Meningococci drive host membrane tubulation to recruit their signaling receptors

Meningococci drive host membrane tubulation to recruit their signaling receptors | 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 Meningococci drive host membrane tubulation to recruit their signaling receptors Mathieu Coureuil, Audrey LAURENT, Kevin SOLLIER, Bruno Saubamea, and 17 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6302813/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 25 Nov, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Once passed into the bloodstream, bacterial pathogens have a limited time to interact with permissive receptors at the surface of host cells. Neisseria meningitidis has developed an extremely effective strategy allowing it to find its receptors in a few seconds. Here, we report that N. meningitidis type IV pili exploit the physical properties of host cells plasma membranes to promote the formation of early tubular membrane structures essential for initial bacterial adhesion. These tubular structures, which form before any signaling events in host cells, concentrate and trap multiple plasma membrane-associated proteins in the vicinity of bacteria, thereby facilitating the selection, interaction and activation of specific adhesion and signaling receptors by bacterial ligands bound to type IV pili. Our results define an additional paradigm for the recruitment of specific receptors by pathogenic bacteria, which depends on the physical property of bacterial pili to induce the formation of tubular plasma membrane structures enriched in integral membrane receptor. Biological sciences/Microbiology/Cellular microbiology Biological sciences/Cell biology/Cell signalling Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 INTRODUCTION Signal activation in pathogen-infected cells, an essential event in the pathophysiology of infection, is currently viewed as a cascade of biochemical reactions elicited by the interaction of pathogen ligands with host cell receptors. In general, soluble ligands of biological fluids are in large excess over cognate plasma membrane receptors, granting a high probability of receptor binding and activation. However, many bacterial ligands are anchored to bacterial surfaces, such as the outer-membrane and the cell wall, or pathogen appendages, and their overall concentration is markedly lower than that of soluble ligands. Consequently, the probability of this type of ligand to eventually meet its receptor is theoretically much lower, especially if cognate receptors are expressed at low density. The challenge of "finding the relevant receptor" is particularly critical for bacteria, as they rely on interactions with host receptors for survival. One example is the model organism Listeria monocytogenes , for which the low probability of establishing a stable interaction with host tissues is balanced by the typically large number of infecting bacteria: 10 to 100 million bacteria are necessary for gut infection 1 or for neuro-invasion, while 10 thousand bacteria are required to infect hosts by intravenous injection 2 . In addition, infectious bacteria present in the bloodstream generally have a limited time window to bind to and infect target tissues before being wiped out by host defense mechanisms. Intriguingly, a poorly understood facilitating mechanism has been developed by Neisseria meningitidis , for which only 5 diplococci are sufficient to cause septicaemia in mice 3 . This extremely low number of bacteria is particularly remarkable since N. meningitidis do not survive the resident immune cells checkpoints in the liver and the bladder. To infect endothelial cells lining blood vessels this pathogen requires stable adhesion to endothelial cells within a time window of a few seconds, despite the continuous shear forces exerted by the blood-flow 4 . N. meningitidis adhesion relies on their type IV pili (T4P) and on T4P-induced endothelial cell plasma membrane remodeling, consisting of long tubular villi-like structures emerging from the cell surface, designed from now on as “tubular membrane structures”. The mechanisms leading to the formation of these key structures, which are mandatory for tissue colonization 5 , and their role in early adhesion remain largely unknown. T4P pili are common organelles made of filamentous polymers, belonging to the type IV filament family, which display a critical role in monoderm and diderm bacteria and in archaea 6 . They are involved in aggregation, biofilm formation, DNA uptake, mechano-sensing, adhesion / interaction with biotic and abiotic surfaces, and are essential for bacterial virulence and pathogenesis 7 – 9 . T4P mediate the interaction with cell receptors and the activation of signaling pathways in host cells by many human pathogens, such as P. aeruginosa 10 , Neisseria meningitidis 11 , 12 , Neisseria gonorrhoeae 13 , Clostridium difficile 14 , enteropathogenic or enterotoxigenic E. coli (EPEC and ETEC) 15 – 17 and Streptococcus sanguinis 18 . Only a few studies were conducted to elucidate the mechanisms of this interaction and identify T4P molecular targets. Most of them pointed out the property of T4P to interact with glycans present on host cell receptors 10 , 19 – 21 or carried by mucins of the mucus layer protecting epithelial cells 22 . T4P are dynamic structures, and their biogenesis relies on complex multi-protein machinery. Although the nomenclature differs between bacterial species 23 , 24 , 16 widely conserved essential proteins contribute to T4P biogenesis in pathogenic meningococci (PilC1 and C2, D, E, F, G, H, I, J, K, M, N, O, P, Q, W). Among them, PilE corresponds to the main constitutive protein (also named pilin) of the pilus fiber, PilC1 and PilC2 to the pilus-tip adhesins 25 , 26 . Only PilC1 is an adhesin for endothelial cells, a PilC1 null mutant only expressing PilC2 being piliated and non-adhesive. T4P are continuously elongated and retracted thanks to PilF and PilT ATPases, respectively 7 – 9 . In addition, the minor pilin PilV is distributed along the fiber and necessary for adhesion and signaling in host cells 19 , 27 . The cycles of elongation/retraction occur stochastically within few seconds and generate mechanical forces of 50–100 piconewtons (pN) for each filament, reaching up to 1 nanonewton for a bundle containing 8 to 10 pili 28 , 29 . These forces are larger than those required to break protein-protein or protein-glycan bonds, which are in the order of a few dozen to a hundred of pN 30 , 31 . Since T4P retraction forces are likely to rapidly disrupt the interaction between T4P and cognate receptors, the firm adhesion of bacteria to host cells must require some additional stabilizing mechanism(s) provided by host-cells. It was proposed, for example, that meningococci might exploit the “one dimensional” wetting property of the plasma membrane to form tubular-like structures along T4P 32 , 33 . Subsequent actin polymerization was proposed to stabilize these membrane structures to which meningococci are anchored 33 . The stable interaction between meningococcal T4P and human endothelial cells relies on the accumulation and mechanical activation of signaling complexes, consisting of CD147 and the β2-adrenergic receptor (β2AR), cross-linked by the cytoskeletal protein α-actinin-4 13,34–36,19,3 . PilV and PilV are both needed to initiate bacterial adhesion: the first interaction of the pilus-tip PilV with host cells is thought to be reinforced by the interaction of the secondary adhesin PilV with plasma membrane receptors. A possible model including all mechanisms above would place T4P/PilC-induced wetting as instrumental in concentrating CD147/β2-AR/α-actinin-4 complexes close to bacterial pili, enabling PilV-receptor interaction and host cell response. Here, we investigated the very early steps of T4P interaction with endothelial cells and demonstrated that the initial T4P-induced host cell tubular membrane structures only result from physical cues, independently of host cell cytoskeletal reorganization and signaling events. These tubular structures, promoted by membrane wetting, non-specifically accumulate multiple plasma membrane-associated proteins, including those specifically involved in meningococcal-induced signaling. Accumulation in these tubular structures allows their interaction with T4P ligands and subsequent activation of the signaling pathways, which ultimately drive the formation of the cellular protrusions - strengthened by cytoskeletal proteins - necessary for bacterial colonization. Wetting, by both increasing the interaction area and concentrating specific