(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



The anterior Hox gene ceh-13 and elt-1/GATA activate the posterior Hox genes nob-1 and php-3 to specify posterior lineages in the C. elegans embryo

['John Isaac Murray', 'Department Of Genetics', 'Perelman School Of Medicine', 'University Of Pennsylvania', 'Philadelphia', 'Pennsylvania', 'United States Of America', 'Elicia Preston', 'Jeremy P. Crawford', 'Division Of Developmental Biology']

Date: 2022-06

Hox transcription factors play a conserved role in specifying positional identity during animal development, with posterior Hox genes typically repressing the expression of more anterior Hox genes. Here, we dissect the regulation of the posterior Hox genes nob-1 and php-3 in the nematode C. elegans. We show that nob-1 and php-3 are co-expressed in gastrulation-stage embryos in cells that previously expressed the anterior Hox gene ceh-13. This expression is controlled by several partially redundant transcriptional enhancers. These enhancers act in a ceh-13-dependant manner, providing a striking example of an anterior Hox gene positively regulating a posterior Hox gene. Several other regulators also act positively through nob-1/php-3 enhancers, including elt-1/GATA, ceh-20/ceh-40/Pbx, unc-62/Meis, pop-1/TCF, ceh-36/Otx, and unc-30/Pitx. We identified defects in both cell position and cell division patterns in ceh-13 and nob-1;php-3 mutants, suggesting that these factors regulate lineage identity in addition to positional identity. Together, our results highlight the complexity and flexibility of Hox gene regulation and function and the ability of developmental transcription factors to regulate different targets in different stages of development.

Hox genes are critical for head-to-tail patterning during embryonic development in all animals. Here we examine the factors that are necessary to turn on two posterior Hox genes, nob-1 and php-3, in the nematode worm, C. elegans. We identified six new transcription factors and three enhancer regions of DNA that can activate expression of nob-1/php-3. Unexpectedly, these activating transcription factor genes included ceh-13, an anterior Hox gene, and elt-1, a regulator of skin development that is briefly expressed in many cells that do not adopt skin fates, including the cells that express nob-1. Furthermore, the cellular defects we observed in ceh-13 and nob-1;php-3 mutant embryos indicate that the early embryonic functions of these Hox genes help determine the identity of cells as well as their position within the embryo. Our findings identify new roles for Hox genes in C. elegans and emphasize the ability of transcription factors to contribute to the diversification of cell types and the adoption of specific cell types at different phases of embryonic development.

Funding: This work was supported by the National Institute of General Medical Sciences of the National Instiutes of Health through grants R35GM127093 (to J.I.M.), R00GM111825 (to A.L.Z.), and F31GM123737 (to J.D.R.). https://www.nigms.nih.gov/ This work was also supported by the Cincinnati Children's Research Foundation grant SPR202247 to A.L.Z. https://www.cincinnatichildrens.org/research/cincinnati The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability: The vast majority of relevant data are within the manuscript and its Supporting Information files. To provide full data availability, all Cell Data (CD) files (containing reporter levels and nuclei positions for all cells at all timepoints) are available from the Dryad database: (URL: https://datadryad.org/stash/dataset/doi:10.5061/dryad.zs7h44j86 ; DOI: 10.5061/dryad.zs7h44j86 ).

Here we analyzed the cis-regulatory control of nob-1/php-3 expression in the early embryo. We find that nob-1/php-3 expression is regulated by several partially redundant distal enhancers, including at least three that drive overlapping patterns during gastrulation. We find that ceh-13 is required for normal timing and levels of nob-1/php-3 expression, by activating at least two of its enhancers. We further identified the GATA family transcription factor gene elt-1, previously known as a specifier of epidermal fate, as one of several additional positive regulators of nob-1/php-3 expression. Detailed analysis of cell positions and cell division timing identifies both position defects and defects in cell division patterns and timing in ceh-13 and nob-1/php-3 mutants. This role in cell identity combined with the positive regulation of a posterior Hox gene by an anterior Hox gene suggest novel roles for Hox genes in early lineage specification.

