Generation of tRNA TU deletion strains and assessment of their essentiality
We systematically generated a set of tRNA deletions in E. coli MG1655 (DE3) by individually deleting each of the 43 tRNA TUs. In each strain, a single tRNA TU was replaced by a kanamycin resistant cassette (KanR) using Lambda Red homologous recombination24.
By considering the redundancy of tRNA genes and the wobble rules for decoding, we predicted the essentiality of each tRNA TU. To validate these predictions, we conducted experimental analysis. If a TU resisted disruption even after three attempts, it was identified as a candidate for an essential TU. On the other hand, if the TU could be disrupted within three attempts, it was classified as non-essential (Fig. 1; Table 1).
We found that 33 TUs could be deleted without the need for complementation on a plasmid, indicating that these TUs are non-essential. As expected, all non-essential tRNA TUs belong to multi-copy tRNA isoacceptor families. However, the remaining 10 TUs could only be deleted in the presence of a complementing tRNA plasmid expressing the TU of interest, suggesting that these 10 TUs are essential for E. coli survival. Complementing tRNA plasmid (for the individual plasmid names see Table S4) is a low-copy plasmid carrying the corresponding wild-type tRNA TU under the control of its natural promoter. As an exception, argU and rrnC TUs could only be successfully removed from the chromosome after their natural promoter had been exchanged by the strong lpp promoter on the complementing plasmid.
Among the essential tRNA TUs, we identified three units that carry the single tRNA gene for each of the following amino acids: tryptophan (trpT (ACG)), histidine (hisR (GUG)), and cysteine (cysT (ACG)). Additionally, we identified five units that belong to multicopy tRNA family containing all anticodons for a specific tRNA family (proM (UGG), thrU (UGU), glyT (UCC), argU (UCU), leuW (UAG), serT (UGA), leuZ (UAA), serV (GCU), metTU (CAU), glnUW (UUG), glnVX (CUG), argQZYV (ACG)). Additionally, lysT-valT-lysW-valZ-lysYZQ could only be deleted in the presence of a complementing plasmid, despite identical valine and lysine tRNAs present in valUXY-lysV, indicating that valUXY-lysV alone could not sustain the growth of E. coli. Surprisingly, the ileX TU also appears to be essential despite the presence of an ileY on the chromosome, which only differs from ileX at nucleotide positions 6 and 67, which do not play an important role in tRNA structure25,26. In all cases, successful deletion of the TU of interest was confirmed by the absence of respective tRNA TU from the locus by colony PCR (cPCR) (Figure S1) and further validated by whole-genome sequencing of the deletion strains.
Generation of tRNA TU deletion strains and determination of their essentiality. (A) Kanamycin resistant gene (KanR) is PCR amplified with appropriate overhangs. In each strain, a single tRNA TU is replaced by the KanR using homologous recombination. ptRNA plasmid is a low-copy plasmid that complements a wild-type tRNA TU targeted for deletion under the control of either its natural or the lpp promoter. Upon deletion of a non-essential tRNA TU, the cell remains viable, and colonies are formed on LB-Kan agar plates. Upon deletion of an essential tRNA TU, the cell becomes inviable, and no colonies form on LB-Kan agar plates. The tRNA TU must be complemented by the plasmid (ptRNA) carrying the corresponding TU before removing the TU from the chromosome. Finally, tRNA knockout strains are confirmed by colony PCR (cPCR) and further validated by whole genome sequencing (WGS). LB-Lysogeny broth, Kan-kanamycin, Cm-chloramphenicol, pKD46-Lambda Red recombinase expression plasmid.
Characterization of cellular fitness of tRNA deletion strains
To examine the cellular fitness of the individual tRNA deletion strains, we compared growth in minimal and rich media in combination with different temperatures (Figure S2). The growth of each knockout strain was qualitatively compared to the growth of the parental strain under the same conditions. Based on these comparisons, the strains were classified as not impaired, slightly impaired, or very impaired.
Most strains with an essential TU removed and sustained with a corresponding complementing tRNA plasmid showed minimal to no growth impairment under specific conditions. This suggests that the complementing tRNA plasmid was generally sufficient for optimal cell growth despite a difference in gene copy number. The majority (21/33) of non-essential tRNA deletion strains exhibited a similar growth phenotype to the parental strain, indicating the robustness of E. coli to tRNA TU deletion (Fig. 2A). However, the removal of valVW, alaWX, or metZWV resulted in a significant growth impairment under specific conditions. These strains were classified as very impaired. Despite retaining at least one copy of each tRNA gene family, these cells were unable to fully compensate for the loss of the tRNA TU by upregulating the remaining copies. However, growth was restored upon complementation with the corresponding tRNA TU on a plasmid (Fig. 2B, C), confirming that the observed growth defect was a result of tRNA loss. These results are consistent with prior studies, in which it was shown that growth impairment caused by tRNA gene copy removal can be rescued by introducing the corresponding wild-type tRNA on the plasmid9,27,28.
Examination of the growth of wild-type (wt) and tRNA deletion mutants under diverse growth conditions. Parental strain and tRNA deletion mutants were grown in the specified medium and temperature and growth was monitored by measuring OD600. (A) The number of strains with a given phenotype in different growth conditions is shown. (B) The growth curves of wt with empty pSEVA271 (wt-KanR; black circles), wt with empty pSEVA271 and pSEVA361 (wt-KanR-CmR; black squares), and very impaired non-essential knockout strains with (grey triangles) and without (red quares) the tRNA complementing plasmid. (C) Doubling times of corresponding strains and conditions were determined in the exponential growth phase. All growth experiments are performed in quadruplicates. An unpaired t-test was used to compare doubling times between tRNA knockout mutants and the wt (ns (non-significant), ** (p-value ≤ 0.01), *** (p-value ≤ 0.001), **** (p-value ≤ 0.0001)). Error bars indicate the standard deviation of the replicates. LB-Lysogeny broth, Kan-kanamycin, Cm-chloramphenicol, M9-minimal medium, Glc-glucose, ev-empty vector.
