factors, or unrelated to neuron morphology or easily quantifiable behavior.
3.5 CEH-60 is not essential for on-food pharyngeal pumping
The C. elegans hermaphrodite muscle system consists of body wall, pharyngeal, anal sphincter, anal depressor, vulval and uterine muscles and a contractile gonadal sheath [82]. Because many CEH-60 targets are involved in muscle contraction and are expressed in the pharyngeal muscle, and because ceh-60 itself is expressed in the PM6 pharyngeal muscle cells, we decided to investigate pharyngeal pumping in ceh-60 mutants by quantifying the pumping rate in the presence of food of adult animals. While eat-2 mutants, which have a well-documented defect in pharyngeal pumping rate [38] indeed show a decreased average pumping rate compared to wild types, ceh-60 mutants do not (Fig 4A). Additionally, we quantified isthmus peristalsis, a contraction of pharynx muscles that carries food from the corpus of the pharynx to the terminal bulb, using fluorescent E. coli OP50 bacteria. We observed no difference between wild type and ceh-60 mutant animals (Fig 4B). Thus, the function of ceh-60 in muscle tissue is still to be determined.
(A) Wild-type animals and ceh-60 mutant animals have similar pumping rates on food, while eat-2 mutants pump significantly slower. Each dot represents one animal (N), imaged thrice for 15 seconds. Pumping rates were averaged per animal. **** p < 0.0001. NS = not significant. N = 6. (B) Wild-type animals and ceh-60 mutants have similar isthmus peristalsis or “gulp” rates. NS = not significant. N ≥ 6.
DamID results (S1 and S2 Tables), ceh-60 expression patterns [10] and Pbx functions in vertebrate muscle development [20–24] all suggest a role for CEH-60 in muscle structure, specifically in the pharynx, yet both pharyngeal contractions measured were not measurably affected in ceh-60 mutants (Fig 4). CEH-60 function in the pharynx may not be apparent when studying on-food behavior, and may require more demanding conditions such as the absence of food, which normally leads to serotonin-dependent enhanced pumping [83].
While our dataset of CEH-60 targets also contains several genes known to be expressed in body wall muscle, CEH-60 itself is absent from this tissue [10, 35]. Because CEH-60 can also act as a repressor of transcription [9], these data could indicate that in the pharynx, CEH-60 actively represses transcription of certain body wall muscle genes. It would be interesting to explore this further in future research.
CEH-60’s function in the pharynx may not be related to muscle structure directly. Recently, it has been shown that the PM6 pharyngeal muscle cells, in which ceh-60 is abundantly expressed and which surround the pharynx grinder, transdifferentiate into secretory cells during lethargus, possibly aiding in the construction of a new grinder [84]. The grinder is an extracellular matrix, constructed during each larval transition phase, coinciding with earlier-reported cyclic expression of ceh-60 [11, 85]. CEH-60 may act in PM6 cells to aid in this cyclic transdifferentiation from muscle cells to secretory cells, or in re-establishment of muscle nature once the extracellular matrix of the grinder is built.
4. Conclusions
Expression patterns and functional information from homologs and paralogs all point towards the possibility for new roles of the transcription factor CEH-60/PBX in C. elegans. Through DamID, we identified gene targets of CEH-60 and hypothesized on its function in neuron development and muscle structure. Specifically, we tested morphology and function of sensory AWC neurons and pharyngeal muscle. While we did not find evidence for CEH-60-related function in the observed assays, our phenotypic analysis was limited to simple behavioral assays and morphological characterization of neurons. This means that roles for CEH-60 in neurons and muscle are still possible, although they may be subtle or hard to uncover because of genetic redundancy. In addition, we compared our DamID results to available ChIP-seq data and conclude that while there is a common core of genes identified, both techniques also identify unique targets.
Supporting information
S1 Fig. Spearman’s rank-order correlation coefficients for dam::ceh-60 and gfp::dam samples.
