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
. 2023 Jun 12;19(6):e11398.
doi: 10.15252/msb.202211398. Epub 2023 Mar 27.

A self-propagating, barcoded transposon system for the dynamic rewiring of genomic networks

Max A English #  1   2 , Miguel A Alcantar #  1   2 , James J Collins  1   2   3   4
Affiliations

A self-propagating, barcoded transposon system for the dynamic rewiring of genomic networks

Max A English et al. Mol Syst Biol. .

Abstract

In bacteria, natural transposon mobilization can drive adaptive genomic rearrangements. Here, we build on this capability and develop an inducible, self-propagating transposon platform for continuous genome-wide mutagenesis and the dynamic rewiring of gene networks in bacteria. We first use the platform to study the impact of transposon functionalization on the evolution of parallel Escherichia coli populations toward diverse carbon source utilization and antibiotic resistance phenotypes. We then develop a modular, combinatorial assembly pipeline for the functionalization of transposons with synthetic or endogenous gene regulatory elements (e.g., inducible promoters) as well as DNA barcodes. We compare parallel evolutions across alternating carbon sources and demonstrate the emergence of inducible, multigenic phenotypes and the ease with which barcoded transposons can be tracked longitudinally to identify the causative rewiring of gene networks. This work establishes a synthetic transposon platform that can be used to optimize strains for industrial and therapeutic applications, for example, by rewiring gene networks to improve growth on diverse feedstocks, as well as help address fundamental questions about the dynamic processes that have sculpted extant gene networks.

Keywords: Tn-Seq; functional genomics; gene regulatory network; laboratory evolution; transposon.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no conflict of interest. JJC is an editorial advisory board member. This has no bearing on the editorial consideration of this article for publication.

Figures

Figure 1
Figure 1. Engineered platform for continuous transposon‐mediated mutagenesis and gene regulatory network rewiring

  1. Our approach to genome evolution can be broken down into three core elements: (i) a mechanism to control transposase expression either from within the transposon (in cis) or from an independent pHelper plasmid (in trans); (ii) a DNA assembly platform to functionalize and barcode transposon variants, allowing in situ network rewiring and lineage tracking, respectively; and (iii) a parallelized approach to evolution focusing on comparing replicate lineages across different transposon variants and environmental conditions. These experiments can be coupled to a modified Tn‐Seq pipeline for genome‐wide insertion site identification.

  2. Transposase expression in trans enables the sequential insertion of transposons functionalized with orthogonal fluorescent reporter cassettes into the genome of E. coli MG1655 cells, as determined by flow cytometry (FITC, 488 nm; mCherry, 561 nm). The data represent ~2.7 × 105 cells from pooled colonies of two sequential transformations.

  3. A comparison of two strategies to titrate transposase expression levels from the pHelper plasmid utilizing either pBAD or pTet. We used relative CFU counts on selective plates as a proxy for insertion frequencies following transformation of chemically competent MDS42‐pHelper cells with the transposon donor suicide vector (R6K origin of replication, n = 2 technical replicates). The inducers (arabinose or ATc) were added after the heat shock, during the rescue in SOC/SOB (1 h, 37°C).

  4. The introduction of transposase and transposon elements into MDS42 cells lacking endogenous MGEs increases the rate of in vivo mutagenesis. As a proxy, we measured the proportion of replicate, bottlenecked cultures that develop resistance to the antibiotic D‐cycloserine (20 μM) via the spontaneous inactivation of the cycA gene (e.g., through transposon insertion; Fehér et al, ; Pósfai et al, ; Umenhoffer et al, 2010). End‐point ODs (40 h, 37°C) were measured for n = 24 colonies each from duplicate transformations, n = 24 colonies for the parental MDS42‐pHelper strain, and n = 48 for “wild‐type” MDS42. The adjusted P‐values are shown above the violin plots and correspond to FDR‐corrected two‐sided Wilcoxon signed‐rank tests between populations cultured with or without ATc. Tn, mariner transposon (kanR); Tn[sfGFP], mariner transposon harboring a green fluorescent protein (sfGFP) gene; Tn[GFP], mariner transposon harboring an mNeonGreen gene; Tnpase, himar1C9 transposase; ATc, anhydrotetracycline; ara, arabinose; Tn‐Seq, transposon insertion sequencing.

