A Novel Prophage-like Insertion Element within yabG Triggers Early Entry into Sporulation in Clostridium botulinum

Sporulation is a finely regulated morphogenetic program important in the ecology and epidemiology of Clostridium botulinum. Exogenous elements disrupting sporulation-associated genes contribute to sporulation regulation and introduce diversity in the generally conserved sporulation programs of endospore formers. We identified a novel prophage-like DNA segment, termed the yin element, inserted within yabG, encoding a sporulation-specific cysteine protease, in an environmental isolate of C. botulinum. Bioinformatic analysis revealed that the genetic structure of the yin element resembles previously reported mobile intervening elements associated with sporulation genes. Within a pure C. botulinum culture, we observed two subpopulations of cells with the yin element either integrated into the yabG locus or excised as a circular DNA molecule. The dynamics between the two observed conformations of the yin element was growth-phase dependent and likely mediated by recombination events. The yin element was not required for sporulation by C. botulinum but triggered an earlier entry into sporulation than in a related isolate lacking this element. So far, the yin element has not been found in any other C. botulinum strains or other endospore-forming species. It remains to be demonstrated what kind of competitive edge it provides for C. botulinum survival and persistence.


Introduction
Some bacteria escape starvation and other adverse conditions by forming resistant endospores through a time-and energy-consuming differentiation process called sporulation, elaborately described in several Bacillus, Clostridium, and related species [1][2][3][4].The process is coordinated and timely controlled by multiple sigma factors and accessory transcriptional regulators.Spores can survive extreme conditions over long periods of time [5,6] and, for pathogenic species, play a major role in the epidemiology of bacterial diseases, such as anthrax caused by Bacillus anthracis [7], Clostridioides difficile infection [8], tetanus caused by Clostridium tetani [9], Bacillus cereus infection [10], and botulism caused by Clostridium botulinum [11].Environmental spores of C. botulinum are of major concern for the food industry due to the risk of food-borne botulism [12] and can also pose direct risks to humans through exposure and colonization of the gut, causing toxicoinfectious botulism in infants [13] or in at-risk adults [14,15].Mobile genetic elements can contribute to the regulation of sporulation.In some of the B. subtilis and C. difficile strains, gene encoding the late-stage mother-cell sigma-factor SigK (σ K ) is disrupted by an intervening prophage-like element, termed the skin element.This element is timely excised during sporulation, resulting in a functional sigK to let sporulation proceed [16][17][18].Other sporulation-related genes interrupted by intervening prophage-like elements include spsM of B. subtilis, required for the maturation of the spore surface layers and interrupted by the SPβ prophage [19]; spoVFB of Bacillus weihenstephanensis coding for the β-subunit of dipicolinic acid synthase and interrupted by the vfbin element [20]; and gerE of Bacillus cereus coding for a transcription factor that works with SigK to control the late stages of the mother cell of gene expression [21].These reports underline a prevalence for mobile genetic elements (prophage) inserted in sporulation-associated genes and suggest direct or indirect modulation of the sporulation cascade, perhaps providing an ecological benefit to the host bacteria.Such bacteria-phage interactions relate to active lysogeny, where the prophage disrupts the target gene expression and its regulation and therefore acts as a regulatory switch (phage-RS) [22].
The yabG is regulated by SigK [23] and codes for a cysteine protease that regulates the assembly of the spore coat by processing several coat proteins in B. subtilis [24,25].YabG also processes two proteins required for spore germination in C. difficile, CspBA and Pre-pro-SleC, as well as a coat protein required for efficient host colonization [26,27].Interestingly, yabG is the only member of the SigK regulon that is conserved across spore formers and part of a genomic signature for sporulation [28].Of note, SigK is known as a late sporulation factor but has also shown an early role in sporulation across several clostridial species [5].In our previous work [29], we sequenced the genome of a C. botulinum strain (V73) isolated from household dust and related to a case of infant botulism.We identified a prophage-like sequence disrupting yabG (hereinafter referred to as the yin element, for the yabG-intervening element).To test the hypothesis that the yin element has a role in the control of sporulation in C. botulinum, we first compared the phenotypes of V73 and its counterpart devoid of the yin element (ST7B).We showed that the yin element was temporally excised from the chromosome and triggered an earlier entry into sporulation in V73 compared with ST7B.This work describes, for the first time, a regulatory prophagelike insertion element within yabG and further illustrates the diversity encountered in the regulation of the sporulation program in endospore formers while raising questions on the ecological benefit of such an intervening element for persistence and survival of the bacteria.

