Learning about mammalian gene regulation from functional enhancer assays in the mouse Alex S. Nord PII: DOI: Reference:
S0888-7543(15)30006-9 doi: 10.1016/j.ygeno.2015.06.008 YGENO 8750
To appear in:
Genomics
Received date: Revised date: Accepted date:
12 February 2015 6 May 2015 8 June 2015
Please cite this article as: Alex S. Nord, Learning about mammalian gene regulation from functional enhancer assays in the mouse, Genomics (2015), doi: 10.1016/j.ygeno.2015.06.008
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Learning about mammalian gene regulation from functional enhancer assays in the mouse
PT
Alex S. Nord
RI
Center for Neuroscience, Department of Neurobiology, Physiology and Behavior, College of Biological Sciences, and Department of Psychiatry and Behavioral Sciences, School of Medicine, University of California, Davis, CA, USA
SC
Corresponding author: Alex S. Nord,
[email protected], 530-754-5022
MA
NU
Abstract
Enhancer biology is emerging as a critical area of research that informs studies of
D
evolution, development, and disease. From early experiments that defined and mapped
TE
the first enhancers to recent enhancer models of human disease, functional experiments in
AC CE P
the mouse have played a central role in revealing enhancer biology. Three decades of in vivo enhancer studies in mouse have laid the groundwork for the current understanding of mammalian enhancers, demonstrating the developmental and tissue-specific activity of enhancers and illuminating general features of enhancer evolution and function. Recent studies offer an emerging perspective on the importance of chromosomal context, combinatorial enhancer activity, and the functional consequences of enhancer sequence variation. This review describes the basic principles of functional testing in mouse, summarizes the contributions these studies have made to our understanding of enhancer biology, and describes limitations and future outlook of in vivo mouse enhancer studies.
ACCEPTED MANUSCRIPT Introduction
PT
Mouse models have advanced our understanding of the in vivo function of enhancer
RI
sequences, providing critical insight into how mammalian enhancer sequences function at the whole organism level, from their activity during development across complex tissues
SC
to their role in evolution and disease. Despite the significant strides in our understanding
NU
of enhancer function, much work remains, and functional studies in the mouse will be critical to our future understanding of the regulatory capacity and functional mechanisms
MA
of enhancers in mammalian systems. This review takes on three primary objectives. First, it describes basic principles of enhancer testing in the mouse. Next, it highlights mouse-
TE
D
based studies that have revealed the roles of enhancers in evolution, development, and disease. Finally, it describes new approaches and opportunities for using mouse-based
AC CE P
approaches to understand the mechanisms of enhancer activity.
Transgenic enhancer assays in mouse
How do enhancer assays work in mouse?
Kothary et al. showed that a construct using a minimal promoter taken from the Hsp68 heat-shock inducible gene paired to the E. coli Beta-galactosidase (lacZ) reporter gene could be used to screen for endogenous elements that drive expression [1,2]. This construct can be delivered via microinjection into fertilized mouse eggs and will stably integrate into the genome, enabling image-based activity screening of transgenic mice
ACCEPTED MANUSCRIPT (Figure 1). This approach can be used to “trap” endogenous regulatory elements via insertion in cis, enabling characterization of regulatory element function. The extension
PT
of this assay to enable candidate enhancer screening by fusing an enhancer sequence
RI
upstream of the minimal promoter and reporter gene has been used extensively to study the function of enhancers in vivo in mouse. In this assay, the enhancer drives expression
NU
thus enabling characterization of activity in vivo.
SC
of the reporter in the same structures as the endogenous version of the enhancer is active,
MA
While powerful, this functional enhancer assay is susceptible to position effects driven by insertion site, so multiple independent transgenic individuals must be screened.
TE
D
Furthermore, the rate of genome integration can be relatively low and screen requires many microinjected eggs to be generated. Thus, testing a candidate sequence in this assay
AC CE P
requires construct cloning and preparation, the collection and microinjection of tens to hundreds of fertilized eggs, surgical reimplantation, and eventual sample collection and histological preparation and scoring. As such, mouse transgenic assays require high levels of technical expertise, are expensive and require substantial time to perform, and have not been amenable to high-throughput analysis. Furthermore, activity differences between enhancers or alleles can be difficult identify due to variation driven by position effects, a limiting factor with regards to using this approach to compare subtle quantitative changes in function. Nonetheless, these experiments provide information that cannot be generated via other methods, enabling characterization of enhancer activity from the single cell to whole organism level, maintaining complex developmental context in a mammalian model system.
ACCEPTED MANUSCRIPT
PT
Results of mouse transgenic assays and the VISTA database
RI
Functional enhancer testing in the mouse has been critical to discovering and validating individual enhancers, as well as to understanding general principles of gene regulation
SC
and enhancer biology. A particular strength is the ability to characterize in vivo function
NU
in the context of complexity of heterogeneous tissues and developmental stages. Examples of early applications include studies of En2 in the central nervous system [3],
MA
Bmp5 across embryonic development [4], and Nkx-2.5 expression in the developing heart [5]. Transgenic enhancer assays in mouse have shown that enhancers drive highly
TE
D
specific and dynamic gene expression patterns that underlie mammalian development.
