{"paper_id":"3d355eec-1738-4f68-aabd-466eb26d3dc4","body_text":"Auto-crosslinking sporesilk fibers promote endospore and Cry toxin clustering | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Auto-crosslinking sporesilk fibers promote endospore and Cry toxin clustering Han Remaut, Mike Sleutel, Adrià Sogues This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6239967/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Mar, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Bacillus thuringiesis infects and kills insect larvae via its spores and associated entomotoxin-containing parasporal bodies (PSBs). We show that endospores and PSBs of B. thuringiensis Sv. Israëlensis are covered in a ‘sporesilk’ matrix that consist of 8 nm wide fibers with a double helical symmetry, formed by protofilaments of stacked alphahelical hairpins. These alpha endospore appendages (‘A-ENA’) are stabilized by up to ten autocatalytic and proximity-induced intermolecular isopeptide bonds per subunit, forming a continuous covalent polymer with remarkable chemical and physical robustness. We show A-ENA functions as B. thuringiensis virulence factor, increasing insecticidal activity by clustering spores and PSBs into an infectious biofilm-like unit. Moreover, we demonstrate the recombinant production and self-assembly of A-ENA nanofibers, and show that the exogenic addition of A-ENA fibers to B. thuringiensis strains natively lacking sporesilks results in spore-PSB clustering and gain of virulence, enabling the rational, non-GMO functionalization of these biological pest control agents. Biological sciences/Microbiology/Bacteria/Bacterial toxins Biological sciences/Microbiology/Biofilms Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy Biological sciences/Microbiology/Bacteria/Bacterial pathogenesis Bacillus thuringiensis spore biofilm cross-alpha amyloid isopeptide bond self-assembly nanofiber endospore appendage insect pathogen delta endotoxin Invertebrate Biological Control Agent (IBCA) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Bacillus thuringiensis (Bt) is an insecticidal, Gram-positive, aerobic endospore-forming bacterium of the Bacillus cereus sensu lato group, found in soil, dust, and on plant surfaces (Aronson et al., 1986 ). Bt spores are used as established biopesticides in agriculture, private residential use and vector-borne disease control due to their remarkable capacity to kill larvae of various insect pests from the Coleoptera, Lepidoptera, Hymenoptera and Diptera orders, including Aedes and Anopheles mosquitos - vectors for zika (Musso et al., 2019 ), dengue (Jansen and Beebe, 2010 ) and malaria ( Phillips et al., 2017 ). Since Bt strains target specific insect groups with minimal impact on humans and non-target organisms (e.g. wildlife, pollinators and other beneficial insects), they can form the basis of biopesticides that offer a safe and eco-friendly alternative to non-discriminatory, environmentally persistent and potentially toxic insecticidal chemicals (Koch et al., 2015 ; McClintock et al., 1995 ). Bt adopts its entomopathogenic lifestyle by the characteristic production of vegetative insecticidal toxins (Vip) (Bravo et al., 2007 ; Schnepf et al., 1998 ) or delta-endotoxins (Cry and Cyt) during sporulation (Bravo et al., 2007 ; Schnepf et al., 1998 ). Bt toxins exhibit a distinctive composition unique to each strain, resulting in a remarkable selectivity for specific insect genera. The toxins are produced as pro-toxins that assemble into pure or mixed crystalline condensates referred to as parasporal bodies or crystals (PSB, (Federici et al., 1990 )) (Fig. 1 a,b). Upon ingestion of Bt spores, the associated pro-toxin crystals dissolve and exert a poreforming activity in the insect gut, followed by intestinal sepsis, starvation and host death. In the process, germinated vegetative Bt cells colonize the insect until available nutrients in the carcass become depleted, then inducing sporulation to ensure further dissemination and survival of adverse conditions until the spores are taken up by e.g. a new host to restart the cycle. Bt was originally registered as a General Use in 1961 but is now found in over 1000 registered American pesticide products (National Pesticide Information Center. NPIC Product Research Online (NPRO): ‘ Bacillus thuringiensis ’. http://npic.orst.edu/NPRO/ ), 157 of which are specific to Bacillus thuringiensis subsp. israelensis (bti; (National Pesticide Information Center. NPIC Product Research Online (NPRO): ‘ israelensis ’. http://npic.orst.edu/NPRO/ )). Bti is a naturally occurring soil bacterium that is used as a larvicide in aquatic and wetland areas where black fly and mosquito-borne disease control is deployed (Charles and De Barjac, 1982 ). Typical commercial formulations (Aquabac©, Teknar©, Bactimos©, and Vectobac© ) consist of a mixture of spores and extracellular PSBs that are produced through a simple fermentation process that leverages the natural growth cycle of Bti where nutrient starvation and high cell density trigger sporulation. These conditions mimic the natural scenario of Bacillus biofilm formation which consists of a complex network of vegetative cells, spores, PSBs and extracellular matrix (ECM) components (e.g. for Bacillus subtilis : TasA fibers, eDNA, poly-γ-glutamic acid, exopolysaccharide) (Arbour et al., 2023 ). Because most studies on Bti tend to focus on PSB composition, larvicidal activity and specificity, the composition of the extracellular milieu of a spore dominated population remains poorly understood despite obvious academic and industrial interest. Here we show that Bti forms a bona fide, spore-based biofilm characterized by a pervasive extrasporal matrix (ESM) that is constructed of 8nm diameter protein nanofibers (dubbed A-ENA; see nomenclature below) that function as molecular tethers between spores and PSBs. Using cryoEM we solve the helical ultrastructure of the A-ENA fibers and show that they are composed of an 11kDa two-helix bundle protein that is extensively covalently cross-linked to its nearest neighbors via iso-peptide bonds. We demonstrate that A-ENA subunits spontaneously and efficiently polymerize in E . coli through a dock-and-lock self-assembly mechanism wherein subunits first rendezvous driven by hydrophobics, followed by enzyme-free, autocatalytic iso-peptide bond formation, yielding robust, micrometer-long covalent, flexible nanofibers. We show that loss of A-ENA breaks up the association of PSBs to Bti spores and attenuates their killing activity towards Chironomus aprilinus larvae. Remarkably, the addition of recombinant A-ENA fibers to Bacillus thuringiensis subsp. Kurstaki (Btk; natively a-ena − ) enhanced spore clustering and toxin retention, resulting in an increased killing activity towards the agricultural pest Trichoplusia ni . Our results show that A-ENA (i) is a virulence factor that exerts its function by ensuring close proximity of the infectious particle (spore) to its toxic payload (PSB) and (ii) is shown to function outside its native Bti context due to its aspecific mode of binding to spore and PSB surfaces, and is therefore expected to see more widespread use in other Bt systems that are naturally devoid of A-ENA but otherwise characterized by the presence of extracellular PSBs. Results BTI spores and PSBs are surrounded by a proteinaceous fiber biofilm To investigate the mesoscopic structure of a Bti biofilm, we dissolved freshly sporulated (72h, LB agar, 30°C) Bti cultures into miliQ and analyzed the bacterial suspension using light (Fig. 1 c,d) and negative stain electron transmission microscopy (nsTEM; Fig. 1 e). The suspension showed the presence of clustered microcolonies that were composed primarily of bacterial spores and parasporal bodies (PSBs), with vegetative cells representing only a small minority (< 1%) of the population (Fig. 1 c,d). Strikingly, both the spores and PSBs were engulfed in a dense extrasporal matrix composed of a fibrous mesh that permeated the entire sample (Fig. 1 e, unstained). The fibers were shown to emanate from both the spore and PSB surface, and were also seen to mediate direct connections between endospores and across endospores and PSBs (Fig. 1 e; Extended Data Fig. 1 a-d). We found numerous cases of large fiber bundles (100nm diameter) composed of dozens of individual fibers, making connection to the PSBs and the spore surface. Closer inspection of the fiber biofilm revealed nanofibers of ~ 8nm diameter and with a distinct double helical patterning in 2D class averages (Fig. 2 a,b; Extended Data Fig. 1 d). Bti PSBs comprise polycrystalline condensates of insecticidal proteins (Cry11Aa1, Cyt1Aa1, Cry4Aa1 and Cry4Ba1) that are packaged in a spherical parasporal sack, and that dissolve under the elevated pH in the insect gut (Rudd et al., 2023 ). When incubated at pH 11 we found that Bti spore preparations retained a biofilm-like organization, with spores embedded in a dense nanofiber matrix (Extended Data Fig. 1 e-i). Remarkably, parasporal sacks remained tightly entangled in the fiber network, but lost their condensate content. Close inspections showed nanofibers tightly associated with the surface of the parasporal sacks. Based on these observations, we hypothesized that the particulate, crystalline nature of the toxins and the fiber-mediated tethering of spores and PSBs ensures the cytotoxic payload remains at a high localized concentration and closely associated with the infectious particles (spore), thus ensuring their co-transmission during host-pathogen interactions. A-ENA is a new family type of Endospore appendages Intrigued by the existence of this fiber net, we proceeded with a cryo electron microscopy (cryoEM) analysis of the Bti exosporal matrix. Freshly resuspended Bti spores were deposited with no further processing onto a quantifoil holey carbon grid and plunge-frozen in liquid ethane, and imaged at 60k magnification using a JEOL CRYO ARM 300 microscope equipped with omega energy filter and a K3 direct electron detector. Similarly to nsTEM, cryoEM revealed a dense network of fibers in proximity to the vitrified endospores (Fig. 2 c). Fragments of isolated fibers were boxed and subjected to 2D alignment and classification. The resulting 2D class averages and Fourrier-Bessel indexing using PyHi (Zhang, 2022 ) confirmed the double helical character of the protein filaments (Fig. 2 d). Using cryosparc v4.4.1, we refined the fiber ultrastructure to a global resolution of 2.1Å, with a helical rise and twist of 5.37Å and 186.4° (Fig. 2 e; Extended Data Fig. 2 a). The cryoEM map revealed a right-handed two-start helix with parallel protofilaments that showed clear continuous main chain density for axially stacked alpha-helical subunits (Fig. 3 a,b,d). The high-resolution cryoEM reconstruction showed identifiable side chain densities that allowed for the de novo building and sequence assignment of an initial atomic model of the constituent subunits (Fig. 2 f). A standard protein BLAST search using the manually built protein sequence yielded protein Q8KNV8 ( https://www.uniprot.org/ ) as the top-ranking candidate for the major subunit. Q8KNV8 is a 101 amino acids protein with no known protein family membership as judged from InterProScan ( https://www.ebi.ac.uk/interpro/ ). Based on the lack of any sequence or structural homology or resemblance to known pili, functional amyloids or the recently characterized endospore appendages (i.e. S- and L-ENA (DUF3992) (Pradhan et al., 2021 ; Sleutel et al., 2024 )), it is clear that the Bti fibers constitute a new family of protein nanofibers referred to as a lpha helical en dospore a ppendages, or A-ENA in short. The genetic locus corresponding to Q8KNV8 is ATN07_33990, and is found on the 128kb pBtoxis-like plasmid (Ben-Dov et al., 1999 ; Berry et al., 2002 ) (pAM65-52-4-128K; RefSeq: NZ_CP013279.1) that carries all the insecticidal d-endotoxins. We identify a local three gene cluster ( a-ena ) consisting of ATN07_33990 ( a-ena ), ATN07_33985 and ATN07_33980 ( a-ena1 ) (Fig. 2 e). ATN07_33985, also referred to as pBt031, encodes a putative two-domain N-acetylmuramoyl-L-alanine amidase with a catalytic cell wall hydrolase domain (Pfam: PF01520) (Mistry et al., 2021 ) and a sporulation related repeat domain (Pfam: PF05036), and has previously been suggested to be involved in mother cell wall lysis similar to the sporulation specific autolysin CwlB in B. subtilis (Berry et al., 2002 ). Looking at the broader genomic context, the a-ena cluster is integrated into a region encoding for numerous d-endotoxins, i.e. upstream are located Cyt2Ba (ATN07_33970) and Cry8Ea (ATN07_33980). Downstream of a-ena , there are three other toxin genes, i.e. Cyt1A (ATN07_34010), Cry2A/C (ATN07_34015) and Cry11a (ATN07_34020). Of final note, the a-ena gene cluster is flanked by ATN07_34000, which has been annotated as a Tn3 family transposase (uniprot: A0A160LKE1) (Fig. 2 g). In the a-ena gene cluster, ATN07_33980 encodes for protein Q8KNV7 which shares 74.0% seq id to Q8KNV8, and is predicted to fold into a two-helix bundle that is structurally homologous to Q8KNV8 (the all-atom RMSD using Alphafold database entries AF-Q8KNV7-F1 and AF-Q8KNV8-F1 is 0.216 Å). Based on this similarity, we refer to Q8KNV8 and Q8KNV7 as A-ENA and A-ENA1. LC-MS/MS de novo sequencing of isolated fibers detected the presence of all three proteins of the local a-ena gene cluster, i.e. A-ENA, A-ENA1 and the N-acetylmuramoyl-L-alanine amidase. To resolve the ambiguity concerning the subunit composition of the ex vivo fibers we carefully inspected the reconstructed cryoEM volume and manually compared the A-ENA and A-ENA1 sequences at key sites with subunit-specific side-chain type (Extended Data Fig. 2 d). Based on that analysis, we unambiguously identified A-ENA as the major subunit of the fibers, but do not rule out that A-ENA1 is present in the fibers at low substoichiometric levels. A-ENA forms cross-α fibrils crosslinked via isopeptide bonds Through careful inspection of the A-ENA cryoEM map, we identified multiple sites of continuous density bridging adjacent A-ENA protomers within and across protofilaments (Fig. 3 , Extended Data Fig. 3 ). After the atomic A-ENA model was built, we identified these sites as intermolecular isopeptide bonds (IPB) composed of a nucleophilic IPB ‘donor’ (i.e. G2 N-terminus, K42, K76, K77 or K84; Fig. 3 a, k) and an electrophilic IPB ‘acceptor’ (i.e. E29, E44, N64, E78, or Q82; Fig. 3 a, k). Adjacent each IPB site, we find glutamic acid residues favorably positioned to act as acid-base catalyst (e.g. E28 Fig. 3 k, Extended Data Fig. 3 c), similar to what is seen for intramolecular IPBs in bacterial pili (Kang and Baker, 2011 ; Kang et al., 2007 ). In the latter, subunit folding brings the triads of IPB donor (K), acceptor (D, E, N or Q) and acid-base catalyst (E or D) in close proximity resulting in autocatalytic IPB formation. In A-ENA, however, IPB donor and acceptor are located on separate subunits, and a knobs-in-holes stacking of adjacent A-ENA subunits results in the formation of proximity-induced intermolecular IPBs (Extended Data Fig. 3 b). In this way, five different IPB pairs (IPB1-5: G2-Q44; K42-N64; K77-E78; K84-E29 and K76-Q82) are formed twice per A-ENA subunit, i.e. ones as donor (→) and ones as acceptor (←), totaling ten IPBs per protomer (Fig. 3 ; Extended Data Fig. 3 ; Table 1 ). Of these, six IPBs connect consecutive subunits within a single protofilament (IPB 2, 3 and 4; as donor towards N-pole: i → i-1 , and as acceptor towards C-pole: i← i + 1 ). Two additional axial IPBs (IPB1) are formed between the subunits i→i-5 as donor, and i←i + 5 as acceptor (Fig. 3 e). Finally, IPB5 forms two isopeptide bonds across the two protofibrils, ( i → i’-1 ) as donor, and (i←i’) as acceptor (Fig. 3 f). Whilst IPB2-5 are formed by knobs-in-holes interactions of alpha helices α1 and α2 of consecutive protomers within (IPB2-4) and adjacent protomers across (IPB5) the protofilaments, IPB1 connects distant subunits, realized by a long N-terminal latch (NTL) constructed of residues 2 to 14 and that is docked via residues M3, P4, I6, P7, L10 and I12 into a hydrophobic groove that runs axially along the A-ENA fiber surface spanning 4 subunits. Table 1 Isopeptide bond types observed in A-ENA fibers IPB Donor Acceptor Catalyst Contact Inter / Intra fibril Number 1 G2* Q44 E41 i -> i-5 intra 2 2 K42 N64 E39 i -> i-1 intra 2 3 K84 E29 E28** i -> i-1 intra 2 4 K77 E78 E28** i -> i-1 intra 2 5 K76 Q82 E86 i -> i’ inter 2 *Donor group for IPB type 1 is the amino group of the N-terminus **E28 functions as the catalyst for IPB types 3 and 4 A-ENA self-assembly and autocatalytic isopeptide bond formation Most bacterial pili and filaments require a dedicated assembly machinery (Lukaszczyk et al., 2019 ), and known polymers with intermolecular isopeptide bonds such as poly-ubiquitin or sortase-mediated pili employ a transpeptidation reaction that requires enzymatic accessory proteins (Kang and Baker, 2011 ). In A-ENA, however, the proximity-induced IPB triads as a result of subunits stacking suggests they may form autocatalytically (Extended Data Fig. 3 ). To evaluate the requirement of accessory proteins, A-ENA and A-ENA1 were recombinantly expressed in the cytoplasm of E. coli BL21. Purified fractions enriched towards either A-ENA or A-ENA1 were shown to contain abundant protein fibers, which both yielded 2D class averages that appeared identical to those obtained from the ex vivo material (Fig. 3 g,h,i; Extended Data Fig. 2 b, c). Testament to the remarkable stability of these fibers is the 30min incubation step at 100°C in the presence of 2% (w/v) sodium dodecyl sulfate (SDS) that we employed to purify the target fibers form cellular proteins. Similarly, A-ENA fibers were recalcitrant to treatment with chaotropic agents (8M urea), exposure to strong acidic (50% (v/v) formic acid) or alkaline conditions (2M NaOH), as well as wet (i.e. autoclaving) or desiccated (200°C oven) high temperature treatment (Extended Data Fig. 4 ). This remarkable stability suggested that IPBs were present in the recombinant fibers as well. We confirmed that hypothesis by solving the cryoEM structures of the highly similar recA-ENA and rec-A-ENA1 fibers to 2.40Å (rise:5.47 Å, twist: 186.36°) and 2.63 Å (rise:5.31 Å, twist: 186.50°) resolution, respectively (Extended Data Fig. 2 ). Indeed, rec-A-ENA fibers exhibited all 5 IPB types observed in the ex vivo A-ENA fibers, totaling 10 IPBs per protomer. Although rec-A-ENA1 fibers are near isomorphous to A-ENA (Extended Data Fig. 2 e), they lacked both the type 1 and 5 IPBs, leading to 6 confirmed IPBs per protomer, with no observable impact on the fiber stability (Fig. 3 j,k). This was unexpected because the IPB-associated residues are conserved in A-ENA1, and the NTL sequence is identical to the A-ENA sequence apart from N5 (A-ENA: T5; Extended Data Fig. 3 a). Possibly, an increased susceptibility to proteolytic attack of the NTL and/or reduced processing of N-terminal methionine (i.e. to liberate the G2 N-terminus) may impair IPB formation for recombinant A-ENA1 in E. coli . Regardless of these minor differences between A-ENA and A-ENA1, these results clearly show that IPB formation is auto-catalytic and independent of e.g. intervention of a transglutaminase (Griffin et al., 2002 ). Moreover, the efficient polymerization in E. coli (yield 1.8g A-ENA fiber / L culture) demonstrates that A-ENA biogenesis is not reliant on external factors (e.g. chaperone, usher, protease, …) and that A-ENA self-assembly is fully encoded in its primary sequence. A multiple sequence alignment of A-ENA like sequences shows the strong conservation of the IPB donor/acceptor pairs, as well as their cognate acid-base catalysts (Extended Data Fig. 3 a,b). Different to intermolecular IPBs that form by transpeptidation (Kang and Baker, 2011 ), A-ENA IPB triads are not part of recognition sequences, but rather take up predefined positions in the alpha 1 and alpha 2 helical wheels, such that packing of the A-ENA subunits brings the IPB partners in close proximity (Extended Data Fig. 3 a,b,c). Looking at the A-ENA axial protomer/protomer interface (e.g. i → i-1 ), we find that the buried interfaces are predominantly hydrophobic (mainly small hydrophobic chains apart from IPB residues; Extended Data Fig. 3 d), and there are no obvious complementary charged surface patches, suggesting protomer docking is predominantly driven by the hydrophobic effect. There is exceptional complementarity between the N- and C-pole surfaces of helices a 1 and a 2 . This leads to a tight ‘knobs-in-holes’ packing that drives the pre-orientation of the nucleophile/electrophile pair and its acid-base catalyst, and results in their electrostatic activation by the exclusion of solvent from the cross-α interfaces, essentially priming the residues for autocatalytic IPB formation. A-ENA and A-ENA1 expression is uniquely associated with the spore extracellular matrix The close association of A-ENA fibers with spores and parasporal bodies (Fig. 1 and S1), and localization of the a-ena and a-ena1 genes on the pAM65-52-4-128K plasmid and in close proximity to the Bti cry toxins (Fig. 2 g) suggest A-ENA expression occurs during sporulation. Using the DBTSBS server (Sierro et al., 2008 ) we identify a putative recognition site (CACCTTTGGAATAGCTGTTACATATAATA; p-value 0.01) for the late stage mother cell sporulation transcription factor SigK 46bp upstream of a-ena , in agreement with the predicted timing of expression given the extracellular, spore-associated localization of A-ENA. In agreement with this analysis, expression of a plasmid-encoded mcherry reporter gene under control of the hypothesized a-ena promotor sequence was found to be uniquely associated with sporulating cells, and absent from vegetative cells (Extended Data Fig. 5 a). No known sigma factor recognition motifs were found upstream of a-ena1 or ATN07_33980 (using a p-value cut-off of 0.01), although separation of the forward oriented a-ena and a-ena1 by the reverse oriented ATN07_33985 would suggest the a-ena genes to have independent promotors. To further test this hypothesis and evaluate the timing, (co-)occurrence and localization of A-ENA and A-ENA1 fibers, we modified a-ena and a-ena1 with C-terminally fused Spy and Snoop-tags, thus enabling their selective labelling with Spy-catcher-GFP and SnoopCatcher-mCherry conjugates, respectively. In agreement with the P a−ena promotor activity seen from the mcherry reporter (Extended Data Fig. 5 a), the presence of A-ENA SpyTag was restricted to sporulating cells and released spores (Extended Data Fig. 5 b,c). Similarly, A-ENA1 SnoopTag was spore-associated and not observed for vegetative cells. However, whilst A-ENA SpyTag fibers were ubiquitous throughout the spore population, A-ENA1 SnoopTag only showed sporadic labelling (i.e < 1% of spores), suggesting a phase variable expression that is independent of a-ena . A-ENA promotes spore clustering and retention of PSBs To shed light on the biological function of A-ENA, a-ena was selectively inactivated by introduction of a premature stop codon in the Bti pAM65-52-4-128K plasmid using CRISPR/Cas9 technology. Selective inactivation of a-ena versus a-ena1 (74% sequence identity) was achieved by means of an unique guide RNA target sequence and confirmed using PCR and Sanger sequencing (Extended Data Fig. 5 d,e). Spores of the resulting Bti Δa-ena lost recognition by a polyclonal serum raised against recombinant A-ENA (Extended Data Fig. 5 f), and lacked discernable sporesilk fibers when inspected by nsEM (Extended Data Fig. 5 d). Conversely, inactivation of a-ena1 did not result in loss of spore-associated A-ENA fibers (Extended Data Fig. 5 d). Thus a-ena expression is essential and sufficient for A-ENA production in the extrasporal matrix, whilst A-ENA1 is only a sporadic, or a minor, dispensable component of the A-ENA-like fibers. Using time-lapse light microscopy, we found no discernable difference in growth rate or sporulation efficiency for WT and Δa-ena Bti (Extended Data Fig. 5 g). To evaluate if a-ena expression would alter PSB content, OD-normalized spore prep fractions were treated at pH11 and analysed by SDS-PAGE (Fig. 4 b, Extended Data Fig. 6 a). SDS-PAGE showed bands at ~ 28, ~70 and ~ 130 kDa, identified, respectively, as the Cyt1A, Cry11Aa1 and Cry4Aa1/Cry4Ab1 toxins by MS fingerprinting, as well as minor bands around 55 and 38 kDa found to represent Cry11Aa1 breakdown products (Extended Data Fig. 6 a). A quantitative comparison of the released bands showed spore prep inputs of Bti WT and Δa-ena mutant showed no discernable difference in the of amounts of Cry toxins. Our microscopy experiments reveal Bti spore preparations consist of microcolonies with clustered spores and PSBs, embedded in a dense extrasporal matrix of A-ENA sporesilks (Fig. 1 , Extended Data Fig. 1 ). During sporulation, the endospore and PSB are formed in the daughter and mother cell compartment, respectively (Fig. 1 a) (Schnepf et al., 1998 ). We therefore asked if A-ENA fibers are implicated in spore clustering and may serve to ensure spores and PSBs remain associated after mother cell lysis. To do so, spore preparation of Bti WT and Δa-ena were subjected to cushioned density centrifugation (1 hr at 100 g or 10.000 g ) with three phases corresponding to a low density buffer or supernatant (‘SN’) layer with a ρ = 1.00g/ml (ρ low ), a medium density layer of 40% Histodenz with a ρ = 1.21 g/ml (ρ medium ), and a high density layer of 60% Histodenz with ρ = 1.32 g/ml (ρ high ). Sedimentation fractions were qualitatively followed by light microscopy (Fig. 4 a), and quantitatively monitored by SDS-PAGE of