(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Differential modularity of the mammalian Engrailed 1 enhancer network directs sweat gland development [1]

['Daniel Aldea', 'Department Of Genetics', 'Perelman School Of Medicine', 'Philadelphia', 'Pennsylvania', 'United States Of America', 'Blerina Kokalari', 'Yuji Atsuta', 'Genetics Department', 'Harvard Medical School']

Date: 2023-02

Abstract Enhancers are context-specific regulators of expression that drive biological complexity and variation through the redeployment of conserved genes. An example of this is the enhancer-mediated control of Engrailed 1 (EN1), a pleiotropic gene whose expression is required for the formation of mammalian eccrine sweat glands. We previously identified the En1 candidate enhancer (ECE) 18 cis-regulatory element that has been highly and repeatedly derived on the human lineage to potentiate ectodermal EN1 and induce our species’ uniquely high eccrine gland density. Intriguingly, ECE18 quantitative activity is negligible outside of primates and ECE18 is not required for En1 regulation and eccrine gland formation in mice, raising the possibility that distinct enhancers have evolved to modulate the same trait. Here we report the identification of the ECE20 enhancer and show it has conserved functionality in mouse and human developing skin ectoderm. Unlike ECE18, knock-out of ECE20 in mice reduces ectodermal En1 and eccrine gland number. Notably, we find ECE20, but not ECE18, is also required for En1 expression in the embryonic mouse brain, demonstrating that ECE20 is a pleiotropic En1 enhancer. Finally, that ECE18 deletion does not potentiate the eccrine phenotype of ECE20 knock-out mice supports the secondary incorporation of ECE18 into the regulation of this trait in primates. Our findings reveal that the mammalian En1 regulatory machinery diversified to incorporate both shared and lineage-restricted enhancers to regulate the same phenotype, and also have implications for understanding the forces that shape the robustness and evolvability of developmental traits.

Author summary Enhancers are regulatory elements in the genome that modulate the expression of protein-coding genes by directing how much, where, or when a given gene is expressed. Accordingly, enhancers are major determinants of mammalian traits and thought to be the predominant drivers of evolutionary change. Here we interrogated the identity and compared the functionality of the enhancers that control the specification of a single, highly variable trait, the density of sweat glands in mammalian skin, by regulating expression of the mammalian Engrailed 1 (En1) gene during development. We find that mammals have evolved two distinct types of enhancers to regulate this single trait: a shared enhancer active in multiple mammalian species that not only controls En1 expression in the skin but also in the brain, and also an enhancer whose activity is restricted to the skin of primates and rapidly evolved on this lineage to affect sweat phenotypes. Our findings implicate differences in the intrinsic properties of enhancers, namely the extent to which their activity is restricted to a specific context, in shaping not only the complexity of the regulatory landscape of a developmental gene but also the means by which that landscape evolves to generate trait variation.

Citation: Aldea D, Kokalari B, Atsuta Y, Dingwall HL, Zheng Y, Nace A, et al. (2023) Differential modularity of the mammalian Engrailed 1 enhancer network directs sweat gland development. PLoS Genet 19(2): e1010614. https://doi.org/10.1371/journal.pgen.1010614 Editor: David R. Beier, Seattle Children’s Research Institute, UNITED STATES Received: September 21, 2022; Accepted: January 13, 2023; Published: February 6, 2023 Copyright: © 2023 Aldea 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: All data are available in the manuscript or the supplementary materials. Funding: Funding for this study was supported by award P30-DK19525 (https://www.nih.gov/) to the University of Pennsylvania Diabetes Research Center for the use of the Transgenic and Chimeric Mouse Facility Core to support work in this study, by award P30-AR069589 (https://www.nih.gov/) to the Penn Skin Biology and Diseases Resource-based Center (SBDRC) Cores A and B to support the work in this study, by award 5T32AR007465 (https://www.nih.gov/), which provided salary support to H. L. D., by awards BCS-1847598 (https://www.nsf.gov/) and R01AR077690 (https://www.nih.gov/) Y.G.K., which both provided salary support to Y.G.K. and funds for the work in this study. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist.

