Transposon mutagenesis in Mycobacterium abscessus identifies an essential penicillin-binding protein involved in septal peptidoglycan synthesis and antibiotic sensitivity

  1. Chidiebere Akusobi
  2. Bouchra S Benghomari
  3. Junhao Zhu
  4. Ian D Wolf
  5. Shreya Singhvi
  6. Charles L Dulberger
  7. Thomas R Ioerger
  8. Eric J Rubin  Is a corresponding author
  1. Department of Immunology and Infectious Diseases, Harvard T.H. Chan School of Public Health, United States
  2. Department of Biology, Northeastern University, United States
  3. Department of Molecular and Cell Biology, University of California, Berkeley, United States
  4. Department of Computer Science and Engineering, Texas A&M University, United States
6 figures, 3 tables and 10 additional files

Figures

Figure 1 with 1 supplement
Overview of transposon sequencing (TnSeq) analysis.

(A) Schematic of TnSeq protocol. Phage carrying Himar1 transposon was used to transduce Mycobacterium abscessus (Mab) subsp. abscessus ATCC 19977 cultures. Over 51,000 independent transposon mutants were generated across three libraries. (B) Location of transposon insertions in Mab genome. Black lines represent the average number of transposon insertions per gene across the three replicates. (C) Breakdown of gene essentiality categories as determined by the hidden Markov model (HMM) in TRANSIT. (D) Essentiality comparison of mutual orthologs between Mab and Mycobacterium tuberculosis (Mtb). (E) Clusters of orthologous group (COG) categories of essential genes in Mab.

Figure 1—figure supplement 1
Correlation of Mycobacterium abscessus (Mab) transposon sequencing (TnSeq) libraries.

Pairwise comparisons of transposon insertions in each gene between two replicate libraries. Pearson correlation was calculated for each comparison and the r2 value is reported.

Figure 2 with 1 supplement
Validation of essential genes.

(A) Growth curve of Mycobacterium abscessus (Mab) strains with sgRNAs targeting canonical essential genes. (B) Growth curve of Mab strains with sgRNAs targeting uniquely essential Mab genes.

Figure 2—figure supplement 1
Validation of uniquely essential Mycobacterium abscessus (Mab) genes.

(A) Transformation of CRISPRi plasmids carrying sgRNAs targeting canonical and (B) uniquely essential genes in Mab. Equal volume of transformations was plated on +/-ATc (anhydrotetracycline) plates.

Figure 3 with 3 supplements
Penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) is uniquely essential in Mycobacterium abscessus (Mab) and non-essential in Mycobacterium tuberculosis (Mtb).

(A) Transposon insertion profile of PBP-lipo across three replicate transposon mutant libraries. Red bars indicate the number of transposon insertions. Horizontal black lines indicate ‘TA’ sites, where the transposon inserts. Genes are schematized with gray arrows, with PBP-lipo colored in blue. Dotted box demarcates lack of insertions in PBP-lipo. (B) Comparison of essentiality for PBP orthologs between Mtb and Mycobacterium smegmatis (Msm). (C) Growth of Mab cultures transformed with CRISPRi plasmid carrying either 2 or 3 sgRNAs targeting PBP-lipo. ‘-’ symbol indicates cultures were grown without anhydrotetracycline (ATc). ‘+’ symbol indicates cultures were grown with anhydrotetracycline (ATc). (D) Growth of wildtype Mtb and PBP-lipo knockout strains. (E) Growth of Mab with native PBP-lipo knockdown complemented by recoded (rc) PBP-lipo, which sgRNAs can no longer bind. PBP-lipo (rc) (S364A) is a recoded and catalytically inactive version of the enzyme.

Figure 3—figure supplement 1
Knockdown of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) impairs Mycobacterium abscessus (Mab) growth.

(A) Schematic of MAB_3167c, encoding PBP-lipo. Colored rectangles indicate sgRNAs-binding sites. (B) Fold change in mRNA expression measured by quantitative PCR (qPCR) for empty sgRNA (negative control) and 1, 2, and 3 sgRNAs targeting PBP-lipo. (C) Growth of Mab cultures transformed with CRISPRi plasmid carrying either 0 sgRNA (empty guide control) or 1 sgRNA targeting PBP-lipo. ‘-’ symbol indicates cultures were grown without anhydrotetracycline (ATc). ‘+’ symbol indicates cultures were grown with ATc. (D) Colony forming unit (CFU) of Mab with 1, 2, or 3 sgRNAs targeting PBP-lipo. Cultures were spotted on 7H10 plates +/-ATc.