the high pH-released fraction to follow PSBs (Fig. 4 b) and by OD 600 to track the fractionation of spores (Fig. 4 c). At low g force, PSBs of WT Bti were primarily retained in the ρ low fraction (i.e. >95%), where they were found as agglutinated spore – PSB clusters. These low density spore-PSB clusters retained about 50% of spores (based on OD 600 ), with the other half found as smaller spore clusters in the ρ medium fraction (with sparse PSBs). In the Δa-ena mutant, however, no agglutinated spore-PSB clusters were observed, and the majority of PSBs migrated into the ρ medium fraction as individual, dispersed particles. Similarly, Δa-ena spores were found as individual particles that migrated into ρ medium fraction. At high g force, PSBs in both WT and Δa-ena Bti fully migrated into the ρ medium fraction, WT as smaller spore-PSB clusters, Δa-ena as dispersed particles. In both cases, about 30% of spores migrated into the ρ high fraction as individual particles. These fractionations indicate the individual PSBs and the majority of spores to have a density between 1.21–1.32 g/ml. In the WT Bti, the extrasporal matrix of A-ENA nanofibers agglutinates spores and PSBs into biofilm-like microcolonies with reduced density. With an average volume of ~ 0.68 µm 3 , spores experience around ~ 0.8 and ~ 80 pN gravitational force at 100 and 10.000 g , respectively. Whilst at 1 pN force a fraction of the spores gets dislodged from the low density biofilm, higher forces (up to 80pN) are required to break up the biofilm into smaller fragments that migrate into the ρ medium fraction. In absence of A-ENA, sporse and PSBs fractionate as individual particles, even at low g force. Spurred on by this striking result, we also followed a gain-of-function approach, where we introduced a-ena in Btk, which naturally lacks pAM65-52-4-128K or A-ENA homologs. To do so, a-ena was cloned with its native promotor into shuttle vector pAS (i.e. modified pJOE899 by removal of cas9 ), forming pAS_ a-ena . α-A-ENA dot blot analysis the respective spore preparations confirmed a lack of A-ENA in Btk WT, and showed the recombinant expression of A-ENA in Btk pAS_ a-ena (Extended Data Fig. 5 f). Light microscopy and nsTEM imaging of Btk WT showed mostly dispersed spores and PSBs (Fig. 4 e, Extended Data Fig. 6 c), with the presence of spore-attached S-ENA fibers (Extended Data Fig. 6 d; (Pradhan et al., 2021 )), but a lack of A-ENA-like fibers or an extensive extrasporal matrix as seen for Bti (Extended Data Fig. 6 d). In sharp contrast, Btk pAS_ a-ena showed a rich network of A-ENA fibers, resulting in the clustering of Btk spores into multibody biofilms (Fig. 4 e, Extended Data Fig. 6 c,d). Moreover, typical bipyramidal Btk toxin crystals (here also referred to as PSB were seen to be enclosed in the Btk pAS_ a-ena biofilms. Large A-ENA bundles could be seen to grace the PSB crystal surface as well as making contact with neighbouring spores, seemingly functioning as a molecular tether mimicking the spore-PSB association seen in Bti. Similar to Bti, we found no significant difference in the PSB formation for WT and Btk pAS_ a-ena (Extended Data Fig. 6 a). Remarkably, a very similar A-ENA dependent spore and PSB clustering can be seen by the exogenous addition of recombinant A-ENA nanofibers (‘+A-ENA’) purified from E. coli (Fig. 4 e, Extended Data Fig. 6 c,d), showing the agglutinating activity does not require A-ENA expression by the sporulating cells, and can be induced post-sporulation. When WT Btk, pAS_ a-ena and + A-ENA spore preps were subjected to cushioned density centrifugation, we observed that the spore – PSB agglutination induced by recombinant expression, or exogenous addition of A-ENA results in the increased retention of spores and PSBs in the low density fraction (Extended Data Fig. 6 b,e). Even at high g -forces, as much as 50% of PSBs were retained in a low density agglutinated biofilm fraction. Moreover, whereas WT Btk spores and PSB separated over a 70% sucrose cushion, as previously reported (Thomas and Ellar, 1983 ), in Btk pAS_ a-ena PSBs would co-pellet with the spores, demonstrating the encapsulation in A-ENA matrix ensured a robust spore – PSB association (Extended Data Fig. 6 f). A-ENA as a virulence factor The genetic association of a-ena and Cry toxin genes on a plasmid and the observation that A-ENA facilitates spore – PSB union suggested to us that A-ENA may represent a virulence factor for Bt insecticidal activity. We therefore compared the survival of Chironomus aprilinus larvae in the presence of buffer or equivalent spore preparations of Bti WT and Δa-ena at a dose of approximately OD 600 0.004. Mortality was monitored over 6 days following a single addition of spores to the aqueous medium. The results showed an increase in the MDK 80 (mean duration of killing 80% of the population) from 3 to 4 days for Bti WT and A-ENA KO, respectively. We next evaluated if recombinant A-ENA would result in a gain of virulence in strains naturally lacking sporesilks. To do so, we exposed two week old cabbage looper larvae ( Trichoplusia ni ; a common pest of cruciferous crops and susceptible to Btk (Estada and Ferre, 1994 )) to Btk WT, Btk pAS_ a-ena and Btk + A-ENA. Resuspended spore preparations were deposited on the solid feeding medium at a dose of 8 µL/cm 2 of OD 600 0.06. Whereas the recombinant expression or exogenous addition of A-ENA in Btk resulted in an MDK80 of 4 days, WT Btk did not show more than 50% lethality after six days of exposure (Fig. 4 g). To ensure that any observed effects were not due to the toxic properties of A-ENA fibers, we included Bti WT in the experiment. As expected from its known killing profile, Bti WT had no effect on the survival rate of T. ni larvae. Moreover, exposure of the larvae to A-ENA fibers showed no killing activity (Fig. 4 g). Thus addition of A-ENA fibers provides a convenient approach for the non-recombinant increase of the spore – toxin clustering in Bt strains lacking native sporesilks, and results in improved pesticidal activity. Phylogenetic analysis suggests biological functions beyond spore/PSB coupling Phylogenetic analysis of A-ENA prevalence across the Firmicutes phylum shows that A-ENA homologues are found across both the Clostridia and Bacillales clades (Fig. 5 ). Due to the genetic association between A-ENA and cyt/cry-toxins in Bti, we focused on the A-ENA prevalence across the Bacillus cereus group sensu lato . A-ENA is found in 496 of 5976 (8%) tested genomes of the Btyper database, and although found in all GTDB Bacillus species, it is predominantly found in B. thuringiensis (156 genomes), B. mycoides (124 genomes), B. wiedmanni (96 genomes) and to a lesser extent in B. anthracis/paranthracis/cereus (39 genomes) (Fig. 5 ). Moreover, out of the 496 genomes containing A-ENA, 351 (71%) genomes contained one or more type of Bt-toxin (Fig. 5 outer ring). If we compare that to the Bt-toxin prevalence across the Btyper database -only 1071 genomes (18%) were Bt positive, showing that A-ENA and Bt are positively correlated, reinforcing the notion that the main biological function of A-ENA is related to the presence of toxins, and by extension parasporal bodies. Interestingly, 350 genomes (70%) carried more than 2 A-ENA-like genes (as is the case for Bti) with 83 (17%) genomes harboring 5 or more homologues (e.g. the Bacillus cereus sensu stricto strain AFS036381 has 19 A-ENA-like genes) (Extended Data Fig. 7a). We performed a gene cluster analysis and found that many A-ENA-like genes are organized in (putative) operons and/or gene clusters (Extended Data Fig. 7b). Looking at these gene clusters in further detail, we discern two types of A-ENA homologues, (i) those constructed of a single A-ENA domain (shown in orange) and (ii) those carrying an additional domain (A-ENA fusion, shown in blue). After performing an interproscan analysis on the dataset of A-ENA homologue sequences, we identify 5 classes of passenger domain types (Extended Data Fig. 7c,d). The gene architecture of the most abundant class (type 1) consists of an A-ENA domain, followed by a collagen-like region and a C1q domain. Other less frequently observed passenger domain types are SdrD_B (PF17210) / OmcB (PF01345) or 4Fe-4S cluster domain (PF13370). We recently showed that collagen-like proteins can form trimeric tip fibrillae (ruffles) on the termini of S- and L-type ENA fibers found on Bacillus cereus and showed that these ruffles are involved in spore-spore clustering (Pradhan et al., 2021 ; Sleutel et al., 2024 )). We reasoned that a similar mechanism could be at play for the A-ENA/C1q homologues and modelled a heterohexamer of WP_277491740.1 (putative A-ENA ruffle) of a Bacillus cereus strain SIBC65 isolated from raw bulk milk in complex with WP_000075910.1 (single A-ENA domain) which is the presumed major subunit of an A-ENA type pilus (Extended Data Fig. 8a,b). AF3 predicts an A-ENA / ruffle complex wherein the A-ENA domains of the ruffle dock onto the single domain A-ENA protomers thereby mimicking the homotypic A-ENA contacts in the fiber (Extended Data Fig. 8c). The residues involved in IPB formation are conserved in both WP_277491740.1 and WP_000075910.1 (Extended Data Fig. 8d), suggesting that the ruffle is seamlessly integrated into the fiber or covalently tethered to the terminus of the A-ENA fiber (Extended Data Fig. 8e). Clues to a putative function of A-ENA ruffles can be found for Pasteuria ramosa , an endoparasitic bacterium of aquatic microcrustaceans (e.g. water flea Daphnia ). Recently, Huessy et al demonstrated that the collagen-like protein Pcl7 of P. ramosa functions as an adhesin to mediate spore attachment to the epithelium of the oesophagus of the host Daphnia (Huessy et al., 2024 ). Here we show that although Pcl7 is a CLP lacking an N-terminal A-ENA domain, it is embedded in a 3 gene operon flanked by two additional CLPs (ORFs 01218, 01220) one of which (01218) has an N-terminal A-ENA domain (Extended Data Fig. 8f,g). We also identify another 3 gene cluster (ORFs 00694, 00695 and 00696) which encode for 3 A-ENA-like genes. AF3 predicts a heterohexameric structure of the three CLPs in complex with the 3 putative A-ENA major subunits that is topologically similar to the ruffle described above (Extended Data Fig. 8h,i). Discussion The natural entomopathogenic properties of Bacillus thuringiensis (Bt) strains have led to world-wide adoption of Bt as an industrial biopesticide and vector control agent. During the sporulation process, Bt secretes parasporal bodies (PSB) that are crystalline agglomerates of different insecticidal pro-toxins that synergistically target specific insect groups. Typically, PSBs are present in the extracellular milieu, and are therefore subject to environmental dispersal and separation from the infectious body, i.e. the spore. Although some strains such as Bacillus sphaericus ( Kalfon et al., 1984 ) have solved this potential issue by producing PSBs that are embedded within the interstitial space between the spore body and the exosporium, most commonly used commercial strains such as B.t. subsp Kurstaki and B.t. subsp Israelensis produce extrasporal PSBs, raising the question how simultaneous uptake of spore and PSB are ensured during infection. Co-transmission of spores and PSBs is a likely requirement for efficient establishment of the natural Bt lifecycle considering cytotoxicity is predominantly mediated by the PSB, giving ingested spores the opportunity to germinate and proliferate on the nutrients supplied by the insect remains. Insect exposure to PSBs alone will lead to insect death but no subsequent proliferation and dispersion of Bt titers, whereas ingestion of PSB deprived spore suspensions will have minimal cytotoxic effect. Here we showed that Bti has solved this peril by producing a proteinaceous extracellular matrix that functions as a molecular glue holding spores and PSBs in tight unison within a spore biofilms (Fig. 6 ). A robust spore-PSB connection is particularly logical for Bti given the aquatic lifestyle of its natural hosts, e.g. mosquito larvae. The absence of such a connection is likely to lead to separation of both entities in a liquid environment due to dilution effects. This is reflected in the difference between the survival rates of Chironomus aprilinus larvae exposed to WT Bti and Bti Da-ena . For the latter we showed that the A-ENA matrix is absent, leading to poor retention of PSBs during spore purification and a marked increase of the C. aprilinus survival rate on days 2–3 (50%,30%) compared to wild-type spores (30%,20%). Note that our assay probes for the mortality during initial exposure and is therefore not sensitive to secondary, long-term effects such as further Bti proliferation and dissemination after the first infection cycle, which would occur in a natural setting (e.g. large body of water). Similarly, the effects of the physical spore-PSB coupling will be more pronounced in turbulent and/or larger reaction volumes (here we work in 3mL) where the dilution factor and subsequent probability of spore-PSB decoupling would be larger for the A-ENA mutant strain. Intrigued by the results for Bti, we turned to Btk and found that its spore biofilms are devoid of A-ENA, with a clear disconnect between the spores and the bipyramidal toxin crystals. Recombinant expression of A-ENA in Btk led to the instalment of an A-ENA based matrix (similar to Bti) capturing spores and toxin crystals into a cohesive network. This translated into a decreased survival of Trichoplusia ni when challenged with Btk spores that either recombinantly expressed A-ENA or were mixed with exogenously added A-ENA fibers purified from E. coli . Given the low sequence identity between the toxins produced by Bti and Btk, it would seem that aspecific interactions drive the binding of A-ENA fibers to the surface of the Btk crystals. This points towards a general promiscuity of A-ENA mediated interactions with a diverse range of surfaces (spore, PSB) amplified through avidity (display density of patchy, sticky surfaces on A-ENA fibers combined with fiber super bundling). It is tantalizing to speculate that A-ENA could lead to a similar gain of function in other PSB producing Bacillus thuringiensis strains devoid of A-ENA, and function as a generalist virulence factor. A phylogenetic analysis revealed that A-ENA orthologues are predominantly found in entomopathogenic Bacillus strains, and correlates with the presence of toxin genes in the genome. This tentatively generalizes the PSB-retention function seen for Bti across the group of A-ENA carrying Bt strains. Interestingly, a Pfam analysis also showed that A-ENA orthologues also exist as natural fusion products consisting of an N-terminal A-ENA domain, and a C-terminal collagen-like / C1q region. We recently showed that L-type ENAs tethered to the exosporium of a clinical Bacillus paranthracis isolate are decorated at their distal terminus with a tip fibrillum that is encoded by the collagen-like protein L-BclA, which also carries a C-terminal C1q domain (Sleutel et al., 2024 ). Moreover, we showed that L-BclA mediates spore-spore aggregation by binding to an as of yet unidentified receptor on the spore surface. Here, through AF3 modelling we show that genes with an A-ENA/collagen-like/C1q architecture might serve similar roles as L-BclA, in that the A-ENA moiety mediates insertion in or at an A-ENA terminus, while the C-terminal C1q domain represents the functional moiety, e.g. host-pathogen interactions. This indeed appears to be the case for the collagen-like protein Pcl7 of Pasteuria ramosa which mediates spore attachment to the oesophagus of its host Daphnia ( Huessy et al., 2024 ). Pcl7 is integrated into a triplet gene cluster flanked by two other collagen-like proteins, one of which carries an A-ENA motif. AF3 confirms the predicted P. ramosa A-ENA fiber / collagen fibrillum complex. Taken altogether, it seems plausible that the (short) hair-like appendages (Duneau et al., 2011 ) that coat the surface of P. ramosa is (in part) formed by A-ENA-like fibrils decorated with adhesive Pcl7 tip fibrillae. Based on our analysis, this might be a recurring feature across the family of A-BclA (i.e. A-ENA + collagen/C1q) carrying strains. Next to the biological implications, A-ENA fibers have remarkable material properties. Although the chemical and physical robustness of spore derived self-assembly structures has been demonstrated for numerous systems (Pradhan et al., 2021 ) (Henriques and Moran, 2007 ; Sleutel et al., 2024 ; Terry et al., 2017 ), those structures are either susceptible to depolymerization under reducing conditions or unfold under extreme conditions (e.g. 100% (w/v) formic acid for L-ENA). For A-ENA, we could not identify an unfolding or depolymerization transition point, making it one of the most resilient (natural) protein structures studied to date – we hypothesize that for A-ENA the limiting stability will be the strength of the peptide bond. We attribute this remarkable stability to the large number of covalent crosslinks combined with the compact size of the protomers. Normalized per molecular weight, this translates to 0.9 (non-reducible) covalent bonds per kilodalton for A-ENA. Other ultra-stable systems have significantly lower cross-linking densities, e.g. 0.2 IPBs / kDa for the Spy filaments of Streptococcus pyogenes (Kang and Baker, 2011 ) and 0.26 and 0.16 S-S bonds / kDa for the L-type and S-type ENAs, respectively of Bacillus paranthracis . The stability of A-ENA might be a biological requirement that follows from the environmental conditions that spores are exposed to. Spores are non-motile dormant structures, subject to a wide range of conditions as a result of environmental dispersion, that may include desiccation, UV radiation, high temperatures, proteolytic attack, and shear stress. By producing a durable net that guarantees co-migration of the PSBs together with the spore right up until the moment of ingestion by the host, Bti has developed a way to optimize for virulence and with it, the opportunity for further propagation. Resource availability CryoEM maps and coordinates for A-ENA, recA-ENA and recA-ENA1 have been deposited at the EMDB and PDB with following, respective, accession numbers: EMD-52871 and PDB-ID 9IHH; EMD-52872 and PDB-ID 9IHJ; EMD-52882 and PDB-ID 9Q82. Methods Cryo-EM sample preparation, data collection and helical reconstruction Bacillus thuringiensis serovar israelensis was inoculated in 10mL LB and grown over night at 37°C under shaking conditions. Next, 500µl of the preculture was spread out onto a square LB agar plate and incubated for 72h at 30°C under stagnant conditions. The resulting confluent growth was harvested and resuspended in miliQ, centrifuged for 5min at 20.000 rcf, the supernatant was discarded and the spore pellet was resuspended in miliQ. The resulting washed spore suspension was used as is for cryoEM grid preparation. High-resolution cryo-EM datasets were collected using Quantifoil™ R2/1 300 copper mesh holey carbon grids. Grids were glow-discharged at 5 mA plasma current for 30 s in an ELMO (Agar Scientific) glow-discharger. A Gatan CP3 cryo-plunger set at − 176°C and relative humidity of 90% was used to prepare the cryo-samples. 3 µL sample of the spore suspension or of the purified recombinant fibers was applied on the holey grid and incubated for 60 seconds. The sample was double sided blotted using Whatman type 2 paper for 3 s and plunge-frozen into precooled liquid ethane at − 176°C. High-resolution movies were recorded at 300 kV on a JEOL Cryoarm300 microscope equipped with an in-column Ω energy filter (operated at slit width of 14 eV) automated with SerialEM 3.0.850. The movies were captured with a K3 direct electron detector run in counting mode at a magnification of 60K with a calibrated pixel size of 0.71 Å/pix, and a total exposure of 60 e/Å 2 over 60 frames. A total of 2817, 3771 and 4160 movies were collected on the ex vivo fibers, the recombinant A-ENA and the recombinant A-ENA1 fibers, respectively. Raw tiffs were imported into Cryosparc v4.5.3 (Punjani et al., 2017 ) for further processing. Movies were motion-corrected using Patch Motion Correction and defocus values were determined using Patch CTF. Exposures were curated and fiber segments were picked using the filament tracer and extracted with a box size of 150 x 150 pixels at a pixel resolution of 1.42 Å/pix. After several rounds of 2D classification, curated particles were used in a helical refinement job using initial estimates for the helical rise and twist values derived from the 2D class average images. After an initial round of helical refinement, particles were recentred and re-extracted at 0.766 Å/pix (300 x 300pix) and used as input for a second round of helical refinement, followed by local and global CTF refinement, and a final helical refinement. The resulting high-resolution volume and particle stack were used for reference-based motion correction after particle duplicate removal (310k particles) followed by a final helical refinement. Reported map statistics were based on these maps for the recombinant A-ENA and A-ENA1 structures. For the ex vivo dataset, we additionally performed reference based motion correction followed by one more round of helical refinement. Figures were created using maps that were sharpened by cryosparc using a B-factor of -77.6Å 2 , -116.9 Å 2 , -114.5Å 2 for ex vivo , rec A-ENA and rec A-ENA1, respectively, automatically determined by cryosparc. For ex vivo A-ENA, the initial atomic model was built either manually de novo or using ModelAngelo (Jamali et al., 2024 ) without providing an input sequence (build_no_seq). PSI-blast analysis using the sequence from the built model as a query, identified Q8KNV8 (A-ENA) as the top ranking hit. A second round of model building was performed with ModelAngelo now using the A-ENA sequence as input (build) which was iteratively refined in real-space using Coot (Emsley et al., 2010 ) and Phenix (Liebschner et al., 2019 ). For model building in the recombinant A-ENA map, a single protomer of the ex vivo structure was docked and iteratively real-space refined in Coot. Next, the map was populated with remaining chains by applying helical symmetry in ChimeraX (sym command) and refined in Phenix. For model building in the recombinant A-ENA1 map, we used ModelAngelo (with A-ENA1 sequence provided) and refined the model as outlined above. MolProbity (Williams et al., 2018 ) was used to evaluate the quality of all the filament models. The refinement statistics are shown in Supplementary Table 2. Determination of the absolute helical hands of the maps were determined from the hand of an a-helix. Alternatively, we also performed ModelAngelo test runs on either hand which would lead to either a highly segmented model consisting of many chains, or a filament model wherein chains would fully map onto individual protomers The ‘ModelAngelo’ hand assignment agreed with hand-assignment following the handedness of the a-helices. Bacterial strains and growth conditions. Escherichia coli Stellar™ was used for cloning and was grown in Luria-Bertani (LB) broth or agar plates at 37°C supplemented with 50 µg/ml kanamycin or 100 µg/ml ampicillin when required. For protein production, E. coli BL21 (DE3) was grown in TB broth supplemented with 100 µg/ml ampicillin and placed at 30°C for A-ENA expression after induction. E. coli Dam-/Dcm- (New England Biolabs) was used to obtain unmethylated plasmids in order to increase the efficiency of transformation in Bt. Bti and Btk were grown in LB at 30°C with 150 rpm shaking and supplemented with 25 µg/ml kanamycin when required. For sporulation, 1 ml of overnight culture was plated on LB media and placed at 30°C for a week. Spores were scrapped and diluted in miliQ water and stored at 4°C. The strains used in this study are listed in Supplementary Table 1. Cloning for recombinant protein expression in E. coli Cloning of untagged A-ENA, A-ENA1 and variants were produced in this study with the purpose of cytoplasmic overexpression in E. coli . Water-boiled colonies of Bti were used as a source of gDNA. As a general cloning strategy, all PCR fragments containing the coding sequence were cloned into a linearized pASK-IBA3plus vector (using primers 321 and 322) by Gibson assembly using NEBuilder HiFi DNA Assembly Cloning Kit (New England Biolabs). The cloning of mCherry:SnoopCatcher (plasmid A185) was carried out using Gibson assembly. The mCherry gene was amplified from a plasmid containing mCherry in the lab collection, using primers p693 and p694. The SnoopCatcher fragment was synthetically ordered as a GeneBlock from IDT. Both fragments were subsequently cloned into the linearized pASK-IBA3plus vector. Cloning of sfGFP:SpyCatcher (plasmid A142) was done by amplifying sfGFP from a plasmid containing sfGFP in the lab collection, using primers p503 and p504. The SpyCatcher fragment was synthetically ordered as a GeneBlock from IDT. Both fragments were subsequently cloned into the linearized pASK-IBA3plus vector. All primers and plasmids used in this study are listed in Supplementary Table 1. Cloning for a-ena-STOP and Δa-ena1 using CRISPR/Cas9 based deletion plasmid Plasmid A266 and A190 were used to knockout a-ena and a-ena1 respectively from B. thuringiensis serovar isrealensis . Both plasmids are derived from the plasmid pJOE8999 (Altenbuchner, 2016 ) purchased from Bacillus Genetic Stock Center (BGSC). Oligonucleotides for sgRNA construction were designed using the online sgRNA design tool Cas-Designer (Park et al., 2015 ). First, pJOE8999 was linearized with primers p551 + p552. A266 homology regions were cloned with p588 + p591 and three STOP codon