Understanding how cis-regulatory function is encoded in DNA sequence using massively parallel reporter assays and designed sequences

Abstract

Genome-scale methods have identified thousands of candidate cis-regulatory elements (CREs), but methods to directly assay the regulatory function of these elements on a comparably large scale have not been available. The inability to directly test and perturb the regulatory activity of large numbers of DNA sequences has hindered efforts to discover how cis-regulatory function is encoded in genomic sequence. Recently developed massively parallel reporter gene assays combine next generation sequencing with high-throughput oligonucleotide synthesis to offer the capacity to test and mutationally perturb thousands of specifically chosen or designed cis-regulatory sequences in a single experiment. These assays are the basis of recent studies that include large-scale functional validation of genomic CREs, exhaustive mutational analyses of individual regulatory sequences, and tests of large libraries of synthetic CREs. The results demonstrate how massively parallel reporter assays with libraries of designed sequences provide the statistical power required to address previously intractable questions about cis-regulatory function.

Keywords: Enhancers; Genomics; Massively parallel reporter assays; cis-regulation.

Fig. 1

Constructing massively parallel reporter libraries with co-transcribed barcodes. A. Co-transcribed sequence barcode (BC) in the 3′ UTR of the reporter gene uniquely identifies the cis-regulatory element (CRE) driving reporter expression. In massively parallel reporter assays, reporter gene activity is detected by performing RNA-seq on the transcribed barcodes. The reporter gene itself is not measured in this version of the assay. B. Quantification of reporter activity. A pooled reporter library is transfected into a population of cells, followed by barcode sequencing of the RNA and DNA fractions. Reporter expression is determined by the number RNA sequence reads per barcode. To account for variable representation of different reporter constructs in the library, RNA barcode reads are normalized by DNA barcode reads. C. Construction of a reporter library using barcoded oligonucleotides synthesized on programmable microarrays. Barcoded oligonucleotides are cloned into a plasmid backbone (cloning step 1), following which a minimal promoter and reporter gene are cloned between the barcode (BC) and the cis-regulatory element (CRE). D. Construction of a reporter library using non-barcoded, randomly generated oligonucleotides. Oligonucleotides are cloned into a plasmid backbone (cloning step 1), followed by random cloning of barcodes into the plasmid constructs (cloning step 2). Barcode-oligonucleotide pairings are determined by sequencing, following which a minimal promoter and reporter gene are cloned between the cis-regulatory sequence and the barcode (cloning step 3).