(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

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



A shared ancient enhancer element differentially regulates the bric-a-brac tandem gene duplicates in the developing Drosophila leg

['Henri-Marc G. Bourbon', 'Center For Integrative Biology', 'Molecular Cellular', 'Developmental', 'Mcd', 'Biology Unit', 'Federal University Of Toulouse', 'Toulouse', 'Mikhail H. Benetah', 'Emmanuelle Guillou']

Date: 2022-05

Gene duplications and transcriptional enhancer emergence/modifications are thought having greatly contributed to phenotypic innovations during animal evolution. Nevertheless, little is known about how enhancers evolve after gene duplication and how regulatory information is rewired between duplicated genes. The Drosophila melanogaster bric-a-brac (bab) complex, comprising the tandem paralogous genes bab1 and bab2, provides a paradigm to address these issues. We previously characterized an intergenic enhancer (named LAE) regulating bab2 expression in the developing legs. We show here that bab2 regulators binding directly the LAE also govern bab1 expression in tarsal cells. LAE excision by CRISPR/Cas9-mediated genome editing reveals that this enhancer appears involved but not strictly required for bab1 and bab2 co-expression in leg tissues. Instead, the LAE enhancer is critical for paralog-specific bab2 expression along the proximo-distal leg axis. Chromatin features and phenotypic rescue experiments indicate that LAE functions partly redundantly with leg-specific regulatory information overlapping the bab1 transcription unit. Phylogenomics analyses indicate that (i) the bab complex originates from duplication of an ancestral singleton gene early on within the Cyclorrhapha dipteran sublineage, and (ii) LAE sequences have been evolutionarily-fixed early on within the Brachycera suborder thus predating the gene duplication event. This work provides new insights on enhancers, particularly about their emergence, maintenance and functional diversification during evolution.

During animal evolution, de novo emergence and rewiring of transcriptional enhancers have contributed to morphological innovations. However, how enhancers regulate distinctly gene duplicates and are evolutionary-fixed remain largely unknown. The Drosophila bric-a-brac (bab) locus, comprising the tandemly-duplicated genes bab1 and bab2, provides a good paradigm to address these issues. In this study, genetic analyses show a partial co-regulation of both genes in the developing leg depending on tissue-specific transcription factors known to bind an intergenic enhancer. Genome editing reveals that this enhancer is shared by both genes and is also critically required for bab2-specific expression. Chromatin features and phenotypic rescue experiments indicate the existence of partly-redundant limb-specific regulatory information within the bab1 transcription unit. Phylogenomics analyses among Diptera indicate that the Drosophila bab locus originates from duplication of a singleton gene within the Brachycera lineage. Lastly, we show that whereas bab1 promoter and leg enhancer sequences have been well conserved after the duplication event, bab2 promoter and other bab enhancers have evolved more recently in the Cyclorrhapha sublineage. This work brings some new insights about (i) how a single enhancer can drive specificity among tandem gene duplicates, and (ii) how enhancers evolutionary adapt with distinct cognate gene promoters.

Funding: Research in the HMB-MB laboratory was supported by grants from the Association pour la Recherche sur le Cancer (ARC PJA 20141201932) to MB, the Agence Nationale de la Recherche (ANR-16 CE12-0021-01) to MB, and institutional basic support from of the Centre National de Recherche Scientifique (CNRS) to HMB and Toulouse III University to HMB. Research in the G.C. laboratory was supported by a grant from the European Research Council (Advanced Grant 3DEpi) and by the CNRS (for GC and FB). MHB obtained a PhD fellowship from the French « Ministère de L’Enseignement Supérieur et de la Recherche », LHMV from the Mexican CONACYT and AB from a CNRS- Conseil Régional Midi-Pyrénées co-financing and then from the ARC. V.L. was supported by a doctoral fellowship from the Laboratory of Excellence EpiGenMed and the ARC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2022 Bourbon 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.

Here, we show that bab1 expression in the developing distal leg depends on the Rotund, Bowl and C15 proteins, three transcription factors known to regulate directly bab2 expression, by binding to dedicated LAE sequences [ 21 , 29 ]. LAE excision by CRISPR/Cas9-mediated genome editing indicates that this enhancer is required but not sufficient for both bab1 and bab2 regulation and, more unexpectedly, is required also for their differential expression along the P-D leg axis. Phylogenomics analyses indicate that LAE sequences have been fixed early on during dipteran evolution, well before emergence of the bab complex in the Cyclorrhapha sublineage. This work illuminates how a transcriptional enhancer from tandem gene duplicates underwent evolutionary changes to diversify their respective tissue-specific gene expression pattern.