Bacterial Growth and DNA Isolation
C. botulinum isolates V73 and ST7B are associated with an infant botulism case and share a recent common ancestor [29,30].Both strains were cultured anaerobically at 37 • C overnight in trypticase-peptone-glucose-yeast-extract (TPGY) broth, and genomic DNA was extracted using Wizard ® Genomic DNA Purification Kit (Promega, Madison, WI, USA), as per the manufacturer's instructions.The DNA concentration was determined using the Qubit fluorometric assay (Thermo Scientific, Waltham, MA, USA).

Genome Sequencing and Bioinformatic Analysis
The genome of C. botulinum V73 was sequenced using the PacBio sequencer (Pacific Biosciences, Menlo Park, CA, USA).PacBio library preparation, genome assembly, and polishing were carried out as previously described [29].The complete and closed genome of V73 was obtained using both PacBio sequencing reads (this work) and Illumina paired-end reads (SRR11288493) [29].Genome assembly of C. botulinum V73 was analyzed using the Microbial Genomes Atlas (MiGA) web interface [31].Prophages were identified using the Phaster online analysis tool [32].The genome of V73 was annotated using various tools, including PATRIC [33], RASTtk [34], and Prokka [35].Gene annotation of the yin element was also conducted using the Phyre2 web portal [36].Protein-structure prediction and models of the complexes between recombination directionality factors (RDFs) and their associated recombinases were generated using AlphaFold [37].AlphaFold2 was used for structural modeling of individual proteins and complexes [37].We first obtained models for SpoIVCA, SprA (B.subtilis), and CD1231 (C.difficile) recombinases and their cognate directionality factors SkrB, SprB, and CD1234 and for the yin-encoded recombinase (HEQ52_18460).The models were obtained by uploading the sequences to the open access AlphaFold2 Colab notebook (https://colab.research.google.com/github/deepmind/alphafold/blob/main/notebooks/AlphaFold.ipynb,accessed on 16 August 2022) [38].We also generated models for putative complexes between the recombinases and their known or predicted directionality factors (as above).The interaction was predicted to involve the C-terminal end of the recombinases (Figure S1).We then generated models between the yin recombinase (HEQ52_18460) and the predicted yin-encoded proteins with 250 amino acid residues or fewer.Only one of the selected yin-encoded proteins, HEQ52_18460 (Table S1), was predicted to form a complex with the HEQ52_18485 recombinase.The interaction was predicted to form at the C-terminal region of the recombinase [39].Five models were generated for each protein or protein complex.The confidence of the modeling was assessed by the pLDDT metric and the predicted alignment error (PAE), i.e., uncertainty about the interface.Values of pLDDT > 90 were expected to be highly accurate.Structural representations were generated using PyMOL Molecular Graphics System (Schrödinger, Inc., New York, NY, USA).

Growth Curves
The growth of C. botulinum ST7B and V73 was monitored using the Hidex Sense multiwell plate reader (Hidex Oy, Turku, Finland).Briefly, following two consecutive overnight sub-cultures in TPGY medium at 37 • C in anaerobic conditions, cultures of the two isolates were diluted 1:100 in 5 mL of TPGY broth.Then, 200 µL of the resulting suspension were added per well of a 96-well plate.The plate was then placed into the Hidex Sense plate reader and incubated at 37 • C anaerobically.Optical density at 600 nm was recorded every 15 min for 24 h.The results were based on three biological replicate series consisting of four technical replicates of each isolate.

Toxin Quantification
C. botulinum V73 and ST7B cultures were grown at 37 • C anaerobically in TPGY medium.Each strain was inoculated 1:100 into 5 mL of fresh TPGY medium.Samples were collected for up to 24 h and stored at −80 • C. Botulinum neurotoxin (BoNT) type A1 was quantified using a sandwich-ELISA, as previously described [40,41], in triplicate.

Spore Count Assay and Spore Heat Resistance
C. botulinum V73 and ST7B cultures were grown anaerobically at 37 • C. Total viable cell and heat-resistant spore counts were measured at 48, 72, and 168 h after inoculation, as previously described [42].Spore heat resistance of C. botulinum V73 and ST7B was studied as described in [43], by heating for variable times at 98 • C and counting the survivors.

