Not-so-selfish DNA? Intronic enhancers fine-tune spatiotemporal gene expression

In multicellular organisms, the genomic content of all cells remains similar, yet different cell types acquire different functions due to differences in gene expression. Dynamic gene expressionis achieved by cell- or tissue-specific transcription factors binding to cis-regulatory elements or enhancers. Eukaryotic genes are characterized by the presence of introns and exons; during transcription, the introns are removed or “spliced out” and the exons are joined together to make the mature RNA that encodes the protein. Introns constitute 25% of genomic DNA in humans and were considered as junk or “selfish DNA” until their functional benefits were discovered (Jo and Choi, 2015). Intronic enhancers were first identified in mouse immunoglobulin chains (Banerji et al., 1983) and are characterized by the lack of orientation to the promoter and cell type-specific.

However, only a handful of intronic enhancers have been identified in plants and there is a need for systemic identification and validation to better understand the dynamics of gene expression. In this issue, Meng et al. (2021) performed a genome-wide search for intronic enhancers in Arabidopsis thaliana seedlings and floral tissue. The authors used open chromatin signatures based on DNase-Seq (Zhu et al, 2015) for enhancer prediction followed by validation with transgenic assays and CRISPR/Cas-based genome editing. DNase-Seq identifies DNase I hypersensitive sites (DHSs), genomic regions that are hypersensitive to DNase I, which cleaves nucleosome-depleted DNA (mostly bound by proteins). DHSs can be identified by sequencing. Meng et al. called an intronic enhancer only if 80% of the DHS sequence overlapped with the intron. Using this approach, the authors identified a total of 941 and 1271 candidate intronic enhancers, associated with 806 and 1,069 genes in seedlings and floral tissues, respectively.

The authors found most of the candidate intronic enhancers to be located close to the promoter, with their frequency decreasing after the third intron. Intriguingly, genes with DHSs in both promoter and intronic regions showed higher expression compared to genes with DHSs in only one of these locations. Also, 17.5% of the intronic DHSs in seedling tissue were absent in floral tissue, whereas 39.1% of intronic DHSs in floral tissue were absent in seedling tissue, indicating tissue specificity of these DHSs. The authors also validated 21 intronic DHSs by transgenic assays (see figure). In short, they cloned DHSs (ranging from 173 to 778 bp) into a vector with a minimal 35S promoter driving the GUS reporter gene. The minimal 35S promoter cannot drive GUS expression on its own unless the upstream DHS acts as an enhancer. Using this strategy, the authors confirmed that 15 of the 21 DHSs tested (71%) act as intronic enhancers independent of their orientation to the promoter.

Figure.

Validation of predicted intronic DHSs using transgenic assays. Expression patterns of predicted DHSs in seedlings (left) and flowers (right) with 35s enhancer (35sEn) and 35S promoter (mini 35s) driving GUS expression as a positive and negative control, respectively. The center panel shows a heat map of GUS expression analysis from predicted DHSs in the different tissues (labeled on top) with heat map scalebar ranging from 0% (no expression) to 100% (strong expression). Adapted from Meng et al. (2021), Figure 2.

The authors analyzed in detail the intronic DHS7 in the TRIPTYCHON (TRY) gene. This gene encodes an MYB transcription factor involved in trichome and root hair development and in flowering time. Sequence analysis of DHS7 showed binding sites for the transcription factors SQUAMOSA-PROMOTER BINDING PROTEIN-LIKE 9 (SPL9) and ENHANCED DROUGHT TOLERANCE 1 (EDT1), involved in flowering time and trichome development, respectively. Two independent enhancers (DHS7#1, DHS7#2) were identified within DHS7. The authors validated these intronic enhancers using CRISPR/Cas-based genome editing and confirmed that DHS7#1 and DHS7#2 affect TRY expression in different tissues and developmental stages. They also confirmed that these enhancers do not affect the expression of flanking genes. DHS7#1 and DHS7#2 contain the binding motifs for SPL9 and EDT1, respectively, and the authors further confirmed the function of these two motifs using yeast one-hybrid and luciferase assays.

Another intronic enhancer studied in detail was DHS44, found in the CELLULOSE SYNTHASE LIKE A10 (CSLA10) gene, encoding a protein for the synthesis of ß-linked mannan and glucomannan polysaccharides. Transgenic and deletion analysis using CRISPR/Cas showed that the DHS44 intronic enhancer is required for proper abscission of floral organs, via a specific effect on the regulation of CSLA10 expression in the floral abscission zone.

Thus, this paper identified intronic enhancers in seedling and floral tissues and validated enhancers using both transgenic assays and endogenous gene manipulation assays. The authors assert that the number of DHSs identified in this study is underrepresented, as it is difficult to capture rare cell-type-specific DHSs which are lost in the bulk analysis, and the strict threshold of 80% and the inability of DNase I to access the sterically hindered DNA eliminates some potential intronic enhancers.