https://orcid.org/0000-0001-8617-1414
Figures
Abstract
Premature expression of genes in mobile genetic elements can be detrimental to their bacterial hosts. Tn916, the founding member of a large family of integrative and conjugative elements (ICEs; aka conjugative transposons), confers tetracycline-resistance and is found in several Gram-positive bacterial species. We identified a transcription terminator near one end of Tn916 that functions as an insulator that prevents expression of element genes when Tn916 is integrated downstream from an active host promoter. The terminator blocked expression of Tn916 genes needed for unwinding and rolling circle replication of the element DNA, and loss of the terminator caused a fitness defect for the host cells. Further, we identified an element-encoded antiterminator (named canT for conjugation-associated antitermination) that is essential for transcription of Tn916 genes after excision of the element from the host chromosome. We found that the antiterminator is orientation-specific, functions with heterologous promoters and terminators, is processive and is most likely a cis-acting RNA. Insulating gene expression in conjugative elements that are integrated in the chromosome is likely a key feature of the interplay between mobile genetic elements and their hosts and appears to be critical for the function and evolution of the large family of Tn916-like elements.
Author summary
Horizontal gene transfer allows bacteria to rapidly acquire new traits that can enhance their adaptability to different conditions. Integrative and conjugative elements (ICEs) are mobile genetic elements that reside integrated in a bacterial chromosome and can transfer to another cell via cell-to-cell contact through the element-encoded secretion system. ICEs often confer beneficial traits to their hosts, including antibiotic resistances, symbiotic/pathogenic determinants, metabolic capabilities, and anti-phage defense systems. Tn916, the first-described ICE, was identified based on its ability to transfer tetracycline resistance in the pathogen Enterococcus faecalis, and is found in several Gram-positive species. Once transferred into a new cell, Tn916 integrates into AT-rich sequences, sometimes downstream from a host promoter. We found that Tn916 has a transcription terminator near one end of the element that blocks transcription from an upstream host promoter, thereby protecting cells from detrimental effects of premature expression of element genes. Further, we found that Tn916 has a transcription antitermination system that is essential for expression of element genes after excision from the host chromosome. Our findings highlight the complex layers of transcriptional regulation that have evolved in ICEs, impacting host cell viability and the spread of the element.
Citation: Wirachman ES, Grossman AD (2024) Transcription termination and antitermination are critical for the fitness and function of the integrative and conjugative element Tn916. PLoS Genet 20(12): e1011417. https://doi.org/10.1371/journal.pgen.1011417
Editor: Morten Kjos, Norwegian University of Life Sciences: Norges miljo- og biovitenskapelige universitet, NORWAY
Received: September 4, 2024; Accepted: November 28, 2024; Published: December 9, 2024
Copyright: © 2024 Wirachman, Grossman. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: Funding: Research reported here is based upon work supported, in part, by the National Institute of General Medical Sciences of the National Institutes of Health under award number R35 GM122538 and R35 GM148343 to ADG. Any opinions, findings, and conclusions or recommendations expressed in this report are those of the authors and do not necessarily reflect the views of the National Institutes of Health. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Horizontal gene transfer helps drive microbial evolution, allowing bacteria to rapidly acquire new genes that can enhance their ability to thrive in different conditions. Integrative and conjugative elements (ICEs) are mobile genetic elements that normally reside integrated in a bacterial chromosome and can transfer to another cell via conjugation (Fig 1A). Once transferred, an ICE can integrate into the chromosome of the new host generating a transconjugant. ICEs often carry cargo genes that confer beneficial traits to their hosts, including antibiotic resistances, symbiotic/pathogenic determinants, metabolic capabilities, anti-phage defense systems, and many others [1–3]. While integrated in the host chromosome, cargo genes are often expressed, but most of the genes needed for the ICE lifecycle are not.
A) ICE life cycle. The chromosome is depicted as a circle with an ICE indicated in red. After excision, the circular double stranded ICE DNA (red) undergoes rolling circle replication. Transfer of linear ssDNA occurs to a recipient cell (shaded). Once transferred, the linear ssDNA is circularized and replicated to become dsDNA and then undergoes rolling circle replication and can be integrated to generate a stable transconjugant. B) Genetic map of Tn916. Gene names (numbers) are indicated below the corresponding genes. The ends of Tn916 are indicated by black lines. Promoters are indicated by bent arrows. Pxis is a weak promoter with poorly conserved -35 and -10 regions [19]. The origin of transfer (oriT916) [75], single strand origin of replication (sso916) [74], and the antiterminator canT are indicated above the map. The orientation of canT is indicated by the arrow above it. Three regions containing transcription terminators (T1, T2 consisting of T2a and T2b, T3) are indicated as red “T’s”. Functional modules of genes are indicated by brackets below the map. C) The excised circular form of Tn916. Following excision, canT is now upstream from the element genes needed for DNA processing and conjugation. Porf7, the main Tn916 promoter that drives expression of the element genes [19] also drives production of canT RNA and the canT RNA inhibits termination at T1, T2 and T3, thereby allowing transcription of the DNA processing and conjugation genes (orf24-orf13). D) RNA sequence of terminator T1. The nucleotide positions of the base of the terminator stem were determined by the ARNold web server [28,29]. The minimum free energy of folding ΔG (of the stem-loop) was calculated using the RNAfold web server [31]. The bolded red AUG at the 3’ end of the loop indicates the start codon of orf24. The changes in the T1 mutant are indicated and were made to preserve the potential ribosome binding site and amino acid sequence of orf24.
Tn916 (~18 kb) (Fig 1B), the first-described ICE, was identified based on its ability to transfer tetracycline resistance in the pathogen Enterococcus faecalis [4,5]. Tn916 and its relatives are found in many Gram-positive species, including Enterococcus, Streptococcus, Staphylococcus, and Clostridium [4–12], and function quite well in Bacillus subtilis [13–18]. As with other ICEs, Tn916 contains genes needed for its lifecycle: recombination (integration, excision); DNA processing (nicking, unwinding, and rolling circle replication); conjugation (a type IV secretion system); and regulation. When Tn916 is integrated in the chromosome, its DNA processing and conjugation genes are not expressed, largely due to the absence of a promoter within the integrated element (Fig 1B). After excision (circularization) of the element, the DNA processing and conjugation genes are expressed from the promoter for orf7 (Porf7) (Fig 1C) [19].
Tn916 integrates into AT-rich genomic regions [20–23] and these can be downstream from a host promoter. Integration in these regions is not random and when analyzed, there are hotspots for integration, and in some cases (e.g., Clostridium difficile CD37) there might be a unique integration site [24,25]. If Tn916 genes are co-directional with a host promoter, then this promoter might drive expression of Tn916 replication and conjugation genes (Fig 1B). Based on analogy to ICEBs1 from B. subtilis [26,27], we postulated that expression of these genes when the element is integrated would be detrimental to host cells, and that Tn916 might have a mechanism to prevent this.
Here, we describe a transcription terminator (T1) near the left end of Tn916 (Fig 1B) that is important for the fitness of host cells by functioning as an insulator to prevent transcription of element genes when Tn916 is integrated downstream from a host promoter. We also discovered a region in Tn916, canT (conjugation-associated antiterminator) that is downstream from Porf7 (Fig 1B and 1C), the promoter required for expression of the genes needed for conjugation [19]. canT prevents termination at element terminators and allows transcription of the DNA processing and conjugation genes (orfs23-13), but only after excision of Tn916 from the host chromosome. canT is essential for horizontal transfer of Tn916.
Results
Identification of functional transcription terminators in Tn916
We identified four predicted intrinsic transcription terminators (T1, T2a, T2b, T3) between the left end through the conjugation genes of Tn916 (Fig 1B) using online tools that predict RNA secondary structure and possible intrinsic terminators [28–31]. Terminator T1 is upstream of orf24, near the left end of Tn916 (Fig 1B and 1D). T2a and T2b (indicated as T2), are adjacent to each other and downstream of orf18 (Figs 1B and S1A) and T3 is within orf15 (Figs 1B and S1B). Because of its location near the left end of Tn916, we focused on T1 as a potential insulator of element gene expression when Tn916 is downstream from a host promoter.
To test its efficiency, we cloned T1 between the IPTG-inducible promoter Pspank and lacZ at an ectopic chromosomal locus, outside of Tn916 (Fig 2A) in Bacillus subtilis. In the presence of T1, expression of lacZ was reduced to <5% of that in the absence of T1 (Fig 2B). Mutations in T1 (T1–) that disrupted the stem and the U-tract of the terminator (Fig 1D) restored expression of lacZ (Fig 2B), indicating that T1 is a functional terminator with an efficiency of >95%.
A) Schematic of lacZ reporter construct with terminator T1 between Pspank and lacZ. B) β-galactosidase specific activities of the lacZ reporters with and without a functional terminator T1, either T1 was absent (Empty, ESW252), intact (T1+, ESW605), or mutated (T1–, ESW606), were measured two hours after induction of Pspank with IPTG. Data presented are averages from three independent experiments with error bars depicting standard error of the mean (mean ± SEM). P-values were calculated by an ordinary one-way ANOVA followed by Dunnett’s multiple comparisons test (* P<0.05, **** P<0.0001) using the GraphPad Prism version 10. Significance comparisons were made against the strain without T1 (Empty). C) Schematic of Pspank-Tn916. Full-length Tn916 is present, but only orf24 through orf20 are indicated in the schematic. D) Relative mRNA levels of the indicated genes from Pspank-Tn916 with an intact (T1+, ESW179) or mutant (T1–, ESW247) terminator T1 were measured one hour after induction of Pspank with IPTG. Data presented are from three independent experiments with error bars depicting standard error of the mean (mean ± SEM). P-values were calculated by two-tailed t-test (** P<0.01, **** P<0.0001) using the GraphPad Prism version 10. Significance comparisons were made for the amount of mRNA from each gene comparing the terminator mutant (T1–) to that from the strain with a functional terminator T1 (T1+).
T1 functions to insulate genes in Tn916 from readthrough transcription from a promoter in the host chromosome
To test the ability of terminator T1 to insulate transcription of Tn916 genes from an upstream promoter in the chromosome, we cloned Pspank upstream of Tn916 with an intact or mutant T1 (Fig 2C) and measured the amount of mRNA of three Tn916 genes (orf23, orf22, and orf20) located downstream of T1 by RT-qPCR. Expression of Tn916 genes was coming solely from the integrated element because the Pspank insertion prevented excision (S2 Fig), likely by altering the region just upstream of the junction between the chromosome and Tn916 that Int is known to bind [32]. After induction of Pspank (addition of IPTG), there were low levels of mRNA from the three genes in wild-type (T1+) Tn916 (Fig 2D). In contrast, the T1 mutant (T1–) had elevated amounts of mRNA from each of the three genes (Fig 2D). Based on these results, we conclude that T1 is a functional terminator that greatly reduces transcription into Tn916 from an upstream promoter in the chromosome.