mutations were introduced with oligos p961 + p962 whereas p709 + p710 and p711 + p712 were used to construct the homology regions of A-ENA1 (A190). In both cases, we used NEBuilder HiFi DNA Assembly Cloning Kit (New England Biolabs) to insert homology regions. Sequencing and insertion were verified with the pair of oligos p525 + p526. Insertion of sgRNA was done by PCR using oligos p594 + p595 (for A266) and p713 + p714 (for A190). Primers p531 + p579 were used to sequence sgRNA region. Verification of the A-ENA STOP mutant was done by PCR and sequencing using oligos p963 + p651 and p964 respectively. Verification of the A-ENA1 mutant was performed by PCR with oligos p717 + p718. Plasmids used in this study are listed in the Supplementary Table 1. Construction and isolation of B. thuringiensis serovar isrealensis mutants The recombinant pJOE8999-based plasmids A266 and A190 described above, were introduced by electroporation into B. thuringiensis serovar isrealensis competent cells. Briefly, B. thuringiensis cells were made electrocompetent by growing in BHI + 0.5% glycerol at 37 ºC until OD 600 reached 0.6. Cells were harvested by centrifugation (5 min at 15 000 xg). Cells were washed 3 times with ice-cold sterile electroporation buffer (1 mM HEPES, 10% glycerol, pH 7.0) and flash-frozen in liquid nitrogen. Up to 1µg of unmethylated plasmid was transformed into 200 µl of competent cells using a 0.2 cm gap cuvette (Biorad) and electroporated using the MicroPulser Electroporator device (Biorad). Cells were transferred into 1ml of recovery buffer (BHI with 10% glycerol, 0.4% glucose, and 10 mM MgCl 2 ) for 1h at 30 ºC before plating in a 25 µg/ml kanamycin plate. Next day, a single colony was inoculated into 10 ml of liquid BHI supplemented with 25 µg/ml kanamycin and incubated with 180 rpm shaking for 3h at 37 ºC. Then, Mannose (Sigma) was added to a final concentration of 0.4% w/v in order to induce the expression of the Cas9 nuclease. After 3 hours, serial dilution of the culture was plated on LB agar plates with 25 µg/ml kanamycin + 0.4% w/v Mannose and incubated at 37 ºC overnight. Screening of the knockout was done using colony PCR, WT B. thuringiensis serovar isrealensis was used as a control. Positive colonies were sequenced-verified using the same oligos (Eurofins). To curate the pJOE8999-based plasmid, mutant strains were passaged three times in the absence of antibiotics and plate in LB plates at 42 ºC overnight. The absence of the plasmid was checked by striking the same colony in LB plates with or without Kanamycin. Strains used in this study are listed in Supplementary Table 1. Cloning for recombinant protein expression in B. thuringiensis For ectopic recombinant expression of the different constructs in B. thuringiensis we modified the pJOE8999 plasmid (Altenbuchner, 2016 ) by removal of the coding sequence for the Cas9 nuclease and its promoter by PCR using oligos p622 + p623, resulting in the ‘pAS’ shuttle expression vector. Complementation of a-ena stop and a-ena1 was achieved by expressing these constructs in the pAS vector. the a-ena region, including its promoter, ORF, and terminator, was amplified using primers p683 and p684. Colony PCR and sequencing (using primers p631 and p671) confirmed successful insertion, resulting in plasmid A176. Plasmid A176 was subsequently used to introduce the SpyTAG sequence via primers p681 and p682, yielding plasmid A180. For a-ena1 , the region was amplified using primers p695 and p696, generating plasmid A186. This plasmid was then used as a template to introduce the SnoopTAG sequence using primers p697 and p698, resulting in plasmid A187. To construct a dual-expression vector for a-ena spytag and a-ena1 snooptag under their native promoters, plasmid A180 was linearized using primers p699 and p700. The a-ena snooptag fragment, amplified from plasmid A187 using primers p701 and p702, was then cloned into the linearized A180 backbone. This final step produced plasmid A188. To monitor A-ENA expression, we replaced the a-ena coding sequence (CDS) with mCherry . This was achieved by linearizing plasmid A176 using primers p737 and p738, followed by cloning the mCherry CDS amplified with primers p714 and p742 using plasmid A185 as the template. This process resulted in the assembly of plasmid A197. Plasmids generated are listed in the Supplementary Table 1. Protein expression and purification sfGFP:spyCatcher and mCherry:Snoopcatcher were expressed in E.coli BL21 (DE3) in Terrific Broth (TB) supplemented with 100 µg/mL of Ampicillin at 37°C and induced with 200 µg/L anhydrotetracycline when OD 600 reached 0.5. At this point, the temperature was set at 23°C and cells were left to express overnight. Next day, cells were harvested by centrifugation (20 min at 5,000g) and pellets were kept at -20°C. Frozen pellets were resuspended in 100 ml of lysis buffer (50 mM Hepes pH 8, 300 mM NaCl, 1 mM MgCl2, DNase, lysozyme and EDTA-free protease inhibitor cocktails (ROCHE)) and lysed by sonication at 4°C. The lysate was centrifuged for 25 min at 30000g and 4°C. The cleared supernatant was loaded onto a 5ml Ni-NTA affinity chromatography column (HisTrap FF crude, GE Healthcare). The column was washed with 5 column volumes with Buffer A (300 mM NaCl, 10 mM Hepes pH 8, 10 mM Imidazole) and finally, His-tagged proteins were eluted with a linear gradient of Buffer B (300 mM NaCl, 10 mM Hepes pH8, 0.5 M Imidazole). Eluted protein was concentrated and loaded onto a Superdex 200 16/60 size-exclusion column (GE Life Sciences) that was equilibrated with SEC buffer (25 mM Hepes pH 8 and 150 mM NaCl) at 4°C. The peak corresponding to the protein was concentrated, flash-frozen in small aliquots in liquid nitrogen and stored at − 80°C. After purification, the samples were run on a sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) to evaluate their purity. A-ENA and A-ENA1 were expressed in E. coli BL21 (DE3) in Terrific Broth (TB) supplemented with 100 µg/mL of Ampicillin at 37°C and induced with 200 µg/L anhydrotetracycline when OD 600 reached 0.5. At this point, the temperature was set at 18°C and cells were left to express overnight. Cells were harvested by centrifugation (20 min at 5,000g) and incubated with A-ENA lysis buffer (25 mM Hepes pH8, 2.5 mM EDTA, 250 mM NaCl and 20 mg of Lysozyme) at 37°C for 2 hours under shaking conditions. Next, 1% final SDS concentration was added and the resuspended pellet was boiled for 30 min. A-ENA fibers were pelleted down by centrifugation at 30,000g for 45 minutes. Pellet was resuspended with a homogenizer in 30 ml deionized H 2 O and centrifuged for 45 minutes at 30,000g. This washing step was repeated 4 times. After the last centrifugation step, the pellet was resuspended with 3 ml of deionized H 2 O. Spore preparation and Fractionation Sporulation of Bti and Btk was triggered by nutrient exhaustion on LB-agar plates supplemented with 25 µg/mL of kanamycin when required. A single colony was inoculated in LB media (supplemented with 25 µg/mL of kanamycin when required) and grown overnight at 30°C. Next day, 200 µl of the cell suspension was plated on 245 x 245 mm LB plates (with or without antibiotic). Plates were incubated at 30°C for 7 days under aerobic conditions. Spores were harvested by scraping the LB plates with a cell scraper and resuspended in deionized H 2 O, spore stock was stored at 4°C. To fractionate Bti and Btk spores, the spore suspension was gently vortexed and then a discontinuous Histodenz gradient was prepared by layering 500 µl of 40% (w/v) Histodenz over 500 µl of 60% (w/v) Histodenz in 2 mL microcentrifuge tubes. A total of 500 µl of spore stock (OD₆₀₀ = 42) was carefully layered on top. Two tubes were prepared per strain: one for SDS-PAGE analysis and the other for optical density (OD) measurements and imaging. Samples were centrifuged at 10,000 × g or 100 × g for 1 hour using a tabletop swinging-bucket rotor (S-24-11-AR, Eppendorf). After centrifugation, three fractions (supernatant (SN), 40%, and 60% )were carefully collected into separate tubes. Each fraction was washed 3 times by adding 500 µl of water, vortexed, and centrifuged at 12,000 × g for 5 minutes. The supernatant was removed, and the pellet was resuspended in 100 µl of DX buffer (10 mM CHES pH 9.6, 8M Urea, 1% SDS, 1× protease inhibitor tablet (Roche), 2 mM DTT). Samples were boiled for 10 minutes with securely closed lids and subsequently centrifuged at maximum speed for 10 minutes to pellet insoluble spores. A total of 50 µl of the supernatant was mixed with 12.5 µl of blue loading dye, and 5 µl of the final mixture was loaded onto an SDS-PAGE gel for protein analysis. For OD measurements and imaging, the SN, 40% Histodenz, and 60% Histodenz fractions were collected into separate tubes. Each fraction was washed as descibed above. The supernatant was removed, and the pellet was resuspended in 100 µl of water. OD₆₀₀ measurements were performed by diluting 50 µl of the sample into 950 µl of water (1:20 dilution) before spectrophotometric analysis. Phase contrast and fluorescence microscopy For imaging non-labeled spores, a single colony of Bti or Btk was inoculated into 10 mL of LB medium with the appropriate antibiotic when required and grown overnight at 30°C. The next day, 250 µL of the culture was plated onto LB agar, with or without antibiotics, and incubated for 5 days at 30°C to allow sporulation. The resulting spore lawn was harvested and resuspended in 5 mL of water. A 3 µL aliquot of the spore suspension was transferred onto an LB-agar strip for microscopy. For time-lapse experiments comparing sporulation, t = 0 was defined as the moment cells were plated on the LB agar plate. At each time point, a sample was collected using a sterile loop, resuspended in 30 µL of PBS, and imaged. For dual labeling with SpyCatcher:GFP and SnoopCatcher:mCherry, a strain expressing A-ENA:SpyTag and A-ENA1:SnoopTag under their native promoters was induced to sporulate as described above. After 5 days, the spore lawn was harvested, and the OD₆₀₀ was adjusted to 1. Both SpyCatcher:GFP and SnoopCatcher:mCherry were added to a final concentration of 50 µM and incubated overnight with gentle rotation. The following day, the spore mixture was washed three times with PBS and mounted on a 1% agar slide for imaging. In all cases, Images were acquired in phase contrast and fluorescence mode using a Leica DMi8 inverted microscope (Leica) equipped with a 100×/1.32 oil objective (Leica). Immunodetection Due to the high degree of sequence identity of surface exposed residues between A-ENA and A-ENA1 we anticipated that it would not be feasible to raise polyclonal antibodies that would be specific to either A-ENA or A-ENA1. Given that our cryoEM analysis indicated that ex vivo fibers are predominantly composed of A-ENA, we opted to solely raise antisera against rec A-ENA. For this, a polyclonal mouse antiserum raised against purified rec-A-ENA fibers was ordered from Davids Biotechnologie GmbH (Regensburg, Germany). The following 63 day immunization schedule was followed: immunizations on days 0, 14, 28, 42 and 56. To detect A-ENA on spore samples, we diluted the spore suspension to OD 600 = 1 and blotted 2 µl onto a nitrocellulose membrane. After 5 minutes of drying at room temperature (RT), the membrane was blocked with 5% (w/v) skimmed milk for 30 minutes. The membrane was incubated with A-ENA antibody mix (dilution 1:1000 in TBS-Tween buffer (Tris-HCl pH8 10 mM; NaCl 150 mM; Tween20 0,05% (vol/vol)) and incubated at RT for 1h. After 3 washes of 5 minutes each, the membrane was incubated in the secondary antibody mix (dilution 1:1000) for an extra hour at RT. After 3 washes of 5 minutes each, the membrane was imaged with the LI-COR Odyssey M using the LI-COR Acquisition Software v 2.0.0.86. A-ENA stability assay To test the chemical stability of recombinant A-ENA fibers, fibers were harvested from a concentrated stock solution via centrifugation (20.000rcf, 1h, supernatans discarded). The resulting pellets after centrifugation of 100µl aliquots were resuspended in various conditions: 100µl 2% (w/v) SDS, 8M urea, 2M NaOH or 100 (v/v) formic acid and incubated as indicated on Extended Data Fig. 4 prior to nsTEM sample preparation. To test the physical robustness of recombinant A-ENA fibers, a 100µl aliquot of the fiber stock solution was desiccated at 200°C for 15min in an open glass vial in an oven. The resulting dried material was rehydrated in 100µl miliQ prior to nsTEM imaging. Additionally, a 100µl aliquot was autoclaved for 20min at 121°C. Next, 3µl aliquots of the treated samples were deposited onto Cu-mesh formvar grids, washed twice with 20µl miliQ and incubated for 1min with 2% (w/v) uranyl acetate and blotted dry with Whatmann 2 filter paper. Micrographs were collected with a 120 kV JEOL 1400 microscope equipped with LaB6 filament and TVIPS F416 CCD camera operated at 60.000x magnification. Monitoring the entomopathogenic activity of different Bt strains To evaluate the virulence of Bti strains either expressing or lacking A-ENA (wild-type vs. a-ena knockout), we employed the insect model Chironomus aprilinus . Fresh C. aprilinus larvae were procured from Zooschatz (Berlin, Germany) under batch number 1223153. Bti strains were cultured on LB plates at 30°C for one week, after which the lawn of spores and PSB were harvested and resuspended in deionized water, and OD 600 was measured. To determine the mortality rate, a minimum of 39 C. aprilinus larvae were included in each experimental group, with a final volume of 1 ml of sea salt solution supplemented with a vitamin premix provided by the supplier. Experiments were carried out in 12 well plates with a single individual per well. Spore suspension was added to reach a final OD 600 of 0.02. Phosphate saline buffer (PBS) was used as a negative control. Larvae survival was monitored daily over a 6-day period. The experiment was conducted in biological triplicates under room temperature conditions. For the virulence assessment of Btk strains, we used the insect model Trichoplusia ni . Larvae or eggs were obtained from Frontier Agricultural Sciences (Newark, USA). Btk wild-type and A-ENA-expressing strains ( pA-ENA ) were sporulated on LB plates at 30°C for one week. Spores and PSBs were harvested and processed as described for Bti strains. To determine the mortality rate, a minimum of 16 two week old T. ni larvae per group were individually placed on 2.5 × 2.5 cm pieces of solid food provided by the supplier. Each food piece was spread with a 50 µl aliquot of the harvested spore suspension at an OD₆₀₀ of 0.06. As a negative control, Bti (which has no toxic effect on T. ni ) was included. Larval survival was monitored daily for one week. For experiments involving recombinant A-ENA, it was added to Btk WT samples at a final concentration of 4.18 mg/ml and incubated for at least 3 hours under slow rotation. As an additional control, purified recombinant A-ENA was tested at the same final concentration (4.18 mg/ml) under identical conditions. The study was performed in biological triplicates at room temperature Search for orthologs and homologs of A-ENA To probe for the presence of A-ENA across the Firmicutes phylum, all publicly available Genbank genomes (9880 genomes) belonging to the Bacillus/Clostridium group (taxid 1239; assembly-level complete) were downloaded from the NCBI database. Homo- and orthologs of A-ENA were searched in all assemblies with hmmsearch (Eddy, 2011 ) (881 genomes) using a hidden Markov model that was generated with hmmbuild using a manually curated multiple sequence alignment of A-ENA sequences obtained from a blastp (Johnson et al., 2008 ) search using Q8KNV8 as a query. For this, AlphaFold2 models (either retrieved from the AlpfaFold database (Varadi et al., 2024 ) or predicted using localcolabfold (Mirdita et al., 2022 )) of the corresponding blast sequences were manually inspected and compared to the A-ENA cryoEM structure to remove any non-A-ENA sequences. The inclusion threshold for hmmsearch was set to an E-value of 1e-7. The phylogenetic tree of the Firmicutes phylum was generated using phyloTv2 ( https://phylot.biobyte.de/ ) using the NCBI taxonomy IDs of the corresponding assemblies as node identifiers, and imported into iTOL (Letunic and Bork, 2024 ) for visualization using the online server ( https://itol.embl.de/ ). For the more detailed search of A-ENA homo- and orthologs across the Bacillus cereus group sensu lato , the BtyperDB v1 database (5976 genomes) was downloaded from https://www.btyper.app/ , and hmmsearch was run against all assemblies with an E-value threshold of 1e-7 (496 genomes). The phylogenetic tree described in BtyperDB (Ramnath et al., 2023 ) and the corresponding meta-data table was kindly provided by Laura M. Caroll. Tree Figures were generated using https://microreact.org/ . Gene cluster analysis was performed using cblaster 1.3.18 (Gilchrist et al., 2020 ) using default settings. Protein domain organization of A-ENA homologs with C-terminal fusions was performed with Interproscan 5.67-99.0 (Jones et al., 2014 ). Declarations Resource availability CryoEM maps and coordinates for A-ENA, recA-ENA and recA-ENA1 have been deposited at the EMDB and PDB with following, respective, accession numbers: EMD-52871 and PDB-ID 9IHH; EMD-52872 and PDB-ID 9IHJ; EMD-52882 and PDB-ID 9Q82. Acknowledgements We thank Dirk Reiter and Marcus Fislage at the VIB-VUB Facility for Bio Electron Cryogenic Microscopy (BECM). We’re thankful to Steven Janvier and Didier Vertommen for MS fingerprinting analysis. This work was funded by VIB. A.S was supported by the EMBO (ALTF-709-2021) and the Marie Skłodowska-Curie Actions (MSCA; SLYDIV project). Author contributions M.S. and H.R. designed the project. M.S. performed the cryoEM imaging, data processing and structure determinations. M.S. and A.S. performed nsTEM imaging. A.S. performed light microscopy and all cloning, biological and biochemical assays. M.S. and A.S. performed A-ENA purification. M.S. and H.R. wrote the manuscript with contributions by A.S. Declaration of interests The authors are named as inventors on a patent application filed by VIB, related to the technological use of A-ENA nanofibers. Supplemental information Extended Data Figs 1–8 and Supplementary Table 1-2 References Altenbuchner, J. (2016). Editing of the Bacillus subtilis Genome by the CRISPR-Cas9 System. Appl Environ Microbiol 82 , 5421-5427. Arbour, C.A., Nagar, R., Bernstein, H.M., Ghosh, S., Al-Sammarraie, Y., Dorfmueller, H.C., Ferguson, M.A.J., Stanley-Wall, N.R., and Imperiali, B. (2023). Defining Early Steps in B. subtilis Biofilm Biosynthesis. bioRxiv. Aronson, A.I., Beckman, W., and Dunn, P. (1986). Bacillus thuringiensis and related insect pathogens. Microbiol Rev 50 , 1-24. Ben-Dov, E., Nissan, G., Pelleg, N., Manasherob, R., Boussiba, S., and Zaritsky, A. (1999). Refined, circular restriction map of the Bacillus thuringiensis subsp. israelensis plasmid carrying the mosquito larvicidal genes. Plasmid 42 , 186-191. Berry, C., O'Neil, S., Ben-Dov, E., Jones, A.F., Murphy, L., Quail, M.A., Holden, M.T., Harris, D., Zaritsky, A., and Parkhill, J. (2002). Complete sequence and organization of pBtoxis, the toxin-coding plasmid of Bacillus thuringiensis subsp. israelensis. Appl Environ Microbiol 68 , 5082-5095. Bravo, A., Gill, S.S., and Soberon, M. (2007). Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control. Toxicon 49 , 423-435. Charles, J.-F., and De Barjac, H. (1982). Sporulation et cristallogenese de Bacillus thuringiesnsis Var. Israelensis en microscopie electronique. Annales de Microbiologie 133 . Duneau, D., Luijckx, P., Ben-Ami, F., Laforsch, C., and Ebert, D. (2011). Resolving the infection process reveals striking differences in the contribution of environment, genetics and phylogeny to host-parasite interactions. BMC Biol 9 , 11. Eddy, S.R. (2011). Accelerated Profile HMM Searches. PLoS Comput Biol 7 , e1002195. Emsley, P., Lohkamp, B., Scott, W.G., and Cowtan, K. (2010). Features and development of Coot. Acta crystallographica Section D, Biological crystallography 66 , 486-501. Estada, U., and Ferre, J. (1994). Binding of Insecticidal Crystal Proteins of Bacillus thuringiensis to the Midgut Brush Border of the Cabbage Looper, Trichoplusia ni (Hubner) (Lepidoptera: Noctuidae), and Selection for Resistance to One of the Crystal Proteins. Appl Environ Microbiol 60 , 3840-3846. Federici, B.A., Lüthy, P., and Ibarra, J.E. (1990). Parasporal Body of Bacillus thuringiensis israelensis. In Bacterial Control of Mosquitoes & Black Flies: Biochemistry, Genetics & Applications of Bacillus thuringiensis israelensis and Bacillus sphaericus, H. de Barjac, and D.J. Sutherland, eds. (Dordrecht: Springer Netherlands), pp. 16-44. Gilchrist, C.L.M., Booth, T.J., and Chooi, Y.-H. (2020). cblaster: a remote search tool for rapid identification and visualisation of homologous gene clusters. bioRxiv, 2020.2011.2008.370601. Griffin, M., Casadio, R., and Bergamini, C.M. (2002). Transglutaminases: nature's biological glues. Biochem J 368 , 377-396. Henriques, A.O., and Moran, C.P., Jr. (2007). Structure, assembly, and function of the spore surface layers. Annu Rev Microbiol 61 , 555-588. Huessy, B., Bumann, D., and Ebert, D. (2024). Ectopical expression of bacterial collagen-like protein supports its role as adhesin in host-parasite coevolution. R Soc Open Sci 11 , 231441. Jamali, K., Kall, L., Zhang, R., Brown, A., Kimanius, D., and Scheres, S.H.W. (2024). Automated model building and protein identification in cryo-EM maps. Nature 628 , 450-457. Jansen, C.C., and Beebe, N.W. (2010). The dengue vector Aedes aegypti: what comes next. Microbes Infect 12 , 272-279. Johnson, M., Zaretskaya, I., Raytselis, Y., Merezhuk, Y., McGinnis, S., and Madden, T.L. (2008). NCBI BLAST: a better web interface. Nucleic Acids Res 36 , W5-9. Jones, P., Binns, D., Chang, H.Y., Fraser, M., Li, W., McAnulla, C., McWilliam, H., Maslen, J., Mitchell, A., Nuka, G. , et al. (2014). InterProScan 5: genome-scale protein function classification. Bioinformatics 30 , 1236-1240. Kalfon, A., Charles, J.F., Bourgouin, C., and de Barjac, H. (1984). Sporulation of Bacillus sphaericus 2297: an electron microscope study of crystal-like inclusion biogenesis and toxicity to mosquito larvae. J Gen Microbiol 130 , 893-900. Kang, H.J., and Baker, E.N. (2011). Intramolecular isopeptide bonds: protein crosslinks built for stress? Trends Biochem Sci 36 , 229-237. Kang, H.J., Coulibaly, F., Clow, F., Proft, T., and Baker, E.N. (2007). Stabilizing isopeptide bonds revealed in gram-positive bacterial pilus structure. Science 318 , 1625-1628. Koch, M.S., Ward, J.M., Levine, S.L., Baum, J.A., Vicini, J.L., and Hammond, B.G. (2015). The food and environmental safety of Bt crops. Front Plant Sci 6 , 283. Letunic, I., and Bork, P. (2024). Interactive Tree of Life (iTOL) v6: recent updates to the phylogenetic tree display and annotation tool. Nucleic Acids Res 52 , W78-W82. Liebschner, D., Afonine, P.V., Baker, M.L., Bunkoczi, G., Chen, V.B., Croll, T.I., Hintze, B., Hung, L.W., Jain, S., McCoy, A.J. , et al. (2019). Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr D Struct Biol 75 , 861-877. Lukaszczyk, M., Pradhan, B., and Remaut, H. (2019). The Biosynthesis and Structures of Bacterial Pili. Subcell Biochem 92 , 369-413. McClintock, J.T., Schaffer, C.R., and Sjoblad, R.D. (1995). A comparative review of the mammalian toxicity of Bacillus thuringiensis-based pesticides. Pesticide Science 45 , 95-105. Mirdita, M., Schutze, K., Moriwaki, Y., Heo, L., Ovchinnikov, S., and Steinegger, M. (2022). ColabFold: making protein folding accessible to all. Nat Methods 19 , 679-682. Mistry, J., Chuguransky, S., Williams, L., Qureshi, M., Salazar, G.A., Sonnhammer, E.L.L., Tosatto, S.C.E., Paladin, L., Raj, S., Richardson, L.J. , et al. (2021). Pfam: The protein families database in 2021. Nucleic Acids Res 49 , D412-D419. Musso, D., Ko, A.I., and Baud, D. (2019). Zika Virus Infection - After the Pandemic. N Engl J Med 381 , 1444-1457. Park, J., Bae, S., and Kim, J.S. (2015). Cas-Designer: a web-based tool for choice of CRISPR-Cas9 target sites. Bioinformatics 31 , 4014-4016. Phillips, M.A., Burrows, J.N., Manyando, C., van Huijsduijnen, R.H., Van Voorhis, W.C., and Wells, T.N.C. (2017). Malaria. Nat Rev Dis Primers 3 , 17050. Pradhan, B., Liedtke, J., Sleutel, M., Lindback, T., Zegeye, E.D., K, O.S., Llarena, A.K., Brynildsrud, O., Aspholm, M., and Remaut, H. (2021). Endospore Appendages: a novel pilus superfamily from the endospores of pathogenic Bacilli. EMBO J 40 , e106887. Punjani, A., Rubinstein, J.L., Fleet, D.J., and Brubaker, M.A. (2017). cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat Methods 14 , 290-296. Ramnath, V., Larralde, M., Menchik, P., Buehler, A.J., Harrand, A.S., Chung, T., Wei, X., Raghuram, V., Gourlé, H., Pierneef, R. , et al. (2023). A community-curated, global atlas of Bacillus cereus sensu lato genomes for epidemiological surveillance. bioRxiv, 2023.2012.2020.572685. Rudd, S.R., Miranda, L.S., Curtis, H.R., Bigot, Y., Diaz-Mendoza, M., Hice, R., Nizet, V., Park, H.W., Blaha, G., Federici, B.A. , et al. (2023). The Parasporal Body of Bacillus thuringiensis subsp. israelensis: A Unique Phage Capsid-Associated Prokaryotic Insecticidal Organelle. Biology (Basel) 12 . Schnepf, E., Crickmore, N., Van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R., and Dean, D.H. (1998). Bacillus thuringiensis and its pesticidal crystal proteins. Microbiol Mol Biol Rev 62 , 775-806. Sierro, N., Makita, Y., de Hoon, M., and Nakai, K. (2008). DBTBS: a database of transcriptional regulation in Bacillus subtilis containing upstream intergenic conservation information. Nucleic Acids Res 36 , D93-96. Sleutel, M., Zegeye, E.D., Llarena, A.K., Pradhan, B., Fislage, M., O'Sullivan, K., Van Gerven, N., Aspholm, M., and Remaut, H. (2024). Helical ultrastructure of the L-ENA spore aggregation factor of a Bacillus paranthracis foodborne outbreak strain. Nat Commun 15 , 7514. Terry, C., Jiang, S., Radford, D.S., Wan, Q., Tzokov, S., Moir, A., and Bullough, P.A. (2017). Molecular tiling on the surface of a bacterial spore - the exosporium of the Bacillus anthracis/cereus/thuringiensis group. Mol Microbiol 104 , 539-552. Thomas, W.E., and Ellar, D.J. (1983). Bacillus thuringiensis var israelensis crystal delta-endotoxin: effects on insect and mammalian cells in vitro and in vivo. J Cell Sci 60 , 181-197. Varadi, M., Bertoni, D., Magana, P., Paramval, U., Pidruchna, I., Radhakrishnan, M., Tsenkov, M., Nair, S., Mirdita, M., Yeo, J. , et al. (2024). AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences. Nucleic Acids Res 52 , D368-D375. Williams, C.J., Headd, J.J., Moriarty, N.W., Prisant, M.G., Videau, L.L., Deis, L.N., Verma, V., Keedy, D.A., Hintze, B.J., Chen, V.B. , et al. (2018). MolProbity: More and better reference data for improved all-atom structure validation. Protein Sci 27 , 293-315. Zhang, X. (2022). Python-based Helix Indexer: A graphical user interface program for finding symmetry of helical assembly through Fourier-Bessel indexing of electron microscopic data. Protein Sci 31 , 107-117. Additional Declarations Yes there is potential Competing Interest. The authors are named as inventors on three patent applications filed by VIB, related to the technological use of A-ENA nanofibers. Supplementary Files SupplementaryInformation.pdf Supplementary Information Cite Share Download PDF Status: Published Journal Publication published 11 Mar, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-6239967\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":442715301,\"identity\":\"eb99d623-d1fd-4866-8875-aaa6c288af0a\",\"order_by\":0,\"name\":\"Han Remaut\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzUlEQVRIiWNgGAWjYBACgwMMDMw/wEzmAwyMDQwJxGnhAbHY2BJI1sJjQKSW42cPMPNUbJM3uN/zTZp3B0MePyEt9mfyEph/nLltuOEY7zZp3jMMxZINBB2WY8D8sO02I1DLZmPeNobEDQcIaTn/xoDZ8N9t+w3HeB6DtewnqOUG0BbJhtuJQC2Mj8G2EPKLwY03Bod5jt1OnnkszfDh3DMSxRKEHZZj+Jin5rZt3+HDDw683WGTx99AyBogQDZWggj1o2AUjIJRMAoIAgCiAEhrs4JDyAAAAABJRU5ErkJggg==\",\"orcid\":\"https://orcid.org/0000-0002-9775-4102\",\"institution\":\"Vrije Universiteit Brussel\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Han\",\"middleName\":\"\",\"lastName\":\"Remaut\",\"suffix\":\"\"},{\"id\":442715302,\"identity\":\"ed30db16-96f0-4947-8a0e-051edc7acdb9\",\"order_by\":1,\"name\":\"Mike Sleutel\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-3247-2187\",\"institution\":\"VIB / Vrije Universiteit Brussel\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Mike\",\"middleName\":\"\",\"lastName\":\"Sleutel\",\"suffix\":\"\"},{\"id\":442715303,\"identity\":\"ebf4fb04-ca63-43c6-9e4b-8ab14d31ac80\",\"order_by\":2,\"name\":\"Adrià Sogues\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-5752-6009\",\"institution\":\"Vlaams Instituut voor Biotechnologie\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Adrià\",\"middleName\":\"\",\"lastName\":\"Sogues\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-03-17 01:15:27\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-6239967/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-6239967/v1\",\"draftVersion\":[],\"editorialEvents\":[{\"content\":\"https://doi.org/10.1038/s41467-026-70495-z\",\"type\":\"published\",\"date\":\"2026-03-11T04:00:00+00:00\"}],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":82854479,\"identity\":\"1a09afa2-f53f-49b7-b9e0-38c16480e931\",\"added_by\":\"auto\",\"created_at\":\"2025-05-16 04:30:38\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":988819,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cem\\u003e\\u003cstrong\\u003eBacillus thuringiensis\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e Sv. Israelensis (Bti) spores produce a novel family of Endospore Appendages (ENA). (a) \\u003c/strong\\u003eSchematic drawing of the sporulation\\u003cstrong\\u003e \\u003c/strong\\u003ecycle\\u003cstrong\\u003e \\u003c/strong\\u003eof \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e. (i) Asymmetric cell division results in the formation of a forespore and a larger mother cell. (ii-iii) The mother cell engulfs the forespore and produces the parasporal body (PSB) with insecticidal proteins. (iv) Consecutive protective layers develop around the maturing spore, while the mother cell prepares for lysis. (v) Upon mother cell lysis, the mature spore and PSB are released into the environment. (vi) The cycle is completed as spores germinate under favorable conditions, returning to the vegetative state. Created using BioRender.com. \\u003cstrong\\u003e(b) \\u003c/strong\\u003ePhase contrast microscopy image of a sporulating Bti cells. M: mother cell, S: spore, P: PSB. \\u003cstrong\\u003e(c) \\u003c/strong\\u003eWide-field view of sporulating cells, showing clusters of varying sizes. \\u003cstrong\\u003e(d) \\u003c/strong\\u003eInset of the dotted square in panel (c) showing a small cluster composed of spores and PSBs. \\u003cstrong\\u003e(e)\\u003c/strong\\u003e TEM micrograph of a resuspended Bti spore biofilm, composed of endospores (S) and PSBs (P).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/6e02599b85fee02f8df9600e.png\"},{\"id\":82853881,\"identity\":\"078236cd-01d2-4d60-9a87-b6b91a5e5adb\",\"added_by\":\"auto\",\"created_at\":\"2025-05-16 04:14:38\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":707593,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eIdentification of the A-ENA major subunit through cryoID. (a,b)\\u003c/strong\\u003e nsTEM micrographs of A-ENA filaments emanating from the spore (a) or PSB (b) surfaces. \\u003cstrong\\u003e(c)\\u003c/strong\\u003e Cryo-EM micrograph of single \\u003cem\\u003eex vivo\\u003c/em\\u003e A-ENA filaments. \\u003cstrong\\u003e(d)\\u003c/strong\\u003e 2D class average of A-ENA filaments. (box size 230x230Å)\\u003cstrong\\u003e (e)\\u003c/strong\\u003e 3D cryoEM map of \\u003cem\\u003eex vivo\\u003c/em\\u003e A-ENA fibers, revealing a right-handed, double helical ultrastructure with a helical rise and twist of 5.37Å and 186.4°. \\u003cstrong\\u003e(f)\\u003c/strong\\u003e \\u003cem\\u003eDe novo-\\u003c/em\\u003ebuilt model used for major subunit identification by protein blast search; \\u003cstrong\\u003e(g) \\u003c/strong\\u003eGenetic locus of the \\u003cem\\u003ea-ena \\u003c/em\\u003ethree-gene cluster on the pAM65-52-4-128K plasmid: two homologous genes encoding ‘A-ENA’\\u003cem\\u003e \\u003c/em\\u003eand\\u003cem\\u003e \\u003c/em\\u003e‘A-ENA1’ are interspersed by a putative N-acetylmuramoyl-L-alanine amidase gene. The \\u003cem\\u003ea-ena \\u003c/em\\u003egene cluster is flanked on either side by genes encoding the major d-endotoxins of \\u003cem\\u003eBti\\u003c/em\\u003e: Cry11A, Cry4B, Cyt1A and Cyt2B. Locus accession numbers (below) are to be prepended with “ATN07_”.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/74fa1bf3b2abbadf5785e90a.png\"},{\"id\":82854053,\"identity\":\"11799364-700f-4450-92f8-96e4e0474e7a\",\"added_by\":\"auto\",\"created_at\":\"2025-05-16 04:22:38\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":1299868,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eStructure and protomer crosslinking in A-ENA nanofibrils.\\u003c/strong\\u003e \\u003cstrong\\u003e(a)\\u003c/strong\\u003e Cartoon representation of a single Bti A-ENA monomer as identified in the cryoEM structure of recombinant A-ENA fibrils. A-ENA monomers are composed of an alpha-helical hairpin (a1 and a2) with a 13 residue ‘N-terminal lock’ (NTL). Residues shown in stick representation (residue number shown in bold) partake in intermolecular isopeptide bond (IPB) formation with neighboring A-ENA monomers (recipient residues of the corresponding A-ENA partner subunit are shown next to each highlighted residue, with the arrows running from the nucleophilic to the electrophilic residue); \\u003cstrong\\u003e(b)\\u003c/strong\\u003e Frontal view of Bti A-ENA (90° rotation with respect to panel a) with residues involved in intermolecular isopeptide bonds highlighted and shown in stick representation: inter-protofibrillar contacts of monomer designated i are made with monomer i’ (i.e. within the same plane in protofibril f’; green arrows) and with monomer i’+1 (shifted one unit upwards in protofibril f’); intra-protofibrillar contacts are made with the A-ENA monomers directly above (i+1) or below (i-1) subunit i, as well as with monomer i-5 via residue 2 in the N-terminal lock (blue arrows). As such, each subunit i is implicated in 10 IPBs, crosslinking 7 different protomers within and across the protofibrils \\u003cstrong\\u003e(c)\\u003c/strong\\u003e IPB interaction network in the (rec-)A-ENA fiber; \\u003cstrong\\u003e(d)\\u003c/strong\\u003e Ultrastructure of the A-ENA fibril with both protofibrils f’ and f coloured green and blue, resp., and with a single A-ENA monomer highlighted in grey; \\u003cstrong\\u003e(e)\\u003c/strong\\u003e Close-up with cartoon representation of the subunits in protofibril f (i+1 to i-5), and illustrated as sticks are the residues involved in the intra-protofibrillar IPBs between i/i+1, i/i-1 and i/i-5 (donating and accepting IPB in full and dotted circles, resp.); \\u003cstrong\\u003e(f)\\u003c/strong\\u003e Cartoon representation of the inter-protofibrillar IPBs between i/i’ and i/i’+1; \\u003cstrong\\u003e(g)\\u003c/strong\\u003e Helical cryoEM ultratructures of recombinantly purified A-ENA and A-ENA-1 fibers; \\u003cstrong\\u003e(h)\\u003c/strong\\u003e Pairwise-sequence alignment of A-ENA and A-ENA-1: d: IPB-donor residue; e: IPB-acceptor residue; g: catalytic glutamate. Symbol subscripts denote the IPB-type; \\u003cstrong\\u003e(i)\\u003c/strong\\u003e nsTEM images of A-ENA(-1) fibers recombinantly expressed in \\u003cem\\u003eE. coli. \\u003c/em\\u003eScalebar represents 100nm; \\u003cstrong\\u003e(j)\\u003c/strong\\u003e Structural comparison between rec-A-ENA and A-ENA-1 protomers (rmsd = 0.55Å): residues 1-8 are not resolved in the A-ENA-1 structure; \\u003cstrong\\u003e(k)\\u003c/strong\\u003e IPB types identified in rec A-ENA fibers (IPBs marked with star were not observed in A-ENA-1).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/c512f96765bbe42fd87ecc33.png\"},{\"id\":82853883,\"identity\":\"52c4ff7f-3636-4576-b1a4-ca4985dd10c1\",\"added_by\":\"auto\",\"created_at\":\"2025-05-16 04:14:38\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":528327,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eA-ENA clusters spores and PSBs and acts as a virulence factor. (a) \\u003c/strong\\u003ePhase-contrast microscopy of the different Histodenz fractions (supernatant (SN), 40% and 60%) comparing Bti WT and \\u003cem\\u003eΔa-ena\\u003c/em\\u003e and different centrifugal forces (100 and 10,000 \\u003cem\\u003eg\\u003c/em\\u003e). PSBs are false colored blue, scale bar = 5 mm. \\u003cstrong\\u003e(b) \\u003c/strong\\u003eSDS-PAGE of high-pH dissolved PSBs in the different density fractions. \\u003cem\\u003eInput\\u003c/em\\u003e (before fractionation) shows same starting toxin load. Bands correspond to Cry11Aa (◆),Cry11a fragments (◇) and Cyt1Aa (▼). \\u003cstrong\\u003e(c) \\u003c/strong\\u003eNormalized OD\\u003csub\\u003e600\\u003c/sub\\u003e (i.e. relative spore content) of the different density fractions of Bti WT and \\u003cem\\u003eΔa-ena.\\u003c/em\\u003e \\u003cstrong\\u003e(d) \\u003c/strong\\u003ensTEM images from Bti WT and \\u003cem\\u003eΔa-ena\\u003c/em\\u003e. PSBs and spores are labelled with blue and white arrowheads, respectively. Scale bar is 3 μM. \\u003cstrong\\u003e(d) \\u003c/strong\\u003ePhase contrast images of \\u003cem\\u003eBtk WT \\u003c/em\\u003e(left), or \\u003cem\\u003eBtk\\u003c/em\\u003e recombinantly expressing (Btk \\u003cem\\u003ep-A-ENA\\u003c/em\\u003e;\\u003cem\\u003e \\u003c/em\\u003emiddle) or with exogenously added A-ENA fibers \\u003cem\\u003e(Btk\\u003c/em\\u003e + A-ENA at 4.18 mg/ml; right). PSBs are false colored blue, scale bar = 5 mm. (\\u003cstrong\\u003ef\\u003c/strong\\u003e) Survival curves of \\u003cem\\u003eChironomus aprilinus\\u003c/em\\u003e in the presence of PBS buffer (negative control; grey dotted line) or two \\u003cem\\u003eBti\\u003c/em\\u003e strains: WT (continuous blue line) and \\u003cem\\u003eΔa-ena\\u003c/em\\u003e (dotted blue line). Dead individuals were counted daily during 6 days. For each condition, the spore suspension was diluted in the media to a final OD\\u003csub\\u003e600\\u003c/sub\\u003e of 0.004. \\u003cem\\u003en=39\\u003c/em\\u003e individuals per group. (\\u003cstrong\\u003eg\\u003c/strong\\u003e) (\\u003cem\\u003eleft\\u003c/em\\u003e) Survival curves of \\u003cem\\u003eTrichoplusia ni \\u003c/em\\u003ein the presence of PBS buffer (grey dotted line), WT Btk (continuous pink line) and Btk \\u003cem\\u003epAS_A-ENA\\u003c/em\\u003e (‘pA-ENA’, pink dotted line). Bti WT was included to test for toxicity from A-ENA. \\u003cem\\u003en=24\\u003c/em\\u003e individuals per group. (\\u003cem\\u003eright\\u003c/em\\u003e) Survival curves of \\u003cem\\u003eT. ni \\u003c/em\\u003ein the presence of PBS buffer (grey dotted line), Btk WT with (dashed line) or without (continuous line) addition of recombinant A-ENA fibers (4.18 mg/ml final concentration). In all experiments, the 50 µL of spore - PSB preparations (at OD\\u003csub\\u003e600\\u003c/sub\\u003e of 0.06) were spread on top of the food tray measuring 2.5 cm x 2.5 cm. \\u003cem\\u003en=16\\u003c/em\\u003e individuals per group, followed for six days.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/d47398069f2440c4f49b0cc2.png\"},{\"id\":82853886,\"identity\":\"9b608722-a4d0-4b70-99b4-d62bd995ad62\",\"added_by\":\"auto\",\"created_at\":\"2025-05-16 04:14:38\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":209395,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003ePhylogenetic analysis of A-ENA occurrence.\\u003c/strong\\u003e \\u003cstrong\\u003e(a)\\u003c/strong\\u003e Phylogenetic tree of the \\u003cem\\u003eBacillus cereus sensu lato \\u003c/em\\u003egroup: inner ring: GTDB species (as labelled), second ring: absence/presence (green/yellow) of A-ENA, outer ring: absence/presence (green/yellow) of Bt toxins. An interactive version of this tree can be accessed at https://microreact.org/project/rzLnUcAAPfJDDYFg9vFxdv-btyper-a-ena-prevalence; \\u003cstrong\\u003e(b)\\u003c/strong\\u003ePhylogenetic tree based on NCBI taxonomy ID using phyloTv2 of A-ENA carrying genomes in the Bacillota phylum.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/dc206f01166f9a3148a0d14b.png\"},{\"id\":82853882,\"identity\":\"2432959f-78c6-45f4-966a-5d345574170a\",\"added_by\":\"auto\",\"created_at\":\"2025-05-16 04:14:38\",\"extension\":\"png\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":115800,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eSporulation – spore biofilm cycle of \\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003eBacillus thuringiensis\\u003c/strong\\u003e\\u003c/em\\u003e. \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e alternates between dispersed vegetative cells (gray) and a biofilm-associated spore state that represents a dormant and infectious dissemination stage (green). Under favorable conditions, the bacterium exists as rod-shaped vegetative cells that divide by binary fission (vegetative cycle). Upon nutrient starvation, cells initiate sporulation, forming two compartments: the forespore, which will develop into a mature spore, and the mother cell, where insecticidal protein will assemble into a parasporal crystal (PSB). PSB assembly is accompanied by A-ENA expression and accumulation in the mother cell. Following mother cell lysis, A-ENA facilitates the aggregation of spores and PSBs, and the forming a structured biofilm in which A-ENA serves as the main extrasporal matrix component. Upon ingestion by insect larvae, and toxin crystals dissolve in the larval gut, leading to gut perforation, sepsis and insect death, spores germinate and resume a vegetative proliferation cycle. Created in https://BioRender.com\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/581842303542eba69750e8e2.png\"},{\"id\":107867932,\"identity\":\"4cf0e31e-5726-4cf7-b654-ab70804c2436\",\"added_by\":\"auto\",\"created_at\":\"2026-04-27 07:05:40\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":4737375,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/51be7e57-94ae-46d6-be95-f936f232a9ef.pdf\"},{\"id\":82853890,\"identity\":\"0c52e2d9-c691-4c8a-91ee-f6adda7d803a\",\"added_by\":\"auto\",\"created_at\":\"2025-05-16 04:14:38\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":9941428,\"visible\":true,\"origin\":\"\",\"legend\":\"Supplementary Information\",\"description\":\"\",\"filename\":\"SupplementaryInformation.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6239967/v1/77c6a1cc9065c344c66757bf.pdf\"}],\"financialInterests\":\"\\u003cb\\u003eYes\\u003c/b\\u003e there is potential Competing Interest.\\nThe authors are named as inventors on three patent applications filed by VIB, related to the technological use of A-ENA nanofibers.\",\"formattedTitle\":\"Auto-crosslinking sporesilk fibers promote endospore and Cry toxin clustering\",\"fulltext\":[{\"header\":\"Introduction\",\"content\":\"\\u003cp\\u003e \\u003cem\\u003eBacillus thuringiensis\\u003c/em\\u003e (Bt) is an insecticidal, Gram-positive, aerobic endospore-forming bacterium of the \\u003cem\\u003eBacillus cereus\\u003c/em\\u003e sensu lato group, found in soil, dust, and on plant surfaces (Aronson et al., \\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e1986\\u003c/span\\u003e). Bt spores are used as established biopesticides in agriculture, private residential use and vector-borne disease control due to their remarkable capacity to kill larvae of various insect pests from the Coleoptera, Lepidoptera, Hymenoptera and Diptera orders, including \\u003cem\\u003eAedes\\u003c/em\\u003e and \\u003cem\\u003eAnopheles\\u003c/em\\u003e mosquitos - vectors for zika (Musso et al., \\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e2019\\u003c/span\\u003e), dengue (Jansen and Beebe, \\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e2010\\u003c/span\\u003e) and malaria \\u003cem\\u003e(\\u003c/em\\u003ePhillips et al., \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e2017\\u003c/span\\u003e). Since Bt strains target specific insect groups with minimal impact on humans and non-target organisms (e.g. wildlife, pollinators and other beneficial insects), they can form the basis of biopesticides that offer a safe and eco-friendly alternative to non-discriminatory, environmentally persistent and potentially toxic insecticidal chemicals (Koch et al., \\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e2015\\u003c/span\\u003e; McClintock et al., \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e1995\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eBt adopts its entomopathogenic lifestyle by the characteristic production of vegetative insecticidal toxins (Vip) (Bravo et al., \\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e2007\\u003c/span\\u003e; Schnepf et al., \\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e1998\\u003c/span\\u003e) or delta-endotoxins (Cry and Cyt) during sporulation (Bravo et al., \\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e2007\\u003c/span\\u003e; Schnepf et al., \\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e1998\\u003c/span\\u003e). Bt toxins exhibit a distinctive composition unique to each strain, resulting in a remarkable selectivity for specific insect genera. The toxins are produced as pro-toxins that assemble into pure or mixed crystalline condensates referred to as parasporal bodies or crystals (PSB, (Federici et al., \\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e1990\\u003c/span\\u003e)) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ea,b). Upon ingestion of Bt spores, the associated pro-toxin crystals dissolve and exert a poreforming activity in the insect gut, followed by intestinal sepsis, starvation and host death. In the process, germinated vegetative Bt cells colonize the insect until available nutrients in the carcass become depleted, then inducing sporulation to ensure further dissemination and survival of adverse conditions until the spores are taken up by e.g. a new host to restart the cycle.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eBt was originally registered as a General Use in 1961 but is now found in over 1000 registered American pesticide products (National Pesticide Information Center. NPIC Product Research Online (NPRO): \\u0026lsquo;\\u003cem\\u003eBacillus thuringiensis\\u003c/em\\u003e\\u0026rsquo;. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://npic.orst.edu/NPRO/\\u003c/span\\u003e\\u003cspan address=\\\"http://npic.orst.edu/NPRO/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e), 157 of which are specific to \\u003cem\\u003eBacillus thuringiensis\\u003c/em\\u003e subsp. \\u003cem\\u003eisraelensis\\u003c/em\\u003e (bti; (National Pesticide Information Center. NPIC Product Research Online (NPRO): \\u0026lsquo;\\u003cem\\u003eisraelensis\\u003c/em\\u003e\\u0026rsquo;. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://npic.orst.edu/NPRO/\\u003c/span\\u003e\\u003cspan address=\\\"http://npic.orst.edu/NPRO/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e)). Bti is a naturally occurring soil bacterium that is used as a larvicide in aquatic and wetland areas where black fly and mosquito-borne disease control is deployed (Charles and De Barjac, \\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e1982\\u003c/span\\u003e). Typical commercial formulations (Aquabac\\u0026copy;, Teknar\\u0026copy;, Bactimos\\u0026copy;, and Vectobac\\u0026copy; ) consist of a mixture of spores and extracellular PSBs that are produced through a simple fermentation process that leverages the natural growth cycle of Bti where nutrient starvation and high cell density trigger sporulation. These conditions mimic the natural scenario of \\u003cem\\u003eBacillus\\u003c/em\\u003e biofilm formation which consists of a complex network of vegetative cells, spores, PSBs and extracellular matrix (ECM) components (e.g. for \\u003cem\\u003eBacillus subtilis\\u003c/em\\u003e: TasA fibers, eDNA, poly-γ-glutamic acid, exopolysaccharide) (Arbour et al., \\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2023\\u003c/span\\u003e). Because most studies on Bti tend to focus on PSB composition, larvicidal activity and specificity, the composition of the extracellular milieu of a spore dominated population remains poorly understood despite obvious academic and industrial interest.\\u003c/p\\u003e \\u003cp\\u003eHere we show that Bti forms a bona fide, spore-based biofilm characterized by a pervasive extrasporal matrix (ESM) that is constructed of 8nm diameter protein nanofibers (dubbed A-ENA; see nomenclature below) that function as molecular tethers between spores and PSBs. Using cryoEM we solve the helical ultrastructure of the A-ENA fibers and show that they are composed of an 11kDa two-helix bundle protein that is extensively covalently cross-linked to its nearest neighbors via iso-peptide bonds. We demonstrate that A-ENA subunits spontaneously and efficiently polymerize in \\u003cem\\u003eE\\u003c/em\\u003e. \\u003cem\\u003ecoli\\u003c/em\\u003e through a dock-and-lock self-assembly mechanism wherein subunits first rendezvous driven by hydrophobics, followed by enzyme-free, autocatalytic iso-peptide bond formation, yielding robust, micrometer-long covalent, flexible nanofibers. We show that loss of A-ENA breaks up the association of PSBs to Bti spores and attenuates their killing activity towards \\u003cem\\u003eChironomus aprilinus\\u003c/em\\u003e larvae. Remarkably, the addition of recombinant A-ENA fibers to \\u003cem\\u003eBacillus thuringiensis\\u003c/em\\u003e subsp. \\u003cem\\u003eKurstaki\\u003c/em\\u003e (Btk; natively \\u003cem\\u003ea-ena\\u003c/em\\u003e\\u003csup\\u003e\\u0026minus;\\u003c/sup\\u003e) enhanced spore clustering and toxin retention, resulting in an increased killing activity towards the agricultural pest \\u003cem\\u003eTrichoplusia ni\\u003c/em\\u003e. Our results show that A-ENA (i) is a virulence factor that exerts its function by ensuring close proximity of the infectious particle (spore) to its toxic payload (PSB) and (ii) is shown to function outside its native Bti context due to its aspecific mode of binding to spore and PSB surfaces, and is therefore expected to see more widespread use in other Bt systems that are naturally devoid of A-ENA but otherwise characterized by the presence of extracellular PSBs.