Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Jun 15;149(12):dev200594.
doi: 10.1242/dev.200594. Epub 2022 Jun 30.

Developmental and evolutionary comparative analysis of a regulatory landscape in mouse and chicken

Aurélie Hintermann  1 , Isabel Guerreiro  1 , Lucille Lopez-Delisle  2 , Christopher Chase Bolt  2 , Sandra Gitto  1 , Denis Duboule  1   2   3 , Leonardo Beccari  1
Affiliations

Developmental and evolutionary comparative analysis of a regulatory landscape in mouse and chicken

Aurélie Hintermann et al. Development. .

Abstract

Modifications in gene regulation are driving forces in the evolution of organisms. Part of these changes involve cis-regulatory elements (CREs), which contact their target genes through higher-order chromatin structures. However, how such architectures and variations in CREs contribute to transcriptional evolvability remains elusive. We use Hoxd genes as a paradigm for the emergence of regulatory innovations, as many relevant enhancers are located in a regulatory landscape highly conserved in amniotes. Here, we analysed their regulation in murine vibrissae and chicken feather primordia, two skin appendages expressing different Hoxd gene subsets, and compared the regulation of these genes in these appendages with that in the elongation of the posterior trunk. In the two former structures, distinct subsets of Hoxd genes are contacted by different lineage-specific enhancers, probably as a result of using an ancestral chromatin topology as an evolutionary playground, whereas the gene regulation that occurs in the mouse and chicken embryonic trunk partially relies on conserved CREs. A high proportion of these non-coding sequences active in the trunk have functionally diverged between species, suggesting that transcriptional robustness is maintained, despite considerable divergence in enhancer sequences.

Keywords: Chromatin topology; Development; Enhancers; Evolution; Gene regulation; Placodes; TADs; Teguments.

PubMed Disclaimer

Conflict of interest statement

Competing interests The authors declare no competing or financial interests.