In both young adults (A) and L2 larvae (B), high correlation values are observed within dam::ceh-60 samples (≥0.67 in YA, ≥0.65 in L2). Correlation between gfp::dam and dam::ceh-60 is also high (≥0.52 in YA, ≥0.51 in L2).
https://doi.org/10.1371/journal.pone.0242939.s001
(DOCX)
S1 Table. 872 candidate gene targets of CEH-60 identified in young adult animals through DamID.
Start and stop represent the genomic positions on the specified chromosome (chr) of the specified open reading frame. Candidate gene targets are sorted by genomic position (i.e. chromosome number and start position). Log2FC is calculated as the log2 value of the ratio of average number of dam::ceh-60 reads over average number of gfp::dam reads.
https://doi.org/10.1371/journal.pone.0242939.s002
(DOCX)
S2 Table. 587 candidate gene targets of CEH-60 identified in L2 animals through DamID.
Start and stop represent the genomic positions on the specified chromosome of the specified open reading frame. Candidate gene targets are sorted by genomic position (i.e. chromosome number and start position). Log2FC is calculated as the log2 value of the ratio of average number of dam::ceh-60 reads over average number of gfp::dam reads.
https://doi.org/10.1371/journal.pone.0242939.s003
(DOCX)
S3 Table. Complete gene ontology analysis of 872 DamID gene targets in young adult animals.
Complete list of biological processes overrepresented in CEH-60 DamID targets. Observed and expected columns represent the number of genes associated with the specified ontology term and the expected number in a random collection of genes for the C. elegans genome. Fold enrichment represents the number of times the specified process is overrepresented in the dataset. FDR p value represents the false discovery rate corrected p value for the specified biological process. GO terms were sorted by FDR p value. Gene ontology analysis was carried out with PANTHER 15 using a statistical overrepresentation test for biological processes (complete).
https://doi.org/10.1371/journal.pone.0242939.s004
(DOCX)
Acknowledgments
OP50-dsRed E. coli bacteria were a gift from Dr. Andre Brown (Imperial College London, UK). Some strains were provided by the Caenorhabditis genetics center (University of Minnesota, US).
References
- 1. Longobardi E, Penkov D, Mateos D, Florian G De, Torres M, Blasi F. Biochemistry of the tale transcription factors PREP, MEIS, and PBX in vertebrates. Dev Dyn. 2014; 59–75. pmid:23873833
- 2. Liu J, Fire A. Overlapping roles of two Hox genes and the exd ortholog ceh-20 in diversification of the C. elegans postembryonic mesoderm. Development. 2000;127: 5179–5190. pmid:11060243
- 3. Liu H, Strauss TJ, Potts MB, Cameron S. Direct regulation of egl-1 of programmed cell death by the Hox protein MAB-5 and CEH-20, a C. elegans homolog of Pbx1. Development. 2006;133: 641–650. pmid:16421192
- 4. Van Auken K, Weaver D, Robertson B, Sundaram M, Saldi T, Edgar L, et al. Roles of the homeothorax/Meis/Prep homolog UNC-62 and the Exd/Pbx homologs CEH-20 and CEH-40 in C. elegans embryogenesis. Development. 2002;129: 5255–5268. pmid:12399316
- 5. Jiang Y, Shi H, Liu J. Two Hox cofactors, the Meis/Hth homolog UNC-62 and the Pbx/Exd homolog CEH-20, function together during C. elegans postembryonic mesodermal development. Dev Biol. 2009;334: 535–546. pmid:19643105
- 6. Jiang Y, Shi H, Amin NM, Sultan I, Liu J. Mesodermal expression of the C. elegans HMX homolog mls-2 requires the PBC homolog CEH-20. Mech Dev. 2008;125: 451–461. pmid:18316179
- 7. Hughes S, Brabin C, Appleford PJ, Woollard A. CEH-20/Pbx and UNC-62/Meis function upstream of rnt-1/Runx to regulate asymmetric divisions of the C. elegans stem-like seam cells. Biol Open. 2013;2: 718–727. pmid:23862020
- 8. Zheng C, Jin FQ, Chalfie M. Hox proteins act as transcriptional guarantors to ensure terminal differentiation. Cell Rep. 2015;13: 1343–1352. pmid:26547238
- 9. Dowen RH. CEH-60/PBX and UNC-62/MEIS coordinate a metabolic switch that supports reproduction in C. elegans. Dev Cell. 2019; 1–16. pmid:30956009
- 10. Van de Walle P, Geens E, Baggerman G, Naranjo-Galindo FJ, Askjaer P, Schoofs L, et al. CEH-60/PBX regulates vitellogenesis and cuticle permeability through intestinal interaction with UNC-62/MEIS in Caenorhabditis elegans. PLoS Biol. 2019;17: 1–28. pmid:31675356
- 11. Van Rompay L, Borghgraef C, Beets I, Caers J, Temmerman L. New genetic regulators question relevance of abundant yolk protein production in C. elegans. Sci Rep. 2015; 1–16. pmid:26553710
- 12.