Source data are available online for this figure.
Figure 2
Figure 2. Validating the transposon‐dependent activation of a model, cryptic metabolic operon

  1. The adaptation of wild‐type E. coli isolates to growth on the carbon source arbutin involves the disruption of an H‐NS repressor binding site (site A) upstream of the positive regulator bglG by IS1 or IS5, and the subsequent activation of the structural genes bglF/B/H (Schentz, ; Hall, 1998). We introduced two transposon variants, Tn and Tn‐pOUT, into MDS42‐pHelper cells and cultured 384 parallel populations each derived from unique, single‐colony founders in media with an increasing proportion of arbutin. We observed an alternative insertion‐activation mechanism for Tn‐pOUT transposons at Site B within the operon.

  2. End‐point OD600 measurements (48 h, 1.0 g/l arbutin) for n = 48 replicates per condition.

  3. OD measurements normalized to minimum and maximum values for n = 48 replicates in a single condition (Tn‐pOUT + ATc, (B) orange points) showing the points at which high‐growth phenotypes emerge in 100% MT‐arbutin media (1.0 g/l). The blue line indicates the proportion of MT‐Arb in the growth media.

  4. Growth curves for two conditions (Tn‐pOUT and Tnpase only, both with ATc) in MT‐arbutin media (1.0 g/l), inoculated from the 24‐h time point in 95% MT‐arbutin media from (C). The curves from high‐growth replicates are highlighted in orange (n = 48 replicates each).

  5. Fold‐enrichment values for peaks identified using MACS3, based on aligned Tn‐Seq reads from the start‐point (TB, left) and end‐point (MT‐arbutin, right) cultures for a single founder colony. The initial insertion (yidJ) is the dominant peak before selection and is retained alongside multiple new, high‐intensity peaks postselection including bglG.

  6. By aggregating all high‐growth variants (n = 11) from the experiment in (B) and pairing start‐point and end‐point Tn‐Seq data, we used Bio‐Tradis (Barquist et al, 2016) to identify differentially enriched insertion loci (labeled points). Differential transposon enrichment analysis was performed using a negative binomial generalized linear model with Benjamini–Hochberg correction for multiple hypothesis testing (as implemented in the Bio‐Tradis toolkit).

  7. All (11/11) of these replicates exhibited a sharp insertion peak at a TA site upstream of bglF, suggesting the Tn‐pOUT mediated activation of the bglF/B/H operon. Two representative end‐point traces (teal and purple) are shown from the 11 replicates.

  8. Longitudinal sequencing enables the genome‐wide tracking of transposon movement. After selection, the transposons demonstrate both preservation of their initial insertion location (dashed arrows) and their convergence on the bgl operon (solid arrows). Additional intermediate insertions were also observed (solid, purple arrow near 1.7 Mbp) outside of the bgl operon. The top panel shows representative traces (teal and purple) corresponding to start‐point transposon insertions for two out of 11 replicates. The bottom panel shows end‐point insertions for the same, representative replicates. The insertion plots were generated by counting the number of NGS reads aligning to each position in the E. coli MDS42 genome and normalizing by the total number of reads per replicate. Tn, unmodified mariner transposon; Tn‐pOUT, mariner transposon with an outward‐facing pJ23104 promoter.

Source data are available online for this figure.
Figure 3
Figure 3. Screening for transposon‐dependent carbon source utilization phenotypes

  1. In a preliminary screen of 31 carbon sources, we assessed the adaptation frequencies of triplicate founder colonies for two transposon variants (Tn and Tn‐pOUT with pJ23104). As a negative control, we included three colonies from the MDS42 pHelper parental strain. The heatmap shows the maximum OD across the three replicates for each condition, with the color map scaled to a maximum OD cutoff of 1.0. We identified three distinct groups: no growth (18 conditions), universal growth (10 conditions), and transposon‐dependent growth (3 conditions).