Spore Germination Assay
Spores of C. botulinum ST7B and V73 were prepared as previously described [43].The spores were then heat-activated at 80 • C for 15 min and OD-adjusted to ~0.5 in Tris-HCl (20 mM, pH 7.4) buffer with 50 mM of NaHCO 3 , 50 mM of sodium L-lactate, and either L-alanine (50 mM) or L-cysteine (50 mM).The microtiter plate containing the samples was then placed into the Hidex Sense multi-well plate reader and incubated anaerobically at   C under double orbital shaking.Optical density at 600 nm was recorded every 5 min for 24 h.Three technical replicates per strain per condition were included.

Transmission Electron Microscopy (TEM) Analysis
C. botulinum V73 and ST7B spore samples were purified and prepared for thin sectioning TEM analysis [43].The grids were visualized using a JEM-1400 transmission electron microscope (JEOL Ltd., Tokyo, Japan) at the Electron Microscopy Unit (EMBI, Institute of Biotechnology, University of Helsinki).

Detection and Quantification of the Different Forms of yin
From an overnight culture, C. botulinum ST7B and V73 were inoculated at a 1:100 dilution in TPGY medium and grown anaerobically at 37 • C.After 5, 24, 48, 72, 96, and 120 h of growth, genomic DNA from 1 mL of culture was extracted as described above.Segments of the yin element (amplicon spanning across yabG and the yin element), the intact yabG coding sequence (internal amplicon), and the circular yin element were amplified by PCR using HotStarTaq Master Mix Kit (Qiagen GmbH, Hilden, Germany) (Table S1) and visualized by agarose gel electrophoresis to detect the different yin element conformations over culture time.Finally, some of these amplicons were sent for Sanger sequencing (Macrogen Europe B.V., Amsterdam, The Netherlands).Using quantitative PCR and Maxima SYBR Green qPCR Master Mix (Thermo Scientific, Waltham, MA, USA), we further quantified the different yin element conformations at all time points.We amplified and quantified one fragment of yabG spanning across the whole yin element (intact yabG and excised yin) and one fragment spanning across one of the att sites (chromosomally integrated yin).As a reference, we amplified one segment of yabG that is not altered by the excision of the yin element.

Genome Sequence Accession Numbers
The closed genome of C. botulinum V73 was deposited in NCBI under the accession numbers CP050820 (chromosome) and CP050821 (plasmid).