AC CE P
Whole mount staining can be performed to visualize enhancer activity without sectioning in early developmental embryos. Large-scale enhancer screens in mouse have to date taken advantage of this approach and focused on characterizing whole embryo activity at e11.5. The results from large-scale functional enhancer studies form the backbone of the VISTA enhancer browser (enhancer.lbl.gov) [6] a resource that contains whole mount images for over 2000 mouse and human sequences tested in vivo at e11.5. This resource has assisted the research community to identify enhancers for functional study, to recognize and select cell- or tissue-specific markers [7], and to link enhancer activity with genomic regions implicated in genetic studies of human disease [8,9].
ACCEPTED MANUSCRIPT
Cloning
Candidate enhancer
NU
SC
Microinjection
RI
PT
Enhancer trap
Genome integration enables enhancer-driven expression
TE
D
MA
Reimplantation
AC CE P
Transgenic e11.5 embryo screening to characterize activity
Figure 1. Schematic of transgenic mouse enhancer assays. Left: Enhancer Trap approach uses a minimal reporter (grey) and a reporter gene (blue) construct that, when microinjected into a fertilized mouse egg, inserts randomly in the genome, “trapping” the activity of nearby enhancers (orange and yellow blocks). Right: Enhancer screening using the same construct except with a candidate enhancer cloned upstream of the minimal promoter (red). When inserted in the genome, the candidate enhancer will drive expression of the reporter gene. Other nearby enhancers may also contribute to reporter gene expression, and multiple transgenic embryos must be screened to identify consistent patterns. After microinjection, eggs are surgically reimplanted, and transgenic mice are screened at desired timepoint. Example e11.5 transgenic embryos with enhancer-driven reporter gene expression selected from VISTA enhancer database (enhancer.lbl.gov).
ACCEPTED MANUSCRIPT What proportion of conserved sequences function as enhancers in vivo?
PT
The publication of the mouse and human genomes enabled global comparative genomics
RI
approachs to identifying mammalian regulatory sequences. These analyses highlighted strong evolutionary conservation across many non-coding regions, indicating the
SC
presence of a large number of regulatory elements [10]. Enhancer assays in the mouse
NU
have offered a means to annotate the function of these evolutionarily conserved regions [11-14]. Surprisingly, a large proportion (61%) of initial tested sequences that are human-
MA
fish conserved or contain at least 200bp perfectly conserved between human-mouse-rat drove reproducible expression patterns in specific e11.5 mouse tissues [13],
AC CE P
enhancers.
TE
D
demonstrating that highly conserved regulatory sequences often act as developmental
The 167 tested sequences drove expression across regions of the central nervous system (CNS), heart, and other organ systems as well as regions such as the tail, hind and forelimbs (Figure 2a). The widely varying patterns of gene expression driven by endogenous enhancers at e11.5 was also observed via enhancer trapping in mouse [15], though this approach fails to identify specific regions and can capture the concurrent activity of multiple enhancers in the local regulatory landscape. These studies show that a large proportion of conserved non-coding regions display in vivo enhancer activity, with the noteworthy detail that these strongly conserved sequences appear disproportionally active in early development and in the CNS (62% in the CNS, 38% in all other tissues).
ACCEPTED MANUSCRIPT A study comparing enhancers active across mouse development confirmed this finding that early developmental enhancers show higher relative evolutionary conservation and
PT
that forebrain enhancers active at early developmental stages exhibit particularly high
RI
levels of conservation [16]. In contrast, enhancers active in adult tissue and in the earliest developmental stages showed lower levels of evolutionary conservation, suggesting
SC
lower selective pressure and increased evolutionary novelty at these stages. These studies
NU
and others [17,18] capture shifting evolutionary pressures on regulatory elements across developmental time points and organ systems. The mechanisms underlying this
MA
surprising effect have yet to be explained, and this finding may in part explain the varying reports of enhancer conservation and turnover reported for different tissues and
TE
D
time points.
AC CE P
Mouse enhancer assays complement genome-wide functional predication methods
The development of massively parallel sequencing technology has enabled genome-wide mapping of the physical interactions between proteins and DNA, and this technology has been applied to predict enhancers in the mouse and human genome. Large-scale efforts to map enhancer activity via mapping of histones with particular modifications on tail residues, coactivator binding, transcription factor binding using ChIP-seq, short RNA expression (e.g. CAGE), and open chromatin domains (e.g. DNAse footprinting) have been hugely successful in mapping enhancers and predicting their activity [19-23]. Functional validation assays in mouse have confirmed that these methods are powerful tools to predict enhancers active in mouse embryonic tissues [24] and across time points
ACCEPTED MANUSCRIPT in development [16]. These genome-wide methods also provide insights into combinatorial presence of coactivators or corepressors and epigenomic modifications that
PT
can be tested in functional assays [25]. Functional mouse assays have been used to reveal
RI
the detailed in vivo expression patterns of particular sequences identified via bulk tissue ChIP-seq experiments, and have been used to finely map the expression domains of large
SC
numbers of human and mouse enhancer sequences in the developing forebrain, heart, and
NU
face [26-29] (Figure 2b).
MA
How correlated are genome-wide predictions and transgenic enhancer activity?