Altun ZF, Hall DH. Nervous system, general description. In: WormAtlas. 2011.
- 13. Bargmann CI, Hartwieg E, Horvitz HR. Odorant-selective genes and neurons mediate olfaction in C. elegans. Cell. 1993;74: 515–527. pmid:8348618
- 14. Beverly M, Anbil S, Sengupta P. Degeneracy and neuromodulation among thermosensory neurons contribute to robust thermosensory behaviors in Caenorhabditis elegans. J Neurosci. 2011;31: 11718–11727. pmid:21832201
- 15. Gabel C V., Gabel H, Pavlichin D, Kao A, Clark DA, Samuel ADT. Neural circuits mediate electrosensory behavior in Caenorhabditis elegans. J Neurosci. 2007;27: 7586–7596. pmid:17626220
- 16. Song B, Avery L. The pharynx of the nematode C. elegans: A model system for the study of motor control. Worm. 2013;2: e21833. pmid:24058858
- 17. Waskiewicz AJ, Rikhof H a, Moens CB. Eliminating zebrafish Pbx proteins reveals a aindbrain ground state. Dev Cell. 2002;3: 723–733. pmid:12431378
- 18. Erickson T, Pillay LM, Waskiewicz AJ. Zebrafish Tshz3b negatively regulates hox function in the developing hindbrain. Genesis. 2011;49: 725–742. pmid:21714061
- 19. Zheng C, Diaz-Cuadros M, Chalfie M. Hox genes promote neuronal subtype diversification through posterior induction in Caenorhabditis elegans. Neuron. 2015;88: 514–527. pmid:26539892
- 20. Berkes CA, Bergstrom DA, Penn BH, Seaver KJ, Knoepfler PS, Tapscott SJ. Pbx marks genes for activation by MyoD indicating a role for a homeodomain protein in establishing myogenic potential. Mol Cell. 2004;14: 465–477. pmid:15149596
- 21. Maves L, Waskiewicz AJ, Paul B, Cao Y, Tyler A, Moens CB, et al. Pbx homeodomain proteins direct Myod activity to promote fast-muscle differentiation. Development. 2007;134: 3371–3382. pmid:17699609
- 22. Bryantsev AL, Duong S, Brunetti TM, Chechenova MB, Lovato TAL, Nelson C, et al. Extradenticle and Homothorax control adult muscle fiber identity in Drosophila. Dev Cell. 2012;23: 664–673. pmid:22975331
- 23. Yao Z, Farr GH, Tapscott SJ, Maves L. Pbx and Prdm1a transcription factors differentially regulate subsets of the fast skeletal muscle program in zebrafish. Biol Open. 2013;2: 546–555. pmid:23789105
- 24. Cho OH, Mallappa C, Hernández-Hernández JM, Rivera-Pérez JA, Imbalzano AN. Contrasting roles for MyoD in organizing myogenic promoter structures during embryonic skeletal muscle development. Dev Dyn. 2015;244: 43–55. pmid:25329411
- 25. Steensel B Van, Henikoff S. Identification of in vivo DNA targets of chromatin proteins using tethered Dam methyltransferase. Nat Biotechnol. 2000;18. pmid:10748524
- 26. Vogel MJ, Peric-Hupkes D, van Steensel B. Detection of in vivo protein—DNA interactions using DamID in mammalian cells. Nat Protoc. 2007;2: 1467–1478. pmid:17545983
- 27. Schuster E, McElwee JJ, Tullet JMA, Doonan R, Matthijssens F, Reece-Hoyes JS, et al. DamID in C. elegans reveals longevity-associated targets of DAF-16/FoxO. Mol Syst Biol. 2010;6: 1–6. pmid:20706209
- 28. La Fortezza M, Grigolon G, Cosolo A, Pinduyrin A, Breimann L, Blum H, et al. DamID profiling of dynamic Polycomb-binding sites in Drosophila imaginal disc development and tumorigenesis. Epigenetics and Chromatin. 2018;11: 1–17. pmid:29310712
