Pathogenic variants in CRX have distinct cis-regulatory effects on enhancers and silencers in photoreceptors

Abstract

Dozens of variants in the gene for the homeodomain transcription factor (TF) cone-rod homeobox (CRX) are linked with human blinding diseases that vary in their severity and age of onset. How different variants in this single TF alter its function in ways that lead to a range of phenotypes is unclear. We characterized the effects of human disease-causing variants on CRX cis-regulatory function by deploying massively parallel reporter assays (MPRAs) in mouse retina explants carrying knock-ins of two variants, one in the DNA-binding domain (p.R90W) and the other in the transcriptional effector domain (p.E168d2). The degree of reporter gene dysregulation in these mutant Crx retinas corresponds with their phenotypic severity. The two variants affect similar sets of enhancers, and p.E168d2 has distinct effects on silencers. Cis-regulatory elements (CREs) near cone photoreceptor genes are enriched for silencers that are derepressed in the presence of p.E168d2. Chromatin environments of CRX-bound loci are partially predictive of episomal MPRA activity, and distal elements whose accessibility increases later in retinal development are enriched for CREs with silencer activity. We identified a set of potentially pleiotropic regulatory elements that convert from silencers to enhancers in retinas that lack a functional CRX effector domain. Our findings show that phenotypically distinct variants in different domains of CRX have partially overlapping effects on its cis-regulatory function, leading to misregulation of similar sets of enhancers while having a qualitatively different impact on silencers.

Figure 1.

Study overview. (A) Schematics of wild-type CRX protein and the p.R90W and p.E168d2 pathogenic variants, containing a point DNA-binding domain mutation or resulting in truncation of the transcriptional effector domain, respectively. (B) Outline of the experimental procedure for constructing and testing the CRE libraries. A library of CRE sequences is cloned upstream of a rod photoreceptor-specific Rhodopsin (“Rho”) minimal promoter driving expression of the DsRed fluorescent protein, with each CRE marked by a unique sequence barcode (BC) in the 3′ UTR. Each library was electroporated into retinal explants with the six indicated Crx genotypes. RNA was collected from the retinas, and transcript counts for each element were measured and normalized to abundance in the input DNA library to calculate transcriptional activity.

Figure 2.

CRE activity in retinas with different Crx genotypes. (A,B) Correlations of the transcriptional activity of each CRE with intact CRX motifs (A) or CRE with killed CRX motifs (B) in the indicated genotypes. Superimposed black cross symbols indicate the transcriptional activity of the 100 scrambled control sequences, and a yellow circle shows the activity of the basal Rho promoter alone. Using the scrambled control sequences to establish an empirical null distribution of CRE activity, we performed robust regression using a trimmed mean M-estimator (fit plotted as a dashed line) for each genotype comparison. The outer dotted lines indicate the 80th percentile of the residuals of the library CREs relative to the regression of the scrambled sequences. The outer vertical and horizontal lines indicate the thresholds for the “strong silencer” and “strong enhancer” classes, and the inner horizontal and vertical lines correspond to the basal Rho promoter alone. (C) Activity of the Rho basal promoter construct in each of the indicated genotypes. Each black dot indicates a different unique sequence barcode; the mean across barcodes is shown as a yellow dot. In all panels, transcriptional activity was adjusted for visualization purposes using a fixed pseudocount of 2 × 10⁻⁵; the out-of-bound region is indicated by the light gray stripes.

Figure 3.

Activities of CREs near genes enriched in rod or cone photoreceptors. (A) Distribution of activity classes in +/+ retina for 66 and 117 CREs within 100 kbp of cone- or rod-enriched genes, respectively, as defined by Ruzycki et al. (2015). (B) Fold change relative to +/+ of 66 CREs near cone-enriched genes. (C) Fold change relative to +/+ of 117 CREs near rod-enriched genes. In B and C, rows are sorted by fold change in −/−.

Figure 4.

Identification and characterization of silencers that convert to enhancers in Crx mutant genotypes. (A) Heatmap comparing the transcriptional activity of each CRE across the six Crx genotypes. In the top right (upper arrow), note a group of CREs with generally low activity in all but the E168d2/d2 and −/− genotypes. Along the bottom (lower arrow), note the stepwise reduction in activity with increasing phenotypic severity. CREs are sorted by the ratio of transcriptional activity in the +/+ and −/− genotypes. (B) Classifications across genotypes for CREs classified as strong silencers (left column of each heatmap) or weak silencers (right column) in wild-type retinas. (C) Transcriptional activity of wild-type or motif mutant CREs across genotypes for 234 silencers in +/+ that convert to enhancers in E168d2/d2 (converted), or the remaining 251 silencers in +/+ that do not convert to enhancers in E168d2/d2 (unconverted). A yellow dot indicates the transcriptional activity of the basal promoter alone in each genotype. (D) Predicted CRX occupancy of the same silencer subgroups as in panel C and CREs that act as enhancers in +/+. (E) Information content of the same subgroups as panel D. (*) P < 0.05, (**) P < 0.01, (****) P < 0.0001 via two-sided Mann–Whitney–Wilcoxon test. In panel C, transcriptional activity was adjusted for visualization purposes using a fixed pseudocount of 2 × 10⁻⁵; the out-of-bound region is indicated by the light gray stripes.

Figure 5.

CRE activities analyzed by chromatin annotations at sites of genomic origin. (A) Distribution of CRE activity classes in +/+ retina separated into groups based on chromatin annotations of the site of genomic origin and CRX-dependent ATAC accessibility (Ruzycki et al. 2018). (Group A) H3K4me3⁺ H3K27ac⁺, (Group C) H3K4me3⁻ H3K27ac⁺, (Group D) H3K4me3⁻ H3K27ac⁻. (B) Transcriptional activity of all CREs, separated by chromatin and ATAC group, across all genotypes. (C) Chromatin classification breakdown of E168d2/d2 converted silencers (elements classified as weak or strong silencers in +/+ that become weak or strong enhancers in E168d2/d2). In panel B, transcriptional activity was adjusted for visualization purposes using a fixed pseudocount of 2 × 10⁻⁵; the out-of-bound region is indicated by the light gray stripes.

Figure 6.

A model of differential enhancer and silencer sensitivity to CRX mutants. Silencers, with higher predicted CRX occupancy, engage in effector domain–mediated homotypic interactions that permit CRX to remain bound even when the DBD is weakened by the p.R90W variant. The p.E168d2 variant lacks most of the effector domain and cannot engage in these interactions. Enhancers have lower CRX predicted occupancy and engage in fewer homotypic interactions.