Intriguingly, despite their apparently opposite roles in anterior vs posterior morphogenesis, both ceh-13 and nob-1 are expressed in overlapping posterior lineages during gastrulation and both require the Wnt pathway for this expression [ 24 , 25 ]. ceh-13 is transiently expressed in the progeny of 7 of the 8 posterior sister cells derived from the (largely ectoderm-producing) AB blastomere at the 24-cell stage [ 21 ]. nob-1 is expressed 1–2 cell cycles later, in the posterior daughters or grand-daughters of four of these ceh-13-expressing cells [ 23 ]. In addition, both ceh-13 and nob-1 are expressed in the posterior daughter of the intestinal blastomere E (“Ep”). This raises the question of whether ceh-13 regulates nob-1 in these lineages. Expression of both factors at later stages is regulated by feedback mechanisms; early ceh-13 activity is required for later ceh-13 expression [ 25 ], and early nob-1 negatively regulates later nob-1 expression through the microRNA mir-57 [ 23 ].

Three C. elegans Hox genes, the anterior Hox gene homolog ceh-13/HOX1 and the posterior Hox gene homologs nob-1/HOX9-13 and php-3/HOX9-13, are also expressed in early embryogenesis, during gastrulation [ 5 , 21 – 23 ]. ceh-13 mutants have severe defects in anterior (head) morphology [ 22 ], while mutants lacking both nob-1 and php-3 have severe posterior (tail) defects [ 5 ]. Partial cell lineage tracing of ceh-13 and nob-1;php-3 mutants identified defects in cell position but not in division patterns, leading to the hypothesis that these genes regulate positional identity, rather than lineage identity [ 5 , 6 , 22 ].

The genome of the nematode Caenorhabditis elegans encodes a single set of six Hox genes on Chromosome III, loosely organized into three degenerate “clusters” that each contain two adjacent genes [ 3 – 6 ]. The cluster containing lin-39 and ceh-13 is inverted relative to Hox clusters in other organisms as ceh-13, a homolog of HoxA1/labial, is downstream of lin-39, a homolog of HoxA4/Deformed [ 7 – 9 ]. In the most distal cluster, php-3 lies 220 bp directly downstream of the 3’UTR of nob-1 and while large deletions in either gene are viable, disrupting both loci or the likely cis-regulatory regions upstream of nob-1 results in embryonic lethality, suggesting these genes are redundant and co-regulated during embryonic development [ 5 , 10 ]. In larval stages, these genes are expressed in specific positions along the A-P axis and regulate both positional differences in cell fate and function, similar to their homologs in other animals [ 11 – 19 ], and directly regulate terminal fates [ 10 , 20 ].

Hox genes encode conserved transcription factors famously expressed in specific positions along the anterior-posterior axis during animal development that specify axial position. Mutations in Hox genes cause a wide variety of developmental defects in both model organisms and humans. Hox gene regulation is complex and includes both transcriptional and post-transcriptional control. In most animals, Hox genes are found in genomic clusters, and their expression along the A-P axis is collinear with their genomic position, and often show “posterior dominance,” where posterior Hox genes repress the expression of more anterior Hox genes (Reviewed in [ 1 – 3 ]).