Whereas both bab genes are required for dimorphic abdominal pigmentation traits and somatic gonad specification [ 17 , 26 ], only bab2 is critical for tarsal segmentation [ 15 ]. While bab1 loss-of-function legs are apparently wild-type, a protein null allele (bab AR07 ) removing bab2 (in addition to bab1) gene activity causes shortened legs owing to ts2-5 tarsal fusions as well as P-D homeotic transformations as seen by the appearance of a few up to several ectopic sex comb teeth in ts4, ts3 and ts2 segments, respectively, in males [ 15 ]. While the two bab genes are co-expressed within ts1-4 cells, bab2 is expressed more proximally than bab1 in ts1, and in a graded manner along the P-D leg axis in ts5 [ 15 ]. We previously showed that bab2 expression in distal leg (and antennal) tissues is governed by a 567-basepair (bp) long CRE/enhancer (termed LAE for “Leg and Antennal Enhancer”) which is located in between the bab1-2 transcription units ( Fig 1A ) [ 21 , 29 ]. However, LAE enhancer contribution to bab1 versus bab2 regulation in the developing distal legs remains to be investigated.

Adult T1-3 legs, on the pro-, meso- and meta-thoraces, respectively, are derived from distinct mono-layered epithelial cell sheets, organized as sac-like structures, called leg imaginal discs (hereafter simply referred to as leg discs) [ 30 – 32 ]. Upon completion of the third-instar larval stage (L3), each leg disc is already patterned along the proximo-distal (P-D) axis through regionalized expression of the Distal-less (Dll), Dachshund (Dac) and Homothorax (Hth) transcriptional regulators in the distal (center of the disc), medial and proximal (peripheral) regions, respectively [ 30 ]. The five tarsal (ts1-5) and the single pretarsal (distalmost) segments are patterned through genetic cascades mobilizing transcription factors, notably the distal selector protein Dll and the tarsal Rotund protein as well as nuclear effectors of Notch and Epidermal Growth Factor Receptor (EGFR) signaling, i.e., Bowl and C15, respectively [ 30 , 31 ].

(A) Schematic view of the Dmel bab locus on the 3L chromosomal arm (Chr3L). The tandem bab1 (blue) and bab2 (red) transcription units (filled boxes and broken lines represent exons and introns, respectively), the previously known CRE/enhancers are depicted by filled dots (abdominal DE and AE in dark and light orange, respectively; leg/antennal LAE in dark green and cardiac CE in purple), and the telomere and centromere directions are indicated by arrows. (B) Scheme depicting C15, Bowl and Rn TF activities in regulating bab2 expression as a four-ring pattern within the developing distal leg. (C) Medial confocal view of a wild-type L3 leg disc. Merged Bab1 (cyan) and Bab2 (red) immunostainings, as well as each marker in isolation in (C’) and (C”), respectively, are shown. Positions of bab2-expressing ts1-5 cells and the pretarsal (pt) field are indicated in (C”). Brackets indicate paralog-specific bab2 expression in ts1 and ts5 cells, and blue arrows corresponding cell rows that do not express bab1. (D) Distal confocal view of a homozygous C15 2 mutant L3 leg disc expressing LAE-RFP ZH2A . Merged Bab1 immunostaining (in cyan) and RFP fluorescence (red), and each marker in isolation in (D’) and (D”), are shown. Bab2-expressing mutant pt cells are circled with a dashed line in (D’) and (D”). (E) Medial confocal view of a mosaic L3 leg disc expressing LAE-GFP ZH2A and harboring rotund mutant clones. Merged Bab1 (cyan) immunostaining, GFP (green) and RFP (red) fluorescence, as well as each marker in isolation in (E’), (E”) and (E”‘), respectively, are shown. Mutant clones are detected as black areas, owing to the loss of RFP. The respective ts1-5 fields are indicated in (E). White arrows indicate rotund-/- clones still expressing bab1 and yellow ones those that do not express bab1. (F) Distal confocal view of a mosaic L3 leg disc expressing LAE-RFP ZH2A and harboring bowl mutant clones (GFP-). Merged Bab1 (cyan) immunostaining, RFP (red) and GFP (green) fluorescence, as well as a higher magnification of the boxed area for each marker in isolation in (F’), (F”) and (F”‘), respectively, are shown. Mutant clones are detected as black areas, owing to the loss of GFP. White arrows indicate pretarsal bowl-/- clones ectopically expressing both bab1 and LAE-RFP ZH2A (bab2).