TE
D
In studies comparing signatures such as histone modifications or co-factor binding with enhancer activity in transgenic assays, it is clear that genome-wide predictions offer
AC CE P
imperfect sensitivity and specificity. For example, the co-activator p300 and the histone modification H3K27ac both show relatively high sensitivity and specificity for active enhancers in the developing mouse limb, yet these marks both fail to capture some active enhancers and to predict others that do not drive expression [30]. Estimates of the correlation between signatures such as p300-binding or H3K27ac and in vivo activity identified in transgenic enhancer assays vary by mark and by tissue. In some cases, the correlation can be very high, for example one study had validation rates around 80% for e11.5 brain and limb enhancers predicted using p300 [24]. In comparison, enhancers predicted in heart using a different antibody that targeted p300 and acCBP showed a rate of 62% success in driving heart expression in vivo at e11.5 [26].
ACCEPTED MANUSCRIPT Dynamic H3K27ac, the histone signature most tightly correlated with active enhancers, has not been as deeply examined. One recent study examined 18 predicted e11.5
PT
forebrain enhancers that showed differential H3K27ac across tissues, with 12 driving
RI
reproducible forebrain expression patterns [16]. This 66% validation rate is slightly decreased from the 78% observed for p300, though it is still strongly correlated with
SC
enhancer activity overall [16] and within the range of success rates observed for other
NU
signatures. It can be difficult to compare across experiments that test sets of predicted enhancers, as the majority of in vivo tested elements have been selected based on the
MA
strength of the prediction, only a limited number have been screened, and weaker predictions are likely to have a different rate of success. Studies are ongoing that directly
TE
D
test the correlation between specific proxy individual or combinatorial signatures, such as combinations of tissue-specific histone modifications and DNA methylation patterns, and
AC CE P
function in mouse transgenic assays. The results are expected to clarify interpretations of genome-wide enhancer prediction methods.
Further limitations of the basic in vivo enhancer testing strategy likely also contribute to the difference between predictions made using an epigenomic or transcriptomic proxy for enhancer activity and functional enhancer assays. It is possible that some sequences have endogenous enhancer activity, but show weakened or no activity in transgenic assays where the enhancer is no longer in the native chromatin context. There is variation across different enhancers tested in transgenic assays with regard to activity pattern reproducibility and penetrance across individual transgenic mice that carry the same candidate enhancer, suggesting that enhancer sequences vary in their ability to function
ACCEPTED MANUSCRIPT when inserted randomly in the mouse genome. Understanding both sequence-encoded and chromatin-dependent control of enhancer activity remains a primary goal of in vivo
RI
PT
studies of enhancer function in mouse.
SC
How do human enhancer sequences perform in mouse assays?
NU
One question regarding transgenic mouse studies is whether heterologous human enhancers generate the same activity pattern in the mouse. In an analysis of transgenic
MA
activity of heterologous human heart enhancers in mouse [27], cis-regulatory sequences taken from the human genome produced the predicted heart-associated expression
TE
D
patterns when inserted in the mouse genome. This suggests that there is a high level of conservation in the trans factors present during development, a finding upheld by a recent
AC CE P
large-scale comparative study [31]. Such conservation should enable heterologous screening of human regulatory sequences in the mouse. Despite the generally consistent expression patterns between orthologous mouse and human enhancers, evolutionary differences in transcription factor expression, binding site recognition, and regulatory and chromatin environment will potentially yield subtle but meaningful differences in activity.
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
Figure 2. Expression patterns driven by enhancers in the e11.5 mouse embryo. A. Whole-mount staining performed on transgenic embryos carrying an insertion of an enhancer fused to a minimal promoter and reporter gene. Each embryo shows a discrete activity pattern across various tissues. B. Sectioned telencephalon (forebrain) from forebrain-positive enhancers showing highly-restricted spatial activity patterns. Images courtesy of the VISTA enhancer browser (enhancer.lbl.gov).
ACCEPTED MANUSCRIPT In vivo enhancer studies of sequence variation and endogenous function
RI
PT
What are the in vivo phenotypic effects of sequence variation at enhancers?
Sequence variation within enhancers is an important feature of evolution and human
SC
disease. Studies have looked for non-coding evolutionary novelty in human or primate
NU
lineages, where there is recent sequence divergence within regions exhibiting strong conservation across the vertebrate phylogeny [32,33]. Where these enhancers are active
MA
in morphogenesis or in complex tissues, mouse studies have been particularly informative. For example, human-specific substitutions have been identified that confer
TE
D
changes in developmental limb expression as assayed in a transgenic mouse model [34]. Another example is a human-specific deletion that results in loss of function in an
AC CE P
enhancer that drives activity in sensory vibrissae and penile spines in other species [35]. In vivo enhancer studies have also revealed the differences in function of non-coding sequence variants present across human populations that are associated with complex human traits. For example, the two alleles of a polymorphism associated with blond hair were cloned into mice, revealing that the enhancer drives expression in the hair follicle and that the blond hair allele reduced the activation on the target gene and resulted in differences in pigmentation in a gain-of-function assay [36]. The better mapping of enhancer sequences in humans and close evolutionary relatives [37] and population level sequencing of full human genomes will spur additional studies of enhancers associated with human evolution and the regulation of complex human traits.