Identification of a Prophage-like Intervening Element in yabG
We recently showed that C. botulinum ST7B (genome accession numbers: CP050251, CP050252) and V73 are genetically related and essentially differ by the presence of one prophage based on read mapping using Illumina next-generation sequencing.Beside the presence of the yin element in C. botulinum V73, there were five SNPs/InDels different between the ST7B and V73 genomes: two located in intergenic regions and three within two genes (encoding a MurR/RpiR family transcriptional regulator or a histidine kinase) [29].These SNPs/InDels are unlikely to impact the sporulation phenotype of C. botulinum ST7B and V73 [29].To confirm the previously published Illumina next-generation sequencing results, we closed and annotated the genome of C. botulinum V73.The genome assembly confirmed the presence of a prophage located within yabG (Figure 1).yabG encodes a sporulation-specific protease shown to regulate the assembly of the spore coat in both B. subtilis and C. difficile [23][24][25]27].
The yin element was indeed predicted to be a putatively functional prophage and is genetically related to the class Caudoviricetes associated with Clostridium hosts (NC_029001, NC_007917, NC_028996, NC_028991, and NC_029004) or Brevibacillus hosts (NC_029029).Some of the predicted genes present in insertional conjugative elements (ICE) or prophagemediated DNA rearrangement elements of endospore formers [17][18][19] can also be found in the yin element, as further detailed in Table S2.Within the yin element, the first open reading frame (HEQ52_18485) located downstream of the 5 -end of yabG encodes a putative recombinase (integrase) with significant Pfam-A matches to the resolvase (PF00239, Evalue = 1.0 × 10 −38 ), recombinase (PF07508, E-value = 5.8 × 10 −22 ), and recombinase zinc-beta-ribbon-domain protein families (PF13408, E-value = 4.9 × 10 −13 ) (Pfam 35.0 database [44]).The recombinase HEQ52_18485 shares 27% amino acid identity (100 out of 372 residues) with the SpoIVCA recombinase present in skin of B. subtilis [16] and 25% amino acid identity (87/352) with the resolvase CD630_12340 associated with sporulation and excision of skin in C. difficile [17].Both spoIVCA in B. subtilis and HEQ52_18485 of C. botulinum V73 are located downstream of the 5 -end region of their respective disrupted gene (sigK and yabG) and in a similar orientation.The yin element was indeed predicted to be a putatively functional prophage and is genetically related to the class Caudoviricetes associated with Clostridium hosts (NC_029001, NC_007917, NC_028996, NC_028991, and NC_029004) or Brevibacillus hosts (NC_029029).Some of the predicted genes present in insertional conjugative elements (ICE) or prophage-mediated DNA rearrangement elements of endospore formers [17][18][19] can also be found in the yin element, as further detailed in Table S2.Within the yin element, the first open reading frame (HEQ52_18485) located downstream of the 5′-end of yabG encodes a putative recombinase (integrase) with significant Pfam-A matches to the resolvase (PF00239, E-value = 1.0 × 10 −38 ), recombinase (PF07508, E-value = 5.8 × 10 −22 ), and recombinase zinc-beta-ribbon-domain protein families (PF13408, E-value = 4.9 × 10 −13 ) (Pfam 35.0 database [44]).The recombinase HEQ52_18485 shares 27% amino acid identity (100 out of 372 residues) with the SpoIVCA recombinase present in skin of B. subtilis [16] and 25% amino acid identity (87/352) with the resolvase CD630_12340 associated with sporulation and excision of skin in C. difficile [17].Both spoIVCA in B. subtilis and HEQ52_18485 of C. botulinum V73 are located downstream of the 5´-end region of their respective disrupted gene (sigK and yabG) and in a similar orientation.
In addition to the recombinase, intervening elements typically require a recombination directionality factor (RDF) [18].Using AlphaFold, we identified a putative RDF encoded within the yin element (Figure S1).When compared with the skin element of C. difficile, the putative RDF HEQ52_18460 differed from CD1234 in size (72 amino acids vs. 53 amino acids) and pI (5.5 vs. 4.9).Yet, based on AlphaFold protein-structure prediction and models of the complexes between the recombinase and the RDF of different intervening elements, i.e., SpoIVCA and Skr, CD1231 and CD1234, or SprA and SprB, the RDFs were all different, but models predicted that they interact with the same region of their associated recombinase (Figure S1).We also possibly identified a putative SigK-dependent promoter upstream of HEQ52_18460 (Figure S2), suggesting its regulation by the latesporulation sigma-factor SigK.This is in line with the expression of yabG being under the control of SigK [23].Excision of the yin element would reconstitute yabG under SigK control.The RDF candidate (HEQ52_18460) identified in the yin element remains to be further studied experimentally.To the best of our knowledge, the yin element is unique: None of the 26 isogenic C. botulinum isolates collected over time from the single infant botulism case [29], nor any available C. botulinum or other bacterial genomes deposited in NCBI, In addition to the recombinase, intervening elements typically require a recombination directionality factor (RDF) [18].Using AlphaFold, we identified a putative RDF encoded within the yin element (Figure S1).When compared with the skin element of C. difficile, the putative RDF HEQ52_18460 differed from CD1234 in size (72 amino acids vs. 53 amino acids) and pI (5.5 vs. 4.9).Yet, based on AlphaFold protein-structure prediction and models of the complexes between the recombinase and the RDF of different intervening elements, i.e., SpoIVCA and Skr, CD1231 and CD1234, or SprA and SprB, the RDFs were all different, but models predicted that they interact with the same region of their associated recombinase (Figure S1).We also possibly identified a putative SigK-dependent promoter upstream of HEQ52_18460 (Figure S2), suggesting its regulation by the late-sporulation sigma-factor SigK.This is in line with the expression of yabG being under the control of SigK [23].Excision of the yin element would reconstitute yabG under SigK control.The RDF candidate (HEQ52_18460) identified in the yin element remains to be further studied experimentally.To the best of our knowledge, the yin element is unique: None of the 26 isogenic C. botulinum isolates collected over time from the single infant botulism case [29], nor any available C. botulinum or other bacterial genomes deposited in NCBI, harbored the yin element (as of November 2023).Moreover, no other insertion elements were found to disrupt yabG in the genomes of spore-forming bacteria in public databases.