Transcriptome profiling
To understand the cellular response to the disturbance of tRNA levels caused by the elimination of a tRNA TU, we conducted an extensive transcriptome analysis. We focused on the deletion of alaWX and valVW TUs as their removal resulted in impaired growth in a minimal medium at 37 °C. We hypothesized that there would be an overlap of the up or downregulation of specific genes as a result of disruptions of tRNA levels. In addition, we wanted to examine the overall cellular response from the individual deletions.
We performed RNA-seq and differential gene expression (DGE) analysis to compare the transcriptomes of tRNA mutant strains with wild-type and considered genes that exhibited log2-fold change values ≥ + 1.5 or ≤-1.5 and false discovery rate (FDR) smaller than 0.05 as differentially expressed. Upon removal of the alaWX TU, we identified 530 differentially expressed genes, with 272 genes upregulated and 258 genes downregulated. Conversely, the removal of the valVW TU resulted in only 44 differentially expressed genes, with 37 genes upregulated, and 7 genes downregulated. We found 12 overlapping differentially expressed genes between the two samples (Fig. 3, Table S10-S13).
Within the 12 overlapping genes, 6 genes were involved in pilus assembly and protein folding. These genes included fimC, fimF, fimI, fimD, fimG (from the fimAICDFGH TU), and the gene for the transcriptional regulator fimZ. Four genes, specifically rrlG, rrlC, rrlB, and rrlA from the rrnG, rrnC, rrnB, and rrnA TUs respectively were upregulated. These genes encode 23S rRNAs, which play a crucial role in translation. In addition, flgB (from the flgBCDEFGHIJ TU) that forms the rod of the flagellar basal body, and fliA (from the fliAZ TU) encoding a minor sigma factor (σ28) responsible for the initiation of transcription of several genes involved in motility and flagellar synthesis were downregulated (Table 2). KEGG Orthology-Based Annotation System (KOBAS) pathway analysis of the differentially expressed genes in ∆alaWX and ∆valVW revealed the majority of perturbed pathways related to metabolism and biosynthesis. Notably, the “ribosome pathway” (KEGG pathway ko03010, which includes ribosomal proteins), was enriched in both ∆alaWX and ∆valVW. On the other hand, in ∆valVW only the flagellar assembly pathway was depleted, whereas ∆alaWX exhibited depletion of 12 different pathways (Table 3).
RNA-seq analysis reveals differentially expressed genes in response to perturbed tRNA levels. Wild-type (wt), ∆alaWX, and ∆valVW were grown in M9 + 0.5% glucose at 37 °C. Samples (n = 3) were harvested in the mid-log phase, total RNA was extracted, and RNA sequencing was performed. The differential expression of genes (DEGs) in ∆alaWX (A) and ∆valVW (B) versus wt was analyzed. Red and blue points mark differentially expressed genes respectively (genes that pass thresholds for false discovery rate < 0.05 and log2-fold change ≥ 1.5, ≤ -1.5). Grey points represent genes that did not meet the threshold and are thus categorized as non-differentially expressed genes. (C) The number of DEGs is shown for each strain, the overlap indicates the number of genes that were similarly regulated. These genes and their corresponding regulation are highlighted in (D).
The log2-fold changes of neither the remaining alanine nor the valine tRNA genes met the defined threshold of ≥ + 1.5 or ≤-1.5, thus excluding them from being classified as differentially expressed. Nevertheless, we analyzed these genes carefully, given their potential importance as integral components of the alanine and valine tRNA pools.
Based on the predicted matching pattern between codons and tRNAs, it is apparent that the cognate codon of tRNA alaWX (GGC), which is GCC, can also be translated using the alaV, alaT, and alaU (UGC) tRNAs through Wobble pairing (Figure S3). However, we did not observe any upregulation of the remaining alanine tRNA genes in ∆alaWX; instead, they exhibited slight downregulation (Fig. 4A). On the other hand, the cognate codon of valVW (GAC), which is GUC, can also be translated using the remaining valine tRNAs (valU, valX, valY, valT, valZ (UAC)) through wobbling. Interestingly, the remaining valine genes in ∆valVW were slightly upregulated compared to the wild-type strain (Fig. 4B). However, the expression of these backup genes was not sufficient to fully compensate for the loss of the removed TU in terms of growth in a minimal medium (Fig. 2).
Effects of alaWX and valVW TU deletion on alanine and valine tRNA pools. Wild-type (wt), ∆alaWX and ∆valVW were grown in M9 + 0.5% glucose at 37 °C. Samples (n = 3) were harvested in the mid-log phase, total RNA was extracted, and RNA sequencing was performed. Expression levels of the remaining alanine (A) and valine (B) tRNAs in ∆alaWX and ∆valVW respectively are shown (FDR < 0.0001).
- SEO Powered Content & PR Distribution. Get Amplified Today.
- PlatoData.Network Vertical Generative Ai. Empower Yourself. Access Here.
- PlatoAiStream. Web3 Intelligence. Knowledge Amplified. Access Here.
- PlatoESG. Carbon, CleanTech, Energy, Environment, Solar, Waste Management. Access Here.
- PlatoHealth. Biotech and Clinical Trials Intelligence. Access Here.
- Source: https://www.nature.com/articles/s41598-024-73407-7