Manipulation of the human tRNA pool reveals distinct tRNA sets that act in cellular proliferation or cell cycle arrest
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Different subsets of the tRNA pool in human cells are expressed in different cellular conditions. The ‘proliferation-tRNAs’ are induced upon normal and cancerous cell division, while the ‘differentiation-tRNAs’ are active in non-dividing, differentiated cells. Here we examine the essentiality of the various tRNAs upon cellular growth and arrest. We established a CRISPR-based editing procedure with sgRNAs that each target a tRNA family. We measured tRNA essentiality for cellular growth and found that most proliferation-tRNAs are essential compared to differentiation- tRNAs in rapidly growing cell lines. Yet in more slowly dividing lines, the differentiation-tRNAs were more essential. In addition, we measured the essentiality of each tRNA family upon response to cell cycle arresting signals. Here we detected a more complex behavior with both proliferation-tRNAs and differentiation tRNAs showing various levels of essentiality. These results provide the so-far most comprehensive functional characterization of human tRNAs with intricate roles in various cellular states.
Article activity feed
-
-
###This manuscript is in revision at eLife
The decision letter after peer review, sent to the authors on June 23, 2020, follows.
Summary
Based on their previous work showing that cell proliferation and differentiation are associated with distinct tRNA programs and codon usages, the authors employed a CRISPR-Cas9 based approach to deplete families of tRNAs belonging to "proliferation" and "differentiation" groups and test the effects of such manipulation on the fitness of cells in different proliferative states. Using competition assays, the authors provide evidence that "proliferative" tRNAs are more essential in fast-proliferating cells, while "differentiation" tRNAs exert higher essentiality in slower proliferating cells. The authors also determined the essentiality of investigated tRNAs in senescent and quiescent cells which revealed …
###This manuscript is in revision at eLife
The decision letter after peer review, sent to the authors on June 23, 2020, follows.
Summary
Based on their previous work showing that cell proliferation and differentiation are associated with distinct tRNA programs and codon usages, the authors employed a CRISPR-Cas9 based approach to deplete families of tRNAs belonging to "proliferation" and "differentiation" groups and test the effects of such manipulation on the fitness of cells in different proliferative states. Using competition assays, the authors provide evidence that "proliferative" tRNAs are more essential in fast-proliferating cells, while "differentiation" tRNAs exert higher essentiality in slower proliferating cells. The authors also determined the essentiality of investigated tRNAs in senescent and quiescent cells which revealed more complex patterns. Overall, it was thought that this study is of broad potential interest inasmuch as it suggests that tRNAs have distinct essentiality in different cells and across distinct proliferative states. Moreover, it was found that this constitutes pioneering work wherein the effects of systematically knocking out tRNA genes are directly studied, an important milestone by itself when considering the abundance and variability of isodecoder species and the homology between isoacceptors. Notwithstanding the overall enthusiasm for the potential importance of the study and uniqueness of the approach, it was found that several major issues should be addressed to corroborate author's conclusions as outlined below.
Essential Revisions
It was thought that a number of important controls were missing. The potential off-target effects of CRISPR-Cas9 method need further validation. Fig S1B should be extended in order to clarify which sgRNAs are potentially off-targeting which tRNA. The manuscript would also benefit from experimentally testing the off-target effects of some of the sgRNAs, especially those binding to other tRNA families. To accurately compare HeLa cells with fibroblasts, the authors should determine potential tRNA expression and codon usage differences between them. Moreover, the efficacy of tRNA depletion between the cell lines should be assessed. Figure 5-additional controls should be provided to ascertain that cells are indeed in quiescent and senescent states. In Figure 5A, it should be explained why the 3 day time point was used when in the most of the study it is shown that the strongest effects occur after 7 days of induction.
Some experimental conditions remain unclear. For instance, it is noted that sgRNA plasmids were selected by puromycin, whereby WI38 cells appear to already be puromycin resistant. It is also not clear how were competition assays carried out in cell arrested states. In general, it was thought that the authors should be more specific regarding their read-outs (i.e. specify whether proliferation or survival were monitored).
Several issues were raised apropos statistical analyses. In figures 3C and D, to assess whether tested variables are truly independent, the authors should use a linear regression modelling Relative fitness ~ tRNA expression (in C) and Relative fitness ~ fraction CRISPR targeted tRNAs (in D). In addition, it is not clear why is z-transformation applied in figure 5E? The heatmap summarizes tRNA essentiality, which in figures 3 and 5C, is depicted using an untransformed log2FC. Using z-transformed and untransformed values to estimate the same effects was thought not to be advisable. Finally, the authors should also include the number of biological replicates, types of statistical tests and their outcomes in each figure where applicable, as in some cases these are missing.
Several statements were found not to be adequately supported by the data. For example, the statement: "our results show that some tRNAs are essential specifically for cancerous cells and not in differentiated cells ... (and the next sentence)", was found not to be supported by the presented data. To this end, the authors are advised either to provide data corroborating these conclusions or to tone down their statements. Also, in discussion section, given that this work is the first in systematically knocking out tRNA gene families, some comment on the potential and limitations of the method appears to be warranted.
-
