Identification of regulatory targets of tissue-specific transcription factors: application to retina-specific gene regulation

Abstract

Identification of tissue-specific gene regulatory networks can yield insights into the molecular basis of a tissue's development, function and pathology. Here, we present a computational approach designed to identify potential regulatory target genes of photoreceptor cell-specific transcription factors (TFs). The approach is based on the hypothesis that genes related to the retina in terms of expression, disease and/or function are more likely to be the targets of retina-specific TFs than other genes. A list of genes that are preferentially expressed in retina was obtained by integrating expressed sequence tag, SAGE and microarray datasets. The regulatory targets of retina-specific TFs are enriched in this set of retina-related genes. A Bayesian approach was employed to integrate information about binding site location relative to a gene's transcription start site. Our method was applied to three retina-specific TFs, CRX, NRL and NR2E3, and a number of potential targets were predicted. To experimentally assess the validity of the bioinformatic predictions, mobility shift, transient transfection and chromatin immunoprecipitation assays were performed with five predicted CRX targets, and the results were suggestive of CRX regulation in 5/5, 3/5 and 4/5 cases, respectively. Together, these experiments strongly suggest that RP1, GUCY2D, ABCA4 are novel targets of CRX.

Figure 1

Schematic view of our approach to identification of regulatory target genes of retina-specific TFs.

Figure 2

(A) Correlation of summary scores between the comparisons of retina versus brain and retina versus ‘pooled’; x-axis is the score from the comparison of retina versus ‘pooled’ and the y-axis is from retina versus brain. The line is for perfect correlation and only used for eye guide. (B) False discovery rates for EST, SGE, microarray and integrated data sets; x-axis is the number of significant genes and y-axis is the genes falsely called significant.

Figure 3

Distribution of the positions of binding sites relative to the TSS. Negative values represent upstream regions. Approximately 2100 eukaryotic binding sites, extracted from the TRANSFAC database, were used for the calculation.

Figure 4

Venn diagram for the target genes for CRX, NRL and NR2E3. The binding motif logos for each factor are shown. The numbers in the parentheses represent the total number of predicted targets for each factor.

Figure 5

EMSA analysis of predicted CRX targets genes. Lanes 1, 3, 5, 7, 9, 11, 13 show the indicated free probes without CRX homeodomain (CRX-HD). Lanes 2, 4, 6, 8, 10, 12, 14 contain the indicated probe plus 20 ng of CRX-HD. Mobility shifts are evident for all the genes.

Figure 6

CRX transactivates RP1, GUCY2D and ABCA4 in transient transfection assays. (A) Schematic diagram showing the luciferase reporter constructs carrying upstream regions of RP1 (−86 to +33), GUCY2D (−134 to +64), ARR3 (−297 to +16), BBS4 (−151 to +33) and ABCA4 (−130 to +8) in the pGL2-basic vector. The positions of CRX core binding sites (TAAT) are labeled by crosses. (B) Transient transfection assays. GripTite 293 MSR cells were transfected with 0.2 μg of the indicated luciferase reporter construct shown in (A) and increasing amounts (0, 0.2, 1 or 2 μg) of the CRX expression vector pcDNA3.1/HisC-Crx. The fold stimulation was calculated relative to control transfections without pcDNA3.1/HisC-Crx. Error bars show the standard error, n = 3.

Figure 7

The promoter regions of Rp1, Gucy2d, Abca4 and Arr3 are occupied by CRX in vivo. ChIP analysis was performed on fresh murine retina using oligomer PCR primers corresponding to the upstream regions of the indicated genes. Lane 1, genomic DNA template; lane 2, no DNA control; lane 3, input DNA pre-immunoprecipitation; lane 4, immunoprecipitation with no antibody; lane 5, immunoprecipitation with anti-CRX antibody.