Results

The anterior Hox gene ceh-13 and the posterior Hox gene nob-1 are expressed sequentially in gastrulating embryos To better understand the expression dynamics of the ceh-13, nob-1, and php-3 Hox genes during embryogenesis (Fig 1), we collected 3D time-lapse movies (~1.5 minute temporal resolution) of embryos expressing either ceh-13, nob-1 or php-3 modified at the endogenous locus by CRISPR/Cas9 genome editing to tag the encoded proteins with green fluorescent protein (GFP) at the C-termini. The same embryos also expressed a ubiquitously expressed mCherry-tagged histone transgene to allow for cell lineage tracing. We traced cells from soon after fertilization (4–8 cell stage) through the last round of cell divisions for most cells (bean stage) by using StarryNite automated cell tracking software, and quantified reporter expression in each cell across time [26–29]. Note, each endogenously tagged Hox::GFP fusion line is homozygous viable, fertile, and displays no obvious phenotypes, demonstrating that the fusion proteins largely function like the wild-type proteins. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. ceh-13 and nob-1/php-3 are expressed broadly in an overlapping pattern in the early C. elegans embryo. A) Time-lapse images of transgenic C. elegans embryos carrying two transgenes, a ubiquitous fluorescent histone to mark all nuclei (shown in green) and a reporter of interest (shown in red), which can be a cis-regulatory element driving a fluorescent histone (transcriptional) or a fluorescently tagged transcription factor protein (translational). Image analysis software identifies nuclei, and quantifies reporter intensity within the nuclei, which can be displayed as a lineage tree colored by expression as in (C). B) 3D projections of nuclei expressing endogenously tagged CEH-13::GFP (red) and endogenously tagged NOB-1::GFP and PHP-3::GFP (blue), with overlap shown in magenta, at 120 minutes (50 cell stage), 150 minutes (100 cell stage), and 280 minutes (400 cell stage) post fertilization. Endogenously tagged NOB-1::GFP and PHP-3::GFP were expressed in indistinguishable patterns (S1 Fig), but PHP-3::GFP was slightly brighter so is shown in the following panels. C) Lineage tree through the 100-cell stage, showing early expression of endogenously tagged CEH-13::GFP and PHP-3::GFP, colored as in (B). D) Expressing lineages showing endogenously tagged CEH-13::GFP and/or PHP-3::GFP expression to 350 minutes of development, colors as in (B). Note that CEH-13::GFP precedes PHP-3::GFP and NOB-1::GFP in all lineages except ABp(l/r)papppp (asterisk—expression is consistent in ABprpapppp and variable in ABplpapppp). (E-H) Quantitative detail for highlighted lineages, showing nuclear fluorescence intensity of CEH-13::GFP (fosmid) and NOB-1::GFP transgene reporters across developmental time for the cells leading to ABp(l/r)appaaa (E), ABp(l/r)appppp (F), ABp(l/r)ppppppp (G), and Ep(l/r)p (H). For cell labels, x = (l/r). Nuclear fluorescence intensity is in arbitrary units. Grey bars mark cell divisions. https://doi.org/10.1371/journal.pgen.1010187.g001 Consistent with previous studies using shorter reporters [21], endogenously tagged CEH-13::GFP expression is first seen in eight Wnt-signaled cells (that are the more-posterior daughter after cell division) and their progeny at the 26-cell stage, during early gastrulation (Figs 1B, 1C, and S2). These include the AB lineage-derived cells ABalap, ABalpp, ABarpp, ABplap, ABplpp, ABprap and ABprpp, and the posterior endoderm progenitor Ep. For simplicity, we use the term “lineage” to denote all cells