receptors, represents a general mechanism facilitating the encounter between membrane-associated ligands and cognate receptors in the context of cell-to-cell interactions. RESULTS N. meningitidis- induced tubular membrane structures are enriched in plasma membrane proteins. N. meningitidis induces the formation of tubular membrane structures in human endothelial cells, which are necessary for efficient bacterial colonization 5,4,34 . These structures have been considered to be the result of signaling pathways elicited in host cells by the pathogen and involving ezrin phosphorylation and actin polymerization 37–41,33 .Here, we investigated the key steps of formation of these structures by examining the distribution of a marker, the endogenous membrane curvature-sensitive CD9 tetraspanin 42 in human umbilical vein endothelial (EA.hy926) cells infected in vitro with wild type meningococci (2C4.3 strain). CD9 labeling near bacterial colonies was reminiscent of that observed with the cell-signaling marker ezrin, which accumulates under N. meningitidis colonies (Figure 1A). However, confocal imaging deconvolution revealed a very poor overlap between ezrin and CD9 labeling, the latter being concentrated in thin tubular plasma membrane structures surrounding adherent bacteria (Figure 1A and Figure S1A), which are distant or located above ezrin-positive areas (Figure 1A, bottom). We then compared CD9 and ezrin distribution using Correlated Light (confocal imaging) and (scanning) Electron Microscopy (CLEM) (Figure 1B and Figure S1B). While CD9 was accumulated into tubular membrane structures, ezrin appeared to be principally contained into distinct spotty areas with poor colocalization with CD9. Dense CD9 accumulation under meningococcal colonies has been considered as a facilitator of meningococcal adhesion to epithelial cell, reflecting the accumulation of receptors contributing to adhesion 43 . We therefore investigated whether CD9 and other tetraspanins would play an active role in the formation of the tubular membrane structures. We studied the localization of three plasma membrane tetraspanins CD9, CD81 and CD151 in different types of endothelial cells (EA.hy926, hCMEC/D3 and primary HDBEC) infected with wild type meningococci (Figure S2A-C). As for CD9 in EA.hy926 cells, all three tetraspanins accumulated under N. meningitidis colonies, independently of the endothelial cell type. CD9 knockdown (KD) in EA.hy926 cells had no effect on N. meningitidis induced signaling, assessed by ezrin accumulation (Figure S3A,D) and the same observation was made after silencing CD81 (Figure S3B,E) and CD151 (Figure S3C,F). Consistently, N. meningitidis -induced ezrin accumulation was not decreased by the CRISPR-mediated triple knockout (KO) of CD9, CD81 and CD151 in EA.hy926 cells (Figure S3G, H), whereas the adhesion of wild-type bacteria on these triple KO cells was similar to that of wild-type cells (Figure S3I). Overall, these data indicate that the tetraspanins CD9, CD81 and CD151 were not required for N. meningitidis adhesion, or for N. meningitidis -induced signaling during the infection of endothelial cells. Several other plasma membrane proteins were recruited to bacterial colonies as CD9: the endogenous CD44, some exogenously expressed yellow fluorescent protein (YFP)-tagged G protein coupled receptors (CXCR4, CCR5, and AT1R), the YFP-tagged T lymphocyte cell marker CD4, and the YFP-tagged chemokine scavenger ACKR2 (Figure S4). Instead, other exogenously expressed plasma membrane receptors, such as the GFP-tagged transferrin receptor or the GFP-tagged Toll like receptors TLR2 and TLR4 were not recruited into these tubular membrane structures (Figure S4). Bacterial pili are indispensable appendages for adhesion and signaling in host cells. For meningococcal T4P, the main structural pilin PilE and the adhesive pilins PilV and PilC1 (Figure 1C) contribute to adhesion and signaling (including ezrin accumulation) in endothelial cells. The N. meningitidis non-encapsulated strain expressing the Opa outer membrane adhesins ( SiaD - Opa + ) uses the secondary adhesins Opa proteins in addition to pilins to adhere to host cells, but still requires T4P to induce the formation of tubular membrane structures and signaling. The SiaD - Opa + meningococcal strains promoted CD9 accumulation into tubular plasma membrane structures surrounding adherent bacteria (Figure 1D,E top panels and S2) and ezrin accumulation under colonies 40 . In contrast, the non-piliated derivative Δ PilE mutant and the adhesion-defective derivatives Δ PilV and Δ PilC1 mutant strains adhered to endothelial cells but failed to induce the formation of CD9-positive tubular membrane structures (Figure 1D,E and S2). These data indicate that adhesive T4P are necessary and sufficient to generate CD9-positive tubular membrane structures in endothelial cells. We then investigated the diffusion dynamics of plasma membrane proteins within the tubular membrane structures induced by meningococcal T4P. Fluorescence recovery after photobleaching (FRAP) experiments were conducted in endothelial cells expressing YFP-CD9, infected with the 2C4.3 strain (Figure 2A). YFP-CD9 mobility was significantly impaired in the tubular structures compared to control plasma membrane areas distant from bacterial colonies on the same coverslip (Figure 2B). Mobility of both the N. meningitidis signaling receptor β2AR (β2AR-YFP) and of the T lymphocyte CD4 surface glycoprotein (CD4-YFP), which are not involved in signaling, was similarly reduced in tubular membrane structures, compared to control plasma membrane (Figure 2C, D). This suggests that these structures indiscriminately trap cell surface proteins and receptors, regardless of their potential role in N. meningitidis -induced signaling in host cells. Such a phenomenon might drive the non-selective accumulation of multiple plasma-membrane-associated proteins in these structures, including the adhesion and signaling receptors for meningococci. To determine whether the N. meningitidis signaling receptor β2AR might accumulate into these tubular structures, the fluorescent signal of β2AR-YFP and that of YFP-CD9 and CD4-YFP – both used here as controls – were quantified in N. meningitidis -induced tubular membrane structures relative to that measured in physiological filopodia present in areas of the plasma membrane distant from meningococcal colonies (Figure 2E,F). We reasoned that if membrane shape was the only driver of protein accumulation, the same signal should be observed in physiological filopodia and in bacteria-induced tubular structures around the colonies. The β2AR-YFP appeared accumulated in the tubular membrane structures but not in filopodia (Figure 2F, 4.3 fold and 1.1 fold, respectively). YFP-CD9 and CD4-YFP were also significantly accumulated in N. meningitidis -induced tubular membrane structures (Figure 2F, CD9: 5.65 fold in tubular structures (D) compared to 1.3 fold in filopodia (F); CD4: 4.3 fold in tubular structures compared to 1.1 fold in filopodia), ruling out the possibility that a simple change in membrane shape was the main driver of this enrichment. The signal of the FLIPPER-TR®, a fluorescent hydrophobic lipid membrane probe, was also significantly higher in tubular membrane structures compared to filopodia (2.18 fold and 0.95 fold, respectively), indicative of some plasma membrane accumulation in the focal acquisition plane of tubular membrane structures. It remained, however, significantly lower than that of YFP-CD9 (Figure 2F), suggesting that plasma membrane accumulation on its own accounted only in part for the observed local enrichment of membrane proteins. Overall, the data indicate that in the tubular membrane structures promoted by N. meningitidis T4P, the diffusion of integral membrane proteins is decreased, possibly contributing to the accumulation of key receptors. N. meningitidis -induced initiation of tubular membrane structures in host cells is independent of signaling. Ezrin accumulation under bacteria is a hallmark of N. meningitidis -promoted receptor signaling in host cells and plays an essential role in stabilizing bacterial adhesion and colony growth. We investigated its potential involvement in the formation of the tubular membrane structures promoted by T4P. Since ezrin activation and its plasma membrane relocation require its Thr567 phosphorylation 44 , EA.hy926 cells were treated prior to bacterial infection with NSC668394 45 (from now on, NSC66), an ezrin Thr567 phosphorylation inhibitor, which selectively blocks ezrin accumulation (Figure 3A,B) and its downstream effects. In particular, NSC66 