ACCEPTED MANUSCRIPT In vivo assays offer a powerful tool to test for activity in relevant tissues of enhancers harboring variants associated with human disease, where in vitro models may not capture
PT
the relevant effect. Recent examples include an enhancer near Irf6 active in the
RI
developing face that harbors variants associated with non-syndromic cleft lip [38], a Bcl11a enhancer active in the developing liver harboring variants associated with fetal
SC
hemoglobin [39], and an enhancer active in the FTO locus harboring obesity-associated
NU
variants that were shown to control expression in multiple tissue of the neighboring Irx3 gene [40]. Studies have linked differential enhancer activity in vivo in the mouse to
MA
specific rare mutations found via patient sequencing in cleft lip and autism [41,42]. The list of in vivo mouse experiments linking disease-associated variants to enhancer function
TE
D
is rapidly growing, and this approach will be vital to understanding the function of non-
AC CE P
coding regions associated with disease.
Enhancer deletion studies in mouse can reveal the phenotypic consequences of enhancer loss-of-function beyond changes in gene expression that are often examined in vitro, which may not have ramifications beyond steady state mRNA levels [43]. Many studies have examined the phenotypic consequences resulting from targeted enhancer deletions in the mouse, including at the Hox and Shh loci discussed below. Additional examples include an enhancer of H19/Igf2 that results in decreased size [44], HoxD enhancer leading to changes in vertebral morphology [45], a Fezf2 enhancer that leads to changes in brain structure [46], and craniofacial enhancers that lead to changed facial morphology [29]. Enhancer deletions can further reveal loss-of-function effects linked to human disease loci. Targeted deletion of a Myc enhancer harboring sequence variants associated
ACCEPTED MANUSCRIPT with a variety of human cancers resulted in decreased expression of Myc in intestinal crypts and in increased resistance to induced intestinal tumorigenesis [47]. These studies
PT
reveal that deletions of enhancers can have subtle to profound consequences on
RI
mammalian phenotypes.
SC
A surprising result of loss-of-function studies in mice has been that targeted deletion of a
NU
number of enhancers that are active in vivo, likely target developmentally critical genes, and show exceptional levels of evolutionary conservation resulted in no obvious
MA
phenotype [48]. One explanation for lack of a strong phenotype in this study is that endogenous enhancer redundancy via the presence of secondary or shadow enhancers
TE
D
confers increased robustness [49]. While it is also possible that subtle or inducible phenotypes that may be under strong selective pressure may have been missed, this
AC CE P
finding is striking and suggests that individual enhancer loss-of-function mutations may have minimal measurable effects, even for elements that are under strong purifying selection. This apparent paradox highlights the ground that remains to be covered in filling out our understanding of endogenous enhancer function.
Beyond transgenic assays: combinatorial and context-dependent enhancer activity
Control of gene regulation of several loci has been studied at great depth in the mouse, revealing principles of endogenous enhancer function, topology, and evolution that are not captured by transgenic assays alone. The Hox loci contain genes active in body patterning. Anchoring on regulation of Hoxd13 and using a combination of chromosome
ACCEPTED MANUSCRIPT interaction, epigenomic, and mouse transgenic and deletion analysis, enhancer-promoter interactions were mapped across the regulatory region flanking the gene. It was shown
PT
that full and robust expression requires the presence of the whole region, though
RI
individual fragments show transgenic activity in the mouse [50]. Work in cells and tissues have identified regional interaction maps dividing the genome into Topologically
SC
Associated Domains (TADs) that appear stable across cells and tissues [51,52]. The
NU
HoxD gene cluster is flanked by large regions containing regulatory sequences that are independently active enhancers, and a subset of HoxD genes switch between interaction
MA
with one topological domain in early limb bud formation to interaction with the second during digit formation, indicating that structural association within TADs can control the
AC CE P
promoters.
TE
D
endogenous activity of enhancers [53], presumably via limiting interaction with target
The Shh gene encodes a critical developmental morphogen and has a complex enhancer landscape across a large genomic region. Human point mutations within an enhancer in this locus lead to digit phenotypes [54], and extensive work has been done identifying the local sequence landscape and function of this enhancer [55]. Chromosomal inversion results in Shh loss of endogenous forebrain enhancer activity and gain of an exogenous limb enhancer, resulting in severe brain and limb defects [56]. In a separate experiment, an enhancer trap approach demonstrated that the enhancers in the region act to recapitulate the Shh expression pattern of the heterologous promoter inserted in many positions across the locus [57]. These studies are an example of the general features of enhancer loss and adoption, where changes in the genomic position of enhancers and
ACCEPTED MANUSCRIPT promoters can drive ectopic expression of genes that are not typically regulated by an
PT
enhancer.
RI
These and other studies show enhancer activity can be combinatorial, contributions of individual enhancer elements to endogenous gene expression are difficult to predict, and
SC
enhancer-promoter specificity can be dependent on local chromosome structure. Overall,
NU
regulatory landscapes are complex, with gene regulation and enhancer activity dependent on the function of multiple regulatory elements and local chromatin context and
MA
topology. Studies in mouse have been critical to understanding these concepts at the whole organism level. Dissecting this complexity and the potential cooperative activity of
TE
D
sets of enhancers is a focal area of current enhancer research.