Figures

Fig. 1.
Fig. 1.
Transcription of Hoxd genes in mammalian and avian skin primordia. (A) Schematic of an E12.5 mouse embryo with vibrissae primordia (VPs) represented as red circles. The dorsoventral (D-V) and the caudorostral (C-R) axes are shown as red arrows, with the directions representing the timing of appearance of the VPs. (B) WISH on E12.5 faces. The arrowheads point to the most caudodorsal placode, which is the largest one and the first to appear. The red arrowheads indicate the detection of transcripts, white arrowheads indicate no detection of transcripts. Staining intensity progressively decreases from Hoxd1 to Hoxd8. (C) Schematic of the back of a HH35 chicken embryo with feather placodes indicated as red circles. A, anterior; P, posterior. (D) WISH on HH35 chickens showing the skin of the upper back. The arrowheads point to the neck at the level of the shoulders. HOXD1 is not expressed, whereas HOXD3 and HOXD8 signal intensities are stronger than for HOXD4 and HOXD9. The red arrowhead indicates the detection of transcripts, white arrowheads indicate no detection of transcripts. (E) Schematic representation of a E9.5 or HH20 posterior trunk and tailbud. A, anterior; P, posterior; FS, forming somites; IM, intermediate mesoderm; LM, lateral mesoderm; NT, neural tube; PM, paraxial mesoderm; PSM, presomitic mesoderm; S, somites. (F) WISH on E9.5 mouse posterior trunk and tail bud. Hoxd1 transcription pattern is different from that of Hoxd4 and Hoxd9. (G) WISH on HH20 chicken posterior trunk and tail bud. HOXD1 transcription pattern is different from HOXD4 and HOXD9. The white arrowheads in panels F and G point to the Hoxd1 stripe in the PSM. The black arrowheads show the difference between mouse (F) and chicken (G) in the persistence of Hoxd1 mRNAs in formed somites. Images are representative of at least three different embryos processed in at least two different WISH experiments. Scale bars: 1 mm.
Fig. 2.
Fig. 2.
Capture Hi-C-seq at the mouse and chicken HoxD loci. (A,B) Capture Hi-C-seq heatmaps using dissected posterior trunk cells. Below each heatmap, a CTCF ChIP-seq track is shown, produced from the same material. The similarities in the structural organization of the mouse and the chicken loci are underlined either by the positions of syntenic CNEs at key positions (numbered vertical dashed lines), the domains produced by TAD calling (black bars below) and the presence of a sub-TAD boundary (asterisk) within T-DOM. Genes are represented by empty rectangles, the filled grey rectangle indicates the HoxD cluster. In panel A, E9.5 mouse posterior trunk cells were used and the CHi-C heatmap was mapped on mm10 with a bin size of 5 kb (chr2:73,800,000-75,800,000). In panel B, HH20 chicken posterior trunk cells were used and the CHi-C heatmap was mapped on galGal6 with a bin size of 2.5 kb (chr7:15,790,000-16,700,000) and on an inverted x-axis. The positions of the two TADs (T-DOM and C-DOM) and of both sub-TADs are shown on top. The scales on the x-axes were adjusted to comparable sizes for ease of comparison, yet the chicken locus is more compacted (scale bars are shown on the right).
Fig. 3.
Fig. 3.
Tissue- and gene-specific interactions of dense H3K27ac regions. (A) The top panel shows a ChIP-seq profile of H3K27ac using mouse E12.5 VPs (orange) superimposed over the E12.5 forebrain cells (grey) (mm10, chr2:73,800,000-75,800,000). The lower panel shows a H3K27ac profile produced from dissected chicken HH35 dorsal skin (green), with HH18 brain cells (grey) (galGal6, chr7:15,790,000-16,700,000, inverted x-axis). The highest interacting regions are depicted as D1, D4 and D9 (see also Fig. S3B). The positions of conserved CNEs were used to delimit these regions in both species (vertical dashed lines) and the extents of TADs are shown below (thick black lines). The T-DOM (empty red rectangle) is split in overlapping genomic windows and the densest window (most acetylated region; MAR) is shown as a filled red rectangle. In the mouse VP, the MAR is within the D1 DNA segment, whereas the chicken MAR overlaps with the D4- and D9-regions. (B) The top panel shows a H3K27ac ChIP-seq using mouse E9.5 posterior trunk cells (orange) superimposed over forebrain cells (grey) (mm10, chr2:73,800,000-75,800,000). The lower panel shows the H3K27ac profile obtained from the same sample but dissected from a HH18 chicken embryo (green) with brain cells (grey) as a control (galGal6, chr7:15,790,000-16,700,000, inverted x-axis). Although the mouse posterior trunk MAR (red rectangle) coincides with the D4 segment, the chicken MAR counterpart also covers D4 and a small part of D9.
Fig. 4.
Fig. 4.