treatment inhibits sustained bacterial adhesion after 2h of infection, which depends on the β2AR signaling pathway and on ezrin accumulation under colonies 33,36,19 (Figure 3C). Then, using CD9-associated fluorescence as a quantitative marker, we investigated the potential contribution of ezrin accumulation to plasma membrane tubulation. To take into account the decrease of bacterial adhesion 2h after infection, we quantified CD9 labeling intensity below bacterial colonies and normalized each value with the respective intensity of the DAPI staining which was used as a proxy of the colony size. Interestingly, the inhibition of ezrin phosphorylation (Figure 3D) did not affect the level of CD9 accumulation around wild type colonies when normalized on their size, indicating that initial N. meningitidis -promoted tubular membrane structures develop independently of ezrin activation and accumulation. T4P retraction, which depends on the PilT ATPase activity and is turned off in the N. meningitidis Δ pilT strain 46 , contributes to generating the mechanical forces that activate the T4P-induced b2AR signaling leading to ezrin recruitment in host cells 19 . CD9 accumulation (Figure 3D) and the formation of tubular membrane structures (Figure 3E,F) persisted, although reduced, in cells infected with Δ pilT mutant bacteria compared to wild type 2C4.3 meningococci. Consistent with the observation above, NSC66 did not further decrease CD9 accumulation to Δ pilT mutant bacteria (Figure 3D). Moreover, CD9 mobility in tubular deformations, whether promoted by wild type or by Δ pilT bacteria, was comparable in FRAP experiments (Figure 3G). Based on these results, the cellular process leading to the formation of early tubular membrane structures appeared to be independent on T4P-dependent receptor signaling or ezrin phosphorylation and activation. Of note, the persistence of some tubular membrane structures in the absence of any T4P retraction and ezrin accumulation raises the hypothesis that some purely physical, signaling-independent and T4P retraction-independent mechanism might be involved in this phenomenon. Tubulation of the plasma membrane along T4P is maintained when biochemical activity in host cells is inhibited. The shape of the apical membrane in endothelial cells is dynamically controlled by the cytoskeleton and by receptor-driven signaling. To impede any biochemical reaction that would interfere with plasma membrane dynamics, we examined the formation of N. meningitidis -induced tubular membrane structures in cells prefixed 15 min with 4% paraformaldehyde (PFA). Upon infection of PFA-treated cells with wild type meningococci, some tubular membrane structures containing endogenous CD9 still grew at the interface with bacteria (Figure 4A), although much less numerous (Figure 4D,E) and shorter (Figure 4C,E) than in living cells. As expected, no ezrin accumulation was observed in prefixed cells (Figure 4A-D). Surprisingly, upon infection with the Δ pilT mutant strain, the accumulation of CD9 around bacterial colonies was much more abundant in PFA-fixed cells than in living cells, and comparable to that observed in living cells infected with the 2C4.3 strain (Figure 4A,B). CD9 diffusion was similarly impaired by PFA fixation, whatever was the infecting strain, both in tubular structures and plasma membrane filopodia (Figure 4H,I). Thus, the larger CD9 accumulation in Δ pilT -induced tubular membrane structures of fixed cells (Figure 4J, about two-fold higher CD9 median intensity in tubular membrane structures, than in filopodia) is likely due to membrane accumulation and not to decreased diffusion as in non-fixed cells. Moreover, in fixed cells infected with Δ pilT bacteria, not only the maximal length of tubular membrane structures was greater than that measured in fixed or living cells infected with the wild type strain (Figure 4C), their number was also higher than in living cells, or fixed cells infected with the wild type strain (Figure 4 compare G with E and F). Overall, these findings indicate that the tubular membrane structures growing along T4P do not require any cell signaling but are limited/interrupted by pilus retraction. Moreover, in living cells this membrane tubulation along immobile T4P is restrained by some fixation sensitive host cell factor(s). To extend our observations in a more representative experimental setup, purified plasma membrane (PM) sheets 47,48 of EA.hy926 cells were prepared from a monolayer of living endothelial cells and then attached to the bottom of Ibidi µ-slides coated with anti-clathrin antibodies. The topology of clathrin-coated zones (flat lattices and coated pits) at the inner side of the membrane facilitates the attachment of PM sheets with the extracellular side oriented upwards. This attachment, which is limited to a few anchoring points, is thought to allow the lipid fraction of the plasma membrane to move freely, while biochemical reactions such as ezrin phosphorylation and actin polymerization are abolished 47,48 . PM sheets were then incubated with wild type or Δ pilT bacteria and the maximal length of tubular structures was determined (Figure 4K, L and S5A). Consistent with the findings in fixed cells, CD9 accumulation was hardly detected below wild type bacteria but clearly visible below Δ pilT colonies, with the maximal length of tubular membrane structures induced by Δ pilT bacteria being similar to that observed on fixed cells (2.59 µm and 3.09 µm, respectively). Actin polymerization inhibits tubulation of the plasma membrane induced by N. meningitidis T4P. From the data above, it appears that some host cell factors limit the formation of plasma tubular membrane structures promoted by immobile T4P. To address the possibility that this might be due to the reaction of the cortical cytoskeleton to a local increase in membrane tension, we examined whether the inhibition of membrane tubulation along T4P would require ATP as energy source. Endothelial cells were incubated with 10mM 2-deoxy-d-glucose (2-DG) to deplete cellular ATP (Figure S5B), before meningococcal infection. The length of CD9-labelled tubular membrane structures was then estimated using microscopy orthogonal sections (Figure 5A,B). ATP depletion, at levels sufficient to inhibit ezrin accumulation (Figure 5A), significantly reduced the length of protrusions in cells infected with the wild type 2C4.3 strain (Figure 5B), confirming the requirement of host cell signaling in the formation of mature protrusions in and around the colony. However, reminiscent of what was observed on PM sheets or upon PFA fixation of endothelial cells (see Figure 4), ATP depletion markedly enhanced the length of tubular protrusions induced by the Δ pilT strain infection (Figure 5A, B). Actin polymerization, typically observed under bacterial colonies in infected endothelial cells 36,40 , requires ATP. Pre-treatment of endothelial cells with the actin polymerization inhibitors Cytochalasin D (CytD) and Latrunculin B (LatB) enhanced the maximum length of tubular membrane structures promoted by Δ pilT bacteria to the same level of low ATP conditions (Figure 5B,C). Altogether, our findings indicate that the signaling-independent growth of tubular membrane structures is counteracted by actin cytoskeleton remodeling in host cells. N. meningitidis can induce tubular membrane structures in cells incompetent for meningococcal-promoted signaling. To further validate the hypothesis that the tubular membrane structures induced by T4P involve a purely physical mechanism, additional studies were conducted in cell types, which are not competent for meningococcal infection and meningococcal-induced signaling or ezrin recruitment. Kidney epithelial HEK cells cannot be infected by wild type N. meningitidis because bacteria cannot adhere to their surface and, in case of surrogate adhesion, bacteria do not induce signaling in the absence of exogenous b2AR and b-arrestins 34 . Surrogate adhesion of a N. meningitidis SiaD - Opa + strain can be reconstituted in HEK cells by expressing hCEACAM1, the adhesion receptor of opacity proteins (Opa) 49,40 . In hCEACAM-expressing HEK cells, CD9 - used as a marker of the tubular structures induced by T4P - accumulated in and around adherent SiaD - Opa + Δ pilT colonies (Figure 6A, left panels). Control non-piliated Δ pilE bacteria failed to induce any CD9 accumulation (Figure 6A, middle panels), consistent with the absolute requirement of T4P in this process. SiaD - Opa + Δ pilT colonies also induced the accumulation of CD9 in canine MDCK hCEACAM cells, in which no meningococcal-promoted signaling can