Figure 3—figure supplement 2
Growth of wildtype Mycobacterium smegmatis (Msm) and penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) knockout strains.
Figure 3—figure supplement 3
MAB_3167c encodes a functional penicillin-binding protein (PBP) in Mycobacterium abscessus (Mab).

(A) Bocillin FL staining of PBP-lipo-strep. Wildtype Mab (wt) and a PBP-lipo-strep-tagged constitutive strain (CE) were stained with bocillin FL (B), a fluorescent penicillin analog and FITC (F) as a negative control. (Left) Blot shows PBPs bound to bocillin FL. (Right) Blot shows ɑ-strep Western blot for PBP-lipo-strep. (B) Western blot of lipoprotein (LP) and non-lipoprotein (non-LP) fractions of Mab. PBP-lipo-strep was detected in both fractions.

Figure 3—figure supplement 3—source data 1

Western blots detecting presence of PBP-lipo-strep.

https://cdn.elifesciences.org/articles/71947/elife-71947-fig3-figsupp3-data1-v1.zip
Figure 4 with 2 supplements
Knockdown of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) disrupts cell morphology.

(A) Microscopy images of PBP-lipo knockdown cultures. Mycobacterium abscessus (Mab) strains carrying CRISPRi plasmids with either 0, 1, 2, or 3 sgRNAs targeting PBP-lipo. Arrows indicate sites of multiple septa formation. (B) Cell lengths of uninduced and induced strains. Measurements were obtained by GEMATRIA and MOMIA image analysis pipelines (Zhu et al., 2021). Student’s t test used to calculate the statistical difference in mean cell lengths. ***p<0.0001, *p<0.05.

Figure 4—figure supplement 1
Morphology of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) knockdown cells.

(A) Number of branched cells per 100 cells visualized under light microscopy. ****p<0.001, ***p=0.001. (B) Cell width measurements of the 0, 1, 2, and 3 sgRNA PBP-lipo knockdown strains. ‘-’ symbolizes strains grown without anhydrotetracycline (ATc). ‘+’ indicates strains grown with ATc. PBP-lipo was induced for knockdown for 18 hr and cultures were visualized on the microscope and analyzed using the GEMATRIA and MOMIA programs (Zhu et al., 2021).

Figure 4—figure supplement 2
Knockout of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) does not alter morphology of Mycobacterium smegmatis (Msm) and Mycobacterium tuberculosis (Mtb) cells.

(A) (Left) Microscopy images of wildtype Msm and ΔPBP-lipo strain. (Right) Quantification of cell length. (B) (Left) Microscopy images of wildtype Mtb and ΔPBP-lipo mutant. (Right) Quantification of cell length. All morphological measurements were conducted using the GEMATRIA and MOMIA pipelines (Zhu et al., 2021).

Figure 5 with 1 supplement
Knockdown of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) disrupts formation of FtsZ rings.

(A) Microscopy images of N-terminally tagged mRFP-PBP-lipo and C-terminally tagged FtsZ-mNeonGreen. ‘Merged’ images show overlay of red and green channels. (B) (Left) Demograph of mRFP-PBP-lipo. (Right) Quantification of septal fluorescence signal across increasing cell lengths. (C) (Left) Demograph of mRFP-PBP-lipo and FtsZ-mNeonGreen. (Right) Fluorescence signal arranged by increasing cell length. (D) Images of GFP-FtsZ expressed from its natural promoter in the setting of PBP-lipo knockdown.

Figure 5—figure supplement 1
mRFP-PBP-lipo is a functional fusion protein.

(A) Schematic of mRFP-PBP-lipo-strep fusion protein. (B) Western blot using ɑ-strep antibody on strains expressing PBP-lipo-strep and mRFP-PBP-lipo-strep. (C) Growth curve showing complementation with mRFP-PBP-lipo fusion proteins.

Figure 5—figure supplement 1—source data 1

Western blot detecting presence of mRFP-PBP-lipo-strep.

https://cdn.elifesciences.org/articles/71947/elife-71947-fig5-figsupp1-data1-v1.zip
Figure 6 with 7 supplements
Penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) network in Mycobacterium abscessus (Mab) does not exist in Mycobacterium smegmatis (Msm).