Introduction Eccrine sweat glands are specialized exocrine appendages of mammalian skin and are a major evolutionary innovation of this phylogenic class [1,2]. The main function of eccrine glands is to secrete hypotonic water onto the skin surface in response to specific neurological stimuli [2,3]. In the ancestral and most common state among mammals, eccrine glands develop only in the palmar/plantar (volar) skin of the ventral autopod (the distal-most segment of the limb) [1–3]. Here, eccrine glands are stimulated by cues from the limbic system and their secretions regulate frictional contact and grip [2–7]. In this context, eccrine glands are described as effectors of the fight or flight response, however, recent evidence also suggests that variation in volar eccrine glands is correlated with differential climbing ability in rodents [3,4,8,9]. Eccrine gland distribution has been expanded to nearly the entire body surface in humans and other catarrhine primates (the monkeys of Africa and Asia, and the anthropoid apes) [1,2,10]. Though structurally indistinguishable from their volar counterparts, generalized eccrine glands are activated by signals from the thermosensory regions of the hypothalamus and function in temperature regulation [2,3,10–15]. The physiological importance of this regional expansion is paramount in humans, who have by far the greatest eccrine gland density of any primate [2,10,16], and who are reliant on the vaporization of eccrine sweat as the main means for cooling the body [2,3,5]. Given this extensive diversification, the study of how eccrine gland phenotypes evolved along mammalian lineages provides an exceptional inroad to understand the broader principles through which evolution generates natural variation and functional novelty. The ability of any region of the skin to build eccrine glands relies on coordinated and reciprocal signaling between the deepest (basal) layer of the skin ectoderm from which eccrine glands arise and the underlying dermal mesenchyme [7,17–21]. In both humans and mice, expression of the transcription factor Engrailed 1 (En1) in the basal ectoderm is a universal hallmark of all skin in which eccrine glands develop [17,22,23]. Moreover, the focal upregulation of En1 expression is the earliest, specific signature of eccrine gland placodes, thickenings of the ectoderm from which eccrine glands derive [17,22]. In mice, disruption of En1 expression in the basal ectoderm at the time of placode formation leads to dose-dependent decreases in the number of volar eccrine glands formed [22–24]. Moreover, natural differences in volar eccrine gland number between inbred mouse strains are primarily the product of strain-specific variation in the quantitative levels of ectodermal En1 [22]. These findings illustrate the critical role of precise spatial, temporal, and quantitative regulation of the En1 locus in the generation of eccrine phenotypes. For En1, as for the majority of protein coding loci, gene expression is coordinated by regulatory elements, or enhancers, that define where, when, and how much the En1 promoter is activated. Notably, En1 is essential for the development of multiple traits independent of its role in eccrine gland formation. These include patterning of the vertebrate midbrain/hindbrain, the formation of the cerebellum, and the dorso-ventral patterning of the embryonic limb bud [22,23,25,26]. Accordingly, control of En1 expression is likely to involve multiple enhancers that have, at least in part, non-overlapping, context-specific roles. Consistent with this, a recent study implicated a lncRNA-containing region distal to the En1 promoter in mice and humans that is required for En1 expression in the ectoderm during dorso-ventral limb patterning, but which is dispensable for En1 expression in the brain at the same stage [27]. Regulation of the En1 locus in the skin ectoderm is also subject to differential temporal control as evidenced by our finding that En1 upregulation in mouse strains with high eccrine gland density coincides with the period of placode formation in the volar ectoderm [22]. However, during earlier developmental stages, including during dorso-ventral patterning, the volar skin ectoderm of mouse strains with high and with low eccrine gland densities shows equivalent levels of En1 [22]. The importance of context-specific EN1 regulation to eccrine phenotypes is evident in our own species. Humans have the most dramatically derived eccrine phenotype of all mammals, having evolved an eccrine gland density that is on average ten times that of other catarrhines [2,10]. We recently uncovered that this adaptive elaboration is underlain by the evolution of an ectodermal EN1 enhancer, ECE18 (hg38 Chr2:118309555–118310531) [24]. The human homolog of ECE18 (hECE18) overlaps the human accelerated region (HAR) 2xHAR20, and has undergone rapid evolution on the human lineage [24,28,29]. Following the split from chimpanzees, successive mutation of the human ECE18 homolog (hECE18) has led to the accumulation of ten derived base substitutions that interact epistatically to render hECE18 the most quantitatively potent of all primate enhancer homologs [24]. Analysis of the endogenous capabilities of the hECE18 enhancer in human basal skin cells (keratinocytes) revealed that this element is required for EN1 expression in this context [24]. The ability of hECE18 to promote the formation of more eccrine glands by upregulating ectodermal En1 in a hECE18 mouse knock-in provided functional evidence that this element modifies EN1 expression to induce our species’ derived eccrine gland phenotype [24]. Intriguingly, the contribution of ECE18 to ectodermal EN1 regulation appears to be restricted to primates. Quantitative comparisons in mouse and human keratinocytes revealed that ECE18 homologs of species outside of the primate order have little to no enhancer activity [24]. Moreover, knock-out of the endogenous ECE18 enhancer in mice produces no effect on ectodermal En1 expression or on eccrine phenotypes in the volar skin of these animals [24]. Whether such species have evolved entirely different enhancers to modulate En1 in eccrine gland formation or there is differential redundancy in the activity of En1 enhancers in lineages outside of primates, is unclear. We therefore set out to determine if other enhancers have evolved to regulate ectodermal En1 expression in the context of eccrine gland development and to define their relationship to ECE18 in this respect.