Phenotypic Impact of the yin Element on Sporulation and Toxin Production
The yin element is located within a sporulation gene in C. botulinum V73.Based on previously described prophage elements regulating gene expression, we considered the possibility that the yin element could play a similar regulatory role on yabG expression through timed self-excision.We investigated the impact of the yin element on the phenotype of C. botulinum V73 against its yin-devoid counterpart C. botulinum ST7B.C. botulinum V73 and ST7B cultures showed similar growth until entry into the stationary phase (Figure 2), suggesting that the yin element does not impact the overall physiology of the vegetative cells or cultures.type of C. botulinum V73 against its yin-devoid counterpart C. botulinum ST7B.C. botulinum V73 and ST7B cultures showed similar growth until entry into the stationary phase (Figure 2), suggesting that the yin element does not impact the overall physiology of the vegetative cells or cultures.
However, the growth curves of V73 and ST7B showed distinct profiles in the stationary phase, where a prominent drop in the optical density of the V73 culture likely reflected a higher number of cells entering sporulation compared with that of ST7B (Figure 2).Phase-contrast microscopy of the C. botulinum V73 and ST7B cultures showed that the V73 culture initiated sporulation earlier and yielded higher free-spore counts than the ST7B culture (Figures 3, 4, and S3).Accordingly, after 240 h, the V73 culture consisted mostly of free spores, whereas in the ST7B culture, several cell types were still represented (Figures 3 and 4).Thus, the distinct growth profiles of ST7B and V73 in stationary phases were likely due to different numbers of cells undergoing sporulation (Figure 2).In regard to the earlier entry into sporulation by C. botulinum V73 compared with ST7B, possible expression control of yabG by SigK and by other regulators remains to be further explored and understood, especially since SigK also showed an early role in sporulation across several clostridial species [5].However, the growth curves of V73 and ST7B showed distinct profiles in the stationary phase, where a prominent drop in the optical density of the V73 culture likely reflected a higher number of cells entering sporulation compared with that of ST7B (Figure 2).
Phase-contrast microscopy of the C. botulinum V73 and ST7B cultures showed that the V73 culture initiated sporulation earlier and yielded higher free-spore counts than the ST7B culture (Figures 3, 4 and S3).Accordingly, after 240 h, the V73 culture consisted mostly of free spores, whereas in the ST7B culture, several cell types were still represented (Figures 3 and 4).Thus, the distinct growth profiles of ST7B and V73 in stationary phases were likely due to different numbers of cells undergoing sporulation (Figure 2).In regard to the earlier entry into sporulation by C. botulinum V73 compared with ST7B, possible expression control of yabG by SigK and by other regulators remains to be further explored and understood, especially since SigK also showed an early role in sporulation across several clostridial species [5].
previously described prophage elements regulating gene expression, we considered the possibility that the yin element could play a similar regulatory role on yabG expression through timed self-excision.We investigated the impact of the yin element on the phenotype of C. botulinum V73 against its yin-devoid counterpart C. botulinum ST7B.C. botulinum V73 and ST7B cultures showed similar growth until entry into the stationary phase (Figure 2), suggesting that the yin element does not impact the overall physiology of the vegetative cells or cultures.
However, the growth curves of V73 and ST7B showed distinct profiles in the stationary phase, where a prominent drop in the optical density of the V73 culture likely reflected a higher number of cells entering sporulation compared with that of ST7B (Figure 2).Phase-contrast microscopy of the C. botulinum V73 and ST7B cultures showed that the V73 culture initiated sporulation earlier and yielded higher free-spore counts than the ST7B culture (Figures 3, 4, and S3).Accordingly, after 240 h, the V73 culture consisted mostly of free spores, whereas in the ST7B culture, several cell types were still represented (Figures 3 and 4).Thus, the distinct growth profiles of ST7B and V73 in stationary phases were likely due to different numbers of cells undergoing sporulation (Figure 2).In regard to the earlier entry into sporulation by C. botulinum V73 compared with ST7B, possible expression control of yabG by SigK and by other regulators remains to be further explored and understood, especially since SigK also showed an early role in sporulation across several clostridial species [5].The sporulation master regulator Spo0A controls BoNT production in C. botulinum Group II type E [45].Due to the apparent link between sporulation and toxinogenesis in C. botulinum Group II, we examined whether the presence of the yin element, through impacting sporulation, may also result in distinct levels of BoNT production between ST7B and V73.Based on ELISA quantification, the V73 culture produced a slightly higher level of botulinum neurotoxin (BoNT) than ST7B after 24 h (Figure S4).However, the difference was marginal and may be explained by the faster release of BoNT through efficient sporulation (Figures 3 and 4) in the V73 cultures compared with ST7B cultures.Hence, the yin element is not likely to play a direct role in BoNT production or its regulation in the conditions tested.
impacting sporulation, may also result in distinct levels of BoNT production between ST7B and V73.Based on ELISA quantification, the V73 culture produced a slightly higher level of botulinum neurotoxin (BoNT) than ST7B after 24 h (Figure S4).However, the difference was marginal and may be explained by the faster release of BoNT through efficient sporulation (Figures 3 and 4) in the V73 cultures compared with ST7B cultures.Hence, the yin element is not likely to play a direct role in BoNT production or its regulation in the conditions tested.S3).