be observed (Figure 6A, right panels). As shown by FRAP experiments in endothelial cells, YFP-CD9 mobility was decreased in infected HEK cell membrane structures surrounding bacteria, compared to control distant plasma membrane areas (Figure 6B, compare with Figure 2B). Furthermore, CD9 accumulation around bacteria was maintained in PFA-fixed HEK cells (Figure 6C), recapitulating what was observed in endothelial cells. These findings support the hypothesis that adhesive N. meningitidis T4P generate a purely physical process, independent of b2AR signaling, driving plasma membrane tubulation. PilC1-dependent adhesion is the initiating event of membrane tubulation The data accumulated so far did not determine what component of T4P could be the initial driver of the tubulation process in host cells. The PilV pilin and the tip-located PilC1 adhesin are both required for adhesion, CD9 recruitment and membrane tubulation (Figure 1 D,E). PilV interacts with CD147 and b2AR receptors and activates T4P retraction-induced signaling in host cells, suggesting that PilV is mainly involved in meningococcal adhesion after receptors accumulation in tubular membrane structures. Although also involved in adhesion, PilC1 has no known signaling properties. PilC1 thus appears as a strong candidate for the initiation of membrane tubulations in host cells. The genetic deletion of PilC1 in wild type or D pilT strains (D C1 and D TC1 mutants, respectively) significantly inhibited meningococcal adhesion in both intact and PFA-fixed cells (Figure 7A). The preserved expression of PilV in T4P of D TC1 bacteria was therefore not sufficient for maintaining normal adhesion (Figure 7A) and could not rescue membrane tubulation in SiaD - Opa + Δ C1 bacteria (Figure 1D,E). Also, the enhanced adhesion observed in fixed cells infected by the D pilT strain was abolished by the absence of PilC1 (Figure 7 A), despite the advantage provided by PilT suppression in terms of prolonged T4P contact with host cells plasma membrane and T4P length (see Figure 4). All together, these data demonstrate a predominant role of PilC1 as the inducer of tubular membrane structures in host cells at the initial steps of adhesion. DISCUSSION We uncovered the mechanism adopted by N. meningitidis to facilitate the encounter with cognate host-cells receptors despite the initial low multiplicity of infection usually observed with this pathogen. Bacterial T4P interact through the pilus tip adhesin PilC1 with host cells, inducing an early signaling-independent host plasma membrane tubulation (Figure S6, step 1). These tubular membrane structures extend the interaction surface between meningococcal ligands, such as PilV molecules with potential receptors. The plasma membrane-associated receptors found in these tubular structures exhibit reduced mobility, compared to areas of the plasma membrane distant from infecting bacteria. The increased local concentration of potential adhesion and signaling receptors in the vicinity of bacterial ligands eventually enable PilV to bind to CD147 / β2AR oligomers during the short time frame (few seconds) of pili immobility, before PilT-dependent retraction. PilV interaction with CD147 facilitates additional PilV molecules distributed along the pilus to bind to glycan chains of β2AR signalling receptors (Figure S6, step 2). The mechanical forces resulting from both T4P retraction and blood flow, applied on the β2AR via PilV, trigger signaling in host cells. The subsequent signaling- and ATP-dependent enrichment of the tubular membrane structures with ezrin and actin polymers allows their enlargement and reinforcement. The cellular protrusions resulting from this process stabilize meningococci at the cell surface, permitting the interaction of additional T4P ligands with additional receptors, and then the progressive growth of a bacterial colony despite the forces exerted by the blood flow (Figure S6, step 3). The ability of the plasma membrane to form early tubular structures along meningococcal T4P was previously observed, but its functional relevance in the context of N. meningitidis pathophysiology had not been addressed so far. The formation of these tubular structures was described as a physical process of one-dimensional “wetting” occurring when membranes interact with nanofibers 32 . The physical nature of this phenomenon is supported here by the demonstration that it is independent of meningococcus-induced signaling, it does not require ATP as energy source and can be reproduced in fixed cells or with purified plasma membrane sheets. Our data also point out the central role of the PilC1 pilus-tip adhesin docking to the plasma membrane, in the initiation of the one-dimensional wetting. These early tubular structures, promoted by T4P, display several properties that differentiate them from common plasma membrane filopodia. They accumulate multiple integral membrane proteins, including the tetraspanins CD9, CD81 and CD151, G protein-coupled and chemokine scavenger receptors, and the single transmembrane domain proteins CD4, CD44 and CD147. Using CD9 as a marker of these tubular structures, we could determine that its local enrichment around bacteria might not only depend on membrane accumulation - the tubular formation itself - but also on its markedly decreased mobility. Indeed, once in the tubular structure, membrane proteins can move along the walls of the tube, with a lower probability of exiting from its basis, keeping them trapped. Our results (see Fig. 2 F) also suggest an active receptor accumulation phenomenon in living cells that has yet to be determined. Moreover, it is not clear how other integral membrane proteins, such as the transferrin receptor or Toll like receptors are excluded from these structures. Several factors regulate the size and the lifetime of these T4P-promoted tubular structures in living cells. As mentioned above, PilT-dependent pilus retraction limit their duration to a few seconds 29 , unless signaling events promoted in host cells stabilize and turn them into larger cellular protrusions. In this context, ezrin recruitment and phosphorylation play an essential role for adhesion and protrusion stabilization 2h after infection. The role of cortical actin polymerization is more complex, since it has a variable effect depending on the duration of T4P interaction with the host cell plasma membrane. On one hand, its inhibition by ATP depletion in cells infected by the wild type strain (which retracts its T4P) reduces the maximal size of tubular structures. On the other hand, in cells infected with the Δ pilT strain, the inhibition of actin polymerization, either indirectly through ATP depletion or directly using cytochalasin D or latrunculin B, markedly enhances the length of tubular structures. In addition to Neisseria family bacteria, many other species express fimbriae or pili for adhesion to and infection of host cells. Interestingly, pili-tip adhesin proteins, such as Pseudomonas PilY1, uropathogenic E. coli FimH and Streptococcus pneumoniae RrgA are similarly exposed and directly engaged in the interaction with host cells by targeting specific receptors or extracellular matrix (ECM) components. Since the pili-dependent formation of tubular membrane structures appears to be a physical process of one-dimensional wetting resulting from the interaction of membranes with nanofibers (i.e. pili), it is plausible that the mechanism of cognate receptor recruitment in these tubular structures for N. meningitidis extends to other bacterial species. Mammalian cells can also produce nanofibers, including actin-based structures (filipodia, microvilli) and ECM components (collagen, fibronectin), which play a role in cell signaling and adhesion. It is not excluded that these structures might induce equivalent membrane tubular processes in adjacent/interacting cells. METHODS Cell lines and culture conditions Cell lines and media The EA.hy926 cell line (ATCC #CRL-2922) was obtained from the American Type Culture Collection (ATCC). EA.hy926 cells were grown in Dulbecco's Modified Eagle Medium (DMEM; Gibco) supplemented with 10% decomplemented fetal calf serum (FCS; Gibco) and 1% penicillin–streptomycin. Human Cerebral Microvascular Endothelial Cells hCMEC/D3s 50 are fully differentiated brain endothelial cell derived from human brain capillaries. hCMEC/D3 were grown onto cultrex rat collagen type I-coated dishes (Bio-techne) in Endothelial Cell Basal Medium-2 (Lonza) supplemented with 5% of FCS, 1.4 µM hydrocortisone (Lonza), 5 µg/mL ascorbic acid (Lonza), 1 ng/mL b-FGF (Lonza), at 37°C in 5% CO2. Primary Human Dermal