Predicting an antitermination mechanism in Tn916
Transcription of Tn916 genes needed for conjugation is normally driven by Porf7 after excision and circularization of the element from the chromosome [19] (Fig 1C). Because terminator T1 is between the promoter Porf7 and conjugation genes after the element excises and is in the circular form, we postulated that there would be an antitermination mechanism to enable transcription of the Tn916 genes essential for conjugative transfer. Previous work found that a transposon (Tn5) insertion in codon 41 (of 83 total codons) of the predicted gene orf5 eliminated conjugation, leading to the inference that orf5 might encode a protein essential for conjugation [33,34]. However, no transcripts from the putative orf5 have been detected [19,35]. Experiments described below demonstrate that the putative orf5 protein product is not required for expression of Tn916 genes or for conjugation. Rather, there is an overlapping sequence now called canT (conjugation-associated antitermination) that functions in the opposite orientation from the putative orf5 and is essential for transcription antitermination and conjugation following excision of the Tn916 from the chromosome (Fig 1C).
The putative gene orf5 does not encode a protein product necessary for conjugation
We made two different mutations that should prevent the production of an orf5-encoded protein. 1) We changed the predicted start codon from AUG to AUA [orf5(M1I)], which should prevent translation of orf5 but preserve the amino acid sequence of the overlapping xis. 2) We changed codon 30 (of 83 total) to a stop codon [GCA to UAA, orf5(A30*)]. In both cases, the conjugation efficiency of Tn916 was indistinguishable from that of an otherwise wild-type element (Fig 3A). These results indicate that the role of the putative orf5 in conjugation is not as a protein-coding gene. More likely, the initial Tn5 insertion in the putative orf5 [33,34] disrupted a region that overlaps orf5 and this region is required for conjugation and likely functions as RNA.
A) Conjugation efficiencies. Donors contained wild-type Tn916 (WT, CMJ253), Tn916 with a mutation in the start codon of the putative orf5 [orf5(M1I), ESW104], Tn916 with a premature stop codon in the putative orf5 [orf5(A30*), ESW420], Tn916 with a deletion [canT(Δ162), ESW120] or 3 bp changes [canT(S3), ESW98] in canT, the Tn916 canT(S3) mutant with wild-type canT expressed in trans [canT(S3)+Pspank-canT, ESW771], and wild-type Tn916 with wild-type canT expressed in trans [WT+Pspank-canT, ESW781]. Conjugation efficiencies were calculated as the number of transconjugants per input donor. Data presented are from three independent experiments with error bars depicting standard error of the mean (mean ± SEM). All three independent mating assays of canT(Δ162) resulted in conjugation efficiencies that are below the limit of detection. P-values were calculated by an ordinary one-way ANOVA followed by Dunnett’s multiple comparisons test (*** P<0.001) using the GraphPad Prism version 10. Significance comparisons were made against the WT strain. B) Schematic of canT alleles. Parts of the 3’ end of orf8 and the 5’ end of xis are indicated inside of gray arrows. The 452-nucleotide canT region is shown inside of the yellow arrow. The region deleted (162 nucleotides) in canT(Δ162) is highlighted in white. The nucleotides changed in canT(S3) are underlined and indicated with stars, with the mutant sequence directly below the wild-type sequence. Prior to determining that the putative orf5 is not relevant as an open reading frame, we made the S3 mutation that would change the predicted tyrosine residue that was predicted computationally to be a potential site for phosphorylation. C) Relative mRNA levels normalized to the number of circular Tn916 per cell. Tn916 alleles include: wild-type (WT, CMJ253), canT(Δ162) (ESW120), canT(S3) (ESW98), canT and terminator T1 double mutant [canT(S3) T1–, ESW630]. Data presented are from three independent experiments with error bars depicting standard error of the mean (mean ± SEM). P-values were calculated by an ordinary one-way ANOVA followed by Dunnett’s multiple comparisons test (* P<0.5, ** P<0.01, *** P<0.001) using the GraphPad Prism version 10. Significance comparisons were made against the WT strain for each gene.
Identification of a region in Tn916 that is required for transcription of conjugation genes
We found that the region between the 3’ end of orf8 and xis was required for conjugation and expression of several Tn916 genes. A deletion that removed 162 bp [canT(Δ162)] (Fig 3B, white-highlighted) and a 3 bp substitution (ATA to TGC) [canT(3S)] (Fig 3B) both caused an approximately 100-fold drop in conjugation (Fig 3A). In contrast to the effect on conjugation, the excision frequency of the mutants was not decreased (WT: 1.68 ± 0.06%; canT(Δ162): 2.04 ± 0.05%; canT(3S): 1.82 ± 0.06%), indicating that the region altered in these mutants is normally required for conjugation but not excision, similar to the phenotypes described for the Tn5 insertion in the putative orf5 [33,34].
We also found that this region was required for expression of genes downstream from the terminator T1. We measured the amount of mRNA for orf23, orf22, orf21, and orf20 per copy of circular Tn916, thereby normalizing for excision and copy number of the excised element. The amount of mRNA for these genes in the two different canT mutants was greatly reduced (Fig 3C), indicating that the normal function of canT is to enable expression of these genes. Together, these results show that canT affects expression of genes downstream from T1 in the Tn916 circle, but is not required for excision.
If canT enabled expression of these genes by allowing transcription to read through T1 (as opposed, for example, to containing a strong promoter), then inactivation of T1 should restore gene expression in the absence of canT. Indeed, we found that loss of T1 restored expression of orfs23, 22, 21, and 20 in the canT(3S) mutant. Levels of mRNA from all four orfs were increased at least four-fold in the canT(S3) T1– double mutant compared to the canT(S3) single mutant (Fig 3C). Based on these results, we conclude that canT functions as an antiterminator in Tn916, enabling transcription to read through terminator T1.
canT is sufficient for antitermination and functions processively and in cis
canT functions as an antiterminator in the absence of other Tn916 genes.
We cloned a 452 bp fragment (Fig 3B, yellow arrow) that contains wild-type or mutant canT between Pspank and T1 in the lacZ reporter described above (Figs 2A and 4A). Expression of lacZ with the canT(Δ162) and canT(S3) mutants was <5% of that with wild-type canT (Fig 4B). These results indicate that canT functions to allow transcription to read through T1, consistent with the results with an intact Tn916.
A) Schematic of lacZ reporter construct with canT alleles and terminator(s): T1 (B); T2 (C); T3 (D); gatB terminator (E); rrnB T1, rrnB T2, and λ t0 (F); between Pspank and lacZ. canT alleles included wild-type (WT), (Δ162), (S3), wild-type but in the opposite orientation (Flipped). B-F) β-galactosidase specific activities from the indicated strains following two hours induction of Pspank with IPTG. Data for each are from three independent experiments with error bars depicting standard error of the mean (mean ± SEM). P-values were calculated by an ordinary one-way ANOVA followed by Dunnett’s multiple comparisons test (**** P<0.0001) using the GraphPad Prism version 10. Significance comparisons were made against the WT strain. B) T1 from Tn916 with canT alleles: WT (ESW398), Δ162 (ESW408), S3 (ESW407), Flipped (ESW440). C) T2 from Tn916 with canT alleles: WT (ESW492), Δ162 (ESW481), S3 (ESW623). D) T3 from Tn916 with canT alleles: WT (ESW493), Δ162 (ESW482), S3 (ESW626). E) Terminator from gatB with canT alleles WT (ESW421), Δ162 (ESW473), S3 (ESW449). F) Three terminators, T1 and T2 from E. coli rrnB, and t0 from phage lambda with canT alleles WT (ESW783), Δ162 (ESW785), S3 (ESW784).
canT functions in cis.
We wished to determine if there was a trans-acting factor that was disrupted in the canT mutant, or if canT functioned in cis. We found that a functional canT in the context of the reporter described above was unable to complement a Tn916 canT mutant. That is, there was no detectable conjugation of the Tn916 canT mutant in strains with a functional canT located elsewhere in the chromosome (Fig 3A, canT(S3) + Pspank-canT). Additionally, conjugation of wild type Tn916 (canT+) was normal in strains with the functional canT located elsewhere, indicating that expression of canT in trans did not inhibit its activity in the native context (Fig 3A, WT + Pspank-canT). Based on the results above, we conclude that canT acts in cis to cause transcription antitermination.
We also found that canT function is orientation-specific, consistent with an antiterminator that functions as RNA. We cloned the 452 bp fragment that contains wild-type canT in the opposite (flipped) orientation from what is normally found in Tn916 (Fig 4A). There was little to no expression of lacZ from the reporter (Fig 4B), indicating that canT function is orientation-specific. We note that clones with wild-type canT in each orientation contain the putative orf5, which overlaps canT. In the construct with the opposite orientation, the putative orf5 is oriented in the sense direction with Pspank and hence orf5 should be transcribed. These findings reinforce the conclusions above that if orf5 encodes a protein product, it is not involved in antitermination. Based on these results, we conclude that canT functions as an antiterminator, is orientation-specific, no other Tn916 genes are needed for its function, and that it is most likely active as RNA.
In attempts to identify smaller fragments that retained canT function, we cloned fragments of varying lengths into the reporter (S3 Fig), as described for the 452 bp fragment. We found that a 400 bp fragment had antitermination activity (S3C Fig). However, the canT(S3) mutation in this fragment had little or no effect on antitermination activity (S3C Fig). In other words, the properties of the 400 bp fragment did not mimic those of the intact Tn916 (Fig 3C) or the 452 bp fragment in the reporter (Figs 4 and S3B). Some shorter DNA fragments also had antiterminator activity (e.g., S3D, S3E and S3G Fig), although we did not test the effects of the canT(S3) mutation on these fragments. Based on these results, we decided to use the 452 bp fragment in further experiments as it provided high antitermination activity and accurately reflected the effects of the canT(S3) mutation observed in the intact Tn916. Below, we also describe the predicted secondary structures of some of the fragments with antitermination activity.
canT functions as an antiterminator for T2 and T3 of Tn916.
T2 and T3 were cloned individually into the canT-lacZ reporters in place of T1 (Fig 4A). Similar to results with T1, expression of lacZ was high in the presence of wild-type canT and much lower with the mutant canT (Fig 4C and 4D).
canT functions as an antiterminator for heterologous terminators and is processive.
