Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2018 Jun 26;9(1):2475.
doi: 10.1038/s41467-018-04899-x.

Pooled CRISPR interference screening enables genome-scale functional genomics study in bacteria with superior performance

Tianmin Wang  1 , Changge Guan  1 , Jiahui Guo  1 , Bing Liu  2 , Yinan Wu  1 , Zhen Xie  3   4 , Chong Zhang  5   6 , Xin-Hui Xing  1   4
Affiliations
Comparative Study

Pooled CRISPR interference screening enables genome-scale functional genomics study in bacteria with superior performance

Tianmin Wang et al. Nat Commun. .

Abstract

To fully exploit the microbial genome resources, a high-throughput experimental platform is needed to associate genes with phenotypes at the genome level. We present here a novel method that enables investigation of the cellular consequences of repressing individual transcripts based on the CRISPR interference (CRISPRi) pooled screening in bacteria. We identify rules for guide RNA library design to handle the unique structure of prokaryotic genomes by tiling screening and construct an E. coli genome-scale guide RNA library (~60,000 members) accordingly. We show that CRISPRi outperforms transposon sequencing, the benchmark method in the microbial functional genomics field, when similar library sizes are used or gene length is short. This tool is also effective for mapping phenotypes to non-coding RNAs (ncRNAs), as elucidated by a comprehensive tRNA-fitness map constructed here. Our results establish CRISPRi pooled screening as a powerful tool for mapping complex prokaryotic genetic networks in a precise and high-throughput manner.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Overview of this work. a Proof-of-concept demonstration of the CRISPRi pooled screening for high-throughput functional genomics in E. coli. An sgRNA library targeting genes of interest is synthesized on a DNA microarray. Oligonucleotides are amplified and cloned into expression plasmids, transformed into E. coli expressing dCas9 protein, resulting in cell libraries. The cell libraries are grown under selective and control conditions. NGS libraries are constructed based on the extracted plasmids to determine the log2 change of each sgRNA between the selective and control conditions (sgRNA fitness). The sgRNA fitness distribution (red histogram) of each gene is compared with that of control sgRNAs (no target site in the E. coli genome; gray histogram) to evaluate the extent to which this gene is associated with relevant phenotypes (selective conditions). In the first part of this work, we designed a tiling sgRNA library composed of 2281 members targeting 44 genes with known phenotypes to evaluate the activities of these sgRNAs. b The absolute values of the Z scores for each sgRNA targeting the true positive genes were extracted, and the distribution of each group (categorized via position in the ORFs) against all 468 sgRNAs was quantified by a two-tailed MWU test. Triple asterisks indicate P < 0.01. For clarity, only sgRNAs with absolute Z scores of 0–4 were plotted. c The minimal number of sgRNAs per gene for reliable hit-gene calling. Results are shown for sampling of 3, 5, 10, 15, 20, and 30 sgRNAs for each gene (16 true positive genes in Library I). Two algorithms (position (Supplementary Fig. 4, see Methods) and random) were applied to determine the priority of sgRNA selection during sampling. Results are presented as box plots of the −Log10(P value) (MWU test) for the genes recalculated with the sampled sgRNA subset. Dashed line refers to P = 0.01. d In the second part of this work, we designed a genome-scale sgRNA library for E. coli based on the rules learned from the tiling experiment and performed screening experiments to test our methods
Fig. 2
Fig. 2
Distribution uniformity of the genome-wide sgRNA library. Comparison of distribution uniformity between the genome-scale sgRNA library and a benchmark Tn-seq transposon insertion dataset along the E. coli chromosome. The proportion of event number (either sgRNA or transposon insertion densities) within each window over all events across the entire E. coli MG1655 chromosome is shown per 10-kb sliding window
Fig. 3
Fig. 3