\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e\\n \\u003ch2\\u003eBTI spores and PSBs are surrounded by a proteinaceous fiber biofilm\\u003c/h2\\u003e\\n \\u003cp\\u003eTo investigate the mesoscopic structure of a Bti biofilm, we dissolved freshly sporulated (72h, LB agar, 30\\u0026deg;C) Bti cultures into miliQ and analyzed the bacterial suspension using light (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ec,d) and negative stain electron transmission microscopy (nsTEM; Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ee). The suspension showed the presence of clustered microcolonies that were composed primarily of bacterial spores and parasporal bodies (PSBs), with vegetative cells representing only a small minority (\\u0026lt;\\u0026thinsp;1%) of the population (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ec,d). Strikingly, both the spores and PSBs were engulfed in a dense extrasporal matrix composed of a fibrous mesh that permeated the entire sample (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ee, unstained). The fibers were shown to emanate from both the spore and PSB surface, and were also seen to mediate direct connections between endospores and across endospores and PSBs (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ee; Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ea-d). We found numerous cases of large fiber bundles (100nm diameter) composed of dozens of individual fibers, making connection to the PSBs and the spore surface. Closer inspection of the fiber biofilm revealed nanofibers of ~\\u0026thinsp;8nm diameter and with a distinct double helical patterning in 2D class averages (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ea,b; Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ed). Bti PSBs comprise polycrystalline condensates of insecticidal proteins (Cry11Aa1, Cyt1Aa1, Cry4Aa1 and Cry4Ba1) that are packaged in a spherical parasporal sack, and that dissolve under the elevated pH in the insect gut (Rudd et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2023\\u003c/span\\u003e). When incubated at pH 11 we found that Bti spore preparations retained a biofilm-like organization, with spores embedded in a dense nanofiber matrix (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ee-i). Remarkably, parasporal sacks remained tightly entangled in the fiber network, but lost their condensate content. Close inspections showed nanofibers tightly associated with the surface of the parasporal sacks. Based on these observations, we hypothesized that the particulate, crystalline nature of the toxins and the fiber-mediated tethering of spores and PSBs ensures the cytotoxic payload remains at a high localized concentration and closely associated with the infectious particles (spore), thus ensuring their co-transmission during host-pathogen interactions.\\u003c/p\\u003e\\n\\u003c/div\\u003e\\n\\u003ch3\\u003eA-ENA is a new family type of Endospore appendages\\u003c/h3\\u003e\\n\\u003cp\\u003eIntrigued by the existence of this fiber net, we proceeded with a cryo electron microscopy (cryoEM) analysis of the Bti exosporal matrix. Freshly resuspended Bti spores were deposited with no further processing onto a quantifoil holey carbon grid and plunge-frozen in liquid ethane, and imaged at 60k magnification using a JEOL CRYO ARM 300 microscope equipped with omega energy filter and a K3 direct electron detector. Similarly to nsTEM, cryoEM revealed a dense network of fibers in proximity to the vitrified endospores (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ec). Fragments of isolated fibers were boxed and subjected to 2D alignment and classification. The resulting 2D class averages and Fourrier-Bessel indexing using PyHi (Zhang, \\u003cspan class=\\\"CitationRef\\\"\\u003e2022\\u003c/span\\u003e) confirmed the double helical character of the protein filaments (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ed). Using cryosparc v4.4.1, we refined the fiber ultrastructure to a global resolution of 2.1\\u0026Aring;, with a helical rise and twist of 5.37\\u0026Aring; and 186.4\\u0026deg; (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ee; Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ea). The cryoEM map revealed a right-handed two-start helix with parallel protofilaments that showed clear continuous main chain density for axially stacked alpha-helical subunits (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ea,b,d). The high-resolution cryoEM reconstruction showed identifiable side chain densities that allowed for the \\u003cem\\u003ede novo\\u003c/em\\u003e building and sequence assignment of an initial atomic model of the constituent subunits (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ef). A standard protein BLAST search using the manually built protein sequence yielded protein Q8KNV8 (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.uniprot.org/\\u003c/span\\u003e\\u003c/span\\u003e) as the top-ranking candidate for the major subunit. Q8KNV8 is a 101 amino acids protein with no known protein family membership as judged from InterProScan (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.ebi.ac.uk/interpro/\\u003c/span\\u003e\\u003c/span\\u003e). Based on the lack of any sequence or structural homology or resemblance to known pili, functional amyloids or the recently characterized endospore appendages (i.e. S- and L-ENA (DUF3992) (Pradhan et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2021\\u003c/span\\u003e; Sleutel et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e)), it is clear that the Bti fibers constitute a new family of protein nanofibers referred to as \\u003cspan type=\\\"Underline\\\" class=\\\"Underline\\\" name=\\\"Emphasis\\\"\\u003ea\\u003c/span\\u003elpha helical \\u003cspan type=\\\"Underline\\\" class=\\\"Underline\\\" name=\\\"Emphasis\\\"\\u003een\\u003c/span\\u003edospore \\u003cspan type=\\\"Underline\\\" class=\\\"Underline\\\" name=\\\"Emphasis\\\"\\u003ea\\u003c/span\\u003eppendages, or A-ENA in short.\\u003c/p\\u003e\\n\\u003cp\\u003eThe genetic locus corresponding to Q8KNV8 is ATN07_33990, and is found on the 128kb pBtoxis-like plasmid (Ben-Dov et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e1999\\u003c/span\\u003e; Berry et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2002\\u003c/span\\u003e) (pAM65-52-4-128K; RefSeq: NZ_CP013279.1) that carries all the insecticidal d-endotoxins. We identify a local three gene cluster (\\u003cem\\u003ea-ena\\u003c/em\\u003e) consisting of ATN07_33990 (\\u003cem\\u003ea-ena\\u003c/em\\u003e), ATN07_33985 and ATN07_33980 (\\u003cem\\u003ea-ena1\\u003c/em\\u003e) (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ee). ATN07_33985, also referred to as pBt031, encodes a putative two-domain N-acetylmuramoyl-L-alanine amidase with a catalytic cell wall hydrolase domain (Pfam: PF01520) (Mistry et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2021\\u003c/span\\u003e) and a sporulation related repeat domain (Pfam: PF05036), and has previously been suggested to be involved in mother cell wall lysis similar to the sporulation specific autolysin CwlB in \\u003cem\\u003eB. subtilis\\u003c/em\\u003e (Berry et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2002\\u003c/span\\u003e). Looking at the broader genomic context, the \\u003cem\\u003ea-ena\\u003c/em\\u003e cluster is integrated into a region encoding for numerous d-endotoxins, i.e. upstream are located Cyt2Ba (ATN07_33970) and Cry8Ea (ATN07_33980). Downstream of \\u003cem\\u003ea-ena\\u003c/em\\u003e, there are three other toxin genes, i.e. Cyt1A (ATN07_34010), Cry2A/C (ATN07_34015) and Cry11a (ATN07_34020). Of final note, the \\u003cem\\u003ea-ena\\u003c/em\\u003e gene cluster is flanked by ATN07_34000, which has been annotated as a Tn3 family transposase (uniprot: A0A160LKE1) (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eg).\\u003c/p\\u003e\\n\\u003cp\\u003eIn the \\u003cem\\u003ea-ena\\u003c/em\\u003e gene cluster, ATN07_33980 encodes for protein Q8KNV7 which shares 74.0% seq id to Q8KNV8, and is predicted to fold into a two-helix bundle that is structurally homologous to Q8KNV8 (the all-atom RMSD using Alphafold database entries AF-Q8KNV7-F1 and AF-Q8KNV8-F1 is 0.216 \\u0026Aring;). Based on this similarity, we refer to Q8KNV8 and Q8KNV7 as A-ENA and A-ENA1. LC-MS/MS \\u003cem\\u003ede novo\\u003c/em\\u003e sequencing of isolated fibers detected the presence of all three proteins of the local \\u003cem\\u003ea-ena\\u003c/em\\u003e gene cluster, i.e. A-ENA, A-ENA1 and the N-acetylmuramoyl-L-alanine amidase. To resolve the ambiguity concerning the subunit composition of the \\u003cem\\u003eex vivo\\u003c/em\\u003e fibers we carefully inspected the reconstructed cryoEM volume and manually compared the A-ENA and A-ENA1 sequences at key sites with subunit-specific side-chain type (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ed). Based on that analysis, we unambiguously identified A-ENA as the major subunit of the fibers, but do not rule out that A-ENA1 is present in the fibers at low substoichiometric levels.\\u003c/p\\u003e\\n\\u003ch3\\u003eA-ENA forms cross-\\u0026alpha; fibrils crosslinked via isopeptide bonds\\u003c/h3\\u003e\\n\\u003cp\\u003eThrough careful inspection of the A-ENA cryoEM map, we identified multiple sites of continuous density bridging adjacent A-ENA protomers within and across protofilaments (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e, Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e). After the atomic A-ENA model was built, we identified these sites as intermolecular isopeptide bonds (IPB) composed of a nucleophilic IPB \\u0026lsquo;donor\\u0026rsquo; (i.e. G2 N-terminus, K42, K76, K77 or K84; Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ea, k) and an electrophilic IPB \\u0026lsquo;acceptor\\u0026rsquo; (i.e. E29, E44, N64, E78, or Q82; Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ea, k). Adjacent each IPB site, we find glutamic acid residues favorably positioned to act as acid-base catalyst (e.g. E28 Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ek, Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ec), similar to what is seen for intramolecular IPBs in bacterial pili (Kang and Baker, \\u003cspan class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e; Kang et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2007\\u003c/span\\u003e). In the latter, subunit folding brings the triads of IPB donor (K), acceptor (D, E, N or Q) and acid-base catalyst (E or D) in close proximity resulting in autocatalytic IPB formation. In A-ENA, however, IPB donor and acceptor are located on separate subunits, and a knobs-in-holes stacking of adjacent A-ENA subunits results in the formation of proximity-induced intermolecular IPBs (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eb). In this way, five different IPB pairs (IPB1-5: G2-Q44; K42-N64; K77-E78; K84-E29 and K76-Q82) are formed twice per A-ENA subunit, i.e. ones as donor (\\u0026rarr;) and ones as acceptor (\\u0026larr;), totaling ten IPBs per protomer (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e; Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e; Table \\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). Of these, six IPBs connect consecutive subunits within a single protofilament (IPB 2, 3 and 4; as donor towards N-pole: \\u003cem\\u003ei\\u003c/em\\u003e\\u0026rarr;\\u003cem\\u003ei-1\\u003c/em\\u003e, and as acceptor towards C-pole: i\\u0026larr;\\u003cem\\u003ei\\u0026thinsp;+\\u0026thinsp;1\\u003c/em\\u003e). Two additional axial IPBs (IPB1) are formed between the subunits i\\u0026rarr;i-5 as donor, and i\\u0026larr;i\\u0026thinsp;+\\u0026thinsp;5 as acceptor (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ee). Finally, IPB5 forms two isopeptide bonds across the two protofibrils, (\\u003cem\\u003ei\\u003c/em\\u003e\\u0026rarr;\\u003cem\\u003ei\\u0026rsquo;-1\\u003c/em\\u003e) as donor, and (i\\u0026larr;i\\u0026rsquo;) as acceptor (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ef). Whilst IPB2-5 are formed by knobs-in-holes interactions of alpha helices \\u0026alpha;1 and \\u0026alpha;2 of consecutive protomers within (IPB2-4) and adjacent protomers across (IPB5) the protofilaments, IPB1 connects distant subunits, realized by a long N-terminal latch (NTL) constructed of residues 2 to 14 and that is docked via residues M3, P4, I6, P7, L10 and I12 into a hydrophobic groove that runs axially along the A-ENA fiber surface spanning 4 subunits.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003ctable id=\\\"Tab1\\\" border=\\\"1\\\"\\u003e\\n \\u003ccaption language=\\\"En\\\"\\u003e\\n \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 1\\u003c/div\\u003e\\n \\u003cdiv class=\\\"CaptionContent\\\"\\u003e\\n \\u003cp\\u003eIsopeptide bond types observed in A-ENA fibers\\u003c/p\\u003e\\n \\u003c/div\\u003e\\n \\u003c/caption\\u003e\\n \\u003cthead\\u003e\\n \\u003ctr\\u003e\\n \\u003cth align=\\\"left\\\" style=\\\"width: 6.772%;\\\"\\u003e\\n \\u003cp\\u003eIPB\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\" style=\\\"width: 10.6095%;\\\"\\u003e\\n \\u003cp\\u003eDonor\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eAcceptor\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eCatalyst\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eContact\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eInter / Intra fibril\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eNumber\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/thead\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 6.772%;\\\"\\u003e\\n \\u003cp\\u003e1\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 10.6095%;\\\"\\u003e\\n \\u003cp\\u003eG2*\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eQ44\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eE41\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003ei -\\u0026gt; i-5\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eintra\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e2\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 6.772%;\\\"\\u003e\\n \\u003cp\\u003e2\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 10.6095%;\\\"\\u003e\\n \\u003cp\\u003eK42\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eN64\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eE39\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003ei -\\u0026gt; i-1\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eintra\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e2\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 6.772%;\\\"\\u003e\\n \\u003cp\\u003e3\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 10.6095%;\\\"\\u003e\\n \\u003cp\\u003eK84\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eE29\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eE28**\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003ei -\\u0026gt; i-1\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eintra\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e2\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 6.772%;\\\"\\u003e\\n \\u003cp\\u003e4\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 10.6095%;\\\"\\u003e\\n \\u003cp\\u003eK77\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eE78\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eE28**\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003ei -\\u0026gt; i-1\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eintra\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e2\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 6.772%;\\\"\\u003e\\n \\u003cp\\u003e5\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\" style=\\\"width: 10.6095%;\\\"\\u003e\\n \\u003cp\\u003eK76\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eQ82\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eE86\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003ei -\\u0026gt; i\\u0026rsquo;\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003einter\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e2\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n \\u003ctfoot\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd colspan=\\\"7\\\"\\u003e*Donor group for IPB type 1 is the amino group of the N-terminus\\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd colspan=\\\"7\\\"\\u003e**E28 functions as the catalyst for IPB types 3 and 4\\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tfoot\\u003e\\n\\u003c/table\\u003e\\n\\u003ch3\\u003e\\u003cbr\\u003e\\u003c/h3\\u003e\\n\\u003cdiv class=\\\"Heading\\\"\\u003e\\u003cstrong\\u003eA-ENA self-assembly and autocatalytic isopeptide bond formation\\u003c/strong\\u003e\\u003c/div\\u003e\\n\\u003cp\\u003eMost bacterial pili and filaments require a dedicated assembly machinery (Lukaszczyk et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2019\\u003c/span\\u003e), and known polymers with intermolecular isopeptide bonds such as poly-ubiquitin or sortase-mediated pili employ a transpeptidation reaction that requires enzymatic accessory proteins (Kang and Baker, \\u003cspan class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e). In A-ENA, however, the proximity-induced IPB triads as a result of subunits stacking suggests they may form autocatalytically (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e). To evaluate the requirement of accessory proteins, A-ENA and A-ENA1 were recombinantly expressed in the cytoplasm of \\u003cem\\u003eE. coli\\u003c/em\\u003e BL21. Purified fractions enriched towards either A-ENA or A-ENA1 were shown to contain abundant protein fibers, which both yielded 2D class averages that appeared identical to those obtained from the \\u003cem\\u003eex vivo\\u003c/em\\u003e material (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eg,h,i; Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eb, c). Testament to the remarkable stability of these fibers is the 30min incubation step at 100\\u0026deg;C in the presence of 2% (w/v) sodium dodecyl sulfate (SDS) that we employed to purify the target fibers form cellular proteins. Similarly, A-ENA fibers were recalcitrant to treatment with chaotropic agents (8M urea), exposure to strong acidic (50% (v/v) formic acid) or alkaline conditions (2M NaOH), as well as wet (i.e. autoclaving) or desiccated (200\\u0026deg;C oven) high temperature treatment (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e). This remarkable stability suggested that IPBs were present in the recombinant fibers as well. We confirmed that hypothesis by solving the cryoEM structures of the highly similar recA-ENA and rec-A-ENA1 fibers to 2.40\\u0026Aring; (rise:5.47 \\u0026Aring;, twist: 186.36\\u0026deg;) and 2.63 \\u0026Aring; (rise:5.31 \\u0026Aring;, twist: 186.50\\u0026deg;) resolution, respectively (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e). Indeed, rec-A-ENA fibers exhibited all 5 IPB types observed in the \\u003cem\\u003eex vivo\\u003c/em\\u003e A-ENA fibers, totaling 10 IPBs per protomer. Although rec-A-ENA1 fibers are near isomorphous to A-ENA (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ee), they lacked both the type 1 and 5 IPBs, leading to 6 confirmed IPBs per protomer, with no observable impact on the fiber stability (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ej,k). This was unexpected because the IPB-associated residues are conserved in A-ENA1, and the NTL sequence is identical to the A-ENA sequence apart from N5 (A-ENA: T5; Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ea). Possibly, an increased susceptibility to proteolytic attack of the NTL and/or reduced processing of N-terminal methionine (i.e. to liberate the G2 N-terminus) may impair IPB formation for recombinant A-ENA1 in \\u003cem\\u003eE. coli\\u003c/em\\u003e.\\u003c/p\\u003e\\n\\u003cp\\u003eRegardless of these minor differences between A-ENA and A-ENA1, these results clearly show that IPB formation is auto-catalytic and independent of e.g. intervention of a transglutaminase (Griffin et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2002\\u003c/span\\u003e). Moreover, the efficient polymerization in \\u003cem\\u003eE. coli\\u003c/em\\u003e (yield 1.8g A-ENA fiber / L culture) demonstrates that A-ENA biogenesis is not reliant on external factors (e.g. chaperone, usher, protease, \\u0026hellip;) and that A-ENA self-assembly is fully encoded in its primary sequence. A multiple sequence alignment of A-ENA like sequences shows the strong conservation of the IPB donor/acceptor pairs, as well as their cognate acid-base catalysts (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ea,b). Different to intermolecular IPBs that form by transpeptidation (Kang and Baker, \\u003cspan class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e), A-ENA IPB triads are not part of recognition sequences, but rather take up predefined positions in the alpha 1 and alpha 2 helical wheels, such that packing of the A-ENA subunits brings the IPB partners in close proximity (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ea,b,c). Looking at the A-ENA axial protomer/protomer interface (e.g. \\u003cem\\u003ei\\u003c/em\\u003e\\u0026rarr;\\u003cem\\u003ei-1\\u003c/em\\u003e), we find that the buried interfaces are predominantly hydrophobic (mainly small hydrophobic chains apart from IPB residues; Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ed), and there are no obvious complementary charged surface patches, suggesting protomer docking is predominantly driven by the hydrophobic effect. There is exceptional complementarity between the N- and C-pole surfaces of helices a\\u003csub\\u003e1\\u003c/sub\\u003e and a\\u003csub\\u003e2\\u003c/sub\\u003e. This leads to a tight \\u0026lsquo;knobs-in-holes\\u0026rsquo; packing that drives the pre-orientation of the nucleophile/electrophile pair and its acid-base catalyst, and results in their electrostatic activation by the exclusion of solvent from the cross-\\u0026alpha; interfaces, essentially priming the residues for autocatalytic IPB formation.\\u003c/p\\u003e\\n\\u003ch3\\u003eA-ENA and A-ENA1 expression is uniquely associated with the spore extracellular matrix\\u003c/h3\\u003e\\n\\u003cp\\u003eThe close association of A-ENA fibers with spores and parasporal bodies (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e and S1), and localization of the \\u003cem\\u003ea-ena\\u003c/em\\u003e and \\u003cem\\u003ea-ena1\\u003c/em\\u003e genes on the pAM65-52-4-128K plasmid and in close proximity to the Bti \\u003cem\\u003ecry\\u003c/em\\u003e toxins (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eg) suggest A-ENA expression occurs during sporulation. Using the DBTSBS server (Sierro et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2008\\u003c/span\\u003e) we identify a putative recognition site (CACCTTTGGAATAGCTGTTACATATAATA; p-value 0.01) for the late stage mother cell sporulation transcription factor SigK 46bp upstream of \\u003cem\\u003ea-ena\\u003c/em\\u003e, in agreement with the predicted timing of expression given the extracellular, spore-associated localization of A-ENA. In agreement with this analysis, expression of a plasmid-encoded \\u003cem\\u003emcherry\\u003c/em\\u003e reporter gene under control of the hypothesized \\u003cem\\u003ea-ena\\u003c/em\\u003e promotor sequence was found to be uniquely associated with sporulating cells, and absent from vegetative cells (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ea). No known sigma factor recognition motifs were found upstream of \\u003cem\\u003ea-ena1\\u003c/em\\u003e or ATN07_33980 (using a p-value cut-off of 0.01), although separation of the forward oriented \\u003cem\\u003ea-ena\\u003c/em\\u003e and \\u003cem\\u003ea-ena1\\u003c/em\\u003e by the reverse oriented ATN07_33985 would suggest the \\u003cem\\u003ea-ena\\u003c/em\\u003e genes to have independent promotors. To further test this hypothesis and evaluate the timing, (co-)occurrence and localization of A-ENA and A-ENA1 fibers, we modified \\u003cem\\u003ea-ena\\u003c/em\\u003e and \\u003cem\\u003ea-ena1\\u003c/em\\u003e with C-terminally fused Spy and Snoop-tags, thus enabling their selective labelling with Spy-catcher-GFP and SnoopCatcher-mCherry conjugates, respectively. In agreement with the P\\u003csub\\u003ea\\u0026minus;ena\\u003c/sub\\u003e promotor activity seen from the \\u003cem\\u003emcherry\\u003c/em\\u003e reporter (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ea), the presence of A-ENA\\u003csup\\u003eSpyTag\\u003c/sup\\u003e was restricted to sporulating cells and released spores (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eb,c). Similarly, A-ENA1\\u003csup\\u003eSnoopTag\\u003c/sup\\u003e was spore-associated and not observed for vegetative cells. However, whilst A-ENA\\u003csup\\u003eSpyTag\\u003c/sup\\u003e fibers were ubiquitous throughout the spore population, A-ENA1\\u003csup\\u003eSnoopTag\\u003c/sup\\u003e only showed sporadic labelling (i.e\\u0026thinsp;\\u0026lt;\\u0026thinsp;1% of spores), suggesting a phase variable expression that is independent of \\u003cem\\u003ea-ena\\u003c/em\\u003e.\\u003c/p\\u003e\\n\\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e\\n \\u003ch2\\u003eA-ENA promotes spore clustering and retention of PSBs\\u003c/h2\\u003e\\n \\u003cp\\u003eTo shed light on the biological function of A-ENA, \\u003cem\\u003ea-ena\\u003c/em\\u003e was selectively inactivated by introduction of a premature stop codon in the Bti pAM65-52-4-128K plasmid using CRISPR/Cas9 technology. Selective inactivation of \\u003cem\\u003ea-ena\\u003c/em\\u003e versus \\u003cem\\u003ea-ena1\\u003c/em\\u003e (74% sequence identity) was achieved by means of an unique guide RNA target sequence and confirmed using PCR and Sanger sequencing (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ed,e). Spores of the resulting Bti \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e lost recognition by a polyclonal serum raised against recombinant A-ENA (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ef), and lacked discernable sporesilk fibers when inspected by nsEM (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ed). Conversely, inactivation of \\u003cem\\u003ea-ena1\\u003c/em\\u003e did not result in loss of spore-associated A-ENA fibers (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ed). Thus \\u003cem\\u003ea-ena\\u003c/em\\u003e expression is essential and sufficient for A-ENA production in the extrasporal matrix, whilst A-ENA1 is only a sporadic, or a minor, dispensable component of the A-ENA-like fibers. Using time-lapse light microscopy, we found no discernable difference in growth rate or sporulation efficiency for WT and \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e Bti (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eg). To evaluate if \\u003cem\\u003ea-ena\\u003c/em\\u003e expression would alter PSB content, OD-normalized spore\\u003csup\\u003eprep\\u003c/sup\\u003e fractions were treated at pH11 and analysed by SDS-PAGE (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eb, Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ea). SDS-PAGE showed bands at ~\\u0026thinsp;28, ~70 and ~\\u0026thinsp;130 kDa, identified, respectively, as the Cyt1A, Cry11Aa1 and Cry4Aa1/Cry4Ab1 toxins by MS fingerprinting, as well as minor bands around 55 and 38 kDa found to represent Cry11Aa1 breakdown products (Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ea). A quantitative comparison of the released bands showed spore\\u003csup\\u003eprep\\u003c/sup\\u003e inputs of Bti WT and \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e mutant showed no discernable difference in the of amounts of Cry toxins.