We cloned the terminator located downstream of B. subtilis gatB into the canT-lacZ reporters (Fig 4A). Expression of lacZ was high with wild-type canT and quite low with the canT mutants (Fig 4E). We also cloned three terminators, the two from E. coli rrnB (rrnB T1, rrnB T2) and the phage lambda terminator t0, between canT and lacZ (Fig 4A). There was expression of lacZ with wild-type canT, but no detectable expression with the canT mutants (Fig 4F). Together, our results indicate that canT functions on T1, T2, and T3 of Tn916, acts on heterologous terminators, functions in cis, and acts processively.
Predicted secondary structures of canT RNA.
Based on the results above, it is most likely that canT RNA functions an antiterminator, analogous to other RNA antiterminators [36–41]. The known RNA antiterminators all contain predicted stem-loop structures and are sometimes complicated [36–41]. We used the ViennaRNA package [30] to predict canT RNA secondary structures for the wild-type 452-nucleotide sequence (S4A Fig) and three different variants that are inactive (S4B–S4D Fig) and one variant that is active (S4E Fig).
The 452-nucleotide sequence that has canT antitermination activity is predicted to fold into a complex structure with multiple regions of base pairing and stem-loops (S4A Fig). The left and right parts (as drawn in S4A Fig, green and blue boxes, respectively) are each predicted to have at least three complex stem-loops. Both the canT(Δ162) (S4B Fig) and canT(S3) (S4C Fig) mutants of the 452-nucleotide canT were defective in antitermination and were missing structures on the left side of the wild-type canT (S4A Fig, green box), perhaps indicating that these structures play a role in antitermination. However, the predicted secondary structure of the antitermination-defective 291-nucleotide sequence (S3F and S4D Figs) contains the sequences that are missing or altered in the canT(Δ162) and canT(S3) mutants. This indicates that this predicted structure (S4D Fig, green box) is not sufficient to cause antitermination. Additionally, a 341-nucleotide sequence that has antitermination activity (S3G Fig) is predicted to have parts of the predicted secondary structures (S4E Fig) that are also predicted to be present in sequences that are not active (S4B–S4D Fig). Based on these observations, we conclude that interpreting antitermination activity from secondary structure predictions of canT RNA is complicated. We suspect that whether or not the secondary structure predications are accurate, there is an RNA tertiary structure that is essential for antitermination activity. We have not further pursued sequence or structural analyses of this region.
Terminator T1 confers a fitness benefit to cells with Tn916 integrated downstream from a strong promoter
We found that T1 of Tn916 conferred a fitness benefit to the host when Tn916 was downstream from and co-directional with a strong promoter in the host chromosome. Initially, we inserted the strong inducible promoter Pxis from ICEBs1 upstream from Tn916 (Fig 5A). Pxis is repressed by ImmR and is derepressed following cleavage of ImmR by the protease ImmA which is activated by RapI [42–44]. We induced transcription from Pxis by expressing rapI from a xylose-inducible promoter (Pxyl-rapI) (Methods). Transcription from Pxis had little or no effect on the cell viability when Tn916 contained a functional T1. That is, the number of viable cells four hours after derepression of Pxis (the addition of xylose to induce Pxyl-rapI) was virtually the same as in cells grown similarly but without xylose (Fig 5B). In contrast, in cells with a mutant T1 (T1-), there was a two-fold reduction in CFUs after four hours of expression from Pxis (+xylose) compared to no expression (no xylose) (Fig 5B). Based on these results, we conclude that terminator T1 is important for the fitness of host cells that contain Tn916 downstream from a strong promoter.
A) Schematic of Pxis-Tn916 either with intact (T1+, ESW270) or mutant (T1–, ESW271) terminator T1, or with a mutant terminator combined with loss-of-function mutations in orf20 (T1– Δ20, ESW316), orf23-22 (T1– Δ23–22, ESW363), oriT (T1– ΔoriT, ESW414). Full-length Tn916 is present, but only orf24 until orf20 are indicated in the schematic. Pxis used here is the strong promoter from ICEBs1 that is repressed by the ICEBs1 repressor ImmR. ImmR is inactivated by the metalloprotease ImmA following expression of RapI [42–44]. rapI was expressed from a xylose-inducible promoter (Pxyl-rapI) following addition of xylose to cells. B) Relative viability of Pxis-Tn916 strains shown in (A) was measured four hours after induction of Pxis and calculated as the number of CFUs following induction, divided by that in the uninduced culture grown in parallel (a value of “1” indicates there is no change in CFUs with induction). Data presented are from at least four independent experiments with error bars depicting standard error of the mean (mean ± SEM). P-values were calculated by an ordinary one-way ANOVA followed by Dunnett’s multiple comparisons test (**** P<0.0001) using the GraphPad Prism version 10. Significance comparisons were made against the T1+ strain. C) Schematic of Tn916 gene transcription when the element is inserted downstream of a host promoter. Genes necessary for DNA replication, including orf23-22 (green) and orf20 (blue), are not transcribed when terminator T1 is intact (T1+), but are transcribed when T1 is mutated (T1–). Expression of these genes from integrated Tn916 leads to DNA unwinding and rolling circle replication occurring in the integrated element, which causes a fitness defect to the host cells. D) Cartoon of repeated rolling circle replication from the Tn916 oriT that is integrated in the chromosome (modified from [26]). The relaxase ORF20 (blue circles) nicks the origin of transfer, oriT (gray bar), that also functions as an origin of replication [18] and is covalently attached to the 5’ end of the DNA. Replication extends (dotted line with arrow) from the free 3’-OH and regenerates a functional oriT that is a substrate for nicking by another molecule of the relaxase. ORF23-22 (green circles) act as helicase processivity factors, helping the host-encoded helicase, PcrA (not drawn), unwind the DNA [18,45]. The rest of the replication machinery (not drawn) is composed of host-encoded proteins.
By analogy to findings with ICEBs1 [26,27], we suspected that the decrease in CFUs of cells with the Tn916 T1 mutant was due to autonomous rolling circle replication of Tn916 that remained integrated in the chromosome. Tn916 has three genes (orf23, 22, 20) and a site (oriT) that are required for DNA unwinding, autonomous rolling circle replication [18], and conjugation. orf20 encodes the relaxase that nicks the element at the origin of transfer, oriT (that also functions as an origin of replication), to initiate DNA unwinding and rolling circle replication. orf22 and orf23 both encode helicase processivity factors that help the host-encoded helicase PcrA unwind the element DNA after nicking [18,45]. Deletions of orf20, orf23-22, or oriT all alleviated the fitness defect caused by expression from Pxis into Tn916 in the absence of a functional T1 (Fig 5A and 5B). Based on these results, we conclude that the terminator T1 in Tn916 is important for preventing expression of the DNA unwinding and replication genes in Tn916 when the element is downstream from a strong promoter and that loss of this terminator results in a fitness defect that is due to DNA unwinding and-or autonomous rolling circle replication of Tn916 while it is in the host chromosome (Fig 5C and 5D).
Isolating cells with Tn916 integrated downstream of a host promoter
We wished to determine if T1 also contributed to host fitness when Tn916 was integrated downstream from endogenous host promoters. To isolate Tn916 insertions downstream from active host promoters, we cloned lacZ near the left end of Tn916, upstream of T1 (Tn916-lacZ) (Fig 6A) and used a strain with this element as a donor for conjugation. Transconjugants were selected on plates with tetracycline and X-gal and blue colonies were picked and verified for the presence of Tn916 and expression of lacZ. Several strains were chosen for further analyses and the sites of integration were determined by arbitrary PCR and sequencing (Methods). Strains with an insertion in sdpA, sunT, fadR, and adaB-ndhF (Fig 6B) were chosen for further analyses. We removed lacZ from each insertion and then introduced mutations that inactivate terminator T1. As a control, we used an insertion in yvgT-bdbC in which the Tn916 replication and conjugation genes were not co-directional with a host promoter (Fig 6B). We also used the initial Tn916-lacZ alleles (lacZ+ T1+) to monitor transcription from the host promoters.
A) Schematic of Tn916-lacZ integrated downstream from an active host promoter. lacZ was cloned within Tn916, upstream of terminator T1 such that lacZ is expressed when Tn916 integrates downstream of an active host promoter. Tn916-lacZ (not downstream from an active promoter, so phenotypically LacZ–) in strain (ESW261) was used as a donor used to isolate Tn916 insertions in transconjugants that are downstream of an active host promoter. B) Schematic and maps of the Tn916 insertions used for pairwise competition assays. Tn916 integrated at sdpA (i), sunT (ii), fadR (iii), and adaB-ndhF (iv) are located downstream of active promoters. Tn916 at yvgT-bdbC (v) is oriented opposite the direction of transcription of the host genes. C) Fitness of T1– cells relative to T1+ cells calculated as Malthusian parameters ratio of T1– to T1+ (Methods). Data presented from at least three independent experiments with error bars depicting standard error of the mean (mean ± SEM). P-values were calculated by an ordinary one-way ANOVA followed by Dunnett’s multiple comparisons test (**P<0.01, ****P<0.0001) using the GraphPad Prism version 10. Significance comparisons were made against the lacZ control. D-J) Population dynamics over the course of the pairwise competition assays. The initial ratio of lacZ+ and lacZ−cells was ~1:1 for (D). The initial ratio of T1+ to T1– cells was ~1:1 for (E,F,I,J) and ~1:10 for (G,H). At least three independent experiments were done for each competition and individual data points are shown. The x-axis (generations) represents the number of population doublings. Circles, replicate 1; squares, replicate 2; triangles, replicate 3; inverted triangles, replicate 4. Black, T1+; red, T1–. Competition assay results of lacZ+ and lacZ−cells: (D) (ESW617 vs ESW616). Competition assay results of T1+ vs T1– cells: (E) sdpA (ESW675 vs ESW693), (F) sunT (ESW677 vs ESW695), (G) fadR (ESW724 vs ESW725), (H) adaB-ndhF (ESW722 vs ESW723), and (J) yvgT-bdbC (ESW713 vs ESW714). (I) T1+ vs T1– ΔoriT cells of Tn916 integrated at adaB-ndhF (ESW722 vs ESW728).
Fitness benefits of terminator T1 in Tn916 insertions that are downstream from host promoters
We found that T1 contributed to host fitness when Tn916 was integrated downstream from endogenous host promoters. To measure the fitness benefit conferred by T1, we did pairwise competition assays between strains with and without a functional T1 (T1+ versus T1–). The two strains were mixed in an appropriate ratio (~1:1 or ~1:10), spotted on LB agar for 24 hours, harvested, and then a dilution of the mix was spotted again on fresh LB agar and repeated for up to five days. The total cell number increased approximately 300- to 1,000-fold each 24-hour cycle, the equivalent of 8–10 cell doublings, and totaling approximately 50 cell doublings over five days. The proportion of each strain in the mixed population was determined every 24 hours, and the relative fitness (w) of T1– to T1+ cells was calculated. Because lacZ had been removed from the Tn916 insertions, we used lacZ located elsewhere in the chromosome as a fitness-neutral marker in T1+ cells (Fig 6C and 6D) to distinguish them from the T1– cells.