(A) Growth curve of 2 sgRNA strain targeting PBP-lipo and 2 sgRNA strain constitutively expressing PBP-lipoMsm. (B) Microscopy images of 2 sgRNA and PBP-lipo complement strains. (C) Genetic synergy of PBP-lipo with other PBPs in Mab. Solid bar represents colony forming unit (CFU) of strains where a single PBP was knocked down. Open bar represents strains where PBP-lipo was knocked down in combination with the listed PBP. (D) Genetic synergy of PBPs and PBP-lipo in Msm. CRISPRi plasmids carrying sgRNAs targeting each of the PBPs in Msm were transformed into wildtype Msm (closed bar) and the ΔPBP-lipo mutant (open bar).

Figure 6—figure supplement 1
Structural comparison of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) across mycobacteria.

(A) Protein alignment of Mycobacterium abscessus (Mab), Mycobacterium smegmatis (Msm), Mycobacterium tuberculosis (Mtb), M. bovis, and M. leprae PBP-lipo homologs. Orange box indicates five additional amino acids ‘RGPAL’ present in Mab’s PBP-lipo. Residues highlighted in yellow indicate significant differences in amino acid identity between Mab and Mtb. (B) In silico generated model of Mab and Mtb PBP-lipo structures. Schematized β-lactam is depicted binding the predicted active site of PBP-lipo. Residues in orange highlight the additional ‘RPGAL’ amino acids present in Mab. Residues in yellow correspond to amino acid differences between Mab and Mtb PBP-lipo.

Figure 6—figure supplement 2
Cell length of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) knockdown and complement strains.
Figure 6—figure supplement 3
Genetic interactions of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) and PBPs.

(A) Genetic synergy of PBP-lipo/PbpB and PBP-lipo/DacB1 in liquid culture. (B) Growth curve of PBP-lipo knockdown overexpressing PbpB. (C) (Left) Microscopy images of 2 sgRNA PBP-lipo knockdown and PbpB complement strains. (Right) Cell lengths of 2 sgRNA PBP-lipo knockdown cells and PbpB complement strain.

Figure 6—figure supplement 4
DacB1 localizes to the septum of Mycobacterium abscessus (Mab) and co-localizes with penicillin-binding protein and hypothetical lipoprotein (PBP-lipo).

(A) Western blot using ɑ-strep antibody on strain expressing DacB1-GFP-strep. Estimated MW of DacB1-GFP is 72 kDa. (B) Microscopy images of DacB1 C-terminally tagged with GFP. White arrows point to septal localization of DacB1. (C) Microscopy images of N-terminally tagged mRFP-PBP-lipo and C-terminally tagged DacB1-GFP. ‘Merged’ images show overlay of red and green channels. (D) (Left) Demograph of mRFP-PBP-lipo and DacB1-mNeonGreen and (right) fluorescence signal arranged by increasing cell length.

Figure 6—figure supplement 4—source data 1

Western blot detecting presence of DacB1-GFPmut3.

https://cdn.elifesciences.org/articles/71947/elife-71947-fig6-figsupp4-data1-v1.zip
Figure 6—figure supplement 5
Genetic interactions of penicillin-binding proteins (PBPs) in Mycobacterium smegmatis (Msm) and Mycobacterium abscessus (Mab).

(A) Fold repression of genes with genetic synergy in Msm knockout and Mab knockdown. (B) Genetic synergy of Msm PBP-lipo with PBPs that lack homologs in Mab.

Figure 6—figure supplement 6
Knockdown of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) increases rate of calcein accumulation in Mycobacterium abscessus (Mab).

(A) Measurement of calcein signal in cultures that were induced for PBP-lipo knockdown with either 1, 2, or 3 sgRNAs. (B) Rate of calcein signal accumulation.

Figure 6—figure supplement 7
Knockdown of penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) impairs growth in clinical isolates.

(A) Colony forming unit (CFU) of Mycobacterium abscessus (Mab) clinical isolate with 3 sgRNA plasmid targeting PBP-lipo. Cultures were spotted on 7H10 plates +/-ATc (anhydrotetracycline). (B) Microscopy images ATCC 19977, T56, and BWH-B strains induced for PBP-lipo knockdown.