Discussion Cis-regulatory modulation of the spatial, temporal, and quantitative expression of the Engrailed 1 locus is crucial to the formation of eccrine sweat glands in mammals [22,24]. Our findings in the mECE20KO mouse model, coupled with the pattern of mECE20 activity in transgenic mice, demonstrate that this enhancer is an essential factor controlling En1 expression in all of these dimensions. While we cannot directly determine the endogenous functional properties of the hECE20 ortholog in human development, the concordance in activity between hECE20 and mECE20 suggests that the role of ECE20 in regulating En1 is broadly conserved among divergent mammals. This is further supported by the high degree of sequence conservation between mouse and human ECE20 homologs, and contrasts with the accelerated divergence of the primate specific ECE18 enhancer [24,28]. In light of this evidence for ECE20 functional conservation and because En1 is the earliest-known factor that is exclusively required for the formation of eccrine glands, ECE20 ranks within the highest tier of the ancestral developmental cascade for making these appendages [17,22]. Future studies to identify the transcription factors that activate ECE20 will thus set the stage for elucidating the mediators that initiate the eccrine developmental program in the ectoderm. An intriguing observation from our study is that the deletion of ECE20 results in a substantial but not a complete loss of En1 expression in the basal ectoderm. Consistent with our previous finding that eccrine gland number is dictated by En1 dosage, eccrine glands are significantly reduced but not entirely absent in mECE20KO animals. The buttressing of En1 expression against the loss of mECE20 may reflect the presence of additional En1 enhancers. Our results showing that both ECE18 and ECE20 human homologs are likely to regulate ectodermal EN1 during development are certainly consistent with such modular regulation of the locus in primates. Candidate regulatory elements that may provide these additional layers of mammalian En1 regulation include the ECE8 (Chr1: 120756823–120757766 [mm10]) and ECE23 (Chr1: 121394405–121395702 [mm10]) elements we have previously identified [24]. The results of our study not only implicate a multi-component network of enhancers that controls ectodermal En1 in mammalian eccrine gland phenotypes, but also highlight the functional compartmentalization of this network with respect to En1 regulation in distinct developmental contexts. A clear indication of this is the finding that mECE20 is required to potentiate En1 expression both in the embryonic brain and also in the skin during eccrine gland development, but not in the limb ectoderm during dorso-ventral patterning. The existence of a separable regulatory module for En1 expression in the latter context is supported by the identification of the Maenli lncRNA-containing region, deletion of which specifically results in a dorsalization phenotype in both human patients and in mice [27]. Studies of spatio-temporal differences in local chromatin configuration, particularly because of the proximity of En1 to the TAD boundary, and of the relative availability of transcription factors that regulate the respective enhancers are needed to determine why En1 engages with different enhancers in different contexts [43]. Considering the developmental importance of maintaining a ventral identity in the limb, it is intriguing to speculate that the evolution of independent modules for ectodermal En1 regulation in dorso-ventral patterning and in eccrine gland density specification could have facilitated the evolution of generalized eccrine glands in catarrhines. Beyond its implications for the complexity of the En1 enhancer network, that ECE20 has pleiotropic effects on En1 expression not only in the skin but also in the brain, provides a potential explanation for the repeated targeting of the ECE18 enhancer during human evolution. Unlike ECE20, ECE18 activity appears to be restricted to the skin ectoderm. As such, our findings suggest that the potential for incurring deleterious exaptations in addition to the effects on eccrine gland density would be comparably lower from mutations in ECE18 than in ECE20. Future studies are needed to fully characterize the function of ECE20 in the brain, however the observed impact of ECE20 disruption on the EN1/FGF8 neural circuit required for maintaining the regional identities of the midbrain-hindbrain is consistent with such a model. By virtue of its context-specific function, the evolution of the primate ECE18 enhancer would allow catarrhines to generate eccrine gland phenotypic variation at a reduced risk of affecting the development and patterning of the brain. Thus, the pleiotropy of ECE20 coupled with the specificity of ECE18 could not only have favored but also constrained the evolution of catarrhine eccrine gland phenotypes to the ECE18 enhancer. Our findings not only implicate an underlying driver of modularity in ectodermal EN1 regulation, but also suggest a basis for the extreme derivation of hECE18 and its importance to the evolution of the singular eccrine phenotype of humans.