derived from a specific blastomere; thus, the “Ep lineage” would refer to all descendants of Ep. The ABalap and ABalpp lineage expression was barely detectable and quickly faded, while the other lineages had more robust and persistent fluorescence, indicating the protein was degraded or maintained in distinct cells. By the 200–350 cell stage, many of the granddaughters of the initially expressing cells have lost CEH-13::GFP expression, while a few cells fated to become blast cells or neurons sustained or increased their expression. Also at this stage, some cells in the D and MS lineages as well as a few additional AB-derived cells show CEH-13::GFP accumulation (Fig 1D). We compared the tagged protein expression pattern to that of endogenous mRNA as measured in a lineage-resolved single cell RNA-seq dataset, and found that the mRNA and tagged protein expression patterns were consistent [30]. In comparison, a rescuing 35 kb CEH-13::GFP fosmid transgene had brighter expression in the cells expressing the endogenously tagged protein, plus additional weak expression detectable in some epidermal precursors from the C lineage (possibly detectable due to higher copy number). A shorter reporter containing only 8.2 kb of upstream sequence lacked expression in the MS lineage, emphasizing the role of distal regulatory elements in regulating C. elegans Hox gene expression (S1D Fig) [31,32]. Next, we defined the embryonic expression patterns of the posterior Hox genes php-3 and nob-1 tagged with GFP at the endogenous locus. As expected, nob-1 and php-3, which are adjacent in the genome with php-3 directly downstream of nob-1, are expressed in identical patterns in both our GFP imaging data (S1A Fig) and in the single-cell RNA-seq data [30]. Expression of both NOB-1::GFP and PHP-3::GFP begins during mid-gastrulation in cells derived from the cells ABplapp, ABplppp, ABprapp, ABprppp, Ep, and the four posterior great-granddaughters of the C blastomere. Expression becomes stronger one cell cycle later in the AB-derived lineages with the exception of that of ABp(l/r)appa. Additional late embryonic expression occurs in the ABp(l/r)papppp lineages starting at the onset of morphogenesis (“bean” stage). NOB-1-expressing lineages give rise to fates including neurons, hypodermis, seam cells, epithelial cells, intestine, and death, and have posterior positions clustered in and near the developing tail at comma stage [33]. Like CEH-13, early NOB-1::GFP expression persists through the terminal cell divisions in only a subset of cells (Fig 1D). A previously described NOB-1::GFP rescuing transgene that includes 9kb of sequence upstream of the nob-1 transcript expresses GFP in a similar pattern, except it is not expressed in the C great-granddaughter lineages, suggesting this expression requires additional regulatory sequences outside of this region (Figs 1D and S1) [23]. Early CEH-13::GFP expression occurs in cells that are distributed broadly along the anterior-posterior axis (Fig 1B), while endogenously tagged NOB-1::GFP and PHP-3::GFP expression is more limited to the posterior of the embryo. In the AB lineage, NOB-1/PHP-3 are expressed exclusively in cells that were the posterior sister after cell division and whose mother expressed CEH-13 (Fig 1D–1H). In addition, these three genes are co-expressed at similar onset times in the posterior intestine lineage derived from Ep. In contrast, in late embryos, high levels of CEH-13::GFP and NOB-1::GFP/PHP-3::GFP are largely mutually exclusive (Figs 1D, S1F, and S1G), consistent with classic models of Hox expression and posterior dominance.