AC CE P
Future development and application of functional mouse enhancer assays
The limitations of current strategies for functional enhancer screening in the mouse revolve around the cost- and resource-intensive nature of experiments in mice and questions of the fidelity of results between transgenic assays and endogenous enhancer activity. There are opportunities for improvement, for example via site-specific enhancer integration, towards the goal of increasing enhancer integration efficiency, reducing position effects, more faithfully recapitulating endogenous activity, and enabling quantitative evaluation of activity levels. While it remains unknown how enhancerpromoter specificity is achieved, use of additional promoters could avoid potential issues with potential enhancer-promoter incompatibility and improve on minimal promoter
ACCEPTED MANUSCRIPT background activity issues. Finally, high-throughput enhancer screening techniques are likely to be adapted to enable characterization of libraries of enhancers in vivo in mouse,
PT
albeit with a likely loss in information content regarding the detailed annotation of
RI
individual enhancers.
SC
Outside of further development of transgenic enhancer methods, the ability to manipulate
NU
sequence or function of an enhancer in vivo in the endogenous chromatin context will enable better understanding of how enhancers function in the complex regulatory
MA
landscapes of mammalian genomes. With the availability of targeted editing tools such as TALENs and Cas9/CRISPR that reduce the time and resources required to generate
TE
D
mouse models [58,59], many more studies will be performed on enhancer functionality in vivo where enhancers are deleted from the genome or specific regulatory variants are
AC CE P
knocked in. Given the widely varying phenotypic consequences of enhancer deletion, an expanded survey of enhancers functioning in different systems has the potential to reveal general characteristics that control the requirement or dispensability of enhancers. TALENs and inactivated chimeric Cas9 proteins have further been fused to regulatory effector domains to target enhancers [60-62], with early efforts showing that epigenetic signatures and transcriptional activity can be modulated by these synthetic transcription factors. While these technologies are still under development, the potential for application in vivo in mouse is promising.
The application of existing technologies for transgenic enhancer assays in combination with complementary approaches to characterize chromatin structure will continue to
ACCEPTED MANUSCRIPT reveal how mammalian regulatory wiring is achieved at the locus-specific and global level. Emerging technologies like Cas9/CRISPR are developing rapidly and likely will
PT
revolutionize enhancer studies as transgenic and sequencing-based technologies did in the
RI
past. Applying these and other tools to in vivo enhancer analysis in the mouse will allow us to answer the next series of outstanding questions facing the field today, refining our
SC
understanding of endogenous gene regulation and the role of enhancers in evolution,
AC CE P
TE
D
MA
NU
development, and disease.
ACCEPTED MANUSCRIPT Acknowledgements
PT
I thank colleagues for guidance, the anonymous reviewers for their help in improving the manuscript, and the VISTA enhancer resource for images used in the figures. ASN is
AC CE P
TE
D
MA
NU
SC
RI
supported by institutional funds from the UC Davis Center for Neuroscience.
ACCEPTED MANUSCRIPT References
[6] [7] [8] [9]
[10] [11] [12] [13] [14]
PT
RI
SC
NU
[5]
MA
[4]
D
[3]
TE
[2]
R. Kothary, S. Clapoff, A. Brown, R. Campbell, A. Peterson, J. Rossant, A transgene containing lacZ inserted into the dystonia locus is expressed in neural tube, Nature. 335 (1988) 435–437. doi:10.1038/335435a0. R. Kothary, S. Clapoff, S. Darling, M.D. Perry, L.A. Moran, J. Rossant, Inducible expression of an hsp68-lacZ hybrid gene in transgenic mice, Development. 105 (1989) 707–714. C. Logan, W.K. Khoo, D. Cado, A.L. Joyner, Two enhancer regions in the mouse En-2 locus direct expression to the mid/hindbrain region and mandibular myoblasts, Development. 117 (1993) 905–916. R.J. DiLeone, L.B. Russell, D.M. Kingsley, An extensive 3' regulatory region controls expression of Bmp5 in specific anatomical structures of the mouse embryo, Genetics. 148 (1998) 401–408. R.D. Searcy, E.B. Vincent, C.M. Liberatore, K.E. Yutzey, A GATA-dependent nkx-2.5 regulatory element activates early cardiac gene expression in transgenic mice, Development. 125 (1998) 4461–4470. A. Visel, S. Minovitsky, I. Dubchak, L.A. Pennacchio, VISTA Enhancer Browser--a database of tissue-specific human enhancers, Nucleic Acids Res. 35 (2007) D88–92. doi:10.1093/nar/gkl822. A. Schmoldt, H.F. Benthe, G. Haberland, Digitoxin metabolism by rat liver microsomes, Biochem. Pharmacol. 24 (1975) 1639–1641. doi:10.1016/j.neuron.2014.04.014. M. Sanchez-Castro, C.T. Gordon, F. Petit, A.S. Nord, P. Callier, J. Andrieux, et al., Congenital Heart Defects in Patients with Deletions Upstream of SOX9, Hum. Mutat. 34 (2013) 1628–1631. doi:10.1002/humu.22449. C.T. Gordon, C. Attanasio, S. Bhatia, S. Benko, M. Ansari, T.Y. Tan, et al., Identification of Novel Craniofacial Regulatory Domains Located far Upstream of SOX9 and Disrupted in Pierre Robin Sequence, Hum. Mutat. 35 (2014) 1011–1020. doi:10.1002/humu.22606. M.G.S. Consortium, R.H. Waterston, K. Lindblad-Toh, E. Birney, J. Rogers, J.F. Abril, et al., Initial sequencing and comparative analysis of the mouse genome, Nature. 420 (2002) 520–562. doi:10.1038/nature01262. M.A. Nobrega, I. Ovcharenko, V. Afzal, E.M. Rubin, Scanning human gene deserts for long-range enhancers, Science. 302 (2003) 413–413. doi:10.1126/science.1088328. S. Prabhakar, F. Poulin, M. Shoukry, V. Afzal, E.M. Rubin, O. Couronne, et al., Close sequence comparisons are sufficient to identify human cis-regulatory elements, Genome Research. 16 (2006) 855–863. doi:10.1101/gr.4717506. L.A. Pennacchio, N. Ahituv, A.M. Moses, S. Prabhakar, M.A. Nobrega, M. Shoukry, et al., In vivo enhancer analysis of human conserved non-coding sequences, Nature. 444 (2006) 499–502. doi:10.1038/nature05295. A. Visel, S. Prabhakar, J.A. Akiyama, M. Shoukry, K.D. Lewis, A. Holt, et al., Ultraconservation identifies a small subset of extremely constrained developmental enhancers, Nat. Genet. 40 (2008) 158–160.