Restoration of Intact yabG by Chromosomal Excision of the yin Element upon Sporulation
C. botulinum V73 and ST7B are phylogenetically related and essentially differ by the presence of the yin element.Sequence analysis of the yabG region of ST7B led us to identify a 14-bp repeat sequence flanking an inverted repeat sequence present at both extremities of the yin element (attR and attL sites), constituting the putative recombination sites recognized by the predicted site-specific recombinase HEQ52_18485 (Figure 1 and Table S2).Work performed on the skin element [17] and the presence of putative attachment sites (att) and site-specific recombinase within the yin element led us to hypothesize that the yin element of V73 is excised, yielding a restored yabG and a circularized yin element.To test this possibility, we designed sets of primers that aimed to amplify (i) a chromosomally integrated yin element, (ii) an intact yabG gene, and (iii) an excised and circularized yin element (Figure 1) and monitored their accumulation along time during sporulation.
The yin element was found chromosomally integrated at all time points in V73 (Figure 5a).Moreover, amplicons corresponding to intact yabG were also observed at any time  S3).

Restoration of Intact yabG by Chromosomal Excision of the yin Element upon Sporulation
C. botulinum V73 and ST7B are phylogenetically related and essentially differ by the presence of the yin element.Sequence analysis of the yabG region of ST7B led us to identify a 14-bp repeat sequence flanking an inverted repeat sequence present at both extremities of the yin element (attR and attL sites), constituting the putative recombination sites recognized by the predicted site-specific recombinase HEQ52_18485 (Figure 1 and Table S2).Work performed on the skin element [17] and the presence of putative attachment sites (att) and site-specific recombinase within the yin element led us to hypothesize that the yin element of V73 is excised, yielding a restored yabG and a circularized yin element.To test this possibility, we designed sets of primers that aimed to amplify (i) a chromosomally integrated yin element, (ii) an intact yabG gene, and (iii) an excised and circularized yin element (Figure 1) and monitored their accumulation along time during sporulation.
The yin element was found chromosomally integrated at all time points in V73 (Figure 5a).Moreover, amplicons corresponding to intact yabG were also observed at any time points in V73 (Figure 5a), suggesting that yabG was reconstituted through chromosomal excision of the yin element at all time points.In line with detecting the intact yabG, we also detected circularized yin (Figure 5a, further discussed below).The observations suggest a heterogeneous population with cells carrying a yin element either integrated in yabG or excised to produce a circular DNA element.Temporal excision of the skin element during sporulation was reported in C. difficile [17,18] and made us hypothesize that the dynamics of the yin element between chromosomal integration and excision into circular DNA could be temporally regulated.Indeed, the C. botulinum V73 population harbored a smaller proportion of cells with a restored yabG at 5 and 24 h than at later times, suggesting that an increasing proportion of bacterial cells harbored an intact yabG upon initiation of sporulation (Figure 5b).Upon excision of the yin element, spliced yabG was restored as an intact yabG gene in V73 (Figure S5).The restored yabG retained the 14-bp repeat (attB) required for recombination, as in ST7B (Figure S6).The attB is also present and identical to the sequence of yabG in other C. botulinum strains, such as CDC_297, A634, and CDC_69096, suggesting that the yin element could integrate into the chromosome at the yabG locus in other C. botulinum strains.
heterogeneous population with cells carrying a yin element either integrated in yabG or excised to produce a circular DNA element.Temporal excision of the skin element during sporulation was reported in C. difficile [17,18] and made us hypothesize that the dynamics of the yin element between chromosomal integration and excision into circular DNA could be temporally regulated.Indeed, the C. botulinum V73 population harbored a smaller proportion of cells with a restored yabG at 5 and 24 h than at later times, suggesting that an increasing proportion of bacterial cells harbored an intact yabG upon initiation of sporulation (Figure 5b).Upon excision of the yin element, spliced yabG was restored as an intact yabG gene in V73 (Figure S5).The restored yabG retained the 14-bp repeat (attB) required for recombination, as in ST7B (Figure S6).The attB is also present and identical to the sequence of yabG in other C. botulinum strains, such as CDC_297, A634, and CDC_69096, suggesting that the yin element could integrate into the chromosome at the yabG locus in other C. botulinum strains.