Several overlapping lineage specific enhancers regulate nob-1 embryonic expression To identify regulators of nob-1 embryonic expression, we tested nearby genomic sequences for embryonic cis-regulatory activity (Fig 2A and 2B). Previous work showed that other C. elegans Hox genes are regulated by distal enhancers located as much as 20kb from a given gene’s promoter [25,31,34], but enhancers for nob-1 and php-3 have not been identified. We took advantage of existing reporters of different lengths to narrow the sequence search space for nob-1 embryonic enhancers (Fig 2C). A transcriptional reporter containing just 5.3 kb of upstream sequence driving histone-mCherry reporter expression, and the rescuing 9kb NOB-1::GFP reporter are expressed in most of the same lineages as endogenous NOB-1::GFP, indicating they contain regulatory sequences sufficient for expression in the full set of expressing lineages. However, while the protein fusion reporters are expressed at similar levels in each lineage, the shorter 5.3kb transcriptional reporter drives much lower fluorescence in the ABp(l/r)ppp lineages compared to other lineages (Fig 2C and 2D). This suggests that additional sequences between -5.3kb and -9kb are required for the full endogenous expression levels in the ABp(l/r)ppp lineage. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Regions upstream of nob-1/php-3 can recapitulate its expression pattern. A) Genome browser view of the nob-1/php-3 locus showing the genes (black), candidate enhancers tested (brown), sequence conservation with other nematodes (grey), and NHR-2 ChIP-seq trace (dark blue) from modENCODE [43] and ATAC-seq traces (red:ABa lineage, green: ABp lineage) from Charest et al, 2020 [66]. B) Schematics of the nob-1 reporter constructs examined, shown in alignment with (A). Enhancers were primarily tested in an orientation 5’ to the pes-10 minimal promoter, but a downstream orientation was also used for the -4.3 kb enhancer. C) Lineage trees colored to show expression patterns for the various reporters and tested enhancers with relevant reproducible activity, using a rainbow color scale to increase visible dynamic range. Major expressing lineages are underlined: ABp(l/r)ap: pink, ABp(l/r)ppp: cyan, Ep: black, Cpapp: orange, ectopic: purple. Note the changes in expression driven by the -4.3 kb enhancer depending on its position relative to the pes-10 minimal promoter. D) Average nuclear fluorescence values for cells in the ABp(l/r)ap (pink) and ABp(l/r)ppp (blue) lineages from at least four embryos, shown on a log scale (arbitrary units). E) Ratio of expression in average ABp(l/r)ap cell to the average ABp(l/r)ppp cell at the 350 cell stage for at least 4 embryos (n = 4–7). F) Lineage trees showing expression driven by the nob-1–5.3 kb upstream region reporter with the -3.4 kb and/or -4.3 kb enhancers are deleted. Lineages where expression is lost are underlined with colors as in (C). https://doi.org/10.1371/journal.pgen.1010187.g002 Although other Caenorhabditis species have large (12-20kb) intergenic regions upstream of nob-1, there is no detectable sequence conservation in the 5.3kb region upstream of nob-1, and only two short conserved stretches (85 and 275 nucleotides long) between 5.3 and 9kb upstream [35–38]. This indicates that any conserved regulatory elements in this region have diverged substantially at the sequence level during Caenorhabditis evolution, limiting utility of evolutionary conservation to identify enhancers. Chromatin marks typically used to identify transcriptional enhancers such as H3K27ac, H3K4me, or chromatin accessibility from whole embryos did not show strong peaks near nob-1 or near 20 additional genes expressed in lineage-specific patterns at the same stage as nob-1 [32,39–41]. We hypothesized that these chromatin signals are diminished in early embryonically expressed genes because the experiments used embryos of mixed stages where later stage nuclei dramatically outnumber nuclei from earlier stages. Therefore, we searched for other factors with published genomic binding patterns that could mark embryonic enhancers. One pattern stood out as preferentially bound near genes expressed with lineage-specific patterns in the early embryo: binding of a NHR-2::GFP fusion protein [42,43]. NHR-2::GFP binds in the intergenic sequences upstream of 19 of these 20 genes, and 16 genes have multiple clustered NHR-2 binding sites (vs. 