AC CE P
[1]
ACCEPTED MANUSCRIPT
[19]
[22] [23] [24] [25]
[26] [27] [28]
AC CE P
[21]
TE
D
[20]
MA
[18]
NU
SC
[17]
PT
[16]
RI
[15]
doi:10.1038/ng.2007.55. S. Ruf, O. Symmons, V.V. Uslu, D. Dolle, C. Hot, L. Ettwiller, et al., Large-scale analysis of the regulatory architecture of the mouse genome with a transposon-associated sensor, Nat. Genet. 43 (2011) 379–386. doi:10.1038/ng.790. A.S. Nord, M.J. Blow, C. Attanasio, J.A. Akiyama, A. Holt, R. Hosseini, et al., Rapid and Pervasive Changes in Genome-wide Enhancer Usage during Mammalian Development, Cell. 155 (2013) 1521–1531. doi:10.1016/j.cell.2013.11.033. O. Bogdanović, A. Fernández Miñán, J.J. Tena, E. de la Calle-Mustienes, C. Hidalgo, I. van Kruysbergen, et al., Dynamics of enhancer chromatin signatures mark the transition from pluripotency to cell specification during embryogenesis, Genome Research. 22 (2012) 2043–2053. doi:10.1101/gr.134833.111. A.B. Stergachis, S. Neph, A. Reynolds, R. Humbert, B. Miller, S.L. Paige, et al., Developmental Fate and Cellular Maturity Encoded in Human Regulatory DNA Landscapes, Cell. 154 (2013) 888–903. doi:10.1016/j.cell.2013.07.020. E.P. Consortium, The ENCODE (ENCyclopedia Of DNA Elements) Project, Science. 306 (2004) 636–640. doi:10.1126/science.1105136. E.P. Consortium, I. Dunham, A. Kundaje, S.F. Aldred, P.J. Collins, C.A. Davis, et al., An integrated encyclopedia of DNA elements in the human genome, Nature. 489 (2012) 57–74. doi:10.1038/nature11247. R. Andersson, C. Gebhard, I. Miguel-Escalada, I. Hoof, J. Bornholdt, M. Boyd, et al., An atlas of active enhancers across human cell types and tissues, Nature. 507 (2014) 455–461. doi:10.1038/nature12787. F. Yue, Y. Cheng, A. Breschi, J. Vierstra, W. Wu, T. Ryba, et al., A comparative encyclopedia of DNA elements in the mouse genome, Nature. 515 (2014) 355–364. doi:10.1038/nature13992. Y. Shen, F. Yue, D.F. McCleary, Z. Ye, L. Edsall, S. Kuan, et al., A map of the cisregulatory sequences in the mouse genome, Nature. 488 (2012) 116–120. doi:10.1038/nature11243. A. Visel, M.J. Blow, Z. Li, T. Zhang, J.A. Akiyama, A. Holt, et al., ChIP-seq accurately predicts tissue-specific activity of enhancers, Nature. 457 (2009) 854–858. doi:10.1038/nature07730. C. Attanasio, A.S. Nord, Y. Zhu, M.J. Blow, S.C. Biddie, E.M. Mendenhall, et al., Tissue-specific SMARCA4 binding at active and repressed regulatory elements during embryogenesis, Genome Research. 24 (2014) 920–929. doi:10.1101/gr.168930.113. M.J. Blow, D.J. McCulley, Z. Li, T. Zhang, J.A. Akiyama, A. Holt, et al., ChIP-Seq identification of weakly conserved heart enhancers, Nat. Genet. 42 (2010) 806–810. doi:10.1038/ng.650. D. May, M.J. Blow, T. Kaplan, D.J. McCulley, B.C. Jensen, J.A. Akiyama, et al., Large-scale discovery of enhancers from human heart tissue, Nat. Genet. 44 (2011) 89–93. doi:10.1038/ng.1006. A. Visel, L. Taher, H. Girgis, D. May, O. Golonzhka, R.V. Hoch, et al., A highresolution enhancer atlas of the developing telencephalon, Cell. 152 (2013)
ACCEPTED MANUSCRIPT
[29]
[35] [36] [37]
[38] [39]
[40]
[41]
[42]
SC
NU
MA
[34]
D
[33]
TE
[32]
AC CE P
[31]
RI
PT
[30]
895–908. doi:10.1016/j.cell.2012.12.041. C. Attanasio, A.S. Nord, Y. Zhu, M.J. Blow, Z. Li, D.K. Liberton, et al., Fine tuning of craniofacial morphology by distant-acting enhancers, Science. 342 (2013) 1241006–1241006. doi:10.1126/science.1241006. J. Cotney, J. Leng, S. Oh, L.E. DeMare, S.K. Reilly, M.B. Gerstein, et al., Chromatin state signatures associated with tissue-specific gene expression and enhancer activity in the embryonic limb, Genome Research. 