Supporting information S1 Fig. Features of the ECE20-containing genomic interval and activity of the ECE20 enhancer in cultured skin cells. (A) Relative genomic position of Engrailed 1 candidate enhancer (ECE) 20 (boxed) and features of the ECE20-containing topologically associated domain (TAD, solid gray rectangle). Genomic positions of additional positive ECEs (red vertical lines) previously reported in Aldea et al. 2021 [24]. PhastCons scores based on alignment of placental mammals are depicted in green [46]. Called peaks from mouse embryonic limb and midbrain for interaction between the En1 promoter (anchor) and the genomic region containing ECE20 based on Capture-C (black, horizontal line), for CTCF enrichment (red and blue triangles show site location and directionality of CTCF sites), and for RAD21 enrichment (blue vertical lines) [33]. (B) Sequence alignment of mammalian genomes centered on ECE20 using mouse genome build mm10 as the base genome [46]. PhastCons scores for each position in the alignment are shown in green. Alignment and PhastCons scores are pulled from the USCS Genome browser (http://genome.ucsc.edu) [46] (C) Quantitative activity of mouse and human ECE20 orthologs in human GMA24F1A cultured keratinocytes, an immortalized human skin cell line that endogenously expressed EN1 [24,51,54]. The fold change in normalized luciferase activity relative to Control (empty vector) is plotted. Each dot represents a biological replicate. In (C) significance is assessed by one-way ANOVA and Tukey-adjusted P-values are reported. ***P<0.001, n.s. not significant. https://doi.org/10.1371/journal.pgen.1010614.s001 (TIF) S2 Fig. Functional characterization of the ECE20 enhancer in ectodermal En1 regulation and eccrine phenotypes. (A) Generation of an ECE20 knock-out mouse (mECE20KO) by CRISPR-Cas9 mediated genome editing. CRISPR-Cas9 target sequence and genotyping strategy are shown. Deletion junctions were confirmed by Sanger sequencing of F1 pups. (B) Fold change in En1 mRNA by qRT-PCR in P2.5 volar forelimb skin of wildtype (WT / WT), and mECE20KO homozygote (mECE20KO/ mECE20KO) mice relative to wildtype. (C) Normalized En1 mRNA allelic ratio in volar forelimb of wildtype at P2.5 of wildtype (C57BL/6JWT / FVB/N) and mECE20KO (C57BL/6J(mECE20KO) / FVB/N) hybrid mice. Ratios were normalized to the allelic ratio in F1 genomic DNA. Each point represents the mean value across three technical replicates of biological samples consisting of pooled P2.5 volar skins from both forelimbs of two or three mice. (D) Location and identity of in silico-predicted DNA binding motifs for transcription factors enriched in skin that are evolutionarily conserved between mouse and human ECE20. Relative expression of the cognate transcription factor (TF expression) and motif score (TF score) are shown. Motifs identified using funMotifs (tissue-specific functional motifs) [47]. (E) Quantification of interfootpad (IFP) eccrine gland number in adult volar forelimbs of WT / WT and mECE20KO/ mECE20KO mice. (F) Quantification of IFP eccrine glands in adult, volar forelimbs of En1 KO / WT; WT / WT and En1 KO / WT; WT / mECE20KO mice. In (B, C) dots represent an individual biological replicate. In (E, F) each point represents the average number of eccrine glands in the IFP across both forelimbs of a mouse. In (B, C) mean (line) with standard deviation are plotted. In (E, F) the median (line) and maximum and minimum are reported for each genotype. In (B, C, E, F) significance assessed by a two-tailed T-test. ***P<0.001, ** P<0.01, * P<0.05. (KO) knock-out. In (B, C) Rlp13a was used as housekeeping transcript for normalization. https://doi.org/10.1371/journal.pgen.1010614.s002 (TIF) S3 Fig. Characterization of ECE20 and ECE18 in the mouse embryonic midbrain-hindbrain and limb-bud. (A) Normalized En1 mRNA allelic ratios in forelimb autopods of wildtype (C57BL/6JWT / FVB/N) and mECE18KO (C57BL/6JmECE18KO / FVB/N) hybrid mice are plotted. (B) Whole-mount in situ hybridization for En1 in wildtype (WT / WT) and mECE20KO/ mECE20KO mice at E10.5. (C) Location and identity of in silico-predicted DNA binding motifs for transcription factors enriched in the brain that are evolutionarily conserved between mouse and human ECE20. Relative expression of the cognate transcription factor (TF expression) and motif score (TF score) are shown. Motifs identified using funMotifs (tissue-specific functional motifs) [47]. (C) Normalized En1 mRNA allelic ratios in midbrain-hindbrain of wildtype (C57BL/6JWT / FVB/N) and mECE18KO (C57BL/6JmECE18KO / FVB/N) hybrid mice. In (A, D) ratios are normalized to the allelic ratio in genomic DNA, and each point represents the mean value across three technical replicates of a pool of three or four mice for embryonic limb-bud in (A), or individual dissections of midbrain-hindbrain in (D) at E10.5. In (A, D) the mean (line) and the standard deviation are reported. In (A, D) significance assessed by a two-tailed T-test. n.s. not significant. (KO) knock-out. https://doi.org/10.1371/journal.pgen.1010614.s003 (TIF) S4 Fig. Generation of ECE18; ECE20 compound knock-out mice. (A) Generation of an ECE18; ECE20 compound knock-out mouse (mECE18KO; mECE20KO) by CRISPR-Cas9 mediated genome editing. CRISPR-Cas9 target sequence and genotyping strategy are shown. Deletion junctions were confirmed by Sanger sequencing of F1 pups. (KO) knock-out. https://doi.org/10.1371/journal.pgen.1010614.s004 (TIF) S1 Table. Primers used to subclone ECE20 orthologs in mouse transgenic assays. https://doi.org/10.1371/journal.pgen.1010614.s005 (DOCX) S1 Data. Raw data main figures. https://doi.org/10.1371/journal.pgen.1010614.s006 (XLSX) S2 Data. Raw data S1–S4 Figs. https://doi.org/10.1371/journal.pgen.1010614.s007 (XLSX)

Acknowledgments We thank Eric Joyce, Iain Mathieson, Pantelis Rompolas, Paola Kuri, Sixia Huang, and Gabriella Rice for helpful discussions on this project and on the manuscript. We thank Constance Cepko and Clifford J. Tabin for training, technical assistance, and support on ultrasound-guided injection. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

[END]
---
[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1010614

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution 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.