Tables

Table 1
Minimum inhibitory concentration (MIC) (µg/ml) of ATCC 19977 +/-penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) knockdown.
Antibiotic-ATc+ATcFold difference in MIC(-ATc/+ATc)
Cell wall
Ampicillin>2562>128×
Amoxicillin>2562>128×
Faropenem81
Cefoxitin82
Vancomycin82
Imipenem>64>64-
Ticarcillin>64>64-
Ceftazidime>64>64-
Cephalexin>64>64-
Ceftazidime>64>64-
D-Cycloserine>64>64-
Ethambutol>64>64-
Isoniazid>512>512-
Ribosome
Clarithromycin160.125128×
Erythromycin320.564×
Amikacin164
Clindamycin>6464-
RNA polymerase
Rifampicin32216×
DNA gyrase
Ofloxacin82
Other
Pyrazinamide>64>64-
Pretomanid>64>64-
Table 2
Minimum inhibitory concentration (MIC) (µg/ml) of Mycobacterium abscessus (Mab) clinical isolates +/-penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) knockdown (cell wall antibiotics).
AmpicillinAmoxicillin
Clinical isolate-ATc+ATc-ATc+ATc
ATCC 19977>2562>2562
T35>256<0.5>256<0.5
T37>256<0.5>256<0.5
T49>2564>2564
T50>2564>2564
T51>2562>2562
T53>2564>2562
T56>256<0.5>256<0.5
BWH-B>2568>2562
BWH-C>2564>2568
BWH-E>2564>2562
Table 3
Minimum inhibitory concentration (MIC) (µg/ml) of Mycobacterium abscessus (Mab) clinical isolates +/-penicillin-binding protein and hypothetical lipoprotein (PBP-lipo) knockdown (rifampicin and clarithromycin).
RifampicinClarithromycin
Clinical isolate-ATc+ATc-ATc+ATc
ATCC 19977322160.125
T3540.2520.25
T3716120.25
T4964420.25
T50321160.25
T51644160.5
T5332180.5
T568120.25
BWH-B321160.25
BWH-C322321
BWH-E161160.125

Additional files

Supplementary file 1

Summary of Mycobacterium abscessus subsp. abscessus ATCC 19977 transposon sequencing (TnSeq) libraries.

https://cdn.elifesciences.org/articles/71947/elife-71947-supp1-v1.docx
Supplementary file 2

List of essential genes in Mycobacterium abscessus, non-essential genes in Mycobacterium tuberculosis (Mtb).

https://cdn.elifesciences.org/articles/71947/elife-71947-supp2-v1.docx
Supplementary file 3

List of essential genes in Mycobacterium abscessus (Mab) with non-essential orthologs in Mycobacterium tuberculosis (Mtb).

https://cdn.elifesciences.org/articles/71947/elife-71947-supp3-v1.docx
Supplementary file 4

Transposon sequencing (TnSeq) summary of Mycobacterium smegmatis (Msm) libraries.

https://cdn.elifesciences.org/articles/71947/elife-71947-supp4-v1.docx
Supplementary file 5

Minimum inhibitory concentration (MIC) (µg/ml) of wildtype mc2155 and ΔPBP-lipo.

https://cdn.elifesciences.org/articles/71947/elife-71947-supp5-v1.docx
Supplementary file 6

Minimum inhibitory concentration (MIC) (µg/ml) of wildtype H37Rv and ΔPBP-lipo.

https://cdn.elifesciences.org/articles/71947/elife-71947-supp6-v1.docx
Supplementary file 7

Bacterial strains used.

https://cdn.elifesciences.org/articles/71947/elife-71947-supp7-v1.docx
Supplementary file 8

Plasmids used.

https://cdn.elifesciences.org/articles/71947/elife-71947-supp8-v1.docx
Supplementary file 9

Primers used.

https://cdn.elifesciences.org/articles/71947/elife-71947-supp9-v1.docx
Transparent reporting form
https://cdn.elifesciences.org/articles/71947/elife-71947-transrepform1-v1.pdf

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Chidiebere Akusobi
  2. Bouchra S Benghomari
  3. Junhao Zhu
  4. Ian D Wolf
  5. Shreya Singhvi
  6. Charles L Dulberger
  7. Thomas R Ioerger
  8. Eric J Rubin
(2022)
Transposon mutagenesis in Mycobacterium abscessus identifies an essential penicillin-binding protein involved in septal peptidoglycan synthesis and antibiotic sensitivity
eLife 11:e71947.
https://doi.org/10.7554/eLife.71947