The ~150-kilobase (kb) long Drosophila melanogaster bric-a-brac (bab) locus, located on the third chromosome (3L arm), comprises two tandemly-duplicated genes ( Fig 1A ), bab1 and bab2, which encode paralogous transcription factors sharing two conserved domains: (i) a Bric-a-brac/Tramtrack/Broad-complex (BTB) domain involved in protein-protein interactions, and (ii) a specific DNA-binding domain (referred to as BabCD, for Bab Conserved Domain), in their amino(N)- and carboxyl(C)-terminal moieties, respectively [ 15 ]. Bab1-2 proteins are co-expressed in many tissues [ 15 , 16 ]. In the larval epidermis, they co-regulate directly yellow expression in a sexually-dimorphic manner, thus controlling adult male versus female body pigmentation traits [ 17 – 20 ]. bab1-2 co-expression in the developing epidermis is partially governed by two CREs which drive reporter gene expression (i) in a monomorphic pattern in the abdominal segments A2-A5 of both sexes (termed AE, for “Anterior Element”), and (ii) in a female-specific pattern in the A5-A7 segments (DE, for “Dimorphic Element”) ( Fig 1A ) [ 18 , 21 ]. In addition to controlling male-specific abdominal pigmentation traits, bab1-2 are required, singly, jointly or in a partially-redundant manner, for embryonic cardiac development, sexually-dimorphic larval somatic gonad formation, salivary glue gene repression, female oogenesis, wing development as well as distal leg (tarsal) and antennal segmentation [ 15 , 17 , 21 – 28 ]. In addition to abdominal AE and DE, two other bab enhancers, termed CE and LAE (see Fig 1A ), have been characterized, which recapitulate bab2 expression in embryonic cardiac cells and developing distal leg (tarsus) as well as antennal cells, respectively [ 21 , 25 , 29 ]. However, while bab1 and bab2 are co-expressed in tarsal cells [ 15 ], contribution of the LAE enhancer to bab1 regulation in the developing leg has not been yet investigated.

Gene duplications have largely contributed to create genetic novelties during evolution [ 1 , 2 ]. Intra-species gene duplicates are referred to as “paralogs”, which eventually diverged functionally during evolution in a phylogenetic manner. Gene family expansion has facilitated phenotypic innovation through (i) acquisition of new molecular functions or (ii) the subdivision of the parental gene function between the duplicate copies [ 3 – 5 ]. Phenotypic novelties are thought having originated from both modifications of protein sequences and evolutionary emergence or modifications of genomic Cis-Regulatory Elements (CREs) or modules, most often dubbed as “enhancer” regions, which regulate gene transcription in a stage-, tissue- and/or cell-type-specific manner [ 6 – 10 ]. While many shared CRE/enhancers have been described in Drosophila for several gene complexes [ 11 – 14 ], how they emerge and are differentially evolving remain largely elusive.