- 29. Vissers JHA, Froldi F, Schröder J, Papenfuss AT, Cheng LY, Harvey KF. The Scalloped and Nerfin-1 transcription factors cooperate to maintain neuronal cell fate. Cell Rep. 2018;25: 1561–1576.e7. pmid:30404010
- 30. Tosti L, Ashmore J, Tan BSN, Carbone B, Mistri TK, Wilson V, et al. Mapping transcription factor occupancy using minimal numbers of cells in vitro and in vivo. Genome Res. 2018;28: 592–605. pmid:29572359
- 31. Lewis JA, Fleming JT. Basic Culture Methods. Methods in Cell Biology. 1995. pp. 3–29. pmid:8531730
- 32. Muñoz-Jiménez C, Ayuso C, Dobrzynska A, Torres A, Askjaer P. An efficient FLP-based toolkit for spatiotemporal control of gene expression in Caenorhabditis elegans. Genetics. 2017;206: 1–16. pmid:28476859
- 33.
Gómez-Salvidar G, Meister P, Askjaer P. DamID analysis of nuclear organization in Caenorhabditis elegans. In: Shackleton S, editor. The nuclear envelope: methods and protocols. Springer; 2016. pp. 341–358. https://doi.org/10.1007/978-1-4939-3530-7
- 34. Sharma R, Ritler D, Meister P. Tools for DNA adenine methyltransferase identification analysis of nuclear organization during C. elegans development. Genesis. 2016;54: 151–159. pmid:26845390
- 35. Cao J, Packer JS, Ramani V, Cusanovich DA, Huynh C, Daza R, et al. Comprehensive single-cell transcriptional profiling of a multicellular organism. Science (80-). 2017;357: 661–667. pmid:28818938
- 36. Schindelin J, Arganda-carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9. pmid:22743772
- 37. Kauffman A, Parsons L, Stein G, Wills A, Kaletsky R, Murphy C. C. elegans positive butanone learning, short-term, and long-term associative memory assays. J Vis Exp. 2011; e2490. pmid:21445035
- 38. Mosbech MB, Kruse R, Harvald EB, Olsen ASB, Gallego SF, Hannibal-Bach HK, et al. Functional loss of two ceramide synthases elicits autophagy-dependent lifespan extension in C. elegans. PLoS One. 2013;8. pmid:23894595
- 39. Van Sinay E, Mirabeau O, Depuydt G, Van Hiel MB, Peymen K, Watteyne J, et al. Evolutionarily conserved TRH neuropeptide pathway regulates growth in Caenorhabditis elegans. Proc Natl Acad Sci. 2017;114: E4065–E4074. pmid:28461507
- 40. Sha K, Gu SG, Pantalena-Filho LC, Goh A, Fleenor J, Blanchard D, et al. Distributed probing of chromatin structure in vivo reveals pervasive chromatin accessibility for expressed and non-expressed genes during tissue differentiation in C. elegans. BMC Genomics. 2010;11: 1–15. pmid:20044946
- 41. Sen SQ, Chanchani S, Southall TD, Doe CQ. Neuroblast-specific open chromatin allows the temporal transcription factor, Hunchback, to bind neuroblast-specific loci. Elife. 2019;8: 1–26. pmid:30694180
- 42. Bloom L, Horvitz HR. The Caenorhabditis elegans gene unc-76 and its human homologs define a new gene family involved in axonal outgrowth and fasciculation. Proc Natl Acad Sci U S A. 1997;94: 3414–3419. pmid:9096408
- 43. Clark SG, Chiu C. C. elegans ZAG-1, a Zn-finger-homeodomain protein, regulates axonal development and neuronal differentiation. Development. 2003;130: 3781–3794. pmid:12835394