\\u003c/p\\u003e\\n \\u003cp\\u003eOur microscopy experiments reveal Bti spore preparations consist of microcolonies with clustered spores and PSBs, embedded in a dense extrasporal matrix of A-ENA sporesilks (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e, Extended Data Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). During sporulation, the endospore and PSB are formed in the daughter and mother cell compartment, respectively (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ea) (Schnepf et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e1998\\u003c/span\\u003e). We therefore asked if A-ENA fibers are implicated in spore clustering and may serve to ensure spores and PSBs remain associated after mother cell lysis. To do so, spore preparation of Bti WT and \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e were subjected to cushioned density centrifugation (1 hr at 100\\u003cem\\u003eg\\u003c/em\\u003e or 10.000\\u003cem\\u003eg\\u003c/em\\u003e) with three phases corresponding to a low density buffer or supernatant (\\u0026lsquo;SN\\u0026rsquo;) layer with a \\u0026rho;\\u0026thinsp;=\\u0026thinsp;1.00g/ml (\\u0026rho;\\u003csup\\u003elow\\u003c/sup\\u003e), a medium density layer of 40% Histodenz with a \\u0026rho;\\u0026thinsp;=\\u0026thinsp;1.21 g/ml (\\u0026rho;\\u003csup\\u003emedium\\u003c/sup\\u003e), and a high density layer of 60% Histodenz with \\u0026rho;\\u0026thinsp;=\\u0026thinsp;1.32 g/ml (\\u0026rho;\\u003csup\\u003ehigh\\u003c/sup\\u003e). Sedimentation fractions were qualitatively followed by light microscopy (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003ea), and quantitatively monitored by SDS-PAGE of the high pH-released fraction to follow PSBs (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eb) and by OD\\u003csub\\u003e600\\u003c/sub\\u003e to track the fractionation of spores (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003ec). At low \\u003cem\\u003eg\\u003c/em\\u003e force, PSBs of WT Bti were primarily retained in the \\u0026rho;\\u003csup\\u003elow\\u003c/sup\\u003e fraction (i.e. \\u0026gt;95%), where they were found as agglutinated spore \\u0026ndash; PSB clusters. These low density spore-PSB clusters retained about 50% of spores (based on OD\\u003csub\\u003e600\\u003c/sub\\u003e), with the other half found as smaller spore clusters in the \\u0026rho;\\u003csup\\u003emedium\\u003c/sup\\u003e fraction (with sparse PSBs). In the \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e mutant, however, no agglutinated spore-PSB clusters were observed, and the majority of PSBs migrated into the \\u0026rho;\\u003csup\\u003emedium\\u003c/sup\\u003e fraction as individual, dispersed particles. Similarly, \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e spores were found as individual particles that migrated into \\u0026rho;\\u003csup\\u003emedium\\u003c/sup\\u003e fraction. At high \\u003cem\\u003eg\\u003c/em\\u003e force, PSBs in both WT and \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e Bti fully migrated into the \\u0026rho;\\u003csup\\u003emedium\\u003c/sup\\u003e fraction, WT as smaller spore-PSB clusters, \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e as dispersed particles. In both cases, about 30% of spores migrated into the \\u0026rho;\\u003csup\\u003ehigh\\u003c/sup\\u003e fraction as individual particles. These fractionations indicate the individual PSBs and the majority of spores to have a density between 1.21\\u0026ndash;1.32 g/ml. In the WT Bti, the extrasporal matrix of A-ENA nanofibers agglutinates spores and PSBs into biofilm-like microcolonies with reduced density. With an average volume of ~\\u0026thinsp;0.68 \\u0026micro;m\\u003csup\\u003e3\\u003c/sup\\u003e, spores experience around ~\\u0026thinsp;0.8 and ~\\u0026thinsp;80 pN gravitational force at 100 and 10.000 \\u003cem\\u003eg\\u003c/em\\u003e, respectively. Whilst at 1 pN force a fraction of the spores gets dislodged from the low density biofilm, higher forces (up to 80pN) are required to break up the biofilm into smaller fragments that migrate into the \\u0026rho;\\u003csup\\u003emedium\\u003c/sup\\u003e fraction. In absence of A-ENA, sporse and PSBs fractionate as individual particles, even at low \\u003cem\\u003eg\\u003c/em\\u003e force.\\u003c/p\\u003e\\n \\u003cp\\u003eSpurred on by this striking result, we also followed a gain-of-function approach, where we introduced \\u003cem\\u003ea-ena\\u003c/em\\u003e in Btk, which naturally lacks pAM65-52-4-128K or A-ENA homologs. To do so, \\u003cem\\u003ea-ena\\u003c/em\\u003e was cloned with its native promotor into shuttle vector pAS (i.e. modified pJOE899 by removal of \\u003cem\\u003ecas9\\u003c/em\\u003e), forming pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e. \\u0026alpha;-A-ENA dot blot analysis the respective spore preparations confirmed a lack of A-ENA in Btk WT, and showed the recombinant expression of A-ENA in Btk pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ef). Light microscopy and nsTEM imaging of Btk WT showed mostly dispersed spores and PSBs (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003ee, Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ec), with the presence of spore-attached S-ENA fibers (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ed; (Pradhan et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2021\\u003c/span\\u003e)), but a lack of A-ENA-like fibers or an extensive extrasporal matrix as seen for Bti (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ed). In sharp contrast, Btk pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e showed a rich network of A-ENA fibers, resulting in the clustering of Btk spores into multibody biofilms (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003ee, Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ec,d). Moreover, typical bipyramidal Btk toxin crystals (here also referred to as PSB were seen to be enclosed in the Btk pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e biofilms. Large A-ENA bundles could be seen to grace the PSB crystal surface as well as making contact with neighbouring spores, seemingly functioning as a molecular tether mimicking the spore-PSB association seen in Bti. Similar to Bti, we found no significant difference in the PSB formation for WT and Btk pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ea). Remarkably, a very similar A-ENA dependent spore and PSB clustering can be seen by the exogenous addition of recombinant A-ENA nanofibers (\\u0026lsquo;+A-ENA\\u0026rsquo;) purified from \\u003cem\\u003eE. coli\\u003c/em\\u003e (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003ee, Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ec,d), showing the agglutinating activity does not require A-ENA expression by the sporulating cells, and can be induced post-sporulation. When WT Btk, pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e and +\\u0026thinsp;A-ENA spore preps were subjected to cushioned density centrifugation, we observed that the spore \\u0026ndash; PSB agglutination induced by recombinant expression, or exogenous addition of A-ENA results in the increased retention of spores and PSBs in the low density fraction (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eb,e). Even at high \\u003cem\\u003eg\\u003c/em\\u003e-forces, as much as 50% of PSBs were retained in a low density agglutinated biofilm fraction. Moreover, whereas WT Btk spores and PSB separated over a 70% sucrose cushion, as previously reported (Thomas and Ellar, \\u003cspan class=\\\"CitationRef\\\"\\u003e1983\\u003c/span\\u003e), in Btk pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e PSBs would co-pellet with the spores, demonstrating the encapsulation in A-ENA matrix ensured a robust spore \\u0026ndash; PSB association (Extended Data Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ef).\\u003c/p\\u003e\\n\\u003c/div\\u003e\\n\\u003ch3\\u003eA-ENA as a virulence factor\\u003c/h3\\u003e\\n\\u003cp\\u003eThe genetic association of \\u003cem\\u003ea-ena\\u003c/em\\u003e and Cry toxin genes on a plasmid and the observation that A-ENA facilitates spore \\u0026ndash; PSB union suggested to us that A-ENA may represent a virulence factor for Bt insecticidal activity. We therefore compared the survival of \\u003cem\\u003eChironomus aprilinus\\u003c/em\\u003e larvae in the presence of buffer or equivalent spore preparations of Bti WT and \\u003cem\\u003e\\u0026Delta;a-ena\\u003c/em\\u003e at a dose of approximately OD\\u003csub\\u003e600\\u003c/sub\\u003e 0.004. Mortality was monitored over 6 days following a single addition of spores to the aqueous medium. The results showed an increase in the MDK\\u003csub\\u003e80\\u003c/sub\\u003e (mean duration of killing 80% of the population) from 3 to 4 days for Bti WT and A-ENA KO, respectively. We next evaluated if recombinant A-ENA would result in a gain of virulence in strains naturally lacking sporesilks. To do so, we exposed two week old cabbage looper larvae (\\u003cem\\u003eTrichoplusia ni\\u003c/em\\u003e; a common pest of cruciferous crops and susceptible to Btk (Estada and Ferre, \\u003cspan class=\\\"CitationRef\\\"\\u003e1994\\u003c/span\\u003e)) to Btk WT, Btk pAS_\\u003cem\\u003ea-ena\\u003c/em\\u003e and Btk\\u0026thinsp;+\\u0026thinsp;A-ENA. Resuspended spore preparations were deposited on the solid feeding medium at a dose of 8 \\u0026micro;L/cm\\u003csup\\u003e2\\u003c/sup\\u003e of OD\\u003csub\\u003e600\\u003c/sub\\u003e 0.06. Whereas the recombinant expression or exogenous addition of A-ENA in Btk resulted in an MDK80 of 4 days, WT Btk did not show more than 50% lethality after six days of exposure (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eg). To ensure that any observed effects were not due to the toxic properties of A-ENA fibers, we included Bti WT in the experiment. As expected from its known killing profile, Bti WT had no effect on the survival rate of \\u003cem\\u003eT. ni\\u003c/em\\u003e larvae. Moreover, exposure of the larvae to A-ENA fibers showed no killing activity (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eg). Thus addition of A-ENA fibers provides a convenient approach for the non-recombinant increase of the spore \\u0026ndash; toxin clustering in Bt strains lacking native sporesilks, and results in improved pesticidal activity.\\u003c/p\\u003e\\n\\u003ch3\\u003ePhylogenetic analysis suggests biological functions beyond spore/PSB coupling\\u003c/h3\\u003e\\n\\u003cp\\u003ePhylogenetic analysis of A-ENA prevalence across the Firmicutes phylum shows that A-ENA homologues are found across both the Clostridia and Bacillales clades (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e). Due to the genetic association between A-ENA and cyt/cry-toxins in Bti, we focused on the A-ENA prevalence across the \\u003cem\\u003eBacillus cereus\\u003c/em\\u003e group \\u003cem\\u003esensu lato\\u003c/em\\u003e. A-ENA is found in 496 of 5976 (8%) tested genomes of the Btyper database, and although found in all GTDB \\u003cem\\u003eBacillus\\u003c/em\\u003e species, it is predominantly found in \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e (156 genomes), \\u003cem\\u003eB. mycoides\\u003c/em\\u003e (124 genomes), \\u003cem\\u003eB. wiedmanni\\u003c/em\\u003e (96 genomes) and to a lesser extent in B. \\u003cem\\u003eanthracis/paranthracis/cereus\\u003c/em\\u003e (39 genomes) (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e). Moreover, out of the 496 genomes containing A-ENA, 351 (71%) genomes contained one or more type of Bt-toxin (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e outer ring). If we compare that to the Bt-toxin prevalence across the Btyper database -only 1071 genomes (18%) were Bt positive, showing that A-ENA and Bt are positively correlated, reinforcing the notion that the main biological function of A-ENA is related to the presence of toxins, and by extension parasporal bodies.\\u003c/p\\u003e\\n\\u003cp\\u003eInterestingly, 350 genomes (70%) carried more than 2 A-ENA-like genes (as is the case for Bti) with 83 (17%) genomes harboring 5 or more homologues (e.g. the \\u003cem\\u003eBacillus cereus sensu stricto\\u003c/em\\u003e strain AFS036381 has 19 A-ENA-like genes) (Extended Data Fig.\\u0026nbsp;7a). We performed a gene cluster analysis and found that many A-ENA-like genes are organized in (putative) operons and/or gene clusters (Extended Data Fig.\\u0026nbsp;7b). Looking at these gene clusters in further detail, we discern two types of A-ENA homologues, (i) those constructed of a single A-ENA domain (shown in orange) and (ii) those carrying an additional domain (A-ENA fusion, shown in blue). After performing an interproscan analysis on the dataset of A-ENA homologue sequences, we identify 5 classes of passenger domain types (Extended Data Fig.\\u0026nbsp;7c,d). The gene architecture of the most abundant class (type 1) consists of an A-ENA domain, followed by a collagen-like region and a C1q domain. Other less frequently observed passenger domain types are SdrD_B (PF17210) / OmcB (PF01345) or 4Fe-4S cluster domain (PF13370). We recently showed that collagen-like proteins can form trimeric tip fibrillae (ruffles) on the termini of S- and L-type ENA fibers found on \\u003cem\\u003eBacillus cereus\\u003c/em\\u003e and showed that these ruffles are involved in spore-spore clustering (Pradhan et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2021\\u003c/span\\u003e; Sleutel et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e)). We reasoned that a similar mechanism could be at play for the A-ENA/C1q homologues and modelled a heterohexamer of WP_277491740.1 (putative A-ENA ruffle) of a \\u003cem\\u003eBacillus cereus\\u003c/em\\u003e strain SIBC65 isolated from raw bulk milk in complex with WP_000075910.1 (single A-ENA domain) which is the presumed major subunit of an A-ENA type pilus (Extended Data Fig.\\u0026nbsp;8a,b). AF3 predicts an A-ENA / ruffle complex wherein the A-ENA domains of the ruffle dock onto the single domain A-ENA protomers thereby mimicking the homotypic A-ENA contacts in the fiber (Extended Data Fig.\\u0026nbsp;8c). The residues involved in IPB formation are conserved in both WP_277491740.1 and WP_000075910.1 (Extended Data Fig.\\u0026nbsp;8d), suggesting that the ruffle is seamlessly integrated into the fiber or covalently tethered to the terminus of the A-ENA fiber (Extended Data Fig.\\u0026nbsp;8e). Clues to a putative function of A-ENA ruffles can be found for \\u003cem\\u003ePasteuria ramosa\\u003c/em\\u003e, an endoparasitic bacterium of aquatic microcrustaceans (e.g. water flea \\u003cem\\u003eDaphnia\\u003c/em\\u003e). Recently, Huessy \\u003cem\\u003eet al\\u003c/em\\u003e demonstrated that the collagen-like protein Pcl7 of \\u003cem\\u003eP. ramosa\\u003c/em\\u003e functions as an adhesin to mediate spore attachment to the epithelium of the oesophagus of the host \\u003cem\\u003eDaphnia\\u003c/em\\u003e (Huessy et al., \\u003cspan class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e). Here we show that although Pcl7 is a CLP lacking an N-terminal A-ENA domain, it is embedded in a 3 gene operon flanked by two additional CLPs (ORFs 01218, 01220) one of which (01218) has an N-terminal A-ENA domain (Extended Data Fig.\\u0026nbsp;8f,g). We also identify another 3 gene cluster (ORFs 00694, 00695 and 00696) which encode for 3 A-ENA-like genes. AF3 predicts a heterohexameric structure of the three CLPs in complex with the 3 putative A-ENA major subunits that is topologically similar to the ruffle described above (Extended Data Fig.\\u0026nbsp;8h,i).\\u003c/p\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eThe natural entomopathogenic properties of \\u003cem\\u003eBacillus thuringiensis\\u003c/em\\u003e (Bt) strains have led to world-wide adoption of Bt as an industrial biopesticide and vector control agent. During the sporulation process, Bt secretes parasporal bodies (PSB) that are crystalline agglomerates of different insecticidal pro-toxins that synergistically target specific insect groups. Typically, PSBs are present in the extracellular milieu, and are therefore subject to environmental dispersal and separation from the infectious body, i.e. the spore. Although some strains such as \\u003cem\\u003eBacillus sphaericus (\\u003c/em\\u003eKalfon et al., \\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e1984\\u003c/span\\u003e) have solved this potential issue by producing PSBs that are embedded within the interstitial space between the spore body and the exosporium, most commonly used commercial strains such as \\u003cem\\u003eB.t.\\u003c/em\\u003e subsp Kurstaki and \\u003cem\\u003eB.t.\\u003c/em\\u003e subsp Israelensis produce extrasporal PSBs, raising the question how simultaneous uptake of spore and PSB are ensured during infection. Co-transmission of spores and PSBs is a likely requirement for efficient establishment of the natural Bt lifecycle considering cytotoxicity is predominantly mediated by the PSB, giving ingested spores the opportunity to germinate and proliferate on the nutrients supplied by the insect remains. Insect exposure to PSBs alone will lead to insect death but no subsequent proliferation and dispersion of Bt titers, whereas ingestion of PSB deprived spore suspensions will have minimal cytotoxic effect.\\u003c/p\\u003e \\u003cp\\u003eHere we showed that Bti has solved this peril by producing a proteinaceous extracellular matrix that functions as a molecular glue holding spores and PSBs in tight unison within a spore biofilms (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003e). A robust spore-PSB connection is particularly logical for Bti given the aquatic lifestyle of its natural hosts, e.g. mosquito larvae. The absence of such a connection is likely to lead to separation of both entities in a liquid environment due to dilution effects. This is reflected in the difference between the survival rates of \\u003cem\\u003eChironomus aprilinus\\u003c/em\\u003e larvae exposed to WT Bti and Bti \\u003cem\\u003eDa-ena\\u003c/em\\u003e. For the latter we showed that the A-ENA matrix is absent, leading to poor retention of PSBs during spore purification and a marked increase of the \\u003cem\\u003eC. aprilinus\\u003c/em\\u003e survival rate on days 2\\u0026ndash;3 (50%,30%) compared to wild-type spores (30%,20%). Note that our assay probes for the mortality during initial exposure and is therefore not sensitive to secondary, long-term effects such as further Bti proliferation and dissemination after the first infection cycle, which would occur in a natural setting (e.g. large body of water). Similarly, the effects of the physical spore-PSB coupling will be more pronounced in turbulent and/or larger reaction volumes (here we work in 3mL) where the dilution factor and subsequent probability of spore-PSB decoupling would be larger for the A-ENA mutant strain.\\u003c/p\\u003e \\u003cp\\u003eIntrigued by the results for Bti, we turned to Btk and found that its spore biofilms are devoid of A-ENA, with a clear disconnect between the spores and the bipyramidal toxin crystals. Recombinant expression of A-ENA in Btk led to the instalment of an A-ENA based matrix (similar to Bti) capturing spores and toxin crystals into a cohesive network. This translated into a decreased survival of \\u003cem\\u003eTrichoplusia ni\\u003c/em\\u003e when challenged with Btk spores that either recombinantly expressed A-ENA or were mixed with exogenously added A-ENA fibers purified from \\u003cem\\u003eE. coli\\u003c/em\\u003e. Given the low sequence identity between the toxins produced by Bti and Btk, it would seem that aspecific interactions drive the binding of A-ENA fibers to the surface of the Btk crystals. This points towards a general promiscuity of A-ENA mediated interactions with a diverse range of surfaces (spore, PSB) amplified through avidity (display density of patchy, sticky surfaces on A-ENA fibers combined with fiber super bundling). It is tantalizing to speculate that A-ENA could lead to a similar gain of function in other PSB producing \\u003cem\\u003eBacillus thuringiensis\\u003c/em\\u003e strains devoid of A-ENA, and function as a generalist virulence factor.\\u003c/p\\u003e \\u003cp\\u003eA phylogenetic analysis revealed that A-ENA orthologues are predominantly found in entomopathogenic \\u003cem\\u003eBacillus\\u003c/em\\u003e strains, and correlates with the presence of toxin genes in the genome. This tentatively generalizes the PSB-retention function seen for Bti across the group of A-ENA carrying Bt strains. Interestingly, a Pfam analysis also showed that A-ENA orthologues also exist as natural fusion products consisting of an N-terminal A-ENA domain, and a C-terminal collagen-like / C1q region. We recently showed that L-type ENAs tethered to the exosporium of a clinical \\u003cem\\u003eBacillus paranthracis\\u003c/em\\u003e isolate are decorated at their distal terminus with a tip fibrillum that is encoded by the collagen-like protein L-BclA, which also carries a C-terminal C1q domain (Sleutel et al., \\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e). Moreover, we showed that L-BclA mediates spore-spore aggregation by binding to an as of yet unidentified receptor on the spore surface. Here, through AF3 modelling we show that genes with an A-ENA/collagen-like/C1q architecture might serve similar roles as L-BclA, in that the A-ENA moiety mediates insertion in or at an A-ENA terminus, while the C-terminal C1q domain represents the functional moiety, e.g. host-pathogen interactions. This indeed appears to be the case for the collagen-like protein Pcl7 of \\u003cem\\u003ePasteuria ramosa\\u003c/em\\u003e which mediates spore attachment to the oesophagus of its host \\u003cem\\u003eDaphnia (\\u003c/em\\u003eHuessy et al., \\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e). Pcl7 is integrated into a triplet gene cluster flanked by two other collagen-like proteins, one of which carries an A-ENA motif. AF3 confirms the predicted \\u003cem\\u003eP. ramosa\\u003c/em\\u003e A-ENA fiber / collagen fibrillum complex. Taken altogether, it seems plausible that the (short) hair-like appendages (Duneau et al., \\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e) that coat the surface of \\u003cem\\u003eP. ramosa\\u003c/em\\u003e is (in part) formed by A-ENA-like fibrils decorated with adhesive Pcl7 tip fibrillae. Based on our analysis, this might be a recurring feature across the family of A-BclA (i.e. A-ENA\\u0026thinsp;+\\u0026thinsp;collagen/C1q) carrying strains.\\u003c/p\\u003e \\u003cp\\u003eNext to the biological implications, A-ENA fibers have remarkable material properties. Although the chemical and physical robustness of spore derived self-assembly structures has been demonstrated for numerous systems (Pradhan et al., \\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e2021\\u003c/span\\u003e) (Henriques and Moran, \\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e2007\\u003c/span\\u003e; Sleutel et al., \\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e; Terry et al., \\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e2017\\u003c/span\\u003e), those structures are either susceptible to depolymerization under reducing conditions or unfold under extreme conditions (e.g. 100% (w/v) formic acid for L-ENA). For A-ENA, we could not identify an unfolding or depolymerization transition point, making it one of the most resilient (natural) protein structures studied to date \\u0026ndash; we hypothesize that for A-ENA the limiting stability will be the strength of the peptide bond. We attribute this remarkable stability to the large number of covalent crosslinks combined with the compact size of the protomers. Normalized per molecular weight, this translates to 0.9 (non-reducible) covalent bonds per kilodalton for A-ENA. Other ultra-stable systems have significantly lower cross-linking densities, e.g. 0.2 IPBs / kDa for the Spy filaments of \\u003cem\\u003eStreptococcus pyogenes\\u003c/em\\u003e (Kang and Baker, \\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e) and 0.26 and 0.16 S-S bonds / kDa for the L-type and S-type ENAs, respectively of \\u003cem\\u003eBacillus paranthracis\\u003c/em\\u003e. The stability of A-ENA might be a biological requirement that follows from the environmental conditions that spores are exposed to. Spores are non-motile dormant structures, subject to a wide range of conditions as a result of environmental dispersion, that may include desiccation, UV radiation, high temperatures, proteolytic attack, and shear stress. By producing a durable net that guarantees co-migration of the PSBs together with the spore right up until the moment of ingestion by the host, Bti has developed a way to optimize for virulence and with it, the opportunity for further propagation.