We found that the loss of a functional T1 caused a fitness defect in the four different Tn916 insertions that were downstream from a host promoter. The proportion of T1– cells in the population dropped from ~50% to ~10% in ~30 doublings for both sdpA::Tn916 and sunT::Tn916 (Fig 6E and 6F), with mean relative fitness of 0.87 and 0.90, respectively (Fig 6C). The proportion of T1– cells in the population dropped from ~90% to ~1% in ~20 doublings for both fadR::Tn916 and adaB-ndhF::Tn916 (Fig 6G and 6H), with mean relative fitness of 0.55 for both (Fig 6C). These results show that T1 can confer a fitness benefit to cells that contain Tn916 downstream from and codirectional with a host promoter.
The fitness defect caused by the loss of T1 in the insertions was due to DNA unwinding and-or autonomous rolling circle replication. We deleted oriT from Tn916 inserted in adaB-ndhF, thereby preventing both DNA unwinding and autonomous replication of Tn916 and found that the proportion of T1+ and T1– ΔoriT cells did not significantly change over the entire competition experiment (Fig 6I), indicating that the fitness defect caused by the loss of T1 was completely suppressed. The mean relative fitness was 1.04 (Fig 6C), in marked contrast to that of the element with a functional oriT (mean relative fitness of 0.55). These results are consistent with those described above for Pxis-Tn916 T1– (Fig 5B).
The fitness effects caused by T1 were dependent on the presence of an upstream host promoter that was co-directional with Tn916 genes. We used a Tn916 insertion that is between yvgT-bdbC and oriented opposite the direction of transcription of the host genes (Fig 6B). There was no significant change in the proportion of cells with Tn916 with T1+ versus T1– during the entire competition (Fig 6J). The mean relative fitness was 0.99 (Fig 6C). Based on these results, we conclude that the terminator T1 of Tn916 is important for the fitness of host cells when the element is downstream from and co-directional with an active host promoter, and that the primary function of T1 is to prevent autonomous rolling circle replication of Tn916 that is in the host chromosome.
We tested whether the fitness effects might be correlated with apparent promoter strengths. We determined promoter activities by measuring β-galactosidase activity during growth in defined minimal liquid medium and after eight hours of growth as a spot on an LB agar plate using the initial Tn916-lacZ insertions described above (Fig 6A and 6B).
During exponential growth in defined minimal medium, fadR had the highest rate of transcription as measured by β-galactosidase synthesis, followed by adaB-ndhF, sunT, and sdpA (S5A Fig). This appears to roughly correlate with the relative effects on fitness in which Tn916 expression from fadR and adaB-ndhF had larger defects than that from sunT and sdpA (Fig 6C). However, after eight hours of growth as spots on an LB agar plate (conditions that are closer to those used in the competition experiments to measure fitness), sunT appeared to have the highest expression, followed by sdpA, fadR, and adaB-ndhF (S5B Fig). The fitness effects observed during the competition experiments (Fig 6C) did not correspond to these levels of β-galactosidase synthesis. Additionally, we compared published transcriptomics data [46] for each of the four loci and the effects on fitness. The published data [46] indicate that during exponential growth, transition to stationary phase, and stationary phase in LB medium, sunT had the highest mRNA levels, followed by fadR, sdpA, and adaB-ndhF, again, not correlating or corresponding to the observed effects on fitness in our competition experiments (Fig 6C).
Based on the published information about transcription from the relevant promoters and our measurements, we believe that quantitatively different fitness effects caused by expression of Tn916 genes from different host promoters are due to a combination of promoter strength, stability of the hybrid mRNAs (host gene and Tn916 genes co-transcribed), and the ability of the hybrid mRNAs to be translated, all during complex and continually changing growth conditions in the mixed communities on a solid surface.
Discussion
We found that Tn916 has a transcription termination-antitermination system that is crucial for element function and host cell fitness. This system includes terminators T1, T2 (T2a + T2b), and T3 and the antiterminator canT. When Tn916 integrates downstream from an active host promoter, T1 insulates expression of genes needed for unwinding and rolling circle replication of the element DNA and preserves host cell fitness. We suspect that the function of terminators T2 and T3 is to terminate spurious transcripts that might come from within the element, analogous to the proposed function of the terminators within the conjugation operon of the B. subtilis conjugative plasmid pLS20 [37].
After excision of Tn916, the promoter Porf7 drives transcription of the DNA processing (nicking, unwinding, replication) and conjugation genes [19] and the terminators between Porf7 and the end of the conjugation operon might be problematic for gene expression. However, the canT antiterminator RNA allows transcription to read through element terminators processively, enabling expression of DNA processing and conjugation genes essential for conjugation once the element is in the circularized form (Fig 1C).
This type of termination-antitermination system appears to be widespread in the large family of Tn916-like elements. We identified predicted intrinsic transcription terminators on the left end of several Tn916-like elements, including Tn2010, Tn5251, Tn5386, Tn5397, Tn5801, Tn6000, Tn6002, Tn6003, Tn6084, and Tn6085a (S6 Fig). Some of these terminators have identical sequences to the Tn916 terminator T1, others have a few variations (S6A Fig), and some have completely different sequences (S6B Fig), but are predicted terminators nonetheless. In elements with terminators identical to Tn916 terminator T1, the 452 bp canT sequence is also present. Some of the elements whose terminator efficiencies are not known have canT-like sequences at the right end and others lack any sequences that are easily recognized as resembling canT. We have not evaluated predicted RNA structures of transcriptions from these regions.
Factors involved in processive transcription antitermination
Processive transcription antitermination can involve protein and-or RNA factors, either acting independently or in combination. For example, the N and Q antitermination proteins of phage lambda bind to an RNA site (nut site for N utilization) or a DNA site near a promoter (qut site for Q utilization), and associate with RNA polymerase to cause antitermination [38,39,47–50]. The conjugative plasmid pLS20 encodes both the protein ConAn1 and the RNA conAn2 that are used in antitermination of the long conjugation operon [37]. Antitermination systems in the lambdoid phage HK022 [40,41,51,52] and the B. subtilis eps operon [36] use RNA elements, put and EAR, respectively, for antitermination, without protein factors, although the possibility of host factor involvement has not been eliminated in the latter. Despite their differences, these factors are known or suspected to alter RNA polymerase such that it is no longer susceptible to most terminators [38,39,50,52–56], and some appear to act over distances up to ~30,000 nucleotides [36,37,57,58]. We suspect that Tn916 canT RNA directly interacts with RNA polymerase to make it resistant to most terminators (Fig 7).
(i) Post-excision, transcription by the RNA polymerase (pink) initiates at Porf7 in the Tn916 circle. (ii) The nascent RNA (wavy line) contains canT RNA (yellow circle). (iii) canT RNA binds to the RNA polymerase, modifying it into a terminator-resistant form, allowing processive antitermination. canT RNA-bound, modified RNA polymerase transcribes and bypasses terminator T1.
Biological roles of processive transcription antitermination
Based on our findings and comparisons with other systems, Tn916 canT likely serves several roles in the biology of Tn916. The presence of canT allows complete transcription of the long DNA processing and conjugation operon, despite the presence of internal terminators (discussed above). It functions to couple element gene expression to excision as canT allows transcripts from Porf7 to read through the transcription terminators, most notably T1, when the element is in the circular form after excision (Fig 1C). Further, the presence of T1 and its function as an insulator to prevent element gene expression when Tn916 is integrated downstream of a host promoter, would prevent Tn916 (in its current form) from functioning as a conjugative element without a mechanism of antitermination. In this way, canT enables Tn916 to have such an insulator and still function.
In addition to canT, Porf7 is critical for the function of Tn916. Both are needed to couple element gene expression to excision. The DNA processing and conjugation operon is not expressed when Tn916 is integrated due to the physical separation of the genes from the main promoter Porf7 [19] (Fig 1B). Following excision, canT allows the transcript initiated from Porf7 to bypass the internal terminators, leading to full expression of the operon in the Tn916 extrachromosomal circle (Fig 1C). The use of antitermination to regulate timing of gene expression also occurs in the lambda phage, in which the N and Q proteins control the switch from immediate-early to delayed-early gene expression [38,39,47,50,59] and early to late gene expression [38,39,48–50,60], respectively.
Mechanisms to prevent autonomous replication of integrated mobile genetic elements
ICEs have different mechanisms for regulating DNA nicking, unwinding and rolling circle replication. These mechanisms function to both couple autonomous replication with excision, and to stop new rounds of replication prior to or concomitant with integration. As described here, Tn916 separates the main promoter and antiterminator from genes needed for DNA nicking, unwinding, and replication and uses a terminator upstream from those genes to prevent chromosomal promoters from reading into the element. This type of mechanism is enabled by the antitermination system in Tn916.
In contrast to Tn916, ICEBs1 uses a transcriptional repressor that controls expression of the excisionase (xis) and the DNA nicking, unwinding, and replication genes [42,61,62]. In this way, the DNA processing genes are expressed only when the excisionase is expressed. Further, repression and depletion of the excisionase is needed for integration, and this happens when xis is repressed, along with the DNA processing genes. This mechanism is similar to that used by some temperate phage, including lambda [63–66].
Our findings with Tn916 highlight the intricate layers of transcriptional regulation in ICEs. We suspect that other ICEs may have similar termination-antitermination mechanisms and that this type of regulation is more widespread and may have a broader role in horizontal gene transfer than previously thought.
Methods
Media and growth conditions
B. subtilis cells were grown shaking at 37°C in either LB medium or MOPS (morpholinepropanesulfonic acid)-buffered 1X S750 defined minimal medium [67] containing 0.1% glutamate, required amino acids (40 μg/ml phenylalanine and 40 μg/ml tryptophan) and either glucose or arabinose (1% w/v) as a carbon source or on LB plates containing 1.5% agar. Escherichia coli cells were grown shaking at 37°C in LB medium for routine strain constructions. Where indicated, tetracycline (2.5 μg/ml) was added to Tn916-containing cells to increase gene expression and excision [19]. Antibiotics were otherwise used at the following concentrations: 5 μg/ml kanamycin, 10 μg/ml tetracycline, 100 μg/ml spectinomycin, 5 μg/ml chloramphenicol, and a combination of erythromycin at 0.5 μg/ml and lincomycin at 12.5 μg/ml to select for macrolide-lincosamide-streptogramin (mls) resistance. 5-bromo-4-chloro-3-indolyl β-D-galactopyranoside (X-gal) was used at a concentration of 120μg/ml.