Blood (Microvascular) Endothelial Cells (HDBEC) (Promocell) are isolated from the dermis of juvenile foreskin and adult skin (different locations). HDBEC were grown onto cultrex rat collagen type I-coated dishes (Bio-techne) in their specific endothelial cell growth medium (PromoCell). HEK-293 (ATCC CRL-1573) cells were grown in flasks coated with 0.01% poly-l-lysine (Sigma) in supplemented with 10% FCS and 1% penicillin–streptomycin. Cells were grown at 37˚C in a humidified incubator under 5% CO2. Transfection and siRNA inhibition EA.hy926 cells were transfected with 1µg of plasmids using Lonza 4D Nucleofector® System (Lonza) according to manufacturer’s instructions and 5.10 5 transfected cells were seeded into µ-Slide 4 Well (Ibidi) 24h prior to infection. For HEK cell, 1.5 × 10 5 cells were seeded in µ-Slide 4 Well and were transfected 24h prior to infection with 200ng of plasmid encoding hCEACAM and YFP-CD9, using EcoTransfect™ (Oz Biosciences) according to manufacturer’s instructions. For relative intensity and accumulation quantification experiments, infections were performed in Opti-MEM (Gibco). The siRNA inhibitions of CD9, CD81 and CD151 were obtained from siRNA pools (ON-TARGETplus Human CD9 (928) siRNA - SMARTpool (Horizon); ON-TARGETplus Human CD81 (975) siRNA - SMARTpool (Horizon); MISSION Pre-designed siRNA − 2 OD, Human CD151, SASI_Hs01_00188401 (Sigma)) and controlled with MISSION® siRNA Universal Negative Control #1 (Sigma). The siRNAs were transfected with Lipofectamine RNAiMAX (Thermofisher) on non-confluent cells. For CD9 siRNA, the transfection was repeated a second time after 3 days. Experiments were performed 3 days after the last transfection. Because tetraspanins are accumulated below meningococcal colonies, the whole amount of protein in cells did not reflect the true level of protein accumulation. Therefore, the inhibitions of tetraspanins expression were controlled by immunofluorescence quantification under bacterial colonies (see the image analysis section below). Production of the CRISPR/Cas9 knockout cell lines Cell line mutation The triple knockout (KO) EA.hy926 cell line was obtained by serial knockout of CD9, then CD81 and then CD151. CD9 and CD81 KO cell lines were obtained using the lentiCRISPR v2 plasmid (gifts from Dr Eric Rubinstein; Addgene plasmid #52961). The CD151 KO cell lines were obtained using the Sigma-Aldrich CRIPRD HSPD0000006331 and HSPD0000006332 plasmids. For each of the three separate tetraspanins we produced two lentiviruses with different guiding sequence. After obtaining each KO cell lines, tetraspanin depletion, adhesion of bacteria and induced signalling was assessed. One cell line was kept for further transduction. Guiding sequences are as follow: CD9sg95 TTGGACTATGGCTCCGATTC & CD9sg314 ATTCGCCATTGAAATAGCTG; CD81sg166 ACACCTTCTATGTAGGTGAG & CD81sg253 AGGAATCCCAGTGCCTGCTG; CD151 n°6331 CTGGTAGTAGGCGTAGGCG & CD151 n°6332 CCAAGCGCTACCACCAGCC. EA.hy926 cells were transduced with lentivirus (MOI 1:1000). 48 h after transduction, cells were treated with 1 µg/ml puromycin to kill non-transduced cells. FACS sorting of tetraspanin-depleted cells was performed 2 to 3 times over the scope of 1 or 2 months to ensure the purity of each CRISPR KO cell lines. Note that we did not select a clonal population but the whole population of the CD9, CD81 and CD151 depleted cells. Cytometry For surface protein expression analysis, cells were washed twice in Phosphate-Buffered-Saline (PBS), trypsinized, fixed in 4% paraformaldehyde (PFA) for 15 minutes, washed twice in PBS and kept at 4°C until staining. For cell sorting, cells were washed twice in PBS, trypsinized, and immediately washed twice in PBS before staining. In both experiments, cells were stained for 30 minutes with the appropriate primary antibody at 4°C in PBS/Bovine Serum Albumin (BSA) 0.1% and washed three times in PBS-BSA before Alexa Fluor-coupled secondary antibody staining for 30 minutes. Cells were then washed twice in PBS-BSA before FACS analysis. Negative controls were assessed with secondary antibody-only staining on the same cells. Data were acquired using a BD LSR Fortessa instrument (BD Biosciences) for fixed samples, and with BD FACSAria II for live cell sorting. Data was analyzed using the FlowJo Software. A minimum of 20 000 cells were acquired for each experiment. Treatments of cells Actin polymerization and ezrin inhibition drugs Treatments with drugs were performed in DMEM + 10% FBS for NSC668394 (Sigma-Aldrich 341216; 30 µM 2h prior to infection, and 3µM throughout infection), wheat germ agglutinin (WGA; Vector Laboratories, Burlingame, CA, USA; 20 µM, 1h prior to infection and maintained throughout infection), Cytochalasin D (Sigma C8273; 2–4 µM, 2h prior to infection and maintained throughout infection), Latrunculin B (Abcam ab141409; 0,1–1 µg/ml, 2h prior to infection and maintained throughout). Controls were assessed with DMSO when relevant. ATP depletion For ATP depletion, treatment was performed in DMEM without glucose (Invitrogen) + 1 mM lactate (as carbon source for bacteria) and 10 mM 2-deoxy-d-glucose (2-DG, Sigma-Aldrich). Control cells were incubated in DMEM without glucose supplemented with 10% FBS. Treatment was performed overnight prior to infection and maintained through the course of the experiment. ATP depletion efficiency was assessed using the commercially available kit Adenosine 5′-triphosphate Bioluminescent Assay (Sigma-Aldrich). Chemical pre-fixation assay Cells were chemically fixed on the day of the infection with 4% paraformaldehyde (PFA, Thermofisher 28908) for 15 minutes and quenched with 50 mM NH 4 Cl for 5 minutes. Cells were then washed several times in PBS and DMEM + 10% FBS prior to infection. Bacterial strains and infection N. meningitidis strains used in this study are derived from serogroup C meningococcal strain 8013, designated as 2C4.3 51 . 2C4.3 is a piliated encapsulated Opa − Opc − variant. Mutant strains Δ pilC1 52 , Δ pilE 53 , Δ pilT 3 , Δ pilV 35 , Δ siaD 40 as well as double mutants were designed in the laboratory. N. meningitidis were grown at 37°C in 5% CO 2 on gonococcal base (GCB) agar (Difco) plates containing 12 µM FeSO4 and Kellogg’s supplements 54 or in Dulbecco's Modified Eagle Medium (DMEM; Gibco) supplemented with 10% decomplemented fetal calf serum (FCS; Gibco). For antibiotic selection of N. meningitidis strains, kanamycin was used at a concentration of 100 µg/ml, chloramphenicol at 6 µg/ml and erythromycin at 2 µg/ml. Before cell infection, bacteria were first sub-cultured to OD600 = 0.1 in pre-warmed cell culture medium and incubated for 2 h at 37°C with agitation and 5% CO 2 . Imaging assay Cells were infected with bacteria at a multiplicity of infection of 30 bacteria per cell for 30 min and washed with media to remove non-adherent bacteria. Infection then proceeded for 2 h and was concluded with two PBS washes and 4% PFA fixation for 15 minutes. Bacterial adhesion assay At least two days before infection, cells were seeded on 2 cm² wells. The day of infection, confluent adherent cells were infected with an estimated number of 10 7 bacteria each for 40 min. The exact number of colony forming units (CFU) in the inoculum was determined by serial dilution and counting of CFU. After 40 min, the cells were washed six times with PBS to remove non-adherent bacteria and adherent bacteria were detached in 500 µL DMEM with 10% FBS by mechanical scratching. To determine the adhesion frequency, adherent bacteria were diluted and spread on agar plates and the CFU were counted the next day. Adhesion frequency was obtained by dividing the adherent bacteria CFU count with the inoculum CFU count. For longer adhesion assay, infection was allowed to proceed for 2 h after the first washes at the 40-minute mark and was concluded similarly to the early adhesion assay. Imaging Immunofluorescence microscopy All incubations were performed at room temperature. Cells were grown and infected on glass coverslips. They were fixed in 4% PFA for 15 minutes, quenched with ammonium chloride 50 mM solution for 5 min and rinsed with PBS with ions. Immunostaining steps were performed with 0.1% BSA blocking and 0.1% Saponin permeabilization at all times. Coverslips were incubated in BSA and Saponin buffer for 20 minutes before staining. Proteins of interest were stained with primary antibodies for 45 min to 1 h. After three washes in PBS, the coverslips were incubated with Alexa Fluor-conjugated secondary antibodies for 45 min. Nuclear DNA and actin were stained with 4′,6′-diamidino-2-phenylindole (DAPI) at 1 µg/ml and Alexa Fluor-conjugated phalloidin (Thermofisher), respectively. After three PBS washes, the coverslips were rinsed in water and mounted in Mowiol for observation. Immunofluorescence imaging was performed either on a Zeiss Spinning Disk microscope (63X, NA 1.4), a Zeiss Apotome fluorescence microscope or a SP8 laser scanning confocal microscope (Leica Microsystems) using a oil immersion objective. When mentioned deconvolution was performed using Huygens software. Scanning Electron Microscopy All incubations were performed at room temperature except when otherwise indicated. Cells were grown and infected on glass coverslips. They were fixed in 0.1M sodium cacodylate (Euromedex) buffer, pH 7.4 containing 2.5% glutaraldehyde (GA, Euromedex) and 1% PFA for 60 min, washed in cacodylate buffer (2x10 min), and then fixed in 1% OsO4 diluted in cacodylate buffer for 45 min at 4°C. After washing in cacodylate buffer (2x10 min), samples were dehydrated in an ascending series of ethanol (30%, 50%, 70%, 95%, 100%, 100%, 100% − 10 min each), followed by Hexamethyldisilazane (HMDS, Sigma-Aldrich)/ethanol (1/1:v/v) for 10 min and HMDS for 10 min. After overnight air drying, each coverslip was placed on a double-sided sticky tape on the top of an aluminum stub and sputter coated with Au/Pd. Images were acquired using a Jeol LV6510 (Jeol, Croissy-sur-Seine, France). Correlative Light Electron Microscopy (CLEM) Cells were grown and infected in 35mm µ–Dishes with a removable gridded glass coverslip bottom (Ibidi 81158). They were first processed for fluorescence imaging as follows. After fixation in 4% PFA + 0.2% GA, they were immunostained similarly to a classic immunofluorescence microscopy. They were then imaged in a SP8 laser scanning confocal microscope (Leica Microsystems) using a X63/1.40 oil immersion objective and 405/410–450, 488/495–530, and 552/560–600 excitation/emission wavelengths (in nm) for the detection of DAPI, AF488, and AF555 respectively. A mosaic image of the whole grid, including the lettered and numbered squares, was acquired in the DAPI channel and used as a map to locate the cells that were then observed at high resolution. For each selected region of interest (ROI), a z-stack with a step of 0.3 µm was acquired and deconvoluted using Huyghens software. Following confocal imaging, cells were fixed and processed for SEM as indicated above. Each ROI previously acquired in the confocal microscope was imaged in the SEM. For correlative image registration between electron microscopy (EM) and fluorescence the Icy (v2.5.2.0) 55 EC-CLEM plugin 56 has been used with the DAPI signal as the referent channel for the registration. FRAP Live experiment data were acquired with a TCS SP8 X confocal microscope (Leica) equipped with a ×100, 1.40 oil HC PL APO objective and a heated stage maintained at 37°C. For FRAP analysis, series of 40 fluorescence measures were captured from 30 images for all experiments from three independent experiments, with time intervals between scans ranging from 100 to 1000 ms. To perform FRAP, 1µm²-ROI were bleached with 10 WLL2 laser pulse iterations at 100% power. Image analysis and fluorescence recovery was made by LAS X software (Leica). Data were normalized with average value of pre-FRAP values and were plotted using Prism9 software (GraphPad), a one-phase decay fit curve was applied to the plotted results. Plasma membrane sheets (PM sheets) preparation The protocol was adapted from 48 , 47 . EA.hy926 cells were separated with 5mM EDTA (ThermoFisher 15575020) on the day of cell passage and seeded on a 6-well plaque pre-coated with Poly-L-Lysine (Sigma P4707). One day after seeding, cells were rinsed twice with PBS and were submitted to sonication in 10 mL cold PBS using a probe sonicator (six pulses of 0.5 s, 20% duty cycle, output control level 2). Cytoplasm released and detached cells in the supernatant were discarded by rinsing twice with HBSS (ThermoFisher 14170088; 5 ml, 5 min). Remaining attached membranes were scrapped and resuspended in 150 µL HBSS with protease cocktail inhibitor (Sigma 4693159001) with cut pipette tips to avoid destroying the collected PM-sheets. PM-sheets were then deposited and stayed overnight into a 8 well µ-Slide (ibidi 80826) previously treated with Poly-D-Lysine (ThermoFisher A3890401) for 1h and then coated with anti-clathrin antibodies (ThermoFisher PA5-143896) overnight. Wells were then washed twice in PBS and adherent PM-sheets were incubated in cell culture media and infected with bacteria and fixed for immunostaining. Image analysis Relative intensity accumulation quantification EA.hy926 cells were transfected and infected as described above. For each condition, 30 confocal images from three independent experiments of filopodia and tubular membrane structures were acquired along plasma membrane. Maximum intensity value were determined from a trace line using ImageJ software. For each image, filopodia and tubular membrane structures values were normalized with the corresponding plasma membrane value and plotted using Prism9 software. Protein accumulation index The accumulation index of CD9 and Ezrin was determined by immunofluorescence volume analysis. Stack images of bacterial colony were acquired with a Zeiss Spinning Disk microscope (63X, NA 1.4). We manually recognized bacterial colonies using Fiji ROI 57 (v2.14.0). Then with a macro designed for this work, each ROI for each image were extracted as a cropped tiff stack image. Then each cropped images channel were isolated with a Fiji macro in order to do a 3D shallow learning pixel classification with ilastik 58 (v1.4.0post1). DAPI was used to recognize bacteria. Bacteria and protein-of-interest volume (Bv and POIv, respectively) were obtained with a last Fiji macro and the accumulation index (AI) was calculated, AI = POIv/Bv. Maximum tubular membrane length quantification The maximum tubular membrane length was assessed for n ≥ 30 colonies among two technical replicates for a total of three experimental replicates for every experimental condition. For each colony, chosen at random, CD9 immunofluorescence was used to spot tubular membrane structure. The distance between the apical plasma membrane focal plane and the tip of tubular structures was estimated with the z-stack tool on a Zeiss Apotome fluorescence microscope at x63 by multiplying the steps amount and step size of 0,24 µm. T4P purification Ammonium sulfate precipitation. N. meningitidis strains were grown overnight on GCB agar plates, scraped off of the plates and resuspended in 2 mL of 20 mM ethanolamine, pH 10.5 supplemented with 1 mM dithiothreitol (DTT) at 4°C. To shear T4P off, the bacteria were vortexed vigorously 3 times for 1 min bursts, returning them to ice for 1 min between each burst. Bacterial cells were removed from the pilus suspension by two successive centrifugations at 10 000 x g for 20 min at 4°C. Cell pellets were resuspended in lysis buffer (50 mM Tris pH 7.5, 25 mM HEPES, 2 mM EDTA, 1% (w/v) SDS - bacterial fraction). The supernatant containing the T4P was collected and supplemented with saturated ammonium sulfate in 20 mM ethanolamine, pH 10.5 at a final concentration of 0.15 M, before overnight agitation at 4°C. Aggregated T4P were pelleted by centrifugation at 17 000 x g for 20 min at 4°C and resuspended overnight at 4°C, in 400 µl of 20 mM ethanolamine, pH 10.5. The T4P solution was centrifuged at 10 000 x g for 20 min at 4°C to remove residual cell debris and the supernatant was concentrated 10-fold using an Amicon 10 kDa MWCO membrane (Merck Millipore). Immunoblotting For analysis of cell lysates, one bacterial loop of each strain of interest – grown overnight in GCB agar + corresponding antibiotics – was resuspended in 500 µl RIPA buffer (50 mM Tris pH 7,5, 150 nM NaCl, 25 nm HEPES, 2 mM EDTA, 1% w/vol SDS). This suspension was heated at 95°C for 5 min. For analysis of T4P, pili were purified as described above. Whole lysate or T4P preparation was deposited on a 12% acrylamide SDS-PAGE gel. After transfer to nitrocellulose membrane, the membrane was incubated for 20 min in a blocking solution (PBS + 0.1% tween-20 + 4% milk) and washed twice with PBS + 0.1% tween-20 (PBST). The membrane was incubated for 1 h in the presence of anti-PilV primary antibody 27 diluted 1:1,000 in PBST, washed three times for 5 min with PBST and then incubated for 45 min with a rabbit anti-IgG secondary antibody coupled to horseradish peroxidase diluted at 1:10,000 in PBST. Proteins were detected by chemiluminescence using the Clarity Western ECL substrate detection kit (Bio-Rad) and a Chemidoc imaging system (Bio-rad). Whole proteins were detected on gel using the stain-free technology from Bio-Rad. Antibodies and staining reagents used in this work FLIPPER-TR® was used at 1µM in OptiMEM (Spirochrome AG, Switzerland) for 30mn before imaging of living infected cells. 