\\u003c/p\\u003e \\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eResource availability\\u003c/h2\\u003e \\u003cp\\u003eCryoEM maps and coordinates for A-ENA, recA-ENA and recA-ENA1 have been deposited at the EMDB and PDB with following, respective, accession numbers: EMD-52871 and PDB-ID 9IHH; EMD-52872 and PDB-ID 9IHJ; EMD-52882 and PDB-ID 9Q82.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"Methods\",\"content\":\"\\u003cp\\u003e\\u003cb\\u003eCryo-EM sample preparation, data collection and helical reconstruction\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003cem\\u003eBacillus thuringiensis\\u003c/em\\u003e serovar \\u003cem\\u003eisraelensis\\u003c/em\\u003e was inoculated in 10mL LB and grown over night at 37\\u0026deg;C under shaking conditions. Next, 500\\u0026micro;l of the preculture was spread out onto a square LB agar plate and incubated for 72h at 30\\u0026deg;C under stagnant conditions. The resulting confluent growth was harvested and resuspended in miliQ, centrifuged for 5min at 20.000 rcf, the supernatant was discarded and the spore pellet was resuspended in miliQ. The resulting washed spore suspension was used as is for cryoEM grid preparation.\\u003c/p\\u003e\\u003cp\\u003eHigh-resolution cryo-EM datasets were collected using Quantifoil\\u0026trade; R2/1 300 copper mesh holey carbon grids. Grids were glow-discharged at 5 mA plasma current for 30 s in an ELMO (Agar Scientific) glow-discharger. A Gatan CP3 cryo-plunger set at \\u0026minus;\\u0026thinsp;176\\u0026deg;C and relative humidity of 90% was used to prepare the cryo-samples. 3 \\u0026micro;L sample of the spore suspension or of the purified recombinant fibers was applied on the holey grid and incubated for 60 seconds. The sample was double sided blotted using Whatman type 2 paper for 3 s and plunge-frozen into precooled liquid ethane at \\u0026minus;\\u0026thinsp;176\\u0026deg;C. High-resolution movies were recorded at 300 kV on a JEOL Cryoarm300 microscope equipped with an in-column Ω energy filter (operated at slit width of 14 eV) automated with SerialEM 3.0.850. The movies were captured with a K3 direct electron detector run in counting mode at a magnification of 60K with a calibrated pixel size of 0.71 \\u0026Aring;/pix, and a total exposure of 60\\u0026nbsp;e/\\u0026Aring;\\u003csup\\u003e2\\u003c/sup\\u003e over 60 frames. A total of 2817, 3771 and 4160 movies were collected on the \\u003cem\\u003eex vivo\\u003c/em\\u003e fibers, the recombinant A-ENA and the recombinant A-ENA1 fibers, respectively. Raw tiffs were imported into Cryosparc v4.5.3 (Punjani et al., \\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e2017\\u003c/span\\u003e) for further processing. Movies were motion-corrected using Patch Motion Correction and defocus values were determined using Patch CTF. Exposures were curated and fiber segments were picked using the filament tracer and extracted with a box size of 150 x 150 pixels at a pixel resolution of 1.42 \\u0026Aring;/pix. After several rounds of 2D classification, curated particles were used in a helical refinement job using initial estimates for the helical rise and twist values derived from the 2D class average images. After an initial round of helical refinement, particles were recentred and re-extracted at 0.766 \\u0026Aring;/pix (300 x 300pix) and used as input for a second round of helical refinement, followed by local and global CTF refinement, and a final helical refinement. The resulting high-resolution volume and particle stack were used for reference-based motion correction after particle duplicate removal (310k particles) followed by a final helical refinement. Reported map statistics were based on these maps for the recombinant A-ENA and A-ENA1 structures. For the \\u003cem\\u003eex vivo\\u003c/em\\u003e dataset, we additionally performed reference based motion correction followed by one more round of helical refinement. Figures were created using maps that were sharpened by cryosparc using a B-factor of -77.6\\u0026Aring;\\u003csup\\u003e2\\u003c/sup\\u003e, -116.9 \\u0026Aring;\\u003csup\\u003e2\\u003c/sup\\u003e, -114.5\\u0026Aring;\\u003csup\\u003e2\\u003c/sup\\u003e for \\u003cem\\u003eex vivo\\u003c/em\\u003e, rec A-ENA and rec A-ENA1, respectively, automatically determined by cryosparc.\\u003c/p\\u003e\\u003cp\\u003eFor \\u003cem\\u003eex vivo\\u003c/em\\u003e A-ENA, the initial atomic model was built either manually \\u003cem\\u003ede novo\\u003c/em\\u003e or using ModelAngelo (Jamali et al., \\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e) without providing an input sequence (build_no_seq). PSI-blast analysis using the sequence from the built model as a query, identified Q8KNV8 (A-ENA) as the top ranking hit. A second round of model building was performed with ModelAngelo now using the A-ENA sequence as input (build) which was iteratively refined in real-space using Coot (Emsley et al., \\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e2010\\u003c/span\\u003e) and Phenix (Liebschner et al., \\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e2019\\u003c/span\\u003e). For model building in the recombinant A-ENA map, a single protomer of the \\u003cem\\u003eex vivo\\u003c/em\\u003e structure was docked and iteratively real-space refined in Coot. Next, the map was populated with remaining chains by applying helical symmetry in ChimeraX (sym command) and refined in Phenix. For model building in the recombinant A-ENA1 map, we used ModelAngelo (with A-ENA1 sequence provided) and refined the model as outlined above. MolProbity (Williams et al., \\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e2018\\u003c/span\\u003e) was used to evaluate the quality of all the filament models. The refinement statistics are shown in Supplementary Table\\u0026nbsp;2. Determination of the absolute helical hands of the maps were determined from the hand of an a-helix. Alternatively, we also performed ModelAngelo test runs on either hand which would lead to either a highly segmented model consisting of many chains, or a filament model wherein chains would fully map onto individual protomers The \\u0026lsquo;ModelAngelo\\u0026rsquo; hand assignment agreed with hand-assignment following the handedness of the a-helices.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eBacterial strains and growth conditions.\\u003c/b\\u003e\\u003cem\\u003eEscherichia coli\\u003c/em\\u003e Stellar\\u0026trade; was used for cloning and was grown in Luria-Bertani (LB) broth or agar plates at 37\\u0026deg;C supplemented with 50 \\u0026micro;g/ml kanamycin or 100 \\u0026micro;g/ml ampicillin when required. For protein production, \\u003cem\\u003eE. coli\\u003c/em\\u003e BL21 (DE3) was grown in TB broth supplemented with 100 \\u0026micro;g/ml ampicillin and placed at 30\\u0026deg;C for A-ENA expression after induction. \\u003cem\\u003eE. coli\\u003c/em\\u003e Dam-/Dcm- (New England Biolabs) was used to obtain unmethylated plasmids in order to increase the efficiency of transformation in Bt. Bti and Btk were grown in LB at 30\\u0026deg;C with 150 rpm shaking and supplemented with 25 \\u0026micro;g/ml kanamycin when required. For sporulation, 1 ml of overnight culture was plated on LB media and placed at 30\\u0026deg;C for a week. Spores were scrapped and diluted in miliQ water and stored at 4\\u0026deg;C. The strains used in this study are listed in Supplementary Table\\u0026nbsp;1.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eCloning for recombinant protein expression in\\u003c/b\\u003e \\u003cb\\u003eE. coli\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eCloning of untagged A-ENA, A-ENA1 and variants were produced in this study with the purpose of cytoplasmic overexpression in \\u003cem\\u003eE. coli\\u003c/em\\u003e. Water-boiled colonies of Bti were used as a source of gDNA. As a general cloning strategy, all PCR fragments containing the coding sequence were cloned into a linearized pASK-IBA3plus vector (using primers 321 and 322) by Gibson assembly using NEBuilder HiFi DNA Assembly Cloning Kit (New England Biolabs).\\u003c/p\\u003e\\u003cp\\u003eThe cloning of mCherry:SnoopCatcher (plasmid A185) was carried out using Gibson assembly. The \\u003cem\\u003emCherry\\u003c/em\\u003e gene was amplified from a plasmid containing \\u003cem\\u003emCherry\\u003c/em\\u003e in the lab collection, using primers p693 and p694. The \\u003cem\\u003eSnoopCatcher\\u003c/em\\u003e fragment was synthetically ordered as a GeneBlock from IDT. Both fragments were subsequently cloned into the linearized pASK-IBA3plus vector. Cloning of sfGFP:SpyCatcher (plasmid A142) was done by amplifying sfGFP from a plasmid containing \\u003cem\\u003esfGFP\\u003c/em\\u003e in the lab collection, using primers p503 and p504. The \\u003cem\\u003eSpyCatcher\\u003c/em\\u003e fragment was synthetically ordered as a GeneBlock from IDT. Both fragments were subsequently cloned into the linearized pASK-IBA3plus vector. All primers and plasmids used in this study are listed in Supplementary Table\\u0026nbsp;1.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eCloning for\\u003c/b\\u003e \\u003cb\\u003ea-ena-STOP\\u003c/b\\u003e \\u003cb\\u003eand\\u003c/b\\u003e \\u003cb\\u003eΔa-ena1\\u003c/b\\u003e \\u003cb\\u003eusing CRISPR/Cas9 based deletion plasmid\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003ePlasmid A266 and A190 were used to knockout \\u003cem\\u003ea-ena\\u003c/em\\u003e and \\u003cem\\u003ea-ena1\\u003c/em\\u003e respectively from \\u003cem\\u003eB. thuringiensis serovar isrealensis\\u003c/em\\u003e. Both plasmids are derived from the plasmid pJOE8999 (Altenbuchner, \\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e2016\\u003c/span\\u003e) purchased from Bacillus Genetic Stock Center (BGSC). Oligonucleotides for sgRNA construction were designed using the online sgRNA design tool Cas-Designer (Park et al., \\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e2015\\u003c/span\\u003e). First, pJOE8999 was linearized with primers p551\\u0026thinsp;+\\u0026thinsp;p552. A266 homology regions were cloned with p588\\u0026thinsp;+\\u0026thinsp;p591 and three STOP codon mutations were introduced with oligos p961\\u0026thinsp;+\\u0026thinsp;p962 whereas p709\\u0026thinsp;+\\u0026thinsp;p710 and p711\\u0026thinsp;+\\u0026thinsp;p712 were used to construct the homology regions of A-ENA1 (A190). In both cases, we used NEBuilder HiFi DNA Assembly Cloning Kit (New England Biolabs) to insert homology regions. Sequencing and insertion were verified with the pair of oligos p525\\u0026thinsp;+\\u0026thinsp;p526. Insertion of sgRNA was done by PCR using oligos p594\\u0026thinsp;+\\u0026thinsp;p595 (for A266) and p713\\u0026thinsp;+\\u0026thinsp;p714 (for A190). Primers p531\\u0026thinsp;+\\u0026thinsp;p579 were used to sequence sgRNA region. Verification of the A-ENA STOP mutant was done by PCR and sequencing using oligos p963\\u0026thinsp;+\\u0026thinsp;p651 and p964 respectively. Verification of the A-ENA1 mutant was performed by PCR with oligos p717\\u0026thinsp;+\\u0026thinsp;p718. Plasmids used in this study are listed in the Supplementary Table\\u0026nbsp;1.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eConstruction and isolation of\\u003c/b\\u003e \\u003cb\\u003eB. thuringiensis serovar isrealensis\\u003c/b\\u003e \\u003cb\\u003emutants\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eThe recombinant pJOE8999-based plasmids A266 and A190 described above, were introduced by electroporation into \\u003cem\\u003eB. thuringiensis serovar isrealensis\\u003c/em\\u003e competent cells. Briefly, \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e cells were made electrocompetent by growing in BHI\\u0026thinsp;+\\u0026thinsp;0.5% glycerol at 37 \\u0026ordm;C until OD\\u003csub\\u003e600\\u003c/sub\\u003e reached 0.6. Cells were harvested by centrifugation (5 min at 15 000 xg). Cells were washed 3 times with ice-cold sterile electroporation buffer (1 mM HEPES, 10% glycerol, pH 7.0) and flash-frozen in liquid nitrogen. Up to 1\\u0026micro;g of unmethylated plasmid was transformed into 200 \\u0026micro;l of competent cells using a 0.2 cm gap cuvette (Biorad) and electroporated using the MicroPulser Electroporator device (Biorad). Cells were transferred into 1ml of recovery buffer (BHI with 10% glycerol, 0.4% glucose, and 10 mM MgCl\\u003csub\\u003e2\\u003c/sub\\u003e) for 1h at 30 \\u0026ordm;C before plating in a 25 \\u0026micro;g/ml kanamycin plate. Next day, a single colony was inoculated into 10 ml of liquid BHI supplemented with 25 \\u0026micro;g/ml kanamycin and incubated with 180 rpm shaking for 3h at 37 \\u0026ordm;C. Then, Mannose (Sigma) was added to a final concentration of 0.4% w/v in order to induce the expression of the Cas9 nuclease. After 3 hours, serial dilution of the culture was plated on LB agar plates with 25 \\u0026micro;g/ml kanamycin\\u0026thinsp;+\\u0026thinsp;0.4% w/v Mannose and incubated at 37 \\u0026ordm;C overnight. Screening of the knockout was done using colony PCR, WT \\u003cem\\u003eB. thuringiensis serovar isrealensis\\u003c/em\\u003e was used as a control. Positive colonies were sequenced-verified using the same oligos (Eurofins). To curate the pJOE8999-based plasmid, mutant strains were passaged three times in the absence of antibiotics and plate in LB plates at 42 \\u0026ordm;C overnight. The absence of the plasmid was checked by striking the same colony in LB plates with or without Kanamycin. Strains used in this study are listed in Supplementary Table\\u0026nbsp;1.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eCloning for recombinant protein expression in\\u003c/b\\u003e \\u003cb\\u003eB. thuringiensis\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eFor ectopic recombinant expression of the different constructs in \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e we modified the pJOE8999 plasmid (Altenbuchner, \\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e2016\\u003c/span\\u003e) by removal of the coding sequence for the Cas9 nuclease and its promoter by PCR using oligos p622\\u0026thinsp;+\\u0026thinsp;p623, resulting in the \\u0026lsquo;pAS\\u0026rsquo; shuttle expression vector. Complementation of \\u003cem\\u003ea-ena stop\\u003c/em\\u003e and \\u003cem\\u003ea-ena1\\u003c/em\\u003e was achieved by expressing these constructs in the pAS vector. the \\u003cem\\u003ea-ena\\u003c/em\\u003e region, including its promoter, ORF, and terminator, was amplified using primers p683 and p684. Colony PCR and sequencing (using primers p631 and p671) confirmed successful insertion, resulting in plasmid A176. Plasmid A176 was subsequently used to introduce the \\u003cem\\u003eSpyTAG\\u003c/em\\u003esequence via primers p681 and p682, yielding plasmid A180.\\u003c/p\\u003e\\u003cp\\u003eFor \\u003cem\\u003ea-ena1\\u003c/em\\u003e, the region was amplified using primers p695 and p696, generating plasmid A186. This plasmid was then used as a template to introduce the \\u003cem\\u003eSnoopTAG\\u003c/em\\u003e sequence using primers p697 and p698, resulting in plasmid A187.\\u003c/p\\u003e\\u003cp\\u003eTo construct a dual-expression vector for \\u003cem\\u003ea-ena\\u003c/em\\u003e\\u003csup\\u003e\\u003cem\\u003espytag\\u003c/em\\u003e\\u003c/sup\\u003e and \\u003cem\\u003ea-ena1\\u003c/em\\u003e\\u003csup\\u003e\\u003cem\\u003esnooptag\\u003c/em\\u003e\\u003c/sup\\u003e under their native promoters, plasmid A180 was linearized using primers p699 and p700. The \\u003cem\\u003ea-ena\\u003c/em\\u003e\\u003csup\\u003e\\u003cem\\u003esnooptag\\u003c/em\\u003e\\u003c/sup\\u003e fragment, amplified from plasmid A187 using primers p701 and p702, was then cloned into the linearized A180 backbone. This final step produced plasmid A188. To monitor A-ENA expression, we replaced the \\u003cem\\u003ea-ena\\u003c/em\\u003e coding sequence (CDS) with \\u003cem\\u003emCherry\\u003c/em\\u003e. This was achieved by linearizing plasmid A176 using primers p737 and p738, followed by cloning the \\u003cem\\u003emCherry\\u003c/em\\u003e CDS amplified with primers p714 and p742 using plasmid A185 as the template. This process resulted in the assembly of plasmid A197. Plasmids generated are listed in the Supplementary Table\\u0026nbsp;1.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eProtein expression and purification\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003esfGFP:spyCatcher and mCherry:Snoopcatcher were expressed in \\u003cem\\u003eE.coli\\u003c/em\\u003e BL21 (DE3) in Terrific Broth (TB) supplemented with 100 \\u0026micro;g/mL of Ampicillin at 37\\u0026deg;C and induced with 200 \\u0026micro;g/L anhydrotetracycline when OD\\u003csub\\u003e600\\u003c/sub\\u003e reached 0.5. At this point, the temperature was set at 23\\u0026deg;C and cells were left to express overnight. Next day, cells were harvested by centrifugation (20 min at 5,000g) and pellets were kept at -20\\u0026deg;C. Frozen pellets were resuspended in 100 ml of lysis buffer (50 mM Hepes pH 8, 300 mM NaCl, 1 mM MgCl2, DNase, lysozyme and EDTA-free protease inhibitor cocktails (ROCHE)) and lysed by sonication at 4\\u0026deg;C. The lysate was centrifuged for 25 min at 30000g and 4\\u0026deg;C. The cleared supernatant was loaded onto a 5ml Ni-NTA affinity chromatography column (HisTrap FF crude, GE Healthcare). The column was washed with 5 column volumes with Buffer A (300 mM NaCl, 10 mM Hepes pH 8, 10 mM Imidazole) and finally, His-tagged proteins were eluted with a linear gradient of Buffer B (300 mM NaCl, 10 mM Hepes pH8, 0.5 M Imidazole). Eluted protein was concentrated and loaded onto a Superdex 200 16/60 size-exclusion column (GE Life Sciences) that was equilibrated with SEC buffer (25 mM Hepes pH 8 and 150 mM NaCl) at 4\\u0026deg;C. The peak corresponding to the protein was concentrated, flash-frozen in small aliquots in liquid nitrogen and stored at \\u0026minus;\\u0026thinsp;80\\u0026deg;C. After purification, the samples were run on a sodium dodecyl sulfate\\u0026ndash;polyacrylamide gel electrophoresis (SDS-PAGE) to evaluate their purity.\\u003c/p\\u003e\\u003cp\\u003eA-ENA and A-ENA1 were expressed in \\u003cem\\u003eE. coli\\u003c/em\\u003e BL21 (DE3) in Terrific Broth (TB) supplemented with 100 \\u0026micro;g/mL of Ampicillin at 37\\u0026deg;C and induced with 200 \\u0026micro;g/L anhydrotetracycline when OD\\u003csub\\u003e600\\u003c/sub\\u003e reached 0.5. At this point, the temperature was set at 18\\u0026deg;C and cells were left to express overnight. Cells were harvested by centrifugation (20 min at 5,000g) and incubated with A-ENA lysis buffer (25 mM Hepes pH8, 2.5 mM EDTA, 250 mM NaCl and 20 mg of Lysozyme) at 37\\u0026deg;C for 2 hours under shaking conditions. Next, 1% final SDS concentration was added and the resuspended pellet was boiled for 30 min. A-ENA fibers were pelleted down by centrifugation at 30,000g for 45 minutes. Pellet was resuspended with a homogenizer in 30 ml deionized H\\u003csub\\u003e2\\u003c/sub\\u003eO and centrifuged for 45 minutes at 30,000g. This washing step was repeated 4 times. After the last centrifugation step, the pellet was resuspended with 3 ml of deionized H\\u003csub\\u003e2\\u003c/sub\\u003eO.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eSpore preparation and Fractionation\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eSporulation of Bti and Btk was triggered by nutrient exhaustion on LB-agar plates supplemented with 25 \\u0026micro;g/mL of kanamycin when required. A single colony was inoculated in LB media (supplemented with 25 \\u0026micro;g/mL of kanamycin when required) and grown overnight at 30\\u0026deg;C. Next day, 200 \\u0026micro;l of the cell suspension was plated on 245 x 245 mm LB plates (with or without antibiotic). Plates were incubated at 30\\u0026deg;C for 7 days under aerobic conditions. Spores were harvested by scraping the LB plates with a cell scraper and resuspended in deionized H\\u003csub\\u003e2\\u003c/sub\\u003eO, spore stock was stored at 4\\u0026deg;C. To fractionate Bti and Btk spores, the spore suspension was gently vortexed and then a discontinuous Histodenz gradient was prepared by layering 500 \\u0026micro;l of 40% (w/v) Histodenz over 500 \\u0026micro;l of 60% (w/v) Histodenz in 2 mL microcentrifuge tubes. A total of 500 \\u0026micro;l of spore stock (OD₆₀₀ = 42) was carefully layered on top. Two tubes were prepared per strain: one for SDS-PAGE analysis and the other for optical density (OD) measurements and imaging. Samples were centrifuged at 10,000 \\u0026times; g or 100 \\u0026times; g for 1 hour using a tabletop swinging-bucket rotor (S-24-11-AR, Eppendorf).\\u003c/p\\u003e\\u003cp\\u003eAfter centrifugation, three fractions (supernatant (SN), 40%, and 60% )were carefully collected into separate tubes. Each fraction was washed 3 times by adding 500 \\u0026micro;l of water, vortexed, and centrifuged at 12,000 \\u0026times; g for 5 minutes. The supernatant was removed, and the pellet was resuspended in 100 \\u0026micro;l of DX buffer (10 mM CHES pH 9.6, 8M Urea, 1% SDS, 1\\u0026times; protease inhibitor tablet (Roche), 2 mM DTT). Samples were boiled for 10 minutes with securely closed lids and subsequently centrifuged at maximum speed for 10 minutes to pellet insoluble spores. A total of 50 \\u0026micro;l of the supernatant was mixed with 12.5 \\u0026micro;l of blue loading dye, and 5 \\u0026micro;l of the final mixture was loaded onto an SDS-PAGE gel for protein analysis.\\u003c/p\\u003e\\u003cp\\u003eFor OD measurements and imaging, the SN, 40% Histodenz, and 60% Histodenz fractions were collected into separate tubes. Each fraction was washed as descibed above. The supernatant was removed, and the pellet was resuspended in 100 \\u0026micro;l of water. OD₆₀₀ measurements were performed by diluting 50 \\u0026micro;l of the sample into 950 \\u0026micro;l of water (1:20 dilution) before spectrophotometric analysis.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003ePhase contrast and fluorescence microscopy\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eFor imaging non-labeled spores, a single colony of \\u003cem\\u003eBti\\u003c/em\\u003e or \\u003cem\\u003eBtk\\u003c/em\\u003e was inoculated into 10 mL of LB medium with the appropriate antibiotic when required and grown overnight at 30\\u0026deg;C. The next day, 250 \\u0026micro;L of the culture was plated onto LB agar, with or without antibiotics, and incubated for 5 days at 30\\u0026deg;C to allow sporulation. The resulting spore lawn was harvested and resuspended in 5 mL of water. A 3 \\u0026micro;L aliquot of the spore suspension was transferred onto an LB-agar strip for microscopy. For time-lapse experiments comparing sporulation, t\\u0026thinsp;=\\u0026thinsp;0 was defined as the moment cells were plated on the LB agar plate. At each time point, a sample was collected using a sterile loop, resuspended in 30 \\u0026micro;L of PBS, and imaged. For dual labeling with SpyCatcher:GFP and SnoopCatcher:mCherry, a strain expressing A-ENA:SpyTag and A-ENA1:SnoopTag under their native promoters was induced to sporulate as described above. After 5 days, the spore lawn was harvested, and the OD₆₀₀ was adjusted to 1. Both SpyCatcher:GFP and SnoopCatcher:mCherry were added to a final concentration of 50 \\u0026micro;M and incubated overnight with gentle rotation. The following day, the spore mixture was washed three times with PBS and mounted on a 1% agar slide for imaging. In all cases, Images were acquired in phase contrast and fluorescence mode using a Leica DMi8 inverted microscope (Leica) equipped with a 100\\u0026times;/1.32 oil objective (Leica).\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eImmunodetection\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eDue to the high degree of sequence identity of surface exposed residues between A-ENA and A-ENA1 we anticipated that it would not be feasible to raise polyclonal antibodies that would be specific to either A-ENA or A-ENA1. Given that our cryoEM analysis indicated that \\u003cem\\u003eex vivo\\u003c/em\\u003e fibers are predominantly composed of A-ENA, we opted to solely raise antisera against rec A-ENA. For this, a polyclonal mouse antiserum raised against purified rec-A-ENA fibers was ordered from Davids Biotechnologie GmbH (Regensburg, Germany). The following 63 day immunization schedule was followed: immunizations on days 0, 14, 28, 42 and 56. To detect A-ENA on spore samples, we diluted the spore suspension to OD\\u003csub\\u003e600\\u003c/sub\\u003e\\u0026thinsp;=\\u0026thinsp;1 and blotted 2 \\u0026micro;l onto a nitrocellulose membrane. After 5 minutes of drying at room temperature (RT), the membrane was blocked with 5% (w/v) skimmed milk for 30 minutes. The membrane was incubated with A-ENA antibody mix (dilution 1:1000 in TBS-Tween buffer (Tris-HCl pH8 10 mM; NaCl 150 mM; Tween20 0,05% (vol/vol)) and incubated at RT for 1h. After 3 washes of 5 minutes each, the membrane was incubated in the secondary antibody mix (dilution 1:1000) for an extra hour at RT. After 3 washes of 5 minutes each, the membrane was imaged with the LI-COR Odyssey M using the LI-COR Acquisition Software v 2.0.0.86.