Results

The tandem bab1-2 gene paralogs are co-regulated in the developing distal leg In addition to the distal selector homeodomain (HD) protein Distal-less, we and others have previously shown that the C15 HD protein (homeoprotein) as well as Rotund and Bowl Zinc-Finger (ZF) transcription factors (TFs) bind dedicated sequences within LAE to ensure precise bab2 expression in four concentric tarsal rings within the leg discs (Fig 1B) [21,29]. bab1-2 are co-expressed in ts2-4 tarsal segments, while bab2 is specifically expressed in ts5 and more proximally than bab1 in ts1, both in a graded manner along the P-D leg axis (Figs 1C and S1A) [15]. Given bab1-2 co-expression in ts1-4, we first asked whether C15, rotund and bowl activities are also controlling bab1 expression in the developing distal leg. To this end, we compared Bab1 expression with that of X-linked reporter genes faithfully reproducing the bab2 expression pattern there [21,29], in homozygous mutant leg discs for a null C15 allele or in genetically-mosaic leg discs harboring rotund or bowl loss-of-function mutant cells (Fig 1D–1F). C15 is specifically activated in the distalmost (center) part of the leg disc giving rise to the pretarsal (pt) segment (see Fig 1B) [33,34]. We have previously shown that the C15 homeoprotein down-regulates directly bab2 to restrict its initially broad distal expression to the tarsal segments [29]. Bab1 expression analysis in a homozygous C15 mutant leg disc revealed that both bab1 and LAE-RFPZH2A (bab2) are similarly de-repressed in the pretarsus (Fig 1C and 1D). In contrast to C15, rotund expression is restricted to the developing tarsal segments [35] and the transiently-expressed Rotund ZF protein contributes directly to bab2 up-regulation in proximal (ts1-2) but has no functional implication in distal (ts3-5) tarsal cells [21]. Immunostaining of genetically-mosaic leg discs at the L3 stage revealed that bab1 is cell-autonomously down-regulated in large rotund mutant clones in ts1-2, but not in ts3-4 segments (Fig 1E), as it is the case for LAE-GFPZH2A reflecting bab2 expression. Lastly, we examined whether the Bowl ZF protein, a repressive TF active in pretarsal but not in most tarsal cells, is down-regulating bab1 expression there [36], like bab2 [29]. Both bab1 and LAE-RFPZH2A (bab2) appeared cell-autonomously de-repressed in bowl loss-of-function pretarsal clones (Fig 1F). In addition to loss-of-function, we also conducted gain-of-function experiments for bowl and rotund. Given Bowl TF instability when overexpressed, bowl gain-of-function has been achieved by down-regulating lines which (i) encodes a related but antagonistic ZF protein destabilizing nuclear Bowl and (ii) is specifically expressed in the tarsal territory [36]. As previously shown for LAE-GFPZH2A (and bab2) expression, nuclear Bowl stabilization in the developing tarsal region appears sufficient to down-regulate cell-autonomously bab1 (S1C Fig). Prolonged expression of the Rotund protein in the entire distal part of the developing leg disc, i.e., tarsal in addition to pretarsal primordia, induces ectopic bab1 expression in the presumptive pretarsal territory, as previously shown for bab2 albeit with some differences in proximalmost GFP+ cells (S1B Fig, differentially-expressing cells are indicated with arrows), thus suggesting differential sensitivity of the two gene duplicates to Rotund TF levels (see discussion). Taken together, these data indicate that the C15, Bowl and Rotund transcription factors, previously shown to interact physically with specific LAE sequences and thus to regulate directly bab2 expression in the developing distal leg, are also controlling bab1 expression there. These results suggest that the limb-specific intergenic LAE enhancer activity regulates directly both bab genes.