4′,6′-diamidino-2-phenylindole (DAPI) was used at 1 µg/ml to stain eukaryotic and prokaryotic DNA. Alexa Fluor-conjugated phalloidin (Thermofisher, A22287) was used at the recommended concentration to stain filamentous actin. TS9 and TS81 59 were used at 1:200 dilution to stain for CD9 and CD81 in cytometry and microscopy assays. TS151 59 and 11B1G4 (generously provided by Leonie Ashman) 60 were used at 1:200 dilution for CD151 staining in cytometry and microscopy respectively. Ezrin was stained with an anti-ezrin antibody at 1:500 dilution (generously provided by Dr. P. Mangeat (CNRS, UMR5539. Montpellier, France). CD44 was stained with proteintech 15675-1-AP antibody at 1:50 dilution. Alexa Fluor-conjugated goat IgG secondary antibodies (Thermofisher, A11029 and A11010) were used at 1:200 dilution in fluorescent microscopy. Clathrin polyclonal antibody (Thermofisher, PA5-143896) was used at 0.5 µg/mL in PM sheets preparation. Anti-PilV antibody 27 and HRP-conjugated goat IgG secondary antibody (Thermofisher, 656120) were used in immunoblotting at 1:10 000 dilution. Statistical analysis Statistical analyses were performed with GraphPad Prism 8 or 9. When needed the data were log transformed. Normality of the distribution of whole data sets were assessed using QQ plot. The variance was assessed with a Brown-Forsythe test. When variance was different and the distribution was normal a Brown-Forsythe and Welch ANOVA or t -test was performed. For non-normal distribution, a non-parametric test was performed (Mann Whitney test or Kruskal-Wallis test). Corrections were indicated in the legend of figures. The H0 hypothesis was rejected for a significance level of p ≤ 0.05. Figures have been created using GraphPad Prism and Adobe Illustrator. Raw data and statistical analysis are listed in the data_and_statistics and supplementary_data_and_statistics files. Declarations ACKNOWLEDGEMENTS This work was supported by the research grant ANR-19-CE14-0045-002 (to NB, PG, SM and MC) and funding from Inserm. AL and MW were supported by a scholarship from Université Paris Cité. KS fellowship was supported by the research grant ANR-19-CE14-0045-002. We warmly thank: Nicolas Biais, Sandrine Bourdoulous, Etienne Morel, Cédric Delevoye and Philippe Chavrier for in-depth discussions; Béatrice Durel, Julie Lesieur, Thomas Guilbert and Pierre Bourdoncle for their expertise in microscopy (conventional, FRAP and FLIM); Jean-Baptiste Manneville for his expertise in biophysics. Image acquisition and image analysis were performed at the Imaging Facility of Structure Fédérative de Recherche (SFR) Necker, INSERM US24/CNRS UAR3633 and the IMAG’IC Facility of the National Infrastructure France BioImaging (ANR-10-INBS-04). Flow cytometry and analysis was performed at the Cytometry facility of SFR Necker, INSERM US24/CNRS UAR3633. Lentiviruses were obtained from the Viral Vectors and Genes Transfer facility (VVTG) platform of SFR Necker, INSERM US24/CNRS UAR3633. Illustrations were created with Biorender.com. declaration of interests The authors declare no competing interests Author contributions Conceptualization, AL, KS, SM, MC; Methodology, AL, KS, BS, NG, GL, ER, SM, MC; Investigation, AL, KS, BS, VM, YW, MW, MR, TH, LL, MC, JM, PG; Validation, HL, EB, SM, MC ; Writing – Original Draft, AL, KS, SM, MC; Writing – Review & Editing, AL, KS, BS, NG, AJ, NB, PG, EB, GL, ER, SM, MC; Visualization, AL, KS, SM, MC; Funding Acquisition, NB, SM, MC; Resources, NG, ER; Supervision, SM, MC. References Farber, J.M., Ross, W.H., and Harwig, J. (1996). Health risk assessment of Listeria monocytogenes in Canada. Int. J. Food Microbiol. 30 , 145–156. https://doi.org/10.1016/0168-1605(96)01107-5. Maudet, C., Kheloufi, M., Levallois, S., Gaillard, J., Huang, L., Gaultier, C., Tsai, Y.-H., Disson, O., and Lecuit, M. (2022). Bacterial inhibition of Fas-mediated killing promotes neuroinvasion and persistence. Nature 603 , 900–906. https://doi.org/10.1038/s41586-022-04505-7. Barnier, J.-P., Euphrasie, D., Join-Lambert, O., Audry, M., Schonherr-Hellec, S., Schmitt, T., Bourdoulous, S., Coureuil, M., Nassif, X., and El Behi, M. (2021). Type IV pilus retraction enables sustained bacteremia and plays a key role in the outcome of meningococcal sepsis in a humanized mouse model. PLoS Pathog. 17 , e1009299. https://doi.org/10.1371/journal.ppat.1009299. Mairey, E., Genovesio, A., Donnadieu, E., Bernard, C., Jaubert, F., Pinard, E., Seylaz, J., Olivo-Marin, J.-C., Nassif, X., and Duménil, G. (2006). Cerebral microcirculation shear stress levels determine Neisseria meningitidis attachment sites along the blood-brain barrier. J. Exp. Med. 203 , 1939–1950. https://doi.org/10.1084/jem.20060482. Mikaty, G., Soyer, M., Mairey, E., Henry, N., Dyer, D., Forest, K.T., Morand, P., Guadagnini, S., Prévost, M.C., Nassif, X., et al. (2009). Extracellular bacterial pathogen induces host cell surface reorganization to resist shear stress. PLoS Pathog. 5 , e1000314. https://doi.org/10.1371/journal.ppat.1000314. Berry, J.L., and Pelicic, V. (2015). Exceptionally widespread nanomachines composed of type IV pilins: the prokaryotic Swiss Army knives. FEMS Microbiol Rev 39 , 134–154. https://doi.org/10.1093/femsre/fuu001. Craig, L., Forest, K.T., and Maier, B. (2019). Type IV pili: dynamics, biophysics and functional consequences. Nat. Rev. Microbiol. 17 , 429–440. https://doi.org/10.1038/s41579-019-0195-4. Ligthart, K., Belzer, C., de Vos, W.M., and Tytgat, H.L.P. (2020). Bridging Bacteria and the Gut: Functional Aspects of Type IV Pili. Trends Microbiol. 28 , 340–348. https://doi.org/10.1016/j.tim.2020.02.003. Pelicic, V. (2023). Mechanism of assembly of type 4 filaments: everything you always wanted to know (but were afraid to ask). Microbiol. Read. Engl. 169 , 001311. https://doi.org/10.1099/mic.0.001311. Bucior, I., Pielage, J.F., and Engel, J.N. (2012). Pseudomonas aeruginosa pili and flagella mediate distinct binding and signaling events at the apical and basolateral surface of airway epithelium. PLoS Pathog 8 , e1002616. https://doi.org/10.1371/journal.ppat.1002616. Coureuil, M., Jamet, A., Bille, E., Lécuyer, H., Bourdoulous, S., and Nassif, X. (2019). Molecular interactions between Neisseria meningitidis and its human host. Cell. Microbiol. 21 , e13063. https://doi.org/10.1111/cmi.13063. Marullo, S., Scott, M.G.H., Enslen, H., and Coureuil, M. (2022). Mechanical Activation of the β2-Adrenergic Receptor by Meningococcus: A Historical and Future Perspective Analysis of How a Bacterial Probe Can Reveal Signalling Pathways in Endothelial Cells, and a Unique Mode of Receptor Activation Involving Its N-Terminal Glycan Chains. Front. Endocrinol. 13 , 883568. https://doi.org/10.3389/fendo.2022.883568. Merz, A.J., Enns, C.A., and So, M. (1999). Type IV pili of pathogenic Neisseriae elicit cortical plaque formation in epithelial cells. Mol. Microbiol. 32 , 1316–1332. https://doi.org/10.1046/j.1365-2958.1999.01459.x. McKee, R.W., Aleksanyan, N., Garrett, E.M., and Tamayo, R. (2018). Type IV Pili Promote Clostridium difficile Adherence and Persistence in a Mouse Model of Infection. Infect. Immun. 86 , e00943-17. https://doi.org/10.1128/IAI.00943-17. Mazariego-Espinosa, K., Cruz, A., Ledesma, M.A., Ochoa, S.A., and Xicohtencatl-Cortes, J. (2010). Longus, a type IV pilus of enterotoxigenic Escherichia coli, is involved in adherence to intestinal epithelial cells. J. Bacteriol. 192 , 2791–2800. https://doi.org/10.1128/JB.01595-09. Aroeti, B., Friedman, G., Zlotkin-Rivkin, E., and Donnenberg, M.S. (2012). Retraction of enteropathogenic E. coli type IV pili promotes efficient host cell colonization, effector translocation and tight junction disruption. Gut Microbes 3 , 267–271. https://doi.org/10.4161/gmic.19814. Oki, H., Kawahara, K., Maruno, T., Imai, T., Muroga, Y., Fukakusa, S., Iwashita, T., Kobayashi, Y., Matsuda, S., Kodama, T., et al. (2018). Interplay of a secreted protein with type IVb pilus for efficient enterotoxigenic Escherichia coli colonization. Proc. Natl. Acad. Sci. U. S. A. 115 , 7422–7427. https://doi.org/10.1073/pnas.1805671115. Raynaud, C., Sheppard, D., Berry, J.-L., Gurung, I., and Pelicic, V. (2021). PilB from Streptococcus sanguinis is a bimodular type IV pilin with a direct role in adhesion. Proc. Natl. Acad. Sci. U. S. A. 118 , e2102092118. https://doi.org/10.1073/pnas.2102092118. Virion, Z., Doly, S., Saha, K., Lambert, M., Guillonneau, F., Bied, C., Duke, R.M., Rudd, P.M., Robbe-Masselot, C., Nassif, X., et al. (2019). Sialic acid mediated mechanical activation of β2 adrenergic receptors by bacterial pili. Nat. Commun. 