22 (2012) 1069–1080. doi:10.1101/gr.129817.111. A.B. Stergachis, S. Neph, R. Sandstrom, E. Haugen, A.P. Reynolds, M. Zhang, et al., Conservation of trans-acting circuitry during mammalian regulatory evolution, Nature. 515 (2014) 365–370. doi:10.1038/nature13972. K.S. Pollard, S.R. Salama, B. King, A.D. Kern, T. Dreszer, S. Katzman, et al., Forces shaping the fastest evolving regions in the human genome, PLoS Genet. 2 (2006) e168. doi:10.1371/journal.pgen.0020168. S. Prabhakar, J.P. Noonan, S. Pääbo, E.M. Rubin, Accelerated evolution of conserved noncoding sequences in humans, Science. 314 (2006) 786–786. doi:10.1126/science.1130738. S. Prabhakar, A. Visel, J.A. Akiyama, M. Shoukry, K.D. Lewis, A. Holt, et al., Human-specific gain of function in a developmental enhancer, Science. 321 (2008) 1346–1350. doi:10.1126/science.1159974. C.Y. McLean, P.L. Reno, A.A. Pollen, A.I. Bassan, T.D. Capellini, C. Guenther, et al., Human-specific loss of regulatory DNA and the evolution of humanspecific traits, Nature. 471 (2011) 216–219. doi:10.1038/nature09774. C.A. Guenther, B. Tasic, L. Luo, M.A. Bedell, D.M. Kingsley, A molecular basis for classic blond hair color in Europeans, Nat. Genet. 46 (2014) 748–752. doi:10.1038/ng.2991. S.K. Reilly, J. Yin, A.E. Ayoub, D. Emera, J. Leng, J. Cotney, et al., Evolutionary genomics. Evolutionary changes in promoter and enhancer activity during human corticogenesis, Science. 347 (2015) 1155–1159. doi:10.1126/science.1260943. F. Rahimov, M.L. Marazita, A. Visel, M.E. Cooper, M.J. Hitchler, M. Rubini, et al., Disruption of an AP-2alpha binding site in an IRF6 enhancer is associated with cleft lip, Nat. Genet. 40 (2008) 1341–1347. doi:10.1038/ng.242. D.E. Bauer, S.C. Kamran, S. Lessard, J. Xu, Y. Fujiwara, C. Lin, et al., An erythroid enhancer of BCL11A subject to genetic variation determines fetal hemoglobin level, Science. 342 (2013) 253–257. doi:10.1126/science.1242088. S. Smemo, J.J. Tena, K.-H. Kim, E.R. Gamazon, N.J. Sakabe, C. Gómez-Marín, et al., Obesity-associated variants within FTO form long-range functional connections with IRX3, Nature. 507 (2014) 371–375. doi:10.1038/nature13138. L. Poitras, M. Yu, C. Lesage-Pelletier, R.B. MacDonald, J.-P. Gagné, G. Hatch, et al., An SNP in an ultraconserved regulatory element affects Dlx5/Dlx6 regulation in the forebrain, Development. 137 (2010) 3089–3097. doi:10.1242/dev.051052. W.D. Fakhouri, F. Rahimov, C. Attanasio, E.N. Kouwenhoven, R.L.F. De Lima,
ACCEPTED MANUSCRIPT
[48] [49] [50] [51] [52] [53]
[54]
[55]
PT
RI
SC
NU
[47]
MA
[46]
D
[45]
TE
[44]
AC CE P
[43]
T.M. Felix, et al., An etiologic regulatory mutation in IRF6 with loss- and gain-of-function effects, Hum. Mol. Genet. 23 (2014) 2711–2720. doi:10.1093/hmg/ddt664. Z. Khan, M.J. Ford, D.A. Cusanovich, A. Mitrano, J.K. Pritchard, Y. Gilad, Primate transcript and protein expression levels evolve under compensatory selection pressures, Science. 342 (2013) 1100–1104. doi:10.1126/science.1242379. P.A. Leighton, J.R. Saam, R.S. Ingram, C.L. Stewart, S.M. Tilghman, An enhancer deletion affects both H19 and Igf2 expression, Genes Dev. 9 (1995) 2079–2089. J. Zákány, M. Gérard, B. Favier, D. Duboule, Deletion of a HoxD enhancer induces transcriptional heterochrony leading to transposition of the sacrum, Embo J. 16 (1997) 4393–4402. doi:10.1093/emboj/16.14.4393. S. Shim, K.Y. Kwan, M. Li, V. Lefebvre, N. Sestan, Cis-regulatory control of corticospinal system development and evolution, Nature. 486 (2012) 74–79. doi:10.1038/nature11094. I.K. Sur, O. Hallikas, A. Vähärautio, J. Yan, M. Turunen, M. Enge, et al., Mice lacking a Myc enhancer that includes human SNP rs6983267 are resistant to intestinal tumors, Science. 