Transcriptional regulation of Hoxd1. (A) The three deletion lines used are shown in red and the transgenic BACs in blue. The extent of the D1- to D9-regions are shown on top (black) as well as the positions of both sub-TADs below (thick black lines). (B) WISH using a Hoxd1 probe on E12.5 mouse embryos. Hoxd1 mRNAs were detected in both the control and the Del(SB2-SB3) lines (red arrowheads), but the signal was absent from Del(attP-SB2) mutant embryos (white arrowhead). (C) X-gal staining of E12.5 mouse embryos carrying randomly integrated BACs. A strong staining was detected in VPs when BACMtx2 was used (red arrowhead), whereas staining was not scored with the other two flanking BAC clones (white arrowheads). Therefore, a region directly downstream the HoxD cluster is necessary and sufficient to activate Hoxd1 transcription in VPs, referred to as D1-region as it coincides with this previously defined region. (D) WISH of Hoxd1 on E9.5 embryos focusing on tail buds. Hoxd1 was detected in forming somites in both the control and the Del(SB2-SB3) lines (red arrowheads), but was absent from Del(attP-SB2) mutant embryos (white arrowhead). (E) X-gal staining of E9.5 mouse embryos carrying various BAC transgenes. Staining was scored in the entire neural tube and paraxial mesoderm with the BACHoxD (black arrowhead). In contrast, embryos carrying BACMtx2 displayed staining in forming somites (red arrowhead), a staining that was absent when BACT1 was used (white arrowhead). (F) Enlargement of the D1-region along with a H3K27ac ChIP-seq using dissected E12.5 VPs (orange, top) and E9.5 posterior trunk cells (orange, bottom) superimposed over E12.5 forebrain (FB) cells (grey) (mm10; chr2:74,747,751-74,916,570). ATAC-seq using E12.5 VPs and E9.5 PTs are shown as a black line with an inverted y-axis to indicate levels of chromatin accessibility. Above the profiles are the three VP-acetylated elements (black arrowheads for H3K27ac-positive elements, asterisks for ATAC-positive elements) and below are MACS2 narrowPeaks for VPs and PTs (orange) and FB (grey). The position of the three EC sequences used as transgenes (tgEC) are shown as numbered black boxes. (G) X-gal staining of VPs from an E12.5 embryo transgenic (tgN) for EC1. (H) X-gal staining of forming somites in an E9.5 embryo transgenic for the EC2 sequence. Red arrowheads in F,G indicate the detection of transcripts in VPs. White arrowheads indicate no transcript detection in VPs. Images are representative of at least two different embryos processed in at least two different WISH experiments. Scale bars: 1 mm.
Fig. 5.
Fig. 5.
Conserved sequences with divergent functions. (A) Bar plots showing the proportion of putative cis-regulatory elements (pCREs) harbouring the H3K27ac mark in one (specific) versus more than one (pleiotropic) tissue, in either mouse (orange) or chicken (green). For simplification purposes, distal pCREs (>2 kb away from the transcription start site) are referred to as enhancers and proximal pCREs (<2 kb away from transcription start site) are referred to as promoters. The numbers of enhancers and promoters (left), and the numbers of conserved enhancers and non-conserved enhancers (right) are indicated. The proportion of specific versus pleiotropic elements is shown for CNE and non-CNE enhancers (right). In both mouse and chicken, enhancers are more specific than promoters and amongst enhancers, conserved elements are found to be acetylated in more tissues than non-conserved elements. (B) Hierarchical clustering obtained using the pheatmap R package. The x-axis indicates enhancer-CNEs, i.e. conserved sequences that are non-coding in mouse and in chicken and overlap with a H3K27ac mark. The y-axis shows the tissues used to obtain MACS2-processed peaks. The values on the heatmap correspond to the enrichment scores of the peaks. Enhancer-CNEs cluster by species and not by tissues. (C) Euler diagrams representing the numbers of enhancer-CNEs in the indicated tissues that are acetylated in mouse (orange), in the corresponding chicken tissue (dark green) or in unrelated chicken tissue (light green), showing that conserved sequences diverge in their regulatory activities. The intersection between the same tissue in mouse and in chicken is outlined in black, and the intersection between a chicken tissue different from the mouse tissue is outlined in white. DFL, distal forelimb; FB, forebrain; PFL, proximal forelimb; PT, posterior trunk.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Abdennur, N. and Mirny, L. A. (2020). Cooler: scalable storage for Hi-C data and other genomically labeled arrays. Bioinformatics 36, 311-316. 10.1093/bioinformatics/btz540 - DOI - PMC - PubMed
    1. Acemel, R. D., Tena, J. J., Irastorza-Azcarate, I., Marlétaz, F., Gómez-Marín, C., de la Calle-Mustienes, E., Bertrand, S., Diaz, S. G., Aldea, D., Aury, J.-M.et al. (2016). A single three-dimensional chromatin compartment in amphioxus indicates a stepwise evolution of vertebrate Hox bimodal regulation. Nat. Genet 48, 336-341. 10.1038/ng.3497 - DOI - PubMed
    1. Afgan, E., Baker, D., van den Beek, M., Blankenberg, D., Bouvier, D., Čech, M., Chilton, J., Clements, D., Coraor, N., Eberhard, C.et al. (2016). The Galaxy platform for accessible, reproducible and collaborative biomedical analyses: 2016 update. Nucleic Acids Res. 44, W3-W10. 10.1093/nar/gkw343 - DOI - PMC - PubMed
    1. Allais-Bonnet, A., Hintermann, A., Deloche, M.-C., Cornette, R., Bardou, P., Naval-Sanchez, M., Pinton, A., Haruda, A., Grohs, C., Zakany, J.et al. (2021). Analysis of polycerate mutants reveals the evolutionary co-option of HOXD1 for horn patterning in bovidae. Mol. Biol. Evol. 38, 2260-2272. 10.1093/molbev/msab021 - DOI - PMC - PubMed
    1. Amândio, A. R., Lopez-Delisle, L., Bolt, C. C., Mascrez, B. and Duboule, D. (2020). A complex regulatory landscape involved in the development of mammalian external genitals. Elife 9, e52962. 10.7554/eLife.52962 - DOI - PMC - PubMed

Publication types

MeSH terms