10 , 4752. https://doi.org/10.1038/s41467-019-12685-6. Le Guennec, L., Virion, Z., Bouzinba-Ségard, H., Robbe-Masselot, C., Léonard, R., Nassif, X., Bourdoulous, S., and Coureuil, M. (2020). Receptor recognition by meningococcal type IV pili relies on a specific complex N-glycan. Proc. Natl. Acad. Sci. U. S. A. 117 , 2606–2612. https://doi.org/10.1073/pnas.1919567117. Shahin, M., Sheppard, D., Raynaud, C., Berry, J.-L., Gurung, I., Silva, L.M., Feizi, T., Liu, Y., and Pelicic, V. (2023). Characterization of a glycan-binding complex of minor pilins completes the analysis of Streptococcus sanguinis type 4 pili subunits. Proc. Natl. Acad. Sci. U. S. A. 120 , e2216237120. https://doi.org/10.1073/pnas.2216237120. Rossy, T., Distler, T., Meirelles, L.A., Pezoldt, J., Kim, J., Talà, L., Bouklas, N., Deplancke, B., and Persat, A. (2023). Pseudomonas aeruginosa type IV pili actively induce mucus contraction to form biofilms in tissue-engineered human airways. PLoS Biol. 21 , e3002209. https://doi.org/10.1371/journal.pbio.3002209. Muir, A., Gurung, I., Cehovin, A., Bazin, A., Vallenet, D., and Pelicic, V. (2020). Construction of a complete set of Neisseria meningitidis mutants and its use for the phenotypic profiling of this human pathogen. Nat. Commun. 11 , 5541. https://doi.org/10.1038/s41467-020-19347-y. Nolan, L.M., Whitchurch, C.B., Barquist, L., Katrib, M., Boinett, C.J., Mayho, M., Goulding, D., Charles, I.G., Filloux, A., Parkhill, J., et al. (2018). A global genomic approach uncovers novel components for twitching motility-mediated biofilm expansion in Pseudomonas aeruginosa. Microb. Genomics 4 , e000229. https://doi.org/10.1099/mgen.0.000229. Rudel, T., Scheurerpflug, I., and Meyer, T.F. (1995). Neisseria PilC protein identified as type-4 pilus tip-located adhesin. Nature 373 , 357–359. Treuner-Lange, A., Chang, Y.-W., Glatter, T., Herfurth, M., Lindow, S., Chreifi, G., Jensen, G.J., and Søgaard-Andersen, L. (2020). PilY1 and minor pilins form a complex priming the type IVa pilus in Myxococcus xanthus. Nat. Commun. 11 , 5054. https://doi.org/10.1038/s41467-020-18803-z. Barnier, J.-P., Meyer, J., Kolappan, S., Bouzinba-Ségard, H., Gesbert, G., Jamet, A., Frapy, E., Schönherr-Hellec, S., Capel, E., Virion, Z., et al. (2021). The minor pilin PilV provides a conserved adhesion site throughout the antigenically variable meningococcal type IV pilus. Proc. Natl. Acad. Sci. U. S. A. 118 , e2109364118. https://doi.org/10.1073/pnas.2109364118. Biais, N., Ladoux, B., Higashi, D., So, M., and Sheetz, M. (2008). Cooperative retraction of bundled type IV pili enables nanonewton force generation. PLoS Biol. 6 , e87. https://doi.org/10.1371/journal.pbio.0060087. Koch, M.D., Fei, C., Wingreen, N.S., Shaevitz, J.W., and Gitai, Z. (2021). Competitive binding of independent extension and retraction motors explains the quantitative dynamics of type IV pili. Proc. Natl. Acad. Sci. 118 , e2014926118. https://doi.org/10.1073/pnas.2014926118. Ferrer, J.M., Lee, H., Chen, J., Pelz, B., Nakamura, F., Kamm, R.D., and Lang, M.J. (2008). Measuring molecular rupture forces between single actin filaments and actin-binding proteins. Proc. Natl. Acad. Sci. 105 , 9221–9226. https://doi.org/10.1073/pnas.0706124105. Schiavone, M., Sieczkowski, N., Castex, M., Trevisiol, E., Dague, E., and François, J.M. (2019). AFM dendritips functionalized with molecular probes specific to cell wall polysaccharides as a tool to investigate cell surface structure and organization. Cell Surf. 5 , 100027. https://doi.org/10.1016/j.tcsw.2019.100027. Charles-Orszag, A., Tsai, F.C., Bonazzi, D., Manriquez, V., Sachse, M., Mallet, A., Salles, A., Melican, K., Staneva, R., Bertin, A., et al. (2018). Adhesion to nanofibers drives cell membrane remodeling through one-dimensional wetting. Nat Commun 9 , 4450. https://doi.org/10.1038/s41467-018-06948-x. Soyer, M., Charles-Orszag, A., Lagache, T., Machata, S., Imhaus, A.F., Dumont, A., Millien, C., Olivo-Marin, J.C., and Dumenil, G. (2014). Early sequence of events triggered by the interaction of Neisseria meningitidis with endothelial cells. Cell Microbiol 16 , 878–895. https://doi.org/10.1111/cmi.12248. Coureuil, M., Lécuyer, H., Scott, M.G.H., Boularan, C., Enslen, H., Soyer, M., Mikaty, G., Bourdoulous, S., Nassif, X., and Marullo, S. (2010). Meningococcus Hijacks a β2-adrenoceptor/β-Arrestin pathway to cross brain microvasculature endothelium. Cell 143 , 1149–1160. https://doi.org/10.1016/j.cell.2010.11.035. Bernard, S.C., Simpson, N., Join-Lambert, O., Federici, C., Laran-Chich, M.-P., Maïssa, N., Bouzinba-Ségard, H., Morand, P.C., Chretien, F., Taouji, S., et al. (2014). Pathogenic Neisseria meningitidis utilizes CD147 for vascular colonization. Nat. Med. 20 , 725–731. https://doi.org/10.1038/nm.3563. Maïssa, N., Covarelli, V., Janel, S., Durel, B., Simpson, N., Bernard, S.C., Pardo-Lopez, L., Bouzinba-Ségard, H., Faure, C., Scott, M.G.H., et al. (2017). Strength of Neisseria meningitidis binding to endothelial cells requires highly-ordered CD147/β2-adrenoceptor clusters assembled by alpha-actinin-4. Nat. Commun. 8 , 15764. https://doi.org/10.1038/ncomms15764. Merz, A.J., and So, M. (1997). Attachment of piliated, Opa- and Opc- gonococci and meningococci to epithelial cells elicits cortical actin rearrangements and clustering of tyrosine-phosphorylated proteins. Infect Immun 65 , 4341–4349. Hoffmann, I., Eugène, E., Nassif, X., Couraud, P.O., and Bourdoulous, S. (2001). Activation of ErbB2 receptor tyrosine kinase supports invasion of endothelial cells by Neisseria meningitidis. J. Cell Biol. 155 , 133–143. https://doi.org/10.1083/jcb.200106148. Lambotin, M., Hoffmann, I., Laran-chich, M.P., Nassif, X., Couraud, P.O., and Bourdoulous, S. (2005). Invasion of endothelial cells by Neisseria meningitidis requires cortactin recruitment by a PI3-Kinase/Rac1 signalling pathway triggered by the lipo-oligosaccharide. J Cell Sci 118 , 3805–3816. Coureuil, M., Mikaty, G., Miller, F., Lécuyer, H., Bernard, C., Bourdoulous, S., Duménil, G., Mège, R.-M., Weksler, B.B., Romero, I.A., et al. (2009). Meningococcal type IV pili recruit the polarity complex to cross the brain endothelium. Science 325 , 83–87. https://doi.org/10.1126/science.1173196. Slanina, H., Hebling, S., Hauck, C.R., and Schubert-Unkmeir, A. (2012). Cell invasion by Neisseria meningitidis requires a functional interplay between the focal adhesion kinase, Src and cortactin. PLoS One 7 , e39613. https://doi.org/10.1371/journal.pone.0039613. Dharan, R., Goren, S., Cheppali, S.K., Shendrik, P., Brand, G., Vaknin, A., Yu, L., Kozlov, M.M., and Sorkin, R. (2022). Transmembrane proteins tetraspanin 4 and CD9 sense membrane curvature. Proc. Natl. Acad. Sci. U. S. A. 119 , e2208993119. https://doi.org/10.1073/pnas.2208993119. Wolverson, P.A., Parreira, I.F., Collins, M.O., Shaw, J.G., and Green, L.R. (2024). Dynamics of the CD9 interactome during bacterial infection of epithelial cells by proximity labelling proteomics. Preprint at bioRxiv, https://doi.org/10.1101/2024.12.13.628358 https://doi.org/10.1101/2024.12.13.628358. Fehon, R.G., McClatchey, A.I., and Bretscher, A. (2010). Organizing the cell cortex: the role of ERM proteins. Nat. Rev. Mol. Cell Biol. 11 , 276–287. https://doi.org/10.1038/nrm2866. Bulut, G., Hong, S.-H., Chen, K., Beauchamp, E.M., Rahim, S., Kosturko, G.W., Glasgow, E., Dakshanamurthy, S., Lee, H.-S., Daar, I., et al. (2012). Small molecule inhibitors of ezrin inhibit the invasive phenotype of osteosarcoma cells. Oncogene 31 , 269–281. https://doi.org/10.1038/onc.2011.245. Maier, B., Koomey, M., and Sheetz, M.P. (2004). A force-dependent switch reverses type IV pilus retraction. Proc Natl Acad Sci U A 101 , 10961–10966. https://doi.org/10.1073/pnas.0402305101. Bonsergent, E., Grisard, E., Buchrieser, J., Schwartz, O., Théry, C., and Lavieu, G. (2021). Quantitative characterization of extracellular vesicle uptake and content delivery within mammalian cells. Nat. Commun. 12 , 1864. https://doi.org/10.1038/s41467-021-22126-y. Bonsergent, E., and Lavieu, G. (2019). Content release of extracellular vesicles in a cell-free extract. FEBS Lett. 593 , 1983–1992. https://doi.org/10.1002/1873-3468.13472. Virji, M., Makepeace, K., Ferguson, D.J., Achtman, M., and Moxon, E.R. (1993). Meningococcal Opa and Opc proteins: their role in colonization and invasion of human epithelial and endothelial cells. Mol Microbiol 10 , 499–510. Weksler, B.B., Subileau, E.A., Perriere, N., Charneau, P., Holloway, K., Leveque, M