Circularization of the yin Element upon Chromosomal Excision
Recombinases mediate the excision and circularization of the skin element in B. subtilis and C. difficile [16,17,46,47].To verify whether the yin element became circular upon excision from the chromosome, we designed primers located at both extremities of the yin element and in divergent orientation.Therefore, PCR products could only be generated upon amplification from the predicted circular yin element.Amplicons were obtained at all time points tested (Figure 5a), indicating that the excised yin element was indeed circular.This is in line with the aforementioned observations that an intact yabG was reconstituted at all time points tested.Sequencing of these amplicons revealed the attP with the same 14-bp repeat sequence flanked by asymmetric overlapping sequences (Figure S7), confirming that the recombination events are indeed mediated through att sites.

Presence of the yin Element Does Not Affect Spore Heat Resistance
Since YabG is involved in the assembly of the spore surface in B. subtilis [24,25] and spore germination in C. difficile [26,27], we hypothesized that the presence/absence of the yin element could control the function of YabG and therefore alter the spore properties.Thermal destruction assays showed that the V73 spores were slightly less heat resistant than spores of ST7B (most heat-sensitive spore subpopulations, depicted in Figure 6 in the first 20 min of heating); however, the difference was not statistically significant for the most heat-resistant spore subpopulations (depicted in Figure 6 after the first 20 min of heating).
excision from the chromosome, we designed primers located at both extremities of the yin element and in divergent orientation.Therefore, PCR products could only be generated upon amplification from the predicted circular yin element.Amplicons were obtained at all time points tested (Figure 5a), indicating that the excised yin element was indeed circular.This is in line with the aforementioned observations that an intact yabG was reconstituted at all time points tested.Sequencing of these amplicons revealed the attP with the same 14-bp repeat sequence flanked by asymmetric overlapping sequences (Figure S7), confirming that the recombination events are indeed mediated through att sites.

Presence of the Yin Element Does Not Affect Spore Heat Resistance
Since YabG is involved in the assembly of the spore surface in B. subtilis [24,25] and spore germination in C. difficile [26,27], we hypothesized that the presence/absence of the yin element could control the function of YabG and therefore alter the spore properties.Thermal destruction assays showed that the V73 spores were slightly less heat resistant than spores of ST7B (most heat-sensitive spore subpopulations, depicted in Figure 6 in the first 20 min of heating); however, the difference was not statistically significant for the most heat-resistant spore subpopulations (depicted in Figure 6 after the first 20 min of heating).Purified spore suspensions were adjusted at OD600nm 1 and heat-treated at 98 °C for different incubation times.D-values were obtained from the linear regression lines for each dataset.We observed the presence of two spore subpopulations, as shown with the two distinct destruction curves fitted on the data of each strain.The heat resistance of V73 spores appeared generally slightly lower than that of ST7B spores; however, the difference was statistically significant (p< 0.05, t-test) only for the most heat-sensitive spore subpopulations depicted during the first 20 min of the experiment.
Overall, the presence of the yin element had therefore no significant impact on the spore heat-resistance properties of V73 and ST7B.In line with this, transmission electron microscopy of purified V73 and ST7B spore preparations did not reveal any obvious structural differences between ST7B and V73 spores (Figure S8).Moreover, the germination patterns of V73 and ST7B spores were comparable (Figure S9).Further comparative proteomic analysis and biochemical characterization of the spores may bring further insights into the role of the yin element and yabG in C. botulinum sporulation.Purified spore suspensions were adjusted at OD 600nm 1 and heat-treated at 98 • C for different incubation times.D-values were obtained from the linear regression lines for each dataset.We observed the presence of two spore subpopulations, as shown with the two distinct destruction curves fitted on the data of each strain.The heat resistance of V73 spores appeared generally slightly lower than that of ST7B spores; however, the difference was statistically significant (p < 0.05, t-test) only for the most heat-sensitive spore subpopulations depicted during the first 20 min of the experiment.
Overall, the presence of the yin element had therefore no significant impact on the spore heat-resistance properties of V73 and ST7B.In line with this, transmission electron microscopy of purified V73 and ST7B spore preparations did not reveal any obvious structural differences between ST7B and V73 spores (Figure S8).Moreover, the germination patterns of V73 and ST7B spores were comparable (Figure S9).Further comparative proteomic analysis and biochemical characterization of the spores may bring further insights into the role of the yin element and yabG in C. botulinum sporulation.