LAE activity regulates both bab1 and bab2 paralogs along the proximo-distal leg axis To test the role of LAE in regulating both bab1 and bab2, we deleted precisely the LAE sequence through CRISPR/Cas9-mediated genome editing (see Materials and Methods) (Fig 2A). Two independent 3L chromosomal deletion events (termed ΔLAE-M1 and -M2; see S2A Fig for deleted DNA sequences) were selected for phenotypic analysis. Both deletion mutants are homozygous viable and give rise to fertile adults with identical fully-penetrant distal leg phenotypes, namely ectopic sex-comb teeth on ts2 (normally only found on ts1) tarsal segment in the male prothoracic (T1) legs (Fig 2B), which are typical of bab2 hypomorphic alleles [15]. The ΔLAE-M1 allele was selected for detailed phenotypic analyses and is below referred to as babΔLAE. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. LAE is differentially required for bab1 and bab2 expression in the developing leg. (A) Schematic view of the Dmel bab locus on the 3L chromosomal arm (Chr3L). The tandem bab1 and bab2 transcription units (filled boxes and broken lines represent exons and introns, respectively) and the intergenic LAE enhancer (in green) are depicted as in Fig 1A. The small CRISPR/Cas9-mediated chromosomal deficiency (babΔLAE) is depicted in beneath (deleted LAE is depicted as a green broken line). (B) Photographs of wild-type (left) and homozygous babΔLAE (right) T1 distal legs from adult males. The regular sex-comb (an array of about 10 specialized bristles on the male forelegs) on distal ts1 is indicated with asterisks, while ectopic sex-comb bristles on distal ts2 from the mutant leg is indicated by an arrow. Note that the five tarsal segments remain individualized in homozygous babΔLAE mutant legs. (C-D) Confocal views of wild-type (C) and homozygous babΔLAE mutant (D) L3 leg discs expressing LAE-GFPZH2A. Merged GFP fluorescence (green), Bab1 (cyan) and Bab2 (red) immunostainings, as well as the two latter in isolation in (C’-D’) and (C”-D”), respectively, are shown. The respective ts1-5 fields are indicated in C”. Brackets in C-C” show positions of GFP+ ts1 and ts5 cells expressing bab2 in a paralog-specific manner (yellow arrows in C’ indicate bab2-expressing GFP+ cell rows neither expressing bab1). Brackets in D-D” show that neither bab2 nor bab1 are expressed in GFP+ ts1 and ts5 mutant cells (as indicated in D” by green arrows). (E) Confocal view of a homozygous babΔLAE mutant L3 leg disc non-expressing the X-linked LAE-GFPZH2A reporter. Note that Bab1-2 are strictly co-expressed in three instead of four cell rings, consistently with the pattern observed in presence of the LAE-GFPZH2A construct. https://doi.org/10.1371/journal.pgen.1010083.g002 First, we quantified bab1 and bab2 mRNAs prepared from dissected wild-type and homozygous babΔLAE mutant leg discs. As shown in S2B Fig, both mRNAs were detected in mutant discs, although bab1 levels were two times lower than wild-type. Second, Bab1-2 expression patterns were analyzed in homozygous babΔLAE leg discs. To identify leg cells that should normally express bab2, we used the LAE-GFPZH2A reporter. In homozygous babΔLAE mutant leg discs, bab2-specific expression in proximalmost ts1 and ts5 cells (see Fig 1C) is no longer observed (Fig 2C and 2D). Furthermore, shared expression of both gene duplicates in distalmost ts1 cells is no longer detectable in babΔLAE mutant discs. Nevertheless, maintenance of bab1-2 co-expression in ts2-4 mutant cells indicates that additional cis-regulatory region(s) acting redundantly with the LAE enhancer must be present within the bab locus on the third chromosome. To exclude possible “transvection” effects of the X-linked LAE-GFPZH2A construct across different chromosomes [37], we also examined Bab1-2 expression patterns in homozygous babΔLAE leg discs in the absence of the LAE-GFPZH2A reporter. As shown in Fig 2E, in the homozygous babΔLAE mutant both bab genes are only (co-)expressed in ts2-4 cells and bab2 remains no longer specifically expressed in ts1 an ts5 cells, ruling out a trans-chromosomal effect of the LAE-GFPZH2A transgene. Taken together, our data indicate that intergenic LAE enhancer activity regulates both bab gene duplicates, being (i) required for bab1-2 co-expression in distal ts1, (ii) dispensable for their co-expression in ts2-4, suggesting the presence of redundant cis-regulatory information and (iii) critically required for bab2-specific tarsal expression both proximally and distally (in ts1 and ts5, respectively). Thus, the LAE enhancer governs both shared and paralog-specific expression of the bab1-2 gene duplicates.