22% and 8% of randomly chosen genes, respectively, p < 0.001; chi-squared test). NHR-2 is a nuclear hormone receptor distantly related to mammalian thyroid and PPAR receptors, and NHR-2::GFP is expressed in most or all somatic cells from the ~50-cell to ~200-cell stages. The functional importance of NHR-2 binding is unclear; nhr-2 RNAi causes embryonic and larval arrest, but partial deletion alleles are viable. However, regardless of its function, we hypothesized that these clustered NHR-2 binding sites might be useful proxies for accessible chromatin and could predict enhancer activity in the early embryo. We tested four NHR-2::GFP-bound regions upstream of nob-1 for enhancer activity by generating transgenic worms expressing histone-mCherry under the control of each candidate enhancer placed upstream of a pes-10 minimal promoter, which drives no consistent embryonic expression on its own (Fig 2A and 2B). We also tested four additional sequences for which we observed no NHR-2::GFP binding but which contained putative binding motifs for POP-1/TCF, which is required for expression of the nob-1 transcriptional reporter [24]. We identified embryonic cells expressing each enhancer reporter by confocal time-lapse imaging and StarryNite. Only three of the regions (located at -3.4 kb, -4.3 kb and -8.3 kb) tested showed enhancer activity that was consistent between embryos and also overlapped with the endogenous NOB-1::GFP expression pattern, suggesting they represent functional enhancers (Figs 2C and S2). All three functional enhancers were identified on the basis of NHR-2 binding. The enhancer located at -3.4 kb (hereafter referred to as the -3.4kb enhancer) recapitulates most of the nob-1 early embryonic expression pattern, with some differences in the C lineage, and drives much weaker expression in ABp(l/r)ppp than in ABp(l/r)app, similar to the nob-1–5.3 kb upstream transcriptional reporter. The enhancer located at -4.3 kb also drives expression in ABp(l/r)app and ABp(l/r)ppp at a very high level (Fig 2D) as well as variable misexpression in cells that do not normally express NOB-1 (Figs 2C and S2A). When this enhancer was cloned downstream of the reporter, it produced less ectopic expression and stronger expression in ABp(l/r)app relative to ABp(l/r)ppp, similar to the -3.4kb enhancer. This confirms the -4.3 kb region contains a bona fide enhancer capable of acting at a distance and suggests that the placement of this enhancer relative to the promoter influences its activity differently in different lineages and may be important for specificity. Embryos carrying multiple copies of the -4.3 kb reporter transgene occasionally displayed the “no backend” phenotype observed in nob-1/php-3 mutant embryos (S2B Fig) [5]. A third nob-1 enhancer (located at -8.3 kb) drives early embryonic expression only in ABp(l/r)ppp, as well as later expression in ABp(l/r)papppp. This most-distal enhancer is included in -9 kb NOB-1::GFP transgene but not the shorter -5.3kb transcriptional reporter, and includes the only substantial stretch of conserved sequence in the nob-1 promoter region. This region thus can explain why ABp(l/r)ppp lineage expression is relatively stronger in the translational reporter compared to the shorter transcriptional reporter. As the -5.3kb nob-1 transcriptional reporter includes both the -3.4kb and -4.3kb enhancers, we deleted each enhancer from this construct to test their necessity for nob-1 expression (Fig 2C). Deleting both enhancers led to a complete loss of early embryonic expression in both AB lineages, suggesting there are no additional enhancers present sufficient for early embryonic expression in these lineages. The deletion of the -3.4kb enhancer did not disrupt reporter fluorescence in the ABp(l/r)app and ABp(l/r)ppp lineages, and in the absence of the other enhancer located at -4.3 kb, this enhancer was not sufficient to drive reporter fluorescence in these lineages. This indicates that while the -3.4kb enhancer may be sufficient to drive expression when placed directly next to the promoter, it is unable to drive expression in the endogenous context. Conversely, the -4.3kb enhancer is a key driver of expression in these lineages. The reporter fluorescence that remains when both enhancers are absent indicates that additional sequences that drive the expression in the cells of the Ep and Cpapp lineages must exist within the -5.3 kb upstream region.