Prior to sample collection for β-galactosidase assay, cells were grown as light lawns on 1.5% agar plates containing 1% w/v glucose, 0.1% w/v monopotassium glutamate, and 1X Spizizen’s salts [2 g/l (NH4)SO4, 14 g/l K2HPO4, 6 g/l KH2PO4, 1 g/l Na3-citrate·2H2O, and 0.2 g/l MgSO4·7H2O] [68]. Cells were resuspended from light lawns and grown at 37°C with shaking in 1X S750 defined minimal glucose medium [67].
Strains, alleles, and plasmids
E. coli strain AG1111 (MC1061 F’ lacIq lacZM15 Tn10) was used as a host for various plasmids. B. subtilis strains (Table 1), except BS49, were derived from JMA222, a derivative of JH642 (trpC2 pheA1) [69,70] that is cured of ICEBs1 [71]. B. subtilis strains were constructed by natural transformation [68] or conjugation as indicated. Key strains and newly reported alleles are summarized below.
ΔcomK::kan in ELC37 replaced most of the comK open reading frame from 47 bp upstream of comK to 19 bp upstream of its stop codon with the kanamycin resistance cassette (kan) from pGK67 [72]. The kan marker was fused with up- and downstream homology regions via isothermal assembly [73] and used for transformation.
Unmarked deletions and mutations in Tn916.
Unmarked deletions and point mutations in Tn916 were generated by a two-step allelic replacement approach. Briefly, DNA flanking the regions to be altered were amplified and inserted by isothermal assembly [73] into the EcoRI and BamHI sites of pCAL1422, a plasmid containing E. coli lacZ and B. subtilis chloramphenicol resistance cassette (cat), as previously described [45,74]. For deletion mutations, there was no DNA between the flanking regions. For point mutations, a fragment containing the desired mutations was assembled between the flanking regions. The resulting plasmids were used to transform the appropriate B. subtilis strains, selecting for integration of the plasmid into the chromosome (chloramphenicol-resistant) by single crossover recombination. Transformants were screened for loss of lacZ and checked by PCR and sequencing for the desired allele. Deletions and point mutations are described below.
- Genes and sites involved in replication. Deletion of genes orf23-22 (encoding helicase processivity factors) extends from immediately after the stop codon of orf24 through the stop codon of orf22. Deletion of orf20 (encoding the relaxase) fuses the first 90 codons of orf20 with the orf20 stop codon, deleting the intervening 306 codons and preserving oriT [75]. Deletion of oriT (ΔoriT) removes 5’- CCCCCCGTAT CTAACAGGGG GG-3’, starting from the nucleotide 41 downstream from the stop codon of orf21.
- Mutations in the putative orf5. orf5(M1I) (ESW104) changes the predicted start codon from AUG to AUA. This mutation preserves the amino acid sequence of the overlapping xis. orf5(A30*) (ESW420) changes codon 30 (of 83 total) from GCA (alanine) to UAA (stop).
- Mutations in canT. Deletion of canT [canT(Δ162), ESW120] removes 162 bp between orf8 and xis, extending from 229 to 390 nucleotides downstream from the stop codon of orf8 (Fig 3B). canT(S3) (ESW98) contains a 3 bp substitution 5’-ATA-3’ to 5’-TGC-3’, starting from the nucleotide 283 downstream from the stop codon of orf8 (Fig 3B).
- Mutations in the terminator T1. Multiple base pair changes were made to inactivate terminator T1 (T1–), including changing three nucleotides in the predicted stem of the stem-loop, two nucleotides near the base of the stem, and two of the U’s that follow the stem in the predicted RNA secondary structure (Fig 1D). The mutations did not alter the ribosome binding site or amino acid sequence of orf24, which overlaps T1.
Pspank-Tn916 and Pxis-Tn916.
Pspank-Tn916 strains (T1+, ESW179; T1–, ESW247) and Pxis-Tn916 strains (T1+, ESW270; T1–, ESW271; T1– Δorf20, ESW316; T1– Δorf23-22, ESW363; T1– ΔoriT, ESW414) were created by cloning, either Pspank or ICEBs1 Pxis promoter, and an MLS resistance cassette (mls) upstream of att(yufKL)::Tn916 and its derivative mutants. DNA flanking the regions to be altered were generated by PCR and assembled flanking the promoter and mls via isothermal assembly [73]. DNA was transformed into B. subtilis selecting for resistance to MLS. Transcription from Pxis was derepressed by overexpression of rapI under the control of a xylose-inducible promoter inserted at the non-essential locus amyE [amyE::(Pxyl-rapI) spc] [44]. RapI causes the protease ImmA to cleave ImmR, the repressor of Pxis, thereby derepressing transcription from Pxis [42,43].
Tn916-lacZ.
Tn916-lacZ (ESW261) has lacZ at the left end of Tn916, upstream of T1 (Fig 6A) and was used to identify insertions that were downstream from an active promoter. Briefly, lacZ and a kanamycin resistance gene (kan) that was flanked by lox sites were inserted 29 bp upstream of orf24 (upstream of T1) by isothermal assembly [73] and introduced into Tn916. The Cre recombinase, expressed from the temperature-sensitive plasmid, pDR244 [76], was then used to remove the lox-flanked kan marker by recombination. Strains were then cured of pDR244 by culturing them on LB agar at 42°C, as previously described [76,77]. lacZ is expressed when integrated Tn916 is integrated downstream from an active promoter.
lacZ reporter for measuring transcription termination and antitermination.
DNA fragments with possible terminators upstream from lacZ were cloned downstream of Pspank in the vector pDR110 (a gift from D. Rudner, integrates by double crossover at amyE; contains Pspank, lacI, spc). Briefly, lacZ was amplified by PCR from pCAL1422 [45] using primers that include either wild-type or mutant T1. The T1 sequence cloned included 6 bp and 15 bp directly upstream and downstream, respectively, of the base of the stem of the terminator hairpin. The PCR products were inserted by isothermal assembly [73] into pDR110, cut with NheI and HindIII and the resulting construct was integrated into B. subtilis at the amyE locus by recombination selecting for resistance to spectinomycin. Reporter strains contained: no terminator (ESW252); T1+ (ESW605); and the T1 mutant (T1–, ESW606).
Antitermination activity was measured using a reporter with a terminator between Pspank and lacZ and cloning additional DNA fragments containing the indicated canT alleles between Pspank and the indicated terminator. The 452 bp DNA fragment with wild-type canT extends from 126 to 577 bp downstream of the stop codon of orf8 (Fig 3B). The strategy for building each construct was similar to that described for the terminators above. Strains contained the canT alleles: WT, Δ162, S3, or flipped, followed by T1 (ESW398, ESW408, ESW407, ESW440, respectively); the WT, Δ162, S3 alleles upstream of T2 (ESW492, ESW481, ESW623, respectively); upstream of T3, (ESW493, ESW482, ESW626, respectively); or upstream of the gatB terminator (ESW421, ESW473, ESW449).
Additionally, each of the three different canT alleles (WT, Δ162, S3) was cloned upstream from an array of three terminators, E. coli rrnB T1, rrnB T2, and the phage lambda λ T0, between Pspank and lacZ and integrated into the chromosome at amyE, essentially as described above, generating strains ESW783 (WT), ESW785 (Δ162), and ESW784 (S3).
β-galactosidase assays
Cells were grown at 37°C in defined minimal medium with shaking. At OD600 ~0.1, IPTG was added (1 mM final concentration) to induce transcription from Pspank and samples were taken two hours post-induction. In other experiments, cells were grown on LB agar plates and harvested after the indicated time of growth. Cells were permeabilized with 15 μl of toluene and β-galactosidase specific activity was determined [(ΔA420 per min per ml of culture per OD600 unit) × 1000] essentially as described [78] after pelleting cell debris.
RT-qPCR to measure gene expression
For reverse transcription reactions, an aliquot of cells was harvested in ice-cold methanol (1:1 ratio) and pelleted. RNA was isolated using Qiagen RNeasy PLUS kit with 10 mg/ml lysozyme. iScript Supermix (Bio-Rad) was used for reverse transcriptase reactions to generate cDNA. Control reactions without reverse transcriptase were performed to assess the amount of DNA present in the RNA samples. RNA was degraded by adding 75% volume of 0.1 M NaOH, incubating at 70°C for 10 min, and neutralizing with an equal volume of 0.1 M HCl.
The relative amounts of cDNA were determined by qPCR using SSoAdvanced SYBR master mix and CFX96 Touch Real-Time PCR system (Bio-Rad). qPCR data were quantified using the standard curve method [79]. Standard curves for these reactions were generated using B. subtilis genomic DNA that contained wild-type Tn916. Primers used to quantify xis, orf23, orf22, orf21, orf20, and the chromosomal locus gyrA are listed in Table 2. The relative mRNA levels of Tn916 genes (as indicated by the Cq values measured by qPCR) were normalized to gyrA after subtracting the signal from control reaction without reverse transcriptase.
qPCR to measure Tn916 excision and circle copy number
qPCR was used to monitor excision (activation) and copy number of circular (excised) Tn916, essentially as described previously [18,80,81]. Briefly, cells containing Tn916 were lysed (40 mg/ml lysozyme) and genomic DNA was prepared using the Qiagen DNeasy kit. qPCR was performed using SsoAdvanced SYBR master mix and the CFX96 Touch Real-Time PCR system (Bio-Rad). qPCR data were quantified using the Pfaffl method [82]. Standard curves for these qPCRs were generated using B. subtilis genomic DNA that contained an empty Tn916 chromosomal attachment site (att1), an ectopic copy of the Tn916 circle attTn916 junction inserted at amyE, and a copy of the nearby locus, mrpG.
Excision frequencies were calculated as the number of copies of the chromosomal site from which Tn916 excised (att1) divided by the number of copies of mrpG (a nearby gene). The average number of copies of the circular Tn916 per cell was calculated as the number of copies of attTn916 divided by the number of copies of mrpG. Primers used to amplify the empty chromosomal attachment site (att1), the attTn916 junction in the circular Tn916, and a region within the nearby gene mrpG were described previously [18] and are listed on Table 2.
Growth and viability assays
Pxis-Tn916 strains were grown in defined minimal medium with 1% arabinose as a carbon source to early exponential phase. At an OD600 of 0.05, the cultures were split and xylose was added (1% final concentration) to one portion to induce transcription from Pxyl, thus expressing rapI (Pxyl-rapI), causing inactivation of ImmR, the repressor of Pxis [42,43]. After four hours, the number of CFUs was determined in induced and non-induced cultures. “Relative viability” was calculated as the number of CFUs present in the induced culture divided by the number of CFUs present in the non-induced culture.