338 (2012) 1360–1363. doi:10.1126/science.1228606. N. Ahituv, Y. Zhu, A. Visel, A. Holt, V. Afzal, L.A. Pennacchio, et al., Deletion of ultraconserved elements yields viable mice, PLoS Biol. 5 (2007) e234. doi:10.1371/journal.pbio.0050234. M. Lagha, J.P. Bothma, M. Levine, Mechanisms of transcriptional precision in animal development, Trends Genet. 28 (2012) 409–416. doi:10.1016/j.tig.2012.03.006. T. Montavon, N. Soshnikova, B. Mascrez, E. Joye, L. Thevenet, E. Splinter, et al., A regulatory archipelago controls Hox genes transcription in digits, Cell. 147 (2011) 1132–1145. doi:10.1016/j.cell.2011.10.023. J.R. Dixon, S. Selvaraj, F. Yue, A. Kim, Y. Li, Y. Shen, et al., Topological domains in mammalian genomes identified by analysis of chromatin interactions, Nature. 485 (2012) 376–380. doi:10.1038/nature11082. B.D. Pope, T. Ryba, V. Dileep, F. Yue, W. Wu, O. Denas, et al., Topologically associating domains are stable units of replication-timing regulation, Nature. 515 (2014) 402–405. doi:10.1038/nature13986. G. Andrey, T. Montavon, B. Mascrez, F. González, D. Noordermeer, M. Leleu, et al., A switch between topological domains underlies HoxD genes collinearity in mouse limbs, Science. 340 (2013) 1234167. doi:10.1126/science.1234167. L.A. Lettice, S.J.H. Heaney, L.A. Purdie, L. Li, P. de Beer, B.A. Oostra, et al., A long-range Shh enhancer regulates expression in the developing limb and fin and is associated with preaxial polydactyly, Hum. Mol. Genet. 12 (2003) 1725–1735. doi:10.1093/hmg/ddg180. L.A. Lettice, I. Williamson, P.S. Devenney, F. Kilanowski, J. Dorin, R.E. Hill, Development of five digits is controlled by a bipartite long-range cisregulator, Development. 141 (2014) 1715–1725. doi:10.1242/dev.095430.
ACCEPTED MANUSCRIPT
[61]
[62]
PT
RI
SC
NU
[60]
MA
[59]
D
[58]
TE
[57]
L.A. Lettice, S. Daniels, E. Sweeney, S. Venkataraman, P.S. Devenney, P. Gautier, et al., Enhancer-adoption as a mechanism of human developmental disease, Hum. Mutat. 32 (2011) 1492–1499. doi:10.1002/humu.21615. E. Anderson, P.S. Devenney, R.E. Hill, L.A. Lettice, Mapping the Shh longrange regulatory domain, Development. 141 (2014) 3934–3943. doi:10.1242/dev.108480. H. Wang, H. Yang, C.S. Shivalila, M.M. Dawlaty, A.W. Cheng, F. Zhang, et al., One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering, Cell. 153 (2013) 910–918. doi:10.1016/j.cell.2013.04.025. H. Yang, H. Wang, C.S. Shivalila, A.W. Cheng, L. Shi, R. Jaenisch, One-Step Generation of Mice Carrying Reporter and Conditional Alleles by CRISPR/Cas-Mediated Genome Engineering, Cell. 154 (2013) 1370–1379. doi:10.1016/j.cell.2013.08.022. E.M. Mendenhall, K.E. Williamson, D. Reyon, J.Y. Zou, O. Ram, J.K. Joung, et al., Locus-specific editing of histone modifications at endogenous enhancers, Nat. Biotechnol. 31 (2013) 1133–1136. doi:10.1038/nbt.2701. I.B. Hilton, A.M. D'Ippolito, C.M. Vockley, P.I. Thakore, G.E. Crawford, T.E. Reddy, et al., Epigenome editing by a CRISPR-Cas9-based acetyltransferase activates genes from promoters and enhancers, Nat. Biotechnol. (2015). doi:10.1038/nbt.3199. N.A. Kearns, H. Pham, B. Tabak, R.M. Genga, N.J. Silverstein, M. Garber, et al., Functional annotation of native enhancers with a Cas9-histone demethylase fusion, Nat. Methods. (2015). doi:10.1038/nmeth.3325.
AC CE P
[56]
ACCEPTED MANUSCRIPT Learning about mammalian gene regulation from functional enhancer assays in the mouse
PT
Alex S. Nord Research Highlights:
AC CE P
TE
D
MA
NU
SC
RI
Mouse transgenic enhancer assays reveal the complex developmental patterns of enhancer activity in vivo. This review describes the transgenic enhancer assay and highlights areas of enhancer biology research where mouse studies have been particularly informative.