Conclusions
We report the presence of a unique intervening mobile genetic element within yabG.This yin element is a prophage and has a unique gene content and order.yabG is conserved among spore-formers and is important for sporulation in at least B. subtilis and C. difficile; the yin element thus conforms to the general pattern that phage-like elements, such as the skin element and others, insert in sporulation-associated genes.It shares a number of common features with other intervening elements described in endospore formers: (i) prophage-like sequence, (ii) localization within a gene involved in sporulation, (iii) circularization of

Figure 2 .
Figure 2. Growth curves of C. botulinum ST7B (blue) and V73 (pink).The experiment consisted of three biological replicates, each made of four technical replicates.Error bars represent minimum and maximum values among technical replicates.

Figure 3 .
Figure 3. Phase-contrast microscopy observations of C. botulinum ST7B and V73 cultures in TPGY up to 240 h.Scale bar shown in each picture is 1 μm.

Figure 2 .
Figure 2. Growth curves of C. botulinum ST7B (blue) and V73 (pink).The experiment consisted of three biological replicates, each made of four technical replicates.Error bars represent minimum and maximum values among technical replicates.

Figure 2 .
Figure 2. Growth curves of C. botulinum ST7B (blue) and V73 (pink).The experiment consisted of three biological replicates, each made of four technical replicates.Error bars represent minimum and maximum values among technical replicates.

Figure 3 .
Figure 3. Phase-contrast microscopy observations of C. botulinum ST7B and V73 cultures in TPGY up to 240 h.Scale bar shown in each picture is 1 μm.

Figure 3 .
Figure 3. Phase-contrast microscopy observations of C. botulinum ST7B and V73 cultures in TPGY up to 240 h.Scale bar shown in each picture is 1 µm.

Figure 4 .
Figure 4. Relative abundance of different cell types in C. botulinum ST7B (a) and V73 (b) over timebased phase-contrast microscopy counting.Detailed counts can be found in Supplementary Materials (TableS3).

Figure 4 .
Figure 4. Relative abundance of different cell types in C. botulinum ST7B (a) and V73 (b) over timebased phase-contrast microscopy counting.Detailed counts can be found in Supplementary Materials (TableS3).

Figure 5 .
Figure 5. Plasticity of the yin element over time.(a) PCR amplification of the yin element integrated into the chromosome, the intact yabG gene, and the circular yin element in C. botulinum ST7B and V73 grown in TPGY medium.DNA-free water was used as a negative control.The amplicons (black bars) are shown near each DNA gel.Legend: Neg., negative control.(b) Quantitative PCR results to determine the relative proportion of restored yabG gene in C. botulinum ST7B and V73 over time.yabG primers span across the att recombination sites, whereas the control primers amplify a region elsewhere in the yabG gene.Error bars represent the standard deviations based on four technical replicates.Legend: pink curve, strain V73; blue curve, strain ST7B.

Figure 5 .
Figure 5. Plasticity of the yin element over time.(a) PCR amplification of the yin element integrated into the chromosome, the intact yabG gene, and the circular yin element in C. botulinum ST7B and V73 grown in TPGY medium.DNA-free water was used as a negative control.The amplicons (black bars) are shown near each DNA gel.Legend: Neg., negative control.(b) Quantitative PCR results to determine the relative proportion of restored yabG gene in C. botulinum ST7B and V73 over time.yabG primers span across the att recombination sites, whereas the control primers amplify a region elsewhere in the yabG gene.Error bars represent the standard deviations based on four technical replicates.Legend: pink curve, strain V73; blue curve, strain ST7B.

Figure 6 .
Figure 6.Thermal destruction of C. botulinum ST7B spores (pink dots) and V73 spores (blue dots).Purified spore suspensions were adjusted at OD600nm 1 and heat-treated at 98 °C for different incubation times.D-values were obtained from the linear regression lines for each dataset.We observed the presence of two spore subpopulations, as shown with the two distinct destruction curves fitted on the data of each strain.The heat resistance of V73 spores appeared generally slightly lower than that of ST7B spores; however, the difference was statistically significant (p< 0.05, t-test) only for the most heat-sensitive spore subpopulations depicted during the first 20 min of the experiment.

Figure 6 .
Figure 6.Thermal destruction of C. botulinum ST7B spores (pink dots) and V73 spores (blue dots).Purified spore suspensions were adjusted at OD 600nm 1 and heat-treated at 98 • C for different incubation times.D-values were obtained from the linear regression lines for each dataset.We observed the presence of two spore subpopulations, as shown with the two distinct destruction curves fitted on the data of each strain.The heat resistance of V73 spores appeared generally slightly lower than that of ST7B spores; however, the difference was statistically significant (p < 0.05, t-test) only for the most heat-sensitive spore subpopulations depicted during the first 20 min of the experiment.