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eA-ENA stability assay\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eTo test the chemical stability of recombinant A-ENA fibers, fibers were harvested from a concentrated stock solution via centrifugation (20.000rcf, 1h, supernatans discarded). The resulting pellets after centrifugation of 100\\u0026micro;l aliquots were resuspended in various conditions: 100\\u0026micro;l 2% (w/v) SDS, 8M urea, 2M NaOH or 100 (v/v) formic acid and incubated as indicated on Extended Data Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e prior to nsTEM sample preparation. To test the physical robustness of recombinant A-ENA fibers, a 100\\u0026micro;l aliquot of the fiber stock solution was desiccated at 200\\u0026deg;C for 15min in an open glass vial in an oven. The resulting dried material was rehydrated in 100\\u0026micro;l miliQ prior to nsTEM imaging. Additionally, a 100\\u0026micro;l aliquot was autoclaved for 20min at 121\\u0026deg;C. Next, 3\\u0026micro;l aliquots of the treated samples were deposited onto Cu-mesh formvar grids, washed twice with 20\\u0026micro;l miliQ and incubated for 1min with 2% (w/v) uranyl acetate and blotted dry with Whatmann 2 filter paper. Micrographs were collected with a 120 kV JEOL 1400 microscope equipped with LaB6 filament and TVIPS F416 CCD camera operated at 60.000x magnification.\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eMonitoring the entomopathogenic activity of different Bt strains\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eTo evaluate the virulence of Bti strains either expressing or lacking A-ENA (wild-type vs. \\u003cem\\u003ea-ena\\u003c/em\\u003e knockout), we employed the insect model \\u003cem\\u003eChironomus aprilinus\\u003c/em\\u003e. Fresh \\u003cem\\u003eC. aprilinus\\u003c/em\\u003e larvae were procured from Zooschatz (Berlin, Germany) under batch number 1223153. Bti strains were cultured on LB plates at 30\\u0026deg;C for one week, after which the lawn of spores and PSB were harvested and resuspended in deionized water, and OD\\u003csub\\u003e600\\u003c/sub\\u003e was measured. To determine the mortality rate, a minimum of 39 \\u003cem\\u003eC. aprilinus\\u003c/em\\u003e larvae were included in each experimental group, with a final volume of 1 ml of sea salt solution supplemented with a vitamin premix provided by the supplier. Experiments were carried out in 12 well plates with a single individual per well. Spore suspension was added to reach a final OD\\u003csub\\u003e600\\u003c/sub\\u003e of 0.02. Phosphate saline buffer (PBS) was used as a negative control. Larvae survival was monitored daily over a 6-day period. The experiment was conducted in biological triplicates under room temperature conditions.\\u003c/p\\u003e\\u003cp\\u003eFor the virulence assessment of \\u003cem\\u003eBtk\\u003c/em\\u003e strains, we used the insect model \\u003cem\\u003eTrichoplusia ni\\u003c/em\\u003e. Larvae or eggs were obtained from Frontier Agricultural Sciences (Newark, USA). \\u003cem\\u003eBtk\\u003c/em\\u003e wild-type and A-ENA-expressing strains (\\u003cem\\u003epA-ENA\\u003c/em\\u003e) were sporulated on LB plates at 30\\u0026deg;C for one week. Spores and PSBs were harvested and processed as described for \\u003cem\\u003eBti\\u003c/em\\u003e strains. To determine the mortality rate, a minimum of 16 two week old \\u003cem\\u003eT. ni\\u003c/em\\u003e larvae per group were individually placed on 2.5 \\u0026times; 2.5 cm pieces of solid food provided by the supplier. Each food piece was spread with a 50 \\u0026micro;l aliquot of the harvested spore suspension at an OD₆₀₀ of 0.06. As a negative control, \\u003cem\\u003eBti\\u003c/em\\u003e (which has no toxic effect on \\u003cem\\u003eT. ni\\u003c/em\\u003e) was included. Larval survival was monitored daily for one week. For experiments involving recombinant A-ENA, it was added to \\u003cem\\u003eBtk\\u003c/em\\u003e WT samples at a final concentration of 4.18 mg/ml and incubated for at least 3 hours under slow rotation. As an additional control, purified recombinant A-ENA was tested at the same final concentration (4.18 mg/ml) under identical conditions. The study was performed in biological triplicates at room temperature\\u003c/p\\u003e\\u003cp\\u003e\\u003cb\\u003eSearch for orthologs and homologs of A-ENA\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003eTo probe for the presence of A-ENA across the Firmicutes phylum, all publicly available Genbank genomes (9880 genomes) belonging to the Bacillus/Clostridium group (taxid 1239; assembly-level complete) were downloaded from the NCBI database. Homo- and orthologs of A-ENA were searched in all assemblies with hmmsearch (Eddy, \\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e) (881 genomes) using a hidden Markov model that was generated with hmmbuild using a manually curated multiple sequence alignment of A-ENA sequences obtained from a blastp (Johnson et al., \\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e2008\\u003c/span\\u003e) search using Q8KNV8 as a query. For this, AlphaFold2 models (either retrieved from the AlpfaFold database (Varadi et al., \\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e) or predicted using localcolabfold (Mirdita et al., \\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e2022\\u003c/span\\u003e)) of the corresponding blast sequences were manually inspected and compared to the A-ENA cryoEM structure to remove any non-A-ENA sequences. The inclusion threshold for hmmsearch was set to an E-value of 1e-7. The phylogenetic tree of the Firmicutes phylum was generated using phyloTv2 (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://phylot.biobyte.de/\\u003c/span\\u003e\\u003cspan address=\\\"https://phylot.biobyte.de/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) using the NCBI taxonomy IDs of the corresponding assemblies as node identifiers, and imported into iTOL (Letunic and Bork, \\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e) for visualization using the online server (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://itol.embl.de/\\u003c/span\\u003e\\u003cspan address=\\\"https://itol.embl.de/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e).\\u003c/p\\u003e\\u003cp\\u003eFor the more detailed search of A-ENA homo- and orthologs across the \\u003cem\\u003eBacillus cereus\\u003c/em\\u003e group \\u003cem\\u003esensu lato\\u003c/em\\u003e, the BtyperDB v1 database (5976 genomes) was downloaded from \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.btyper.app/\\u003c/span\\u003e\\u003cspan address=\\\"https://www.btyper.app/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e, and hmmsearch was run against all assemblies with an E-value threshold of 1e-7 (496 genomes). The phylogenetic tree described in BtyperDB (Ramnath et al., \\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e2023\\u003c/span\\u003e) and the corresponding meta-data table was kindly provided by Laura M. Caroll. Tree Figures were generated using \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://microreact.org/\\u003c/span\\u003e\\u003cspan address=\\\"https://microreact.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e. Gene cluster analysis was performed using cblaster 1.3.18 (Gilchrist et al., \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e2020\\u003c/span\\u003e) using default settings. Protein domain organization of A-ENA homologs with C-terminal fusions was performed with Interproscan 5.67-99.0 (Jones et al., \\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e2014\\u003c/span\\u003e).\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eResource availability\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eCryoEM maps and coordinates for A-ENA, recA-ENA and recA-ENA1 have been deposited at the EMDB and PDB with following, respective, accession numbers:\\u0026nbsp;EMD-52871 and PDB-ID 9IHH;\\u0026nbsp;EMD-52872 and PDB-ID 9IHJ;\\u0026nbsp;EMD-52882 and PDB-ID 9Q82.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eWe thank Dirk Reiter and Marcus Fislage at the VIB-VUB Facility for Bio Electron Cryogenic Microscopy (BECM). We\\u0026rsquo;re thankful to Steven Janvier and Didier Vertommen for MS fingerprinting analysis. This work was funded by VIB. A.S was supported by the EMBO (ALTF-709-2021) and the Marie Skłodowska-Curie Actions (MSCA; SLYDIV project).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthor contributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eM.S. and H.R. designed the project. M.S. performed the cryoEM imaging, data processing and structure determinations. M.S. and A.S. performed nsTEM imaging. A.S. performed light microscopy and all cloning, biological and biochemical assays. M.S. and A.S. performed A-ENA purification. M.S. and H.R. wrote the manuscript with contributions by A.S.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eDeclaration of interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors are named as inventors on a patent application filed by VIB, related to the technological use of A-ENA nanofibers.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eSupplemental information\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eExtended Data Figs 1\\u0026ndash;8 and Supplementary Table 1-2\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\n \\u003cli\\u003eAltenbuchner, J. (2016). Editing of the Bacillus subtilis Genome by the CRISPR-Cas9 System. Appl Environ Microbiol\\u003cem\\u003e\\u0026nbsp;82\\u003c/em\\u003e, 5421-5427.\\u003c/li\\u003e\\n \\u003cli\\u003eArbour, C.A., Nagar, R., Bernstein, H.M., Ghosh, S., Al-Sammarraie, Y., Dorfmueller, H.C., Ferguson, M.A.J., Stanley-Wall, N.R., and Imperiali, B. (2023). Defining Early Steps in B. subtilis Biofilm Biosynthesis. bioRxiv.\\u003c/li\\u003e\\n \\u003cli\\u003eAronson, A.I., Beckman, W., and Dunn, P. (1986). Bacillus thuringiensis and related insect pathogens. Microbiol Rev\\u003cem\\u003e\\u0026nbsp;50\\u003c/em\\u003e, 1-24.\\u003c/li\\u003e\\n \\u003cli\\u003eBen-Dov, E., Nissan, G., Pelleg, N., Manasherob, R., Boussiba, S., and Zaritsky, A. (1999). Refined, circular restriction map of the Bacillus thuringiensis subsp. israelensis plasmid carrying the mosquito larvicidal genes. Plasmid\\u003cem\\u003e\\u0026nbsp;42\\u003c/em\\u003e, 186-191.\\u003c/li\\u003e\\n \\u003cli\\u003eBerry, C., O\\u0026apos;Neil, S., Ben-Dov, E., Jones, A.F., Murphy, L., Quail, M.A., Holden, M.T., Harris, D., Zaritsky, A., and Parkhill, J. (2002). Complete sequence and organization of pBtoxis, the toxin-coding plasmid of Bacillus thuringiensis subsp. israelensis. Appl Environ Microbiol\\u003cem\\u003e\\u0026nbsp;68\\u003c/em\\u003e, 5082-5095.\\u003c/li\\u003e\\n \\u003cli\\u003eBravo, A., Gill, S.S., and Soberon, M. (2007). Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control. Toxicon\\u003cem\\u003e\\u0026nbsp;49\\u003c/em\\u003e, 423-435.\\u003c/li\\u003e\\n \\u003cli\\u003eCharles, J.-F., and De Barjac, H. (1982). Sporulation et cristallogenese de Bacillus thuringiesnsis Var. Israelensis en microscopie electronique. Annales de Microbiologie\\u003cem\\u003e\\u0026nbsp;133\\u003c/em\\u003e.\\u003c/li\\u003e\\n \\u003cli\\u003eDuneau, D., Luijckx, P., Ben-Ami, F., Laforsch, C., and Ebert, D. (2011). Resolving the infection process reveals striking differences in the contribution of environment, genetics and phylogeny to host-parasite interactions. BMC Biol\\u003cem\\u003e\\u0026nbsp;9\\u003c/em\\u003e, 11.\\u003c/li\\u003e\\n \\u003cli\\u003eEddy, S.R. (2011). Accelerated Profile HMM Searches. PLoS Comput Biol\\u003cem\\u003e\\u0026nbsp;7\\u003c/em\\u003e, e1002195.\\u003c/li\\u003e\\n \\u003cli\\u003eEmsley, P., Lohkamp, B., Scott, W.G., and Cowtan, K. (2010). Features and development of Coot. Acta crystallographica Section D, Biological crystallography\\u003cem\\u003e\\u0026nbsp;66\\u003c/em\\u003e, 486-501.\\u003c/li\\u003e\\n \\u003cli\\u003eEstada, U., and Ferre, J. (1994). Binding of Insecticidal Crystal Proteins of Bacillus thuringiensis to the Midgut Brush Border of the Cabbage Looper, Trichoplusia ni (Hubner) (Lepidoptera: Noctuidae), and Selection for Resistance to One of the Crystal Proteins. Appl Environ Microbiol\\u003cem\\u003e\\u0026nbsp;60\\u003c/em\\u003e, 3840-3846.\\u003c/li\\u003e\\n \\u003cli\\u003eFederici, B.A., L\\u0026uuml;thy, P., and Ibarra, J.E. (1990). Parasporal Body of Bacillus thuringiensis israelensis. In Bacterial Control of Mosquitoes \\u0026amp; Black Flies: Biochemistry, Genetics \\u0026amp; Applications of Bacillus thuringiensis israelensis and Bacillus sphaericus, H. de Barjac, and D.J. Sutherland, eds. (Dordrecht: Springer Netherlands), pp. 16-44.\\u003c/li\\u003e\\n \\u003cli\\u003eGilchrist, C.L.M., Booth, T.J., and Chooi, Y.-H. (2020). cblaster: a remote search tool for rapid identification and visualisation of homologous gene clusters. bioRxiv, 2020.2011.2008.370601.\\u003c/li\\u003e\\n \\u003cli\\u003eGriffin, M., Casadio, R., and Bergamini, C.M. (2002). Transglutaminases: nature\\u0026apos;s biological glues. Biochem J\\u003cem\\u003e\\u0026nbsp;368\\u003c/em\\u003e, 377-396.\\u003c/li\\u003e\\n \\u003cli\\u003eHenriques, A.O., and Moran, C.P., Jr. (2007). Structure, assembly, and function of the spore surface layers. Annu Rev Microbiol\\u003cem\\u003e\\u0026nbsp;61\\u003c/em\\u003e, 555-588.\\u003c/li\\u003e\\n \\u003cli\\u003eHuessy, B., Bumann, D., and Ebert, D. (2024). Ectopical expression of bacterial collagen-like protein supports its role as adhesin in host-parasite coevolution. R Soc Open Sci\\u003cem\\u003e\\u0026nbsp;11\\u003c/em\\u003e, 231441.\\u003c/li\\u003e\\n \\u003cli\\u003eJamali, K., Kall, L., Zhang, R., Brown, A., Kimanius, D., and Scheres, S.H.W. (2024). Automated model building and protein identification in cryo-EM maps. Nature\\u003cem\\u003e\\u0026nbsp;628\\u003c/em\\u003e, 450-457.\\u003c/li\\u003e\\n \\u003cli\\u003eJansen, C.C., and Beebe, N.W. (2010). The dengue vector Aedes aegypti: what comes next. Microbes Infect\\u003cem\\u003e\\u0026nbsp;12\\u003c/em\\u003e, 272-279.\\u003c/li\\u003e\\n \\u003cli\\u003eJohnson, M., Zaretskaya, I., Raytselis, Y., Merezhuk, Y., McGinnis, S., and Madden, T.L. (2008). NCBI BLAST: a better web interface. Nucleic Acids Res\\u003cem\\u003e\\u0026nbsp;36\\u003c/em\\u003e, W5-9.\\u003c/li\\u003e\\n \\u003cli\\u003eJones, P., Binns, D., Chang, H.Y., Fraser, M., Li, W., McAnulla, C., McWilliam, H., Maslen, J., Mitchell, A., Nuka, G.\\u003cem\\u003e, et al.\\u003c/em\\u003e (2014). InterProScan 5: genome-scale protein function classification. Bioinformatics\\u003cem\\u003e\\u0026nbsp;30\\u003c/em\\u003e, 1236-1240.\\u003c/li\\u003e\\n \\u003cli\\u003eKalfon, A., Charles, J.F., Bourgouin, C., and de Barjac, H. (1984). Sporulation of Bacillus sphaericus 2297: an electron microscope study of crystal-like inclusion biogenesis and toxicity to mosquito larvae. J Gen Microbiol\\u003cem\\u003e\\u0026nbsp;130\\u003c/em\\u003e, 893-900.\\u003c/li\\u003e\\n \\u003cli\\u003eKang, H.J., and Baker, E.N. (2011). Intramolecular isopeptide bonds: protein crosslinks built for stress? Trends Biochem Sci\\u003cem\\u003e\\u0026nbsp;36\\u003c/em\\u003e, 229-237.\\u003c/li\\u003e\\n \\u003cli\\u003eKang, H.J., Coulibaly, F., Clow, F., Proft, T., and Baker, E.N. (2007). Stabilizing isopeptide bonds revealed in gram-positive bacterial pilus structure. Science\\u003cem\\u003e\\u0026nbsp;318\\u003c/em\\u003e, 1625-1628.\\u003c/li\\u003e\\n \\u003cli\\u003eKoch, M.S., Ward, J.M., Levine, S.L., Baum, J.A., Vicini, J.L., and Hammond, B.G. (2015). The food and environmental safety of Bt crops. Front Plant Sci\\u003cem\\u003e\\u0026nbsp;6\\u003c/em\\u003e, 283.\\u003c/li\\u003e\\n \\u003cli\\u003eLetunic, I., and Bork, P. (2024). Interactive Tree of Life (iTOL) v6: recent updates to the phylogenetic tree display and annotation tool. Nucleic Acids Res\\u003cem\\u003e\\u0026nbsp;52\\u003c/em\\u003e, W78-W82.\\u003c/li\\u003e\\n \\u003cli\\u003eLiebschner, D., Afonine, P.V., Baker, M.L., Bunkoczi, G., Chen, V.B., Croll, T.I., Hintze, B., Hung, L.W., Jain, S., McCoy, A.J.\\u003cem\\u003e, et al.\\u003c/em\\u003e (2019). Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr D Struct Biol\\u003cem\\u003e\\u0026nbsp;75\\u003c/em\\u003e, 861-877.\\u003c/li\\u003e\\n \\u003cli\\u003eLukaszczyk, M., Pradhan, B., and Remaut, H. (2019). The Biosynthesis and Structures of Bacterial Pili. Subcell Biochem\\u003cem\\u003e\\u0026nbsp;92\\u003c/em\\u003e, 369-413.\\u003c/li\\u003e\\n \\u003cli\\u003eMcClintock, J.T., Schaffer, C.R., and Sjoblad, R.D. (1995). A comparative review of the mammalian toxicity of Bacillus thuringiensis-based pesticides. Pesticide Science\\u003cem\\u003e\\u0026nbsp;45\\u003c/em\\u003e, 95-105.\\u003c/li\\u003e\\n \\u003cli\\u003eMirdita, M., Schutze, K., Moriwaki, Y., Heo, L., Ovchinnikov, S., and Steinegger, M. (2022). ColabFold: making protein folding accessible to all. Nat Methods\\u003cem\\u003e\\u0026nbsp;19\\u003c/em\\u003e, 679-682.\\u003c/li\\u003e\\n \\u003cli\\u003eMistry, J., Chuguransky, S., Williams, L., Qureshi, M., Salazar, G.A., Sonnhammer, E.L.L., Tosatto, S.C.E., Paladin, L., Raj, S., Richardson, L.J.\\u003cem\\u003e, et al.\\u003c/em\\u003e (2021). Pfam: The protein families database in 2021. Nucleic Acids Res\\u003cem\\u003e\\u0026nbsp;49\\u003c/em\\u003e, D412-D419.\\u003c/li\\u003e\\n \\u003cli\\u003eMusso, D., Ko, A.I., and Baud, D. (2019). Zika Virus Infection - After the Pandemic. N Engl J Med\\u003cem\\u003e\\u0026nbsp;381\\u003c/em\\u003e, 1444-1457.\\u003c/li\\u003e\\n \\u003cli\\u003ePark, J., Bae, S., and Kim, J.S. (2015). Cas-Designer: a web-based tool for choice of CRISPR-Cas9 target sites. Bioinformatics\\u003cem\\u003e\\u0026nbsp;31\\u003c/em\\u003e, 4014-4016.\\u003c/li\\u003e\\n \\u003cli\\u003ePhillips, M.A., Burrows, J.N., Manyando, C., van Huijsduijnen, R.H., Van Voorhis, W.C., and Wells, T.N.C. (2017). Malaria. Nat Rev Dis Primers\\u003cem\\u003e\\u0026nbsp;3\\u003c/em\\u003e, 17050.\\u003c/li\\u003e\\n \\u003cli\\u003ePradhan, B., Liedtke, J., Sleutel, M., Lindback, T., Zegeye, E.D., K, O.S., Llarena, A.K., Brynildsrud, O., Aspholm, M., and Remaut, H. (2021). Endospore Appendages: a novel pilus superfamily from the endospores of pathogenic Bacilli. EMBO J\\u003cem\\u003e\\u0026nbsp;40\\u003c/em\\u003e, e106887.\\u003c/li\\u003e\\n \\u003cli\\u003ePunjani, A., Rubinstein, J.L., Fleet, D.J., and Brubaker, M.A. (2017). cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat Methods\\u003cem\\u003e\\u0026nbsp;14\\u003c/em\\u003e, 290-296.\\u003c/li\\u003e\\n \\u003cli\\u003eRamnath, V., Larralde, M., Menchik, P., Buehler, A.J., Harrand, A.S., Chung, T., Wei, X., Raghuram, V., Gourl\\u0026eacute;, H., Pierneef, R.\\u003cem\\u003e, et al.\\u003c/em\\u003e (2023). A community-curated, global atlas of \\u003cem\\u003eBacillus cereus sensu lato\\u003c/em\\u003e genomes for epidemiological surveillance. bioRxiv, 2023.2012.2020.572685.\\u003c/li\\u003e\\n \\u003cli\\u003eRudd, S.R., Miranda, L.S., Curtis, H.R., Bigot, Y., Diaz-Mendoza, M., Hice, R., Nizet, V., Park, H.W., Blaha, G., Federici, B.A.\\u003cem\\u003e, et al.\\u003c/em\\u003e (2023). The Parasporal Body of Bacillus thuringiensis subsp. israelensis: A Unique Phage Capsid-Associated Prokaryotic Insecticidal Organelle. Biology (Basel)\\u003cem\\u003e\\u0026nbsp;12\\u003c/em\\u003e.\\u003c/li\\u003e\\n \\u003cli\\u003eSchnepf, E., Crickmore, N., Van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R., and Dean, D.H. (1998). Bacillus thuringiensis and its pesticidal crystal proteins. Microbiol Mol Biol Rev\\u003cem\\u003e\\u0026nbsp;62\\u003c/em\\u003e, 775-806.\\u003c/li\\u003e\\n \\u003cli\\u003eSierro, N., Makita, Y., de Hoon, M., and Nakai, K. (2008). DBTBS: a database of transcriptional regulation in Bacillus subtilis containing upstream intergenic conservation information. Nucleic Acids Res\\u003cem\\u003e\\u0026nbsp;36\\u003c/em\\u003e, D93-96.\\u003c/li\\u003e\\n \\u003cli\\u003eSleutel, M., Zegeye, E.D., Llarena, A.K., Pradhan, B., Fislage, M., O\\u0026apos;Sullivan, K., Van Gerven, N., Aspholm, M., and Remaut, H. (2024). Helical ultrastructure of the L-ENA spore aggregation factor of a Bacillus paranthracis foodborne outbreak strain. Nat Commun\\u003cem\\u003e\\u0026nbsp;15\\u003c/em\\u003e, 7514.\\u003c/li\\u003e\\n \\u003cli\\u003eTerry, C., Jiang, S., Radford, D.S., Wan, Q., Tzokov, S., Moir, A., and Bullough, P.A. (2017). Molecular tiling on the surface of a bacterial spore - the exosporium of the Bacillus anthracis/cereus/thuringiensis group. Mol Microbiol\\u003cem\\u003e\\u0026nbsp;104\\u003c/em\\u003e, 539-552.\\u003c/li\\u003e\\n \\u003cli\\u003eThomas, W.E., and Ellar, D.J. (1983). Bacillus thuringiensis var israelensis crystal delta-endotoxin: effects on insect and mammalian cells in vitro and in vivo. J Cell Sci\\u003cem\\u003e\\u0026nbsp;60\\u003c/em\\u003e, 181-197.\\u003c/li\\u003e\\n \\u003cli\\u003eVaradi, M., Bertoni, D., Magana, P., Paramval, U., Pidruchna, I., Radhakrishnan, M., Tsenkov, M., Nair, S., Mirdita, M., Yeo, J.\\u003cem\\u003e, et al.\\u003c/em\\u003e (2024). AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences. Nucleic Acids Res\\u003cem\\u003e\\u0026nbsp;52\\u003c/em\\u003e, D368-D375.\\u003c/li\\u003e\\n \\u003cli\\u003eWilliams, C.J., Headd, J.J., Moriarty, N.W., Prisant, M.G., Videau, L.L., Deis, L.N., Verma, V., Keedy, D.A., Hintze, B.J., Chen, V.B.\\u003cem\\u003e, et al.\\u003c/em\\u003e (2018). MolProbity: More and better reference data for improved all-atom structure validation. Protein Sci\\u003cem\\u003e\\u0026nbsp;27\\u003c/em\\u003e, 293-315.\\u003c/li\\u003e\\n \\u003cli\\u003eZhang, X. (2022). Python-based Helix Indexer: A graphical user interface program for finding symmetry of helical assembly through Fourier-Bessel indexing of electron microscopic data. Protein Sci\\u003cem\\u003e\\u0026nbsp;31\\u003c/em\\u003e, 107-117.\\u003c/li\\u003e\\n\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":true,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"nature-portfolio\",\"isNatureJournal\":true,\"hasQc\":false,\"allowDirectSubmit\":false,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Nature Portfolio\",\"twitterHandle\":\"\",\"acdcEnabled\":false,\"dfaEnabled\":false,\"editorialSystem\":\"ejp\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false},\"keywords\":\"Bacillus thuringiensis, spore biofilm, cross-alpha amyloid, isopeptide bond, self-assembly, nanofiber, endospore appendage, insect pathogen, delta endotoxin, Invertebrate Biological Control Agent (IBCA)\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-6239967/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-6239967/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e \\u003cem\\u003eBacillus thuringiesis\\u003c/em\\u003e infects and kills insect larvae via its spores and associated entomotoxin-containing parasporal bodies (PSBs). We show that endospores and PSBs of \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e Sv. Isra\\u0026euml;lensis are covered in a \\u0026lsquo;sporesilk\\u0026rsquo; matrix that consist of 8 nm wide fibers with a double helical symmetry, formed by protofilaments of stacked alphahelical hairpins. These alpha endospore appendages (\\u0026lsquo;A-ENA\\u0026rsquo;) are stabilized by up to ten autocatalytic and proximity-induced intermolecular isopeptide bonds per subunit, forming a continuous covalent polymer with remarkable chemical and physical robustness. We show A-ENA functions as \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e virulence factor, increasing insecticidal activity by clustering spores and PSBs into an infectious biofilm-like unit. Moreover, we demonstrate the recombinant production and self-assembly of A-ENA nanofibers, and show that the exogenic addition of A-ENA fibers to \\u003cem\\u003eB. thuringiensis\\u003c/em\\u003e strains natively lacking sporesilks results in spore-PSB clustering and gain of virulence, enabling the rational, non-GMO functionalization of these biological pest control agents.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Auto-crosslinking sporesilk fibers promote endospore and Cry toxin clustering\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-05-16 04:14:33\",\"doi\":\"10.21203/rs.3.rs-6239967/v1\",\"editorialEvents\":[],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"nature-communications\",\"isNatureJournal\":true,\"hasQc\":false,\"allowDirectSubmit\":false,\"externalIdentity\":\"NCOMMS\",\"sideBox\":\"Learn more about [Nature Communications](http://www.nature.com/ncomms/)\",\"snPcode\":\"\",\"submissionUrl\":\"https://mts-ncomms.nature.com/\",\"title\":\"Nature Communications\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"ejp\",\"reportingPortfolio\":\"Nature Communications\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false}}],\"origin\":\"\",\"ownerIdentity\":\"dce67504-1aa2-43ec-b341-a1e51e8506ae\",\"owner\":[],\"postedDate\":\"May 16th, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"published-in-journal\",\"subjectAreas\":[{\"id\":47130408,\"name\":\"Biological sciences/Microbiology/Bacteria/Bacterial toxins\"},{\"id\":47130409,\"name\":\"Biological sciences/Microbiology/Biofilms\"},{\"id\":47130410,\"name\":\"Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy\"},{\"id\":47130411,\"name\":\"Biological sciences/Microbiology/Bacteria/Bacterial pathogenesis\"}],\"tags\":[],\"updatedAt\":\"2026-04-27T07:05:27+00:00\",\"versionOfRecord\":{\"articleIdentity\":\"rs-6239967\",\"link\":\"https://doi.org/10.1038/s41467-026-70495-z\",\"journal\":{\"identity\":\"nature-communications\",\"isVorOnly\":false,\"title\":\"Nature Communications\"},\"publishedOn\":\"2026-03-11 04:00:00\",\"publishedOnDateReadable\":\"March 11th, 2026\"},\"versionCreatedAt\":\"2025-05-16 04:14:33\",\"video\":\"\",\"vorDoi\":\"10.1038/s41467-026-70495-z\",\"vorDoiUrl\":\"https://doi.org/10.1038/s41467-026-70495-z\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-6239967\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-6239967\",\"identity\":\"rs-6239967\",\"version\":[\"v1\"]},\"buildId\":\"8U1c8b4HqxoKbykW_rLl7\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}