Chromatin features predict limb-specific cis-regulatory elements within bab1 Since LAE appeared dispensable for bab1 and bab2 co-expression in ts2-4 cells, our data suggested the existence of other redundant cis-regulatory elements. We sought to identify cis-regulatory information acting redundantly with LAE by taking advantage of available genome-wide chromatin features and High-throughput chromosome conformation Capture (Hi-C) experiments performed from L3 leg or eye-antennal discs (Fig 3). bab1 and bab2 are indeed co-expressed in distal antennal cells within the composite eye-antennal imaginal disc [15]. A topologically-associating domain covering the entire bab locus was detected in Hi-C data from eye-antennal discs (Fig 3A) [38], revealing particularly strong interactions between bab1-2 promoter regions. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Chromatin feature analyses suggest partly-redundant limb-specific regulatory information within the bab1 transcription unit. (A) Hi-C screenshot of a ~160 kb region covering the Dmel bab gene complex. Score scale is indicated on the right (yellow to dark blue from positive to negative). The tandem bab1 and bab2 transcription units as well as the intergenic LAE enhancer are depicted as in Fig 2A. (B) FAIRE-, ATAC- and/or ChIP-Seq profiles from L3 eye-antennal and leg discs. Normalized open chromatin, histone H3 post-translational modifications and Dll binding profiles are shown. The respective locations of the enhancer signature region (ESR), LAE and bab2 promoter sequences are boxed in light blue, green and red, respectively. https://doi.org/10.1371/journal.pgen.1010083.g003 We then used published genome-wide data from Chromatin Immuno-Precipitation (ChIP-Seq) and Formaldehyde-Assisted Isolation of Regulatory Elements (FAIRE-Seq), as well as Assay for Transposase-Accessible Chromatin (ATAC-Seq) experiments [38–41], looking for active enhancer marks (H3K4me1 and H3K27Ac) and nucleosome-depleted chromatin regions (thus accessible to transcription factors), respectively. In the eye-antennal disc active enhancer signatures are mainly associated with a ~15-kb-long genomic region encompassing the bab1 promoter, first exon and part of its first intron (Fig 3B). Note that LAE is also accessible to transcription factors and carries H3K4me1 marks, consistently with its enhancer activity characterized in distal antennal cells [21]. To investigate further the role of this putative enhancer region (hereafter referred to as ESR) within bab1, we analyzed previously-published ChIP-Seq data from L3 leg discs [42] for binding sites for Dll which is critically required to cell-autonomously activate bab1 and bab2 [21,43,44]. In addition to expected binding over LAE [21] and the bab2 promoter, strong Dll binding is also detected throughout ESR, including over the bab1 promoter (Fig 3B). Taken together, we concluded that the bab1 transcription unit is predicted to include uncharacterized limb-specific regulatory information (i.e.., ESR) acting redundantly with the LAE enhancer.

LAE functions together with cis-regulatory elements located within bab1 To validate the existence of regulatory information within the bab1 locus, we performed phenotypic rescue experiments with Bacterial Artificial Chromosome (BAC) constructs covering each about 100 kb of genomic DNA. We have previously shown that a X-linked BAC construct, BAC26B15ZH2A, encompassing bab2 and the downstream intergenic sequence including LAE (see Fig 4A), is able to rescue (i) Bab2 expression in the tarsal primordium and (ii), distal leg phenotypes detected in homozygous animals for the protein null allele babAR07, affecting both Bab1 and Bab2 [15]. Conversely, a mutant BAC26B15 construct (BAC26B15ΔLAEZH2A) inserted at the same genomic landing site (i.e., ZH2A on the X chromosome) and specifically lacking LAE sequence is unable to rescue Bab2 tarsal expression and leg phenotypes of babAR07 mutants (Fig 4B–4D) [21]. These data indicated that (i) in absence of redundant cis-regulatory information, LAE is essential for bab2 expression in the developing tarsus and (ii) the enhancer information redundant with LAE is located outside the genomic region covered by BAC26B15, which only includes the bab1 first exon and thus lacks adjacent intronic ESR sequences. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. The bab1-overlaping BAC69B22 construct includes partially-redundant limb-specific cis-regulatory information. (A) Chromosomal deficiency and BAC constructs covering the bab locus. The tandem gene paralogs and intergenic LAE are depicted as shown in Fig 1A, except that bab2 is depicted in pink instead of red. The babAR07 3L chromosomal deficiency, a protein null allele removing bab1 and bab2 activities, is shown in beneath, with known deleted portion indicated by a dashed line. The two overlapping BAC constructs 69B22 and 26B15, as well as a mutant derivative of the latter specifically-deleted for LAE, are shown in beneath. (B-G) Medial confocal views of wild-type (B, E) and homozygous babAR07 mutant (C-D and F-G) L3 leg discs, harboring singly or combined X-linked BAC construct(s) shown in (A), as indicated above each panel. Bab2 (pink) and Bab1 (cyan) immunostainings are shown. Positions of bab1- and bab2-expressing ts1-4 cells are indicated. Note stochastic bab2 expression in (G). https://doi.org/10.1371/journal.pgen.1010083.g004 To validate the putative regulatory information within the bab1 transcription unit, we have tested the capacity of another BAC, BAC69B22, which overlaps entirely bab1 but lacks LAE (see Fig 4A), to restore Bab1 expression in homozygous babAR07 leg discs. As shown in Fig 4E and 4F, the X-linked BAC69B22ZH2A construct could partially restore bab1 expression in ts2-4 cells, indicating that it contains cis-regulatory information redundant with LAE activity in these tarsal segments. To test the capacity of BAC69B22 sequences to also regulate bab2 expression in ts2-4 cells, we placed BAC69B22ZH2A across BAC26B15ΔLAEZH2A, to allow pairing-dependent trans-interactions (i.e., transvection; both constructs being inserted at the same ZH2A landing site) between the two X chromosomes in females. This configuration partially restored Bab2 expression in ts2-4 cells from babAR07 mutant L3 leg discs, albeit in salt and pepper patterns (Fig 4G), diagnostic of transvection effects [37]. Taken together with our previous chromatin data, these genetic results are consistent with the existence within the 15 kb bab1 ESR of uncharacterized cis-regulatory information capable to drive some bab1 and bab2 expression in distal leg tissues and acting redundantly with the LAE enhancer. The large size and complexity of this region, together with data mining from the literature, suggested that this region includes interspersed regulatory elements whose functional implication in the developing leg and antenna deserves to be studied separately.