Mating assay
Mating assays were performed essentially as described previously [71]. Briefly, donor strains containing Tn916 (tetracycline-resistant) or derivatives were grown in LB medium to early exponential phase. At an OD600 ~0.2, activation of Tn916 was stimulated by addition of tetracycline (2.5 μg/ml final concentration). After one hour, donor strains were mixed in a 1:1 ratio with kanamycin-resistant recipient cells (ELC37) and 5 total OD units of cells were filtered. Mating filters were placed on a 1X Spizizen’s salts [68] 1.5% agar plate at 37°C for one hour. Cells were then harvested off the filter and the number of CFUs of donors (tetracycline-resistant), recipients ELC37 (kanamycin-resistant), and transconjugants (tetracycline/kanamycin-resistant) were determined both pre- and post-mating. Conjugation efficiency is the percentage of transconjugants per donor (using the number of donors determined at the start of mating).
Identification of Tn916 insertions downstream from active host promoters
We used Tn916-lacZ to identify insertions downstream from host promoters. A donor containing Tn916-lacZ (ESW261) was crossed to a recipient without Tn916 (ELC37) and cells were plated on LB agar containing tetracycline to select for transconjugants, kanamycin to kill donors (counterselection), and X-gal to screen for insertions downstream from an active host promoter, as indicated by blue colony color. Approximately 10–15% of the initial transconjugants appeared blue after overnight growth on selective LB agar plates. Blue transconjugant colonies were picked and re-streaked non-selectively on LB agar containing X-gal. After re-streaking, many of the initially light blue colonies appeared white, indicating that there was little or no expression of lacZ. We suspect that the initial light blue appearance was from lacZ expression in the nascent transconjugant, before the element had integrated. We subsequently confirmed that candidates of interest were resistant to tetracycline (encoded by Tn916-lacZ).
Mapping Tn916-lacZ integration sites
Arbitrary PCR was used to map Tn916-lacZ integration sites, as previously described [83,84]. Briefly, blue, tetracycline-resistant colonies were used as a template in a PCR reaction containing arbitrary primers (oELC1003: 5’- GGCACGCGTC GACTAGTACN NNNNNNNNNT GATG-3’) paired with a primer to either the right (oELC1009: 5’- GACATGCTAA TATAGCCATG ACG-3’) or left (oELC1010: 5’-GAAGTATCTT TATATCTTCA CTTTTCAAGG-3’) end of Tn916. Purified PCR products were then amplified using oELC1004 (5’-GGCACGCGTC GACTAGTAC-3’) and oELC1011 (5’-GAACTATTAC GCACATGCAA C-3’) for the right junction or oELC1012 (5’-CGTCGTATCA AAGCTCATTC ATAAG-3’) for the left junction. These PCR products were then sequenced with oELC1011 or oELC1012 and mapped to B. subtilis genome (Genbank accession number CP007800 [70]).
Competition assays and fitness
Strains and growth.
Tn916 T1+ and T1– strains were created by first replacing lacZ from Tn916-lacZ with kan flanked by lox sites and then removing kan. DNA fragments from upstream and downstream of lacZ (in Tn916-lacZ) were assembled by isothermal assembly [73] with either T1+ or T1– and the lox-flanked kan and recombined into each Tn916-lacZ insertion by transformation and selection for resistance to kanamycin. The lox-flanked kan marker was then removed by Cre-mediated recombination (using pDR244 [76] as described above). Strains made include Tn916 inserted in: sdpA (T1+, ESW675; T1–, ESW693), sunT (T1+, ESW677; T1–, ESW695), fadR (T1+, ESW724; T1–, ESW725), adaB-ndhF (T1+, ESW722; T1–, ESW723; T1– ΔoriT, ESW728), yvgT-bdbC (T1+, ESW713; T1–, ESW714).
Because the insertions no longer contained lacZ, we were able to use a constitutively expressed lacZ (Ppen-lacZ) at amyE in the Tn916 T1+ strains to distinguish them from the T1– strains. amyE::[(Ppen-lacZ), spc] (ESW617) was made by PCR amplifying Ppen-lacZ from pCAL1422 [45] and inserting it into BamHI-BlpI-cut pAJW82 (a pDR110-derived vector that is lacking Pspank and lacI) by isothermal assembly [73], and then integrating it into B. subtilis. The cells with Tn916 T1– contained amyE::spc (ESW616) from pAJW82 with no insert. Competition experiments demonstrated that Ppen-lacZ did not affect fitness.
Strains with Tn916 containing a wild-type or mutant T1 were grown in LB medium to early exponential phase. Strains were then mixed at the indicated ratio after adjusting their OD600 to ~0.01 and 50 μl of each competition mixture was spotted on LB agar and grown for 24 hours (1 day) at 37°C. Every 24 hours, spots were resuspended in 1X Spizizen’s salts [68] and diluted to an OD600 of ~0.01 and then 50 μl of this resuspension was spotted on fresh LB agar and grown for another 24 hours at 37°C. This was repeated for a total of 5 days. Under these conditions, cells progress through ~10 doublings per growth cycle on LB agar. At generation (doubling) 0 (initial input) and every 24 hours, the ratio of the two strains in each mixture was determined by serially diluting and plating the appropriate dilution on LB agar containing X-gal and counting the number of lacZ+ (blue; Tn916 T1+) and lacZ−(white; Tn916 T1–) colonies.
Fitness calculations.
We calculated the fitness of the strain containing the mutant relative to that of the strain containing the wild-type terminator T1 using the equation derived from the Malthusian parameter estimate of fitness as described [85].
wT1mut is the fitness of the strain with Tn916 T1– relative to that with Tn916 T1+. NT1mut and NWT are the numbers of CFU/ml of the strains with the mutant and wild-type T1, respectively. i indicates the initial number of CFU/ml of the indicated strain in the mixture and f indicates the CFU/ml after growth of the mixed strains. d, the dilution factor, is the fold-dilution of the cells from the start of the experiment (i) to the time (f) at which fitness was determined. Fitness was determined at a time when the population size was still changing (Fig 6D–6J). The dilution factor used to calculate fitness was ~1,000 (two days of growth) for strains with insertions in sdpA, sunT, adaB-ndhF ΔoriT, and yvgT-bdbC, and the strains used for the lacZ control, and 1 (only one day of growth) for strains with insertions in fadR and adaB-ndhF.
Control competitions were performed to determine the fitness associated with the amyE::[(EMPTY) spc] marker used in Tn916 T1– cells (ESW616) relative to the amyE::[(Ppen-lacZ) spc] marker (ESW617) used in Tn916 T1+ cells. The relative fitness was 1.02 ± 0.02 (mean ± SEM from three independent experiments), indicating that lacZ expression did not affect fitness of host cells (Fig 6C).
Supporting information
S1 Fig. Terminator T2 and T3 of Tn916.
The nucleotide positions of the base of the terminator stem were determined by the ARNold web server [28,29]. The minimum free energy of folding ΔG (of the stem-loop) was calculated using the RNAfold web server [31]. A) Terminator T2a and T2b. Red, bolded UAA indicate the stop codon of orf18. B) Terminator T3.
https://doi.org/10.1371/journal.pgen.1011417.s001
(TIF)
S2 Fig. Pspank-Tn916 cannot excise and conjugate.
A) Number of circular Tn916 per cell (attTn916/mrpG) and B) conjugation efficiencies of wild-type Tn916 (CMJ253), Pspank-Tn916 T1+ (ESW179), and Pspank-Tn916 T1– (ESW247). All strains were grown without tetracycline. Pspank-Tn916 strains were grown continuously with IPTG. Data presented are from one experiment. Mating assays of Pspank-Tn916 T1+ and Pspank-Tn916 T1– resulted in conjugation efficiencies that are below the limit of detection.
https://doi.org/10.1371/journal.pgen.1011417.s002
(TIF)
S3 Fig. Analysis of the antitermination activity of different DNA fragments from the canT region of Tn916.
A) Schematic of lacZ reporter construct with canT alleles and terminator T1 between Pspank and lacZ. B-I) DNA regions tested for antitermination activity. The size of each fragment is shown. The effect caused by canT(S3) was determined for two of the cloned fragments (B,C). β-galactosidase specific activities were measured two hours after induction of Pspank with IPTG and relative specific activities were calculated as the mean β-galactosidase specific activity of each strain divided by the mean β-galactosidase specific activity of the strain with the 452 bp fragment of the wild type canT allele. B) [canT (WT, 452 bp), ESW398] and [canT (S3, 452bp), ESW407]. Data for each are from three independent experiments. C) [canT (WT, 400 bp), ESW437] and [canT (S3, 400bp), ESW459]. Data for each are from three independent experiments. D) [canT (WT, 373 bp), ESW450]. Data presented are from three independent experiments. E) [canT (WT, 349 bp), ESW436]. Data presented are from one experiment. F) [canT (WT, 291 bp), ESW422]. Data presented are from one experiment. G) [canT (WT, 341 bp), ESW545]. Data presented are from three independent experiments. H) [canT (WT, 290 bp), ESW438]. Data presented are from one experiment. I) [canT (WT, 232 bp), ESW423]. Data presented are from one experiment.
https://doi.org/10.1371/journal.pgen.1011417.s003
(TIF)
S4 Fig. Predicted secondary structures of different RNAs from the canT region of Tn916.
RNA secondary structure predictions were generated using the ViennaRNA package [30] in the SnapGene software. Secondary structures shown are calculated to have the lowest predicted free energy. 5’ and 3’ ends of the RNA are indicated. The predicted structures are colored based on estimated confidence on bases being paired or unpaired (red: 90% and greater, yellow: 70–89%, light blue: 50–69%, dark blue: less than 50%). Structures within green and blue boxes are referred to as the left and right sides of the predicted structures. A) Predicted secondary structure of the 452-nucleotide wild-type canT RNA. The location of S3 mutation is indicated (although the structure is that predicted for the wild-type). B) Predicted secondary structure of the mutant canT(Δ162) RNA. The 5’ and 3’ boundaries are the same as those for the 452-nucleotide wild-type canT RNA, but with 162 nucleotides deleted. See Fig 3B for details on the deletion. This allele is inactive in antitermination. C) Predicted secondary structure of the 452-nucleotide mutant canT(S3) RNA. The canT(S3) mutation changes 3 nucleotides (5’-AUA-3’ to 5’-UGC-3’) as shown. This allele is inactive in antitermination. D) Predicted secondary structure of a 291-nucleotide RNA fragment from the canT region. This allele is inactive in antitermination. E) Predicted secondary structure of a 341 nucleotide RNA fragment from the canT region. This fragment has antitermination activity.
https://doi.org/10.1371/journal.pgen.1011417.s004
(TIF)
S5 Fig. Expression of host promoters upstream of four Tn916 insertions.
