Ultraconserved Enhancers Are Required for Normal Development

Abstract

Non-coding "ultraconserved" regions containing hundreds of consecutive bases of perfect sequence conservation across mammalian genomes can function as distant-acting enhancers. However, initial deletion studies in mice revealed that loss of such extraordinarily constrained sequences had no immediate impact on viability. Here, we show that ultraconserved enhancers are required for normal development. Focusing on some of the longest ultraconserved sites genome wide, located near the essential neuronal transcription factor Arx, we used genome editing to create an expanded series of knockout mice lacking individual or combinations of ultraconserved enhancers. Mice with single or pairwise deletions of ultraconserved enhancers were viable and fertile but in nearly all cases showed neurological or growth abnormalities, including substantial alterations of neuron populations and structural brain defects. Our results demonstrate the functional importance of ultraconserved enhancers and indicate that remarkably strong sequence conservation likely results from fitness deficits that appear subtle in a laboratory setting.

Keywords: Arx; brain development; enhancer; gene regulation; hippocampus; in vivo; knockout; neurons; noncoding; ultraconserved.

Figure 1. Selection of ultraconserved sites examined

Bar graph shows the length of ultraconservation for all non-exonic human-rodent ultraconserved sites. Regions containing multiple sequences of perfect human-mouse-rat conservation ≥200bp with less than 1 kb of intervening sequence were combined into a single ultraconserved site, with black or light gray tick marks indicating the length of the individual constitutive ultraconserved elements. Above: representative images showing the activity pattern (blue staining, pink arrows) of each telencephalon enhancer near Arx in an embryonic day 12.5 (E12.5) transgenic mouse embryo (left; Pennacchio et al., 2006) and specifically in the telencephalon in coronal forebrain sections (right; Visel et al., 2013). VISTA enhancer identifiers are indicated in blue text (hs numbers), and transgenic assay reproducibility is indicated in the bottom left corner of each panel. Arx is expressed in both the dorsal and ventral telencephalon at E12.5 (far right). ISH: in situ hybridization. Below: An overview of the Arx genomic locus, located on the X chromosome, showing the position of telencephalon enhancers selected for further study (blue ovals).

Figure 2. Ultraconserved enhancers near Arx are active in Arx-expressing cells

A) To identify the specific cell type each enhancer is active in, we generated E12.5 transgenic embryos containing each one of the enhancers driving a fluorescent reporter gene and performed single-cell RNA-Seq by Drop-Seq on isolated forebrain tissue. B) Cumulative results of single-cell RNA-Seq performed on E12.5 forebrain from all transgenic enhancer-reporter experiments. Each point indicates a unique single cell profiled (4,723 total), and cells were clustered by similarity of RNA expression. Red indicates cells where at least one Arx transcript was captured, and these points are scaled by decile of Arx expression, as shown in the legend. C) Same plot as B, color-coded to indicate cells where one of the ultraconserved telencephalic enhancers drove reporter gene expression. Points indicating mCherry expression are scaled by expression level decile (larger points indicate higher expression). See also Figure S1 and Tables S1–S2.

Figure 3. Single and double knockout of ultraconserved enhancers results in viable mice

A) Schematic diagrams of each enhancer knockout line generated using CRISPR/Cas9 editing in fertilized mouse eggs. Ultraconserved forebrain enhancers are shown as blue ovals, with red crosses indicating enhancer deletion alleles. Schematic locus representation is not to scale. B) In all cases, loss of individual or pairs of ultraconserved enhancers near Arx resulted in viable heterozygous female and hemizygous-null male mice (Figure S2) that did not show any obvious embryonic lethality or loss of fertility (Table S4). n.d.: not determined, n.s.: not significant. See also Figure S2 and Tables S3–S4.

Figure 4. Arx expression is diminished in enhancer deletion mice

A,B) RNA-Seq results for double knockouts of hs119/hs121 (A) and hs122/hs123 (B), showing all genes within a 20 Mb window around Arx. Whole forebrain from E11.5 embryos was profiled for each pair of deletions (n=2 wild-type, n=2 hemizygous-null males for each panel). Each point represents an individual gene, with its position on the X chromosome (x-axis) and significance of deviation from wild-type (y-axis) indicated. Genes with decreased expression in knockout embryos are plotted with −log₁₀(P) scores below 0. Dashed gray lines mark the positions of the enhancers. C) Arx expression (blue staining) profiled by in situ hybridization in coronal sections of E13.5 forebrain from embryos missing the enhancers active in the ventral forebrain. Embryos missing both enhancers in combination (hs119/hs121 KO) have substantial deficits of Arx in the ventral forebrain (black arrows). D) Arx expression in E12.5 forebrain from embryos missing enhancers active in the dorsal forebrain. Embryos missing both enhancers in combination have decreased Arx expression in the dorsal pallium (black arrows) and in the medial pallium (red arrows), with hs122-null embryos having similar changes. Scale bars: 500 μm; LGE: lateral ganglionic eminence, MGE: medial ganglionic eminence; DP: dorsal pallium, MP: medial pallium. See also Figures S3–S4, Table S5.

Figure 5. Loss of ultraconserved ventral forebrain enhancers results in growth and brain abnormalities

A) Growth curves for males wild-type (WT, black) or hemizygous null (KO, red) for the ventral forebrain enhancers. Solid lines show mean weight, with dashed lines indicating standard error of the mean (SEM). Sample sizes are shown on each plot. *, P < 0.05; two-tailed t-test. B) Representative coronal cross-sections of postnatal mouse brains stained for choline acetyltransferase (ChAT, black dots) from a wild-type control and males hemizygous-null (KO) for the enhancers active in the ventral forebrain. The loss of hs121 alone or hs119/hs121 combined results in decreased density of cholinergic neurons in the striatum (red box insets), whereas ChAT+ populations in other structures are unaffected (black box insets). Scale bar: 1 mm. C) Quantification of striatal cholinergic neuron density for ventral enhancer knockouts. Bars indicate group means, while individual points represent biological replicates. ***, P < 0.001; ANOVA. See also Figures S5–S6 and Table S6

Figure 6. Loss of ultraconserved dorsal forebrain enhancers results in abnormal brain development

A) Representative rostral coronal cross-sections of postnatal hippocampus from a wild-type control, along with males hemizygous null for the enhancers active in the dorsal forebrain. Individual loss of hs122 or combined loss of hs122/hs123 results in a smaller dentate gyrus (DG) with disorganized appearance (white arrowheads). CA1/3: hippocampal Cornu Ammonis 1 and 3. Scale bar: 250μm. B) Dentate gyrus length for dorsal forebrain enhancer knockouts normalized to wild-type littermate controls. Bars indicate group means, with individual points showing biological replicates. **, P < 0.01; ***, P < 0.001; two-tailed t-test. See also Figure S7 and Table S6.

Figure 7. Summary of phenotypes resulting from loss of ultraconserved Arx enhancers

n.s.: no significant changes observed.