Sourcing Phage-Encoded Terminators Using ONT-cappable-seq for SynBio Applications in Pseudomonas

Abstract

Efficient transcriptional terminators are essential for the performance of genetic circuitry in microbial SynBio hosts. In recent years, several libraries of characterized strong terminators have become available for model organisms such as Escherichia coli. Conversely, terminator libraries for nonmodel species remain scarce, and individual terminators are often ported over from model systems, leading to unpredictable performance in their new hosts. In this work, we mined the genomes of Pseudomonas infecting phages LUZ7 and LUZ100 for transcriptional terminators utilizing the full-length RNA sequencing technique "ONT-cappable-seq" and validated these terminators in three Gram-negative hosts using a terminator trap assay. Based on these results, we present nine terminators for E. coli, Pseudomonas putida, and Pseudomonas aeruginosa, which outperform current reference terminators. Among these, terminator LUZ7 T50 displays potent bidirectional activity. These data further support that bacteriophages, as evolutionary-adapted natural predators of the targeted bacteria, provide a valuable source of microbial SynBio parts.

Figure 1

ONT-cappable-seq delineates transcriptional boundaries. IGV data track of ONT-cappable-seq data of Pseudomonas phage LUZ7 5 min postinfection is shown. Only the leftmost region of the LUZ7 genome is shown. ONT-cappable-seq can accurately define transcriptional start sites (green arrows) and transcription terminators (red ‘T’) across the genome.

Figure 2

Library of phage terminators for Gram-negative hosts. (A) The efficiency of ONT-cappable-seq-identified phage terminators was analyzed with a terminator trap. The terminator is flanked by an upstream msfGFP reporter (green) and downstream mCherry reporter (red). Quantitative measurements of mCherry levels correspond to the level of read-through from the terminator, while msfGFP levels indicate the influence of the terminator on mRNA stability and translational efficiency of the upstream transcript. (B) The strength of each phage terminator is expressed as termination activity (%), by calculating the ratio of msfGFP and mCherry fluorescence levels, normalizing for the control construct (no terminator), and converting this value to percentage. (C) Termination activity (%) of all tested terminators in E. coli PIR2, P. aeruginosa PAO1, and P. putida KT2440. Bars and error bars display the mean and standard error of four biological replicates, respectively. Bold labels indicate predicted intrinsic terminators, whereas reference terminators are indicated with green bars. (D) Predicted phage terminator type vs in vivo termination activity. Comparison of the in vivo termination activity (%) between the phage terminators that were predicted to be factor-dependent or intrinsic, factor-independent in different bacterial hosts. The termination activity of the predicted intrinsic terminators is significantly higher than that of the putative factor-dependent terminators in all three hosts (Wilcoxon test, p < 0.0001)

Figure 3

LUZ7 T50 is a strong, bidirectional terminator. (A) The termination activity of the top nine phage terminators and reference terminators is assessed with the terminator trap in E. coli PIR2, P. aeruginosa PAO1, and P. putida KT2440 in the sense (gray) and antisense (green) orientations. Bars and error bars display the mean and standard error of four biological replicates, respectively. (B) ONT-cappable-seq data track of the late infection stage transcriptome of LUZ7 showing bidirectional transcriptional termination activity by terminator LUZ7 T50, located between a convergent gene pair. The LUZ7 terminator T50 sequence contains an A-tract and U-tract on either side of the hairpin (green color), suggesting the intrinsic bidirectional nature of the phage terminator.

Figure 4

Terminators with less mCherry read-through show increased msfGFP expression levels. (A) The msfGFP and mCherry expression levels of all phage terminators and reference terminators were assessed with the terminator trap in E. coli PIR2. Bars and error bars display the mean and standard error of four biological replicates, respectively. (B) A significant negative correlation between msfGFP expression levels and mCherry expression levels is observed for the tested terminators in E. coli PIR2. Dots represent the mean normalized fluorescence levels of four biological replicates for each terminator.