The original Tn916-lacZ insertions were used to monitor expression from host promoters under the conditions indicated. In both panel A and B, data presented are from one experiment. A) β-galactosidase activities are plotted as a function of cell density (OD600) during growth in defined liquid glucose medium at 37°C. The slope of each line is the differential rate of synthesis from the indicated promoter. Circles, sdpA::Tn916-lacZ (ESW517); squares, sunT::Tn916-lacZ (ESW562); triangles, fadR::Tn916-lacZ (ESW557); inverted triangles, adaB-ndhF::Tn916-lacZ (ESW561). B) β-galactosidase specific activities of Tn916-lacZ insertion strains grown as spots on LB agar at 37°C. Measurements were taken from spots resuspended in buffer after 8 hours of growth. Strains used were the same as in panel A.
https://doi.org/10.1371/journal.pgen.1011417.s005
(TIF)
S6 Fig. DNA sequence for putative transcription terminators near the left end of Tn916-like elements.
Yellow-highlighted sequences are the predicted stem-loops of the putative terminators near the left end of the indicated elements. None of these have been tested experimentally for terminator activity. A) Sequences that are identical to that of the Tn916 terminator T1 or have a few nucleotide differences in the loop or regions upstream and downstream from the predicted stems. B) Tn5386, Tn5397, Tn5801 have predicted terminators with different sequences than that of Tn916 terminator T1.
https://doi.org/10.1371/journal.pgen.1011417.s006
(TIF)
S1 Data. Underlying raw data for experiments presented in the figures.
The excel spreadsheet contains the underlying data for the experiments presented in each of the figures.
https://doi.org/10.1371/journal.pgen.1011417.s007
(XLSX)
Acknowledgments
We thank Mary E. Anderson for comments on the manuscript, Arielle J. Weinstein for plasmid pAJW82, Laurel D. Wright for previously constructing the Δorf23-22 and Δorf20 alleles used here, and Gabriel T. Vercelli for advice on fitness calculations.
References
- 1. Wozniak RAF, Waldor MK. Integrative and conjugative elements: mosaic mobile genetic elements enabling dynamic lateral gene flow. Nat Rev Microbiol. 2010;8: 552–563. pmid:20601965
- 2. Delavat F, Miyazaki R, Carraro N, Pradervand N, van der Meer JR. The hidden life of integrative and conjugative elements. FEMS Microbiol Rev. 2017;41: 512–537. pmid:28369623
- 3. Johnson CM, Grossman AD. Integrative and conjugative elements (ICEs): what they do and how they work. Annu Rev Genet. 2015;49: 577–601. pmid:26473380
- 4. Franke AE, Clewell DB. Evidence for a chromosome-borne resistance transposon (Tn916) in Streptococcus faecalis that is capable of “conjugal” transfer in the absence of a conjugative plasmid. J Bacteriol. 1981;145: 494–502.
- 5. Franke AE, Clewell DB. Evidence for Conjugal Transfer of a Streptococcus faecalis Transposon (Tn916) from a Chromosomal Site in the Absence of Plasmid DNA. Cold Spring Harb Symp Quant Biol. 1981;45: 77–80. pmid:6271493
- 6. Fitzgerald GF, Clewell DB. A conjugative transposon (Tn919) in Streptococcus sanguis. Infect Immun. 1985;47: 415–420.
- 7. Clewell DB, An FY, White BA, Gawron-Burke C. Streptococcus faecalis sex pheromone (cAM373) also produced by Staphylococcus aureus and identification of a conjugative transposon (Tn918). J Bacteriol. 1985;162: 1212–1220.
- 8.
Clewell DB, Flannagan SE. The conjugative transposons of Gram-positive bacteria. In: Clewell DB, editor. Bacterial conjugation. Boston, MA: Springer US; 1993. pp. 369–393. https://doi.org/10.1007/978-1-4757-9357-4_15
- 9. Roberts AP, Mullany P. A modular master on the move: the Tn916 family of mobile genetic elements. Trends Microbiol. 2009;17: 251–258. pmid:19464182
- 10. Roberts AP, Mullany P. Tn916-like genetic elements: a diverse group of modular mobile elements conferring antibiotic resistance. FEMS Microbiol Rev. 2011;35: 856–871. pmid:21658082
- 11. Santoro F, Vianna ME, Roberts AP. Variation on a theme; an overview of the Tn916/Tn1545 family of mobile genetic elements in the oral and nasopharyngeal streptococci. Front Microbiol. 2014;5. pmid:25368607
- 12. Sansevere EA, Robinson DA. Staphylococci on ICE: overlooked agents of horizontal gene transfer. Mobile Genetic Elements. 2017;7: 1–10. pmid:28932624
- 13. Christie PJ, Korman RZ, Zahler SA, Adsit JC, Dunny GM. Two conjugation systems associated with Streptococcus faecalis plasmid pCF10: identification of a conjugative transposon that transfers between S. faecalis and Bacillus subtilis. J Bacteriol. 1987;169: 2529–2536.
- 14. Ivins BE, Welkos SL, Knudson GB, Leblanc DJ. Transposon Tn916 mutagenesis in Bacillus anthracis. Infect Immun. 1988;56: 176–181.
- 15. Scott JR, Kirchman PA, Caparon MG. An intermediate in transposition of the conjugative transposon Tn916. Proc Natl Acad Sci U S A. 1988;85: 4809–4813. pmid:2838847
- 16. Mullany P, Wilks M, Lamb I, Clayton C, Wren B, Tabaqchali S. Genetic analysis of a tetracycline resistance element from Clostridium difficile and its conjugal transfer to and from Bacillus subtilis. Microbiol. 1990;136: 1343–1349. pmid:2172445
- 17. Roberts AP, Hennequin C, Elmore M, Collignon A, Karjalainen T, Minton N, et al. Development of an integrative vector for the expression of antisense RNA in Clostridium difficile. J Microbiol Methods. 2003;55: 617–624. pmid:14607405
- 18. Wright LD, Grossman AD. Autonomous Replication of the Conjugative Transposon Tn916. J Bacteriol. 2016;198: 3355–3366. pmid:27698087
- 19. Celli J, Trieu-Cuot P. Circularization of Tn916 is required for expression of the transposon-encoded transfer functions: characterization of long tetracycline-inducible transcripts reading through the attachment site. Mol Microbiol. 1998;28: 103–117. pmid:9593300
- 20. Clewell DB, Flannagan SE, Ike Y, Jones JM, Gawron-Burke C. Sequence analysis of termini of conjugative transposon Tn916. J Bacteriol. 1988;170: 3046–3052.
- 21. Scott JR, Bringel F, Marra D, Van Alstine G, Rudy CK. Conjugative transposition of Tn916: preferred targets and evidence for conjugative transfer of a single strand and for a double-stranded circular intermediate. Mol Microbiol. 1994;11: 1099–1108. pmid:8022279
- 22. Cookson AL, Noel S, Hussein H, Perry R, Sang C, Moon CD, et al. Transposition of Tn916 in the four replicons of the Butyrivibrio proteoclasticus B316(T) genome. FEMS Microbiol Lett. 2011;316: 144–151. pmid:21204937
- 23. Mullany P, Williams R, Langridge GC, Turner DJ, Whalan R, Clayton C, et al. Behavior and Target Site Selection of Conjugative Transposon Tn916 in Two Different Strains of Toxigenic Clostridium difficile. Appl Environ Microbiol. 2012;78: 2147–2153. pmid:22267673
- 24. Mullany P, Wilks M, Tabaqchali S. Transfer of Tn916 and Tn916ΔE into Clostridium difficile: demonstration of a hot-spot for these elements in the C. difficile genome. FEMS Microbiol Lett. 1991;79: 191–194.
- 25. Wang H, Roberts AP, Mullany P. DNA sequence of the insertional hot spot of Tn916 in the Clostridium difficile genome and discovery of a Tn916-like element in an environmental isolate integrated in the same hot spot. FEMS Microbiol Lett. 2000;192: 15–20. pmid:11040422
- 26. Menard KL, Grossman AD. Selective pressures to maintain attachment site specificity of integrative and conjugative elements. PLoS Genet. 2013;9: e1003623. pmid:23874222
- 27. McKeithen-Mead SA, Grossman AD. Timing of integration into the chromosome is critical for the fitness of an integrative and conjugative element and its bacterial host. PLoS Genet. 2023;19: e1010524. pmid:36780569
- 28. Gautheret D, Lambert A. Direct RNA motif definition and identification from multiple sequence alignments using secondary structure profiles. J Mol Biol. 2001;313: 1003–1011. pmid:11700055
- 29. Macke TJ, Ecker DJ, Gutell RR, Gautheret D, Case DA, Sampath R. RNAMotif, an RNA secondary structure definition and search algorithm. Nucleic Acids Res. 2001;29: 4724–4735. pmid:11713323
- 30. Lorenz R, Bernhart SH, Höner zu Siederdissen C, Tafer H, Flamm C, Stadler PF, et al. ViennaRNA Package 2.0. Algorithms Mol Biol. 2011;6: 26. pmid:22115189
- 31. Gruber AR, Lorenz R, Bernhart SH, Neuböck R, Hofacker IL. The Vienna RNA Websuite. Nucleic Acids Res. 2008;36: W70–W74. pmid:18424795
- 32. Lu F, Churchward G. Conjugative transposition: Tn916 integrase contains two independent DNA binding domains that recognize different DNA sequences. EMBO J. 1994;13: 1541–1548. pmid:8156992
- 33. Senghas E, Jones JM, Yamamoto M, Gawron-Burke C, Clewell DB. Genetic organization of the bacterial conjugative transposon Tn916. J Bacteriol. 1988;170: 245–249.