ceh-13 and Hox cofactor genes activate nob-1 expression Because CEH-13::GFP expression precedes that of NOB-1::GFP in many lineages, we asked whether ceh-13 is required for nob-1 expression. To evaluate this, we examined nob-1 reporter fluorescence in embryos homozygous for the likely null mutation ceh-13(sw1) [22]. Fluorescence driven by the -5.3kb nob-1 transcriptional reporter is absent in the ABp(l/r)ppp lineage (p < 0.001), 25% lower in the Abp(l/r)ap lineage, and 40% lower in the Ep lineage (p < 0.02) (Fig 3A–3C). Depletion of ceh-13 by RNAi gave similar results with the transcriptional reporter (p < 0.002). To test the relevance of this to the endogenous locus, we measured fluorescence levels of endogenously tagged NOB-1::GFP knock-in allele in ceh-13(sw1) mutant embryos. This fluorescence was reduced in the Ep, Abp(l/r)appa, and Abp(l/r)ppp lineages by 59%, 49%, and 23% respectively (P < 0.001) (Fig 3A, 3D, and 3E). The initial onset of expression of NOB-1::GFP was delayed relative to control embryos in many of these cells (Figs 3E and S3B). Expression of endogenously tagged PHP-3::GFP in ceh-13(sw1) mutant embryos showed similar changes to endogenously tagged NOB-1::GFP (S3C Fig), indicating both genes are regulated by ceh-13, possibly through the same enhancers. In addition, the fact that the nob-1 promoter reporter completely loses ABp(l/r)pp lineage expression in the absence of ceh-13 while the endogenously tagged alleles show only temporal delay and quantitative reduction indicates the importance of additional cis-regulatory sequences outside of the promoter region for nob-1 and php-3 expression. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Expression of cis-regulatory elements upstream of nob-1 and php-3 depends on ceh-13. A) Fold change in expression level relative to mean wild-type control for nob-1 promoter, endogenously tagged NOB-1::GFP knock-in allele, and enhancer reporters in ceh-13 mutant and RNAi conditions. Number of biological replicates for each condition ranges from 3–16, however, since there are, for example, 36 nob-1 cells in the ABp(l/r)ap lineages, the number of points represented by the box plots is much greater. P values determined by Wilcoxon Ranked Sum Test, p<0.05 significant. B) Lineage view of wild-type nob-1–5.3kb upstream reporter (“promoter”) expression in specified lineages. C) -5.3kb nob-1 promoter reporter expression in ceh-13(sw1) null mutant. D) Lineage view of wild-type endogenously tagged NOB-1::GFP expression in specified lineages. E) Expression of endogenously tagged NOB-1::GFP in ceh-13(sw1) null mutant. Underlined lineages and bracketed early cells had specifically significantly lower NOB-1::GFP levels. https://doi.org/10.1371/journal.pgen.1010187.g003 To determine which enhancers require ceh-13, we measured the activity of individual nob-1 enhancer reporters in worms lacking ceh-13 (Figs 3A and S3A). In ceh-13(sw1) mutants, reporter fluorescence driven by the -3.4 kb enhancer decreased significantly in all expressing cells, excluding the C lineage (p < 0.003) with an average reduction of 90% and complete absence of reporter fluorescence in ABp(l/r)ppp. ceh-13 loss also reduces reporter fluorescence driven by the -4.3 kb enhancer in ABp(l/r)ppp (47% decrease) but fluorescence in Abp(l/r)ap is unchanged. This indicates that other factors besides ceh-13 activate the -4.3kb enhancer in the Abp(l/r)ap lineage and suggests that this enhancer may be responsible for the residual expression of the 5.3 kb transcriptional reporter in this lineage in ceh-13 mutant embryos. Fluorescence from the transgene driven by the distal -8.3kb enhancer was also absent in ABp(l/r)ppp in the ceh-13(sw1) mutant (p < 0.002), indicating its activity also requires ceh-13. Motif analysis identified six putative ceh-13-binding motifs in the -3.4 kb enhancer; mutating these sites resulted in a lack of reporter fluorescence in the AB and C lineages (p < 0.02, p < 0.03), but did not affect reporter fluorescence in the Ep lineage indicating ceh-13 regulation of nob-1 may be indirect in the E lineage (Figs 3A and S3B). These results show that multiple ceh-13-dependent enhancers work together to regulate nob-1/php-3 expression in gastrulating embryos. Expression of classically defined Hox targets often