Predicting E. coli essential genes from CRISPRi screening. a Volcano plot of gene fitness and −Log10P value of two-tailed MWU test. Dashed lines represent a threshold (FDR = 0.05) for calling hits based on the screening score (see Methods). Four groups of genes are shown: blue, 313 essential genes from the Keio collection (Supplementary Data 11); gray, non-essential genes from the Keio collection and assigned as true negatives by CRISPRi screening; purple, false-positive “required” genes assigned by CRISPRi screening, potentially because of downstream Keio collection essential genes in operons transcribed as polycistronic mRNA; green, other genes found by CRISPRi screening to significantly inhibit cell growth. b and c ROC curves indicate the performances of different methods in identifying essential genes when considering all 4140 protein-coding genes (b) and 702 protein-coding genes with smaller coding regions (<400 bp) (c). True-positive rates and false-positive rates were calculated using the gold-standard set of essential and nonessential genes from the Keio collection. Shown are ROC curves for CRISPRi screening (55,671 sgRNAs) (red, CRISPRi score), a benchmark Tn-seq experiment (152,018 unique transposon insertions) (yellow, Wetmore et al.), a recently reported Tn-seq dataset with unprecedented-density transposon library size (901,383 members) (purple, Goodall et al.) and a widely used transposon insertion—based genetic footprinting dataset (blue, Gerdes et al.). The dashed line represents a random guess of the essential genes
Fig. 4
Fig. 4
CRISPRi screening maps ncRNA contributions to fitness. Volcano plot of ncRNA-coding gene fitness and −Log10P value of two-tailed MWU test. Dashed lines represent a threshold (FDR = 0.05) for calling hits based on the screening score (see Methods). Three groups of gene (each dot denotes a gene cluster with one or multiple genes, see Methods) are shown: purple, ribosomal RNAs; green, tRNAs; gray, other ncRNAs. Gene names for tRNAs that were not essential to E. coli identified by CRISPRi in rich medium and other ncRNAs reported to confer significant growth defects are given
Fig. 5
Fig. 5
CRISPRi screening dissects the E. coli metabolic network. a Volcano plot of gene fitness in MOPS medium relative to −Log10P value from the two-tailed MWU test. Dashed lines represent the threshold (FDR = 0.05) for calling hits based on the screening score (see Methods). The size of the scatter is proportional to the 1/OD600 value of the relevant gene knockout reported with the Keio collection. b GO enrichment analysis of auxotrophic genes identified by CRISPRi screening in MOPS medium. P values are derived from two-tailed Fisher exact test. c Comparison of gene fitness in MOPS medium with casamino acids (right) or a single carbon source (left) for E. coli. 78 genes responsible for amino acid biosynthesis are highlighted in green, whereas the five genes (trpABCDE) forming a branched pathway leading to tryptophan biosynthesis are in purple. The differences between fitness of amino acid biosynthesis genes and all other genes were quantified via the two-tailed MWU test (P, 10−10.0 with vs. 10−59.7 without casamino acid addition). d Schematic of E. coli amino acid biosynthetic network. Green and purple elements denote amino acid biosynthetic genes and tryptophan biosynthetic genes (trpABCDE), respectively, corresponding to those highlighted in (c)
Fig. 6
Fig. 6
Genes conferring tolerance phenotypes with respect to the toxic chemicals isobutanol (purple) and furfural (green). a, b Volcano plot of gene fitness in MOPS medium supplemented with 4 g/L isobutanol (a; FDR = 0.1, dashed line) and 0.4 g/L furfural (b; FDR = 0.05, dashed line). c Genomic plots of fitness for library variants in the presence of furfural (outer, green), isobutanol (middle, purple) and no supplementation (MOPS medium alone; inner, blue). The bars of each circle indicate the fitness for genes of relevant phenotypes. ncRNAs conferring significant phenotypes (‘+’ for tolerance and ‘−’ for sensitivity) (FDR < 0.05 or |fitness|>3 for auxotrophy in MOPS media and furfural tolerance, FDR < 0.1 or |fitness|>3 for isobutanol tolerance) under these conditions are highlighted
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Baba T, et al. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2006;2:2006.0008. doi: 10.1038/msb4100050. - DOI - PMC - PubMed
    1. Koo BM, et al. Construction and analysis of two genome-scale deletion libraries for Bacillus subtilis. Cell Syst. 2017;4:291. doi: 10.1016/j.cels.2016.12.013. - DOI - PMC - PubMed
    1. Freed EF, et al. Genome-wide tuning of protein expression levels to rapidly engineer microbial traits. ACS Synth. Biol. 2015;4:1244–1253. doi: 10.1021/acssynbio.5b00133. - DOI - PubMed
    1. Garst AD, et al. Genome-wide mapping of mutations at single-nucleotide resolution for protein, metabolic and genome engineering. Nat. Biotechnol. 2016;35:48–55. doi: 10.1038/nbt.3718. - DOI - PubMed
    1. van Opijnen T, Bodi KL, Camilli A. Tn-seq: high-throughput parallel sequencing for fitness and genetic interaction studies in microorganisms. Nat. Methods. 2009;6:767–772. doi: 10.1038/nmeth.1377. - DOI - PMC - PubMed

Publication types

MeSH terms

-