  2. L‐serine utilization was observed in 1/3 of the Tn‐pOUT replicates (EVOL‐1). Further comparison of this isolate to MG1655 and MDS42 demonstrated its high tolerance to L‐serine levels, both with (left panel) and without (right panel) antibiotic selection for the transposon and transposase (n = 8 replicate colonies each).

  3. Detectable growth at even higher concentrations of L‐serine (up to 100 g/l) was observed in the presence of the preferred carbon source glucose (4.0 g/l). The data are from the same eight colonies as (B) grown under different conditions.

  4. A repeated screen focusing on L‐serine increased the number of unique founder replicates to 16. Reproducible evolution of L‐serine tolerance was only observed with the Tn‐pOUT transposon (14/16 replicates). OD measurements were taken after 50 h of growth on the third passage in M9 L‐serine media.

  5. Using Tn‐Seq data from four of these samples at both the start‐ and end‐points, we identified three genes from a single region showing high differential enrichment: pabB, yeaB, and sdaA. Differential transposon enrichment analysis was performed using a negative binomial generalized linear model with Benjamini–Hochberg correction for multiple hypothesis testing (as implemented in the Bio‐Tradis toolkit).

  6. Genome‐wide transposon insertion maps showing convergence from distinct founder peaks (top panel) on a single conserved location, postselection (bottom panel). High‐resolution alignments from two representative samples confirm that insertions near both pabB and yeaB are in the promoter region upstream of sdaA (L‐serine deaminase I, inset).

Source data are available online for this figure.
Figure 4
Figure 4. Modular assembly platform for transposon functionalization and barcode‐based lineage tracking

  1. A modified two‐layer Golden Gate assembly pipeline focusing on modular plasmid assembly (via BbsI) and promoter library insertion (via BsmBI) within the transposon enables rapid prototyping and transposon barcoding.

  2. Comparison of the evolutionary impacts of the two original transposons (Tn, gray; and Tn‐pOUT, pink; n = 24 per condition) with three second‐generation transposons (RB‐TnV2, green; RB‐TnV2_pJEx, teal/purple; RB‐TnV2_pLacO1/pL, blue/orange; n = 9 per condition), across the three carbon sources identified in Fig 3A. Each box represents a replicate culture, with each column derived from the same initial founder colony and each row corresponding to a unique carbon source. Colored cells indicate growth (end‐point OD600 > 0.2). Rows marked with a * symbol indicate contamination from RB‐TnV2_pJEx cultures as determined based on the unique sequencing barcode of each founder.

  3. OD600 measurements for each carbon source were used to track the emergence of growth phenotypes in cultures without ATc (top row) or with ATc (bottom row), and the transposon‐specific inducers CV or IPTG. We included two controls: MDS42 pHelper expressing transposase only, and the parental MDS42 strain (n = 12 per condition).

  4. Computational barcode demultiplexing from pooled sequencing runs for each unique variant‐environment combination enabled the reconstruction of insertion mutant lineages. For a single lineage, condition‐specific insertion spectra evolve from the initial founder insertion(s).

  5. By comparing independent lineages for a single carbon source‐inducer combination, convergent insertion loci emerge (black arrows).

  6. Differential enrichment of RB‐Tn‐Seq reads between paired start‐point and end‐point samples for independent lineages confirms the reproducible insertion sites. For MDS42 cells harboring transposons with a pLacO1 promoter in bMDG + IPTG, the two common peaks in (E) correspond to frdD and yqfB (likely activating ampC and bglA, respectively, inset).

  7. Representative read alignment traces from individual barcoded replicates showing two possible mechanisms of bglA activation under selection for growth on bMGD: upstream insertion of an RB‐TnV2_pLacO1/pL transposon with an outward‐facing pLacO1/pL promoter driving IPTG‐induced expression (upper panel), or direct insertion of an RB‐TnV2_pJEx transposon within a repressor of bglA (Deana et al, 2008) that likely disrupts protein function (lower panel). The accumulation of read alignments in the upper panel indicates that the transposon is oriented such that the outward‐facing promoter is driving expression of bglA.