- 34. Su YA, Clewell DB. Characterization of the left 4 kb of conjugative transposon Tn916: determinants involved in excision. Plasmid. 1993;30: 234–250. pmid:8302931
- 35. Marra D, Scott JR. Regulation of excision of the conjugative transposon Tn916. Mol Microbiol. 1999;31: 609–621. pmid:10027977
- 36. Irnov I, Winkler WC. A regulatory RNA required for antitermination of biofilm and capsular polysaccharide operons in Bacillales. Mol Microbiol. 2010;76: 559–575. pmid:20374491
- 37. Miguel-Arribas A, Val-Calvo J, Gago-Córdoba C, Izquierdo JM, Abia D, Wu LJ, et al. A novel bipartite antitermination system widespread in conjugative elements of Gram-positive bacteria. Nucleic Acids Res. 2021;49: 5553–5567. pmid:33999173
- 38. Goodson JR, Winkler WC. Processive antitermination. Microbiol Spectr. 2018;6: 6.5.05. pmid:30191803
- 39. Weisberg RA, Gottesman ME. Processive antitermination. J Bacteriol. 1999;181: 359–367. pmid:9882646
- 40. Oberto J, Clerget M, Ditto M, Cam K, Weisberg RA. Antitermination of early transcription in phage HK022: absence of a phage-encoded antitermination factor. J Mol Biol. 1993;229: 368–381. pmid:8429552
- 41. King RA, Banik-Maiti S, Jin DJ, Weisberg RA. Transcripts that increase the processivity and elongation rate of RNA polymerase. Cell. 1996;87: 893–903. pmid:8945516
- 42. Bose B, Auchtung JM, Lee CA, Grossman AD. A conserved anti-repressor controls horizontal gene transfer by proteolysis. Mol Microbiol. 2008;70: 570–582. pmid:18761623
- 43. Bose B, Grossman AD. Regulation of horizontal gene transfer in Bacillus subtilis by activation of a conserved site-specific protease. J Bacteriol. 2011;193: 22–29. pmid:21036995
- 44. Berkmen MB, Lee CA, Loveday E-K, Grossman AD. Polar positioning of a conjugation protein from the integrative and conjugative element ICEBs1 of Bacillus subtilis. J Bacteriol. 2010;192: 38–45. pmid:19734305
- 45. Thomas J, Lee CA, Grossman AD. A conserved helicase processivity factor is needed for conjugation and replication of an integrative and conjugative element. PLoS Genet. 2013;9: e1003198. pmid:23326247
- 46. Nicolas P, Mäder U, Dervyn E, Rochat T, Leduc A, Pigeonneau N, et al. Condition-dependent transcriptome reveals high-level regulatory architecture in Bacillus subtilis. Science. 2012;335: 1103–1106. pmid:22383849
- 47. Radding CM, Echols H. The role of the N gene of phage lambda in the synthesis of two phage-specific proteins. Proc Natl Acad Sci U S A. 1968;60: 707–712. pmid:4973489
- 48. Yarnell WS, Roberts JW. The phage λ gene Q transcription antiterminator binds DNA in the late gene promoter as it modifies RNA polymerase. Cell. 1992;69: 1181–1189. pmid:1535556
- 49. Deighan P, Hochschild A. The bacteriophage λQ anti-terminator protein regulates late gene expression as a stable component of the transcription elongation complex. Mol Microbiol. 2007;63: 911–920. pmid:17302807
- 50. Santangelo TJ, Artsimovitch I. Termination and antitermination: RNA polymerase runs a stop sign. Nat Rev Microbiol. 2011;9: 319–329. pmid:21478900
- 51. Sen R, King RA, Mzhavia N, Madsen PL, Weisberg RA. Sequence-specific interaction of nascent antiterminator RNA with the zinc-finger motif of Escherichia coli RNA polymerase. Mol Microbiol. 2002;46: 215–222. pmid:12366844
- 52. Hwang S, Olinares PDB, Lee J, Kim J, Chait BT, King RA, et al. Structural basis of transcriptional regulation by a nascent RNA element, HK022 putRNA. Nat Commun. 2022;13: 4668. pmid:35970830
- 53. Yin Z, Kaelber JT, Ebright RH. Structural basis of Q-dependent antitermination. Proc Natl Acad Sci U S A. 2019;116: 18384–18390. pmid:31455742
- 54. Yin Z, Bird JG, Kaelber JT, Nickels BE, Ebright RH. In transcription antitermination by Qλ, NusA induces refolding of Qλ to form a nozzle that extends the RNA polymerase RNA-exit channel. Proc Natl Acad Sci U S A. 2022;119: e2205278119. pmid:35951650
- 55. Said N, Krupp F, Anedchenko E, Santos KF, Dybkov O, Huang Y-H, et al. Structural basis for λN-dependent processive transcription antitermination. Nat Microbiol. 2017;2: 17062. pmid:28452979
- 56. Krupp F, Said N, Huang Y-H, Loll B, Bürger J, Mielke T, et al. Structural basis for the action of an all-purpose transcription anti-termination factor. Mol Cell. 2019;74: 143–157.e5. pmid:30795892
- 57. Miguel-Arribas A, Martín-María A, Alaerds ECW, Val-Calvo J, Yuste L, Rojo F, et al. Extraordinary long-stem confers resistance of intrinsic terminators to processive antitermination. Nucleic Acids Res. 2023; gkad333. pmid:37125647
- 58. Miguel-Arribas A, Wu LJ, Michaelis C, Yoshida K-I, Grohmann E, Meijer WJJ. Conjugation operons in Gram-positive bacteria with and without antitermination systems. Microorganisms. 2022;10: 587. pmid:35336162
- 59. Patterson TA, Zhang Z, Baker T, Johnson LL, Friedman DI, Court DL. Bacteriophage lambda N-dependent transcription antitermination: competition for an RNA site may regulate antitermination. J Mol Biol. 1994;236: 217–228. pmid:8107107
- 60. Herskowitz I, Signer EE. A site essential for expression of all late genes in bacteriophage λ. J Mol Biol. 1970;47: 545–556. pmid:5418171
- 61. Auchtung JM, Lee CA, Garrison KL, Grossman AD. Identification and characterization of the immunity repressor (ImmR) that controls the mobile genetic element ICEBs1 of Bacillus subtilis. Mol Microbiol. 2007;64: 1515–1528. pmid:17511812
- 62. Auchtung JM, Aleksanyan N, Bulku A, Berkmen MB. Biology of ICEBs1, an integrative and conjugative element in Bacillus subtilis. Plasmid. 2016;86: 14–25. pmid:27381852
- 63. Casjens SR, Hendrix RW. Bacteriophage lambda: early pioneer and still relevant. Virology. 2015;479–480: 310–330. pmid:25742714
- 64. Meyer BJ, Ptashne M. Gene regulation at the right operator (OR) of bacteriophage lambda. III. lambda repressor directly activates gene transcription. J Mol Biol. 1980;139: 195–205. pmid:6447796
- 65. Meyer BJ, Maurer R, Ptashne M. Gene regulation at the right operator (OR) of bacteriophage lambda. II. OR1, OR2, and OR3: their roles in mediating the effects of repressor and cro. J Mol Biol. 1980;139: 163–194. pmid:6447795
- 66. Maurer R, Meyer B, Ptashne M. Gene regulation at the right operator (OR) bacteriophage lambda. I. OR3 and autogenous negative control by repressor. J Mol Biol. 1980;139: 147–161. pmid:6447794
- 67. Jaacks KJ, Healy J, Losick R, Grossman AD. Identification and characterization of genes controlled by the sporulation-regulatory gene spo0H in Bacillus subtilis. J Bacteriol. 1989;171: 4121–4129.
- 68.
Harwood CR, Cutting SM. Molecular biological methods for Bacillus. Chichester, United Kingdom: John Wiley & Sons; 1990.
- 69. Perego M, Spiegelman GB, Hoch JA. Structure of the gene for the transition state regulator, abrB: regulator synthesis is controlled by the spo0A sporulation gene in Bacillus subtilis. Mol Microbiol. 1988;2: 689–699. pmid:3145384
- 70. Smith JL, Goldberg JM, Grossman AD. Complete genome sequences of Bacillus subtilis subsp. subtilis laboratory strains JH642 (AG174) and AG1839. Genome Announc. 2014;2: e00663–14. pmid:24994804
- 71. Auchtung JM, Lee CA, Monson RE, Lehman AP, Grossman AD. Regulation of a Bacillus subtilis mobile genetic element by intercellular signaling and the global DNA damage response. Proc Natl Acad Sci U S A. 2005;102: 12554–12559. pmid:16105942
- 72. Lemon KP, Kurtser I, Grossman AD. Effects of replication termination mutants on chromosome partitioning in Bacillus subtilis. Proc Natl Acad Sci U S A. 2001;98: 212–217.
- 73. Gibson DG, Young L, Chuang R-Y, Venter JC, Hutchison CA, Smith HO. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat Methods. 2009;6: 343–345. pmid:19363495
- 74. Wright LD, Johnson CM, Grossman AD. Identification of a single strand origin of replication in the integrative and conjugative element ICEBs1 of Bacillus subtilis. PLoS Genet. 2015;11: e1005556. pmid:26440206
- 75. Jaworski DD, Clewell DB. A functional origin of transfer (oriT) on the conjugative transposon Tn916. J Bacteriol. 1995;177: 6644–6651.
- 76. Meisner J, Montero Llopis P, Sham L-T, Garner E, Bernhardt TG, Rudner DZ. FtsEX is required for CwlO peptidoglycan hydrolase activity during cell wall elongation in Bacillus subtilis. Mol Microbiol. 2013;89: 1069–1083. pmid:23855774
- 77. Johnson CM, Grossman AD. Identification of host genes that affect acquisition of an integrative and conjugative element in Bacillus subtilis. Mol Microbiol. 2014;93: 1284–1301. pmid:25069588
- 78.
Miller JH. Experiments in molecular genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1972.
- 79.
Bustin SA, Nolan T. Data analysis and interpretation. A-Z of quantitative PCR. La Jolla, CA: International University Line; 2004. pp. 439–492.
- 80. Bean EL, Herman C, Anderson ME, Grossman AD. Biology and engineering of integrative and conjugative elements: construction and analyses of hybrid ICEs reveal element functions that affect species-specific efficiencies. PLoS Genet. 2022;18: e1009998. pmid:35584135
- 81. Bean EL, McLellan LK, Grossman AD. Activation of the integrative and conjugative element Tn916 causes growth arrest and death of host bacteria. PLoS Genet. 2022;18: e1010467. pmid:36279314
- 82. Pfaffl MW. A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Res. 2001;29: e45. pmid:11328886
- 83. Das S, Noe JC, Paik S, Kitten T. An improved arbitrary primed PCR method for rapid characterization of transposon insertion sites. J Microbiol Methods. 2005;63: 89–94. pmid:16157212
- 84. Brophy JAN, Triassi AJ, Adams BL, Renberg RL, Stratis-Cullum DN, Grossman AD, et al. Engineered integrative and conjugative elements for efficient and inducible DNA transfer to undomesticated bacteria. Nat Microbiol. 2018;3: 1043–1053. pmid:30127494
- 85. Lenski RE, Rose MR, Simpson SC, Tadler SC. Long-term experimental evolution in Escherichia coli. I. Adaptation and divergence during 2,000 generations. Am Nat. 1991;138: 1315–1341.
- 86. Browne HP, Anvar SY, Frank J, Lawley TD, Roberts AP, Smits WK. Complete genome sequence of BS49 and draft genome sequence of BS34A, Bacillus subtilis strains carrying Tn916. FEMS Microbiol Lett. 2015;362: 1–4. pmid:25673660
- 87. Haraldsen JD, Sonenshein AL. Efficient sporulation in Clostridium difficile requires disruption of the σK gene. Mol Microbiol. 2003;48: 811–821. pmid:12694623