In vivo measurements reveal a single 5’ intron is sufficient to increase protein expression level in C. elegans

Abstract

Introns can increase gene expression levels using a variety of mechanisms collectively referred to as Intron Mediated Enhancement (IME). To date, the magnitude of IME has been quantified in human cell culture and plant models by comparing intronless reporter gene expression levels to those of intron-bearing reporter genes in vitro (mRNA, Western Blots, protein activity), using genome editing technologies that lacked full control of locus and copy number. Here, for the first time, we quantified IME in vivo, in terms of protein expression levels, using fluorescent reporter proteins expressed from a single, defined locus in Caenorhabditis elegans. To quantify the magnitude of IME, we developed a microfluidic chip-based workflow to mount and image individual animals, including software for operation and image processing. We used this workflow to systematically test the effects of position, number and sequence of introns on two different proteins, mCherry and mEGFP, driven by two different promoters, vit-2 and hsp-90. We found the three canonical synthetic introns commonly used in C. elegans transgenes increased mCherry protein concentration by approximately 50%. The naturally-occurring introns found in hsp-90 also increased mCherry expression level by about 50%. Furthermore, and consistent with prior results examining mRNA levels, protein activity or phenotypic rescue, we found that a single, natural or synthetic, 5 intron was sufficient for the full IME effect while a 3 intron was not. IME was also affected by protein coding sequence (50% for mCherry and 80% for mEGFP) but not strongly affected by promoter 46% for hsp-90 and 54% for the stronger vit-2. Our results show that IME of protein expression in C. elegans is affected by intron position and contextual coding sequence surrounding the introns, but not greatly by promoter strength. Our combined controlled transgenesis and microfluidic screening approach should facilitate screens for factors affecting IME and other intron-dependent processes.