  8. For RB‐TnV2_pLacO1/pL, growth phenotypes in bMDG emerge later in the absence of IPTG (C). Differential enrichment between paired IPTG+ and IPTG− replicates implicates insertions in lacI, likely rescuing the activation potential of the pLacO1/pL transposons (inset: read alignments in the lac operon for a single barcoded replicate population). Differential transposon enrichment analysis was performed using a negative binomial generalized linear model with Benjamini–Hochberg correction for multiple hypothesis testing (as implemented in the Bio‐Tradis toolkit).

  9. A possible model for the establishment of quasi‐constitutive, RB‐TnV2_pLacO1/pL transposon‐mediated gene activation via the disruption of the endogenous repressor lacI. Insertions upstream of bglA and ampC then recapitulate the adaptive insertions seen in the bMDG + IPTG condition (F). RB, random barcode; IR, inverted repeat; CV, crystal violet; IPTG, Isopropyl ß‐D‐1‐thiogalactopyranoside; bMDG, ß‐methyl‐D‐glucoside.

Source data are available online for this figure.
Figure 5
Figure 5. Dynamic environments with sequential carbon source selections drive the emergence of multisite insertions

  1. A schematic of the overall experiment, taking pre‐evolved RB‐TnV2_pJEx strains from the L‐serine + CV culture experiments in Fig 4C (n = 36 total) and reselecting on L‐serine, before transitioning to a new carbon source: ß‐methyl‐D‐glucoside. We performed a similar experiment in parallel with Tn‐pOUT strains pre‐evolved on L‐serine (n = 24).

  2. OD measurements for parallel populations of Tn‐pOUT (gray, n = 24) and RB‐TnV2_pJEx (purple, with CV, n = 36) derived from cultures pre‐evolved on L‐serine (Appendix Figs S9 and S10), and subsequently evolved toward growth on bMDG. In the absence of crystal violet, no growth was observed in the RB‐TnV2_pJEx cultures (teal, n = 36). Gaps indicate reseeding from the prior time point of the crystal violet‐induced cultures (arrows).

  3. Cross‐experiment lineages for two representative barcodes, tracking the distribution of transposon insertions after selection on L‐serine (Fig 4B and Appendix Fig S12) and after further selection on ß‐methyl‐D‐glucoside (B). Insets represent high‐resolution RB‐Tn‐Seq read alignments for the major peaks. RB, random barcode; IR, inverted repeat; CV, crystal violet; IPTG, Isopropyl ß‐D‐1‐thiogalactopyranoside; bMDG, ß‐methyl‐D‐glucoside.

Source data are available online for this figure.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. AbuOun M, Suthers PF, Jones GI, Carter BR, Saunders MP, Maranas CD, Woodward MJ, Anjum MF (2009) Genome scale reconstruction of a salmonella metabolic model: comparison of similarity and differences with a commensal escherichia coli strain. J Biol Chem 284: 29480–29488 - PMC - PubMed
    1. Aguilar C, Escalante A, Flores N, de Anda R, Riveros‐McKay F, Gosset G, Morett E, Bolívar F (2012) Genetic changes during a laboratory adaptive evolution process that allowed fast growth in glucose to an Escherichia coli strain lacking the major glucose transport system. BMC Genomics 13: 385 - PMC - PubMed
    1. Altae‐Tran H, Kannan S, Demircioglu FE, Oshiro R, Nety SP, McKay LJ, Dlakić M, Inskeep WP, Makarova KS, Macrae RK et al (2021) The widespread IS200/IS605 transposon family encodes diverse programmable RNA‐guided endonucleases. Science 374: 57–65 - PMC - PubMed
    1. Ambartsoumian G, D'Ari R, Lin RT, Newman EB (1994) Altered amino acid metabolism in Irp mutants of Escherichia coli K12 and their derivatives. Microbiology 140: 1737–1744 - PubMed
    1. Amster‐Choder O (2005) The bgl sensory system: a transmembrane signaling pathway controlling transcriptional antitermination. Curr Opin Microbiol 8: 127–134 - PubMed

Publication types

Substances