requires Hox cofactors such as those encoded by homothorax or extradenticle. These cofactors form larger TF complexes with Hox factors to increase binding specificity and may also have Hox-independent functions [44–46]. To determine when and where Hox cofactors are expressed in early embryos, we used StarryNite to trace the expression of translational reporters for unc-62, the C. elegans homothorax/Meis ortholog, and of ceh-20 and ceh-40, the orthologs of extradenticle/Pbx. A third extradenticle ortholog, ceh-60, is only expressed in later development (>200 minutes) and does not overlap with ceh-13, so was not investigated further [30]. We found that fosmid translational reporter transgenes for each of these genes show specific and dynamic expression patterns that overlap with each other and with ceh-13 and nob-1 expression (Figs 4A–4D and S5). Notably, all three cofactors are co-expressed with CEH-13 in AB-derived cells that will later express NOB-1. Other CEH-13 and NOB-1 expressing cells also express all three cofactors, except only CEH-20 is expressed in the early E lineage, and all three are lost in the NOB-1-expressing ABp(l/r)appp lineage as the embryo approaches morphogenesis. We conclude that Hox cofactors are expressed in cells where ceh-13 activates nob-1 expression. We tested whether Hox cofactors regulate nob-1 expression by examining nob-1 reporter expression after loss of each gene. We found that depleting unc-62 by RNAi led to significantly less expression of the nob-1–5.3kb reporter in all AB lineages (80% reduction, p = 0.001), with near complete absence of reporter fluorescence in the ABp(l/r)apap and ABp(l/r)ppp lineages (Fig 4E). The -9kb NOB-1:GFP protein reporter transgene showed similarly limited expression, although this was only significant in the ABp(l/r)appa lineage and variable in other lineages (S4B and S4C Fig). In contrast, CEH-13::GFP reporter expression was unchanged after unc-62 RNAi (S4D Fig). These results show that nob-1 is regulated by unc-62. The fact that unc-62 RNAi has a stronger phenotype in the ABp(l/r)apap lineage than loss of ceh-13 suggests that unc-62 may regulate nob-1 independently of ceh-13 in these cells. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Hox co-factors precede nob-1 and regulate its expression. A) Trees of NOB-1::GFP expressing lineages, ABp, E, and C, showing the expression of ceh-13 (fosmid) and nob-1 (GFP transgene) reporters, and fosmid GFP transgene reporters for the Hox co-factor genes, ceh-20, ceh-40 and unc-62. Color thresholds were adjusted for each reporter to show all expressing cells. Highlighted branches are shown in graphs B, C, and D. B-D) Average TF reporter nuclear fluorescence intensity across embryos (n ≥ 2) and for left/right symmetric cells across developmental time for the cells leading to B) ABp(l/r)appaa, C) ABp(l/r)pppaa and D) Cpapp. Fluorescence intensity is in arbitrary units and grey bars mark cell divisions. E,F) Fold change values for -5.3 kb nob-1 promoter expression in untreated (n = 5) and unc-62 RNAi treated (n = 7) (E) or ceh-40(gk159) mutant embryos treated with ges-1 (control, n = 5) or ceh-20 RNAi (n = 5) (F) in specified lineages. Significant changes (p<0.05) marked by asterisk; P values determined by Wilcoxon Ranked Sum Test. https://doi.org/10.1371/journal.pgen.1010187.g004 To investigate the role of the extradenticle orthologs, we examined the nob-1 transcriptional reporter in a ceh-40(gk159) null mutant background in embryos of worms treated with control or ceh-20 RNAi. We found that the combination of ceh-20 RNAi with the ceh-40 mutation resulted in an absence of reporter fluorescence from virtually all expressing cells including the Ep and Cpap lineages as compared to control RNAi (97% decrease, p < 0.004). This indicates that the exd orthologs ceh-20 and ceh-40 are required to activate regulatory elements upstream of nob-1 and that they are at least partially independent of ceh-13 in the cells of the ABp(l/r)ap, Ep, and Cpap lineages. This finding is consistent with previous reports that nob-1 RNAi causes increased lethality in a ceh-40 mutant background relative to wild-type animals [47].

[END]

[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1010187

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.