- 44. Nash B, Colavita A, Zheng H, Roy PJ, Culotti JG. The forkhead transcription factor UNC-130 is required for the graded spatial expression of the UNC-129 TGF-β guidance factor in C. elegans. Genes Dev. 2000;14: 2486–2500. pmid:11018016
- 45. Crump JG, Zhen M, Jin Y, Bargmann CI. The SAD-1 kinase regulates presynaptic vesicle clustering and axon termination. Neuron. 2001;29: 115–129. pmid:11182085
- 46. Wang X, Kweon J, Larson S, Chen L. A role for the C. elegans L1CAM homologue lad-1/sax-7 in maintaining tissue attachment. Dev Biol. 2005;284: 273–291. pmid:16023097
- 47. Jia L, Emmons SW. Genes that control ray sensory neuron axon development in the Caenorhabditis elegans male. Genetics. 2006;173: 1241–1258. pmid:16624900
- 48. Katidou M, Tavernarakis N, Karagogeos D. The contactin RIG-6 mediates neuronal and non-neuronal cell migration in Caenorhabditis elegans. Dev Biol. 2013;373: 184–195. pmid:23123963
- 49. Shen K, Fetter RD, Bargmann CI. Synaptic specificity is generated by the synaptic guidepost protein SYG-2 and its receptor, SYG-1. Cell. 2004;116: 869–881. pmid:15035988
- 50. Hristova M, Birse D, Hong Y, Ambros V. The Caenorhabditis elegans heterochronic regulator LIN-14 is a novel transcription factor that controls the developmental timing of transcription from the insulin/insulin-like growth factor gene ins-33 by direct DNA binding. Mol Cell Biol. 2005;25: 11059–11072. pmid:16314527
- 51. Cassata G, Kagoshima H, Andachi Y, Kohara Y, Dürrenberger MB, Hall DH, et al. The LIM homeobox gene ceh-14 confers thermosensory function to the AFD neurons in Caenorhabditis elegans. Neuron. 2000;25: 587–597. pmid:10774727
- 52. Wolf FW, Hung MS, Wightman B, Way J, Garriga G. vab-8 is a key regulator of posteriorly directed migrations in C. elegans and encodes a novel protein with kinesin motor similarity. Neuron. 1998;20: 655–666. pmid:9581759
- 53. Mcardle K, Allen TS, Bucher EA. Ca2+-dependent muscle dysfunction caused by mutation of the Caenorhabditis elegans troponin T-1 gene. J Cell Biol. 1998;143: 1201–1213. pmid:9832549
- 54. Rushforth AM, White CC, Anderson P. Functions of the Caenorhabditis elegans regulatory myosin light chain genes mlc-1 and mlc-2. Genetics. 1998;150: 1067–1077. pmid:9799259
- 55. Miller DM, Stockdale FE, Karn J. Immunological identification of the genes encoding the four myosin heavy chain isoforms of Caenorhabditis elegans. Proc Natl Acad Sci U S A. 1986;83: 2305–2309. pmid:2422655
- 56. Moerman DG, Plurad S, Waterston RH, Baillie DL. Mutations in the unc-54 myosin heavy chain gene of Caenorhabditis elegans that alter contractility but not muscle structure. Cell. 1982;29: 773–781. pmid:7151169
- 57. Benian GM, Kiff JE, Neckelmann N, Moerman DG, Waterston RH. Sequence of an unusually large protein implicated in regulation of myosin activity in C. elegans. Nature. 1989;342: 45–50. pmid:2812002
- 58. Benian GM, Tinley TL, Tang X, Borodovsky M. The Caenorhabditis elegans gene unc-89, required for muscle M-line assembly, encodes a giant modular protein composed of Ig and signal transduction domains. Encycl Mov Disord. 2010;132: 171–173.
- 59. Spooner PM, Bonner J, Maricq A V., Benian GM, Norman KR. Large isoforms of UNC-89 (obscurin) are required for muscle cell architecture and optimal calcium release in Caenorhabditis elegans. PLoS One. 2012;7. pmid:22768340
- 60. Kagawa H, Gengyo K, Mclachlan AD, Brenner S, Karn J. Paramyosin gene (unc-15) of Caenorhabditis elegans. Molecular cloning, nucleotide sequence and models for thick filament structure. J Mol Biol. 1989;207: 311–333. pmid:2754728
- 61. Trojanowski NF, Raizen DM, Fang-Yen C. Pharyngeal pumping in Caenorhabditis elegans depends on tonic and phasic signaling from the nervous system. Sci Rep. 2016;6: 1–10. pmid:28442746
- 62. Small TM, Gernert KM, Flaherty DB, Mercer KB, Borodovsky M, Benian GM. Three new isoforms of Caenorhabditis elegans UNC-89 containing MLCK-like protein kinase domains. J Mol Biol. 2004;342: 91–108. pmid:15313609
- 63. Matsunaga Y, Hwang H, Franke B, Williams R, Penley M, Qadota H, et al. Twitchin kinase inhibits muscle activity. Mol Biol Cell. 2017;28: 1591–1600. pmid:28428253
- 64. Ardizzi JR, Epstein HF. Immunochemical localization of myosin heavy chain isoforms and paramyosin in developmentally and structurally diverse muscle cell types of the nematode Caenorhabditis elegans. J Cell Biol. 1987;105: 2763–2770. pmid:3320053
- 65. Matsuo K, Koga A, Ihara S. Visualization of endogenous NID-1 and EMB-9 in C. elegans. microPublication Biol. 2019. pmid:32550405
- 66. Gieseler K, Mariol MC, Bessou C, Migaud M, Franks CJ, Holden-Dye L, et al. Molecular, genetic and physiological characterisation of dystrobrevin-like (dyb-1) mutants of Caenorhabditis elegans. J Mol Biol. 2001;307: 107–117. pmid:11243807
- 67. McKeown CR, Han HF, Beckerle MC. Molecular characterization of the Caenorhabditis elegans ALP/Enigma gene alp-1. Dev Dyn. 2006;235: 530–538. pmid:16278882
- 68. Mercer KB, Miller RK, Tinley TL, Sheth S, Qadota H, Benian GM. Caenorhabditis elegans UNC-96 is a new component of M-lines that interacts with UNC-98 and paramyosin and is required in adult muscle for assembly and/or maintenance of thick filaments. Mol Biol Cell. 2007;18: 976–985. pmid:17202407
- 69. Ruksana R, Kuroda K, Terami H, Bando T, Kitaoka S, Takaya T, et al. Tissue expression of four troponin I genes and their molecular interactions with two troponin C isoforms in Caenorhabditis elegans. Genes to Cells. 2005;10: 261–276. pmid:15743415
- 70. Guan G, Wong M-K, Sze Ho VW, An X, Chan L-Y, Tian B, et al. System-level quantification and phenotyping of early embryonic morphogenesis of Caenorhabditis elegans. bioRxiv. 2019.