LAE sequences emerged early on in the Brachycera, thus predating bab gene duplication Having traced back the bab gene duplication raised the question of the evolutionary origin of the LAE enhancer, which regulates both bab1 and bab2 expression [21] (this work). We have previously shown that LAE includes three subsequences highly-conserved among twelve reference Drosophilidae genomes [48], termed CR1-3 (for Conserved Regions 1 to 3; see S4A Fig and S1 Data), of which only two, CR1 and 2, are critical for tissue-specificity [21,29]. The 68 bp CR1 includes contiguous binding sites for Dll and C15 homeoproteins, while the 41 bp CR2 comprises contiguous binding sites for Dll as well as the ZF protein Bowl (S4B and S4C Fig, respectively) [21,29]. To trace back the LAE evolutionary origin, we then systematically searched for homologous CR1-3 sequences (>50% identity) in dipteran genomes for which we identified one or two bab genes. Importantly, conserved LAE sequences have not been yet reported outside drosophilids. Small genomic regions with partial or extensive homologies to the CR1 (encompassing the C15 and Dll binding sites) and CR2 (particularly the Dll and Bowl binding sites) could be detected in all examined Brachycera families but not in any nematoceran (Figs 6B, S4B, and S4C). Contrary to closely-associated CR1-2 homologous sequences, no CR3-related sequence could be identified nearby, in any non-Drosophilidae species. Significantly, homologous LAE sequences are situated (i) in between the tandemly-duplicated paralogs in cyclorrhaphan species for which the entire bab locus sequence was available to us, suggesting an evolutionarily-conserved enhancer role, or (ii) 20 kb upstream of the bab singleton gene in the Asiloidea P. coquilletti (see Fig 6C). Taken together, as summarized in Fig 6B and 6C, these data suggest that a LAE-like enhancer with CR1- and CR2-related elements emerged early on in the Brachycera suborder, 180–200 million years ago, and has been since fixed within or upstream the bab locus in the Cyclorrhapha and Asiloidea superfamilies, respectively.

Unlike LAE, other bab CREs have not been conserved beyond the Cyclorrhapha The broad LAE sequence conservation led us to also trace back the evolutionary origins of the cardiac CE, abdominal anterior AE and sexually-dimorphic DE cis-regulatory elements (see Fig 6A). While CE only regulates bab2, the AE and DE elements are predicted to govern both bab1 and bab2 expression in abdominal cells. Significantly, CE- and DE-related sequences could be only detected within schizophorans (excepted in Calyptratae) (Figs 6B, S5B, and S5C, respectively), whereas AE-related sequences could be readily identified within bab loci from drosophilids (S1 Data) but not from Aschiza, Empidoidea and Nematocera. In conclusion, as summarized in Fig 6B and 6C, contrary to the LAE enhancer which among the Diptera emerged early on in the Brachycera suborder, other so-far identified bab cis-regulatory sequences have not been conserved beyond the Cyclorrhapha. Thus, and unlike the brachyceran LAE (CR1-2) sequences, these data indicate that other shared enhancer sequences (i.e., DE and AE) have been evolutionarily-fixed after the bab1-2 paralog emergence.

[END]

[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1010083

(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.