- 71. Cheetham SW, Gruhn WH, van den Ameele J, Krautz R, Southall TD, Kobayashi T, et al. Targeted DamID reveals differential binding of mammalian pluripotency factors. Development. 2018;145. pmid:30185410
- 72. van Nostrand EL, Sánchez-Blanco A, Wu B, Nguyen A, Kim SK. Roles of the developmental regulator unc-62/Homothorax in limiting longevity in Caenorhabditis elegans. PLoS Genet. 2013;9. pmid:23468654
- 73. Goszczynski B, Captan V V., Danielson AM, Lancaster BR, McGhee JD. A 44 bp intestine-specific hermaphrodite-specific enhancer from the C. elegans vit-2 vitellogenin gene is directly regulated by ELT-2, MAB-3, FKH-9 and DAF-16 and indirectly regulated by the germline, by daf-2/insulin signaling and by the TGF-β/Sma. Dev Biol. 2016;413: 112–127. pmid:26963674
- 74. Orian A, Steensel B Van, Delrow J, Bussemaker HJ, Li L, Sawado T, et al. Genomic binding by the Drosophila Myc, Max, Mad/Mnt transcription factor network. Genes Dev. 2003;17: 1101–1114. pmid:12695332
- 75. Gutierrez-Triana JA, Mateo JL, Ibberson D, Ryu S, Wittbrodt J. iDamIDseq and iDEAR: An improved method and computational pipeline to profile chromatin-binding proteins. Development. 2016;143: 4272–4278. pmid:27707796
- 76. Thiebaud N, Johnson MC, Butler JL, Bell G a., Ferguson KL, Fadool AR, et al. Hyperlipidemic diet causes loss of olfactory sensory neurons, reduces olfactory discrimination, and disrupts odor-reversal learning. J Neurosci. 2014;34: 6970–6984. pmid:24828650
- 77. Fernández-Aranda F, Agüera Z, Fernández-García JC, Garrido-Sanchez L, Alcaide-Torres J, Tinahones FJ, et al. Smell–taste dysfunctions in extreme weight/eating conditions: analysis of hormonal and psychological interactions. Endocrine. 2016;51: 256–267. 9 pmid:26198367
- 78. Mutlu AS, Gao SM, Zhang H, Wang MC. Olfactory specificity regulates lipid metabolism through neuroendocrine signaling in Caenorhabditis elegans. Nat Commun. 2020;11: 1450. pmid:32193370
- 79. Remesal L, Roger-Baynat I, Chirivella L, Maicas M, Brocal-Ruiz R, Pérez-Villalva A, et al. PBX1 acts as terminal selector for olfactory bulb dopaminergic neurons. Development. 2020. pmid:32156753
- 80. Zhang X, Zhang Y. DBL-1, a TGF-β, is essential for Caenorhabditis elegans aversive olfactory learning. Proc Natl Acad Sci U S A. 2012;109: 17081–17086. pmid:23019581
- 81. Schultz RD, Bennett EE, Ellis EA, Gumienny TL. Regulation of extracellular matrix organization by BMP signaling in Caenorhabditis elegans. PLoS One. 2014;9. pmid:25013968
- 82. Altun ZF, Hall DH. Muscle system, introduction. WormAtlas. 2009.
- 83. Horvitz HR, Chalfie M, Trent C, Sulston JE, Peter D. Serotonin and octopamine in the nematode Caenorhabditis elegans. Science (80-). 1982;216: 1012–1014. pmid:6805073
- 84. Sparacio AP, Trojanowski NF, Snetselaar K, Nelson MD, Raizen DM. Teething during sleep: Ultrastructural analysis of pharyngeal muscle and cuticular grinder during the molt in Caenorhabditis elegans. PLoS One. 2020;15: 1–24. pmid:32433687
- 85. Hendriks G, Gaidatzis D, Aeschimann F, Großhans H. Extensive Oscillatory Gene Expression during C. elegans Larval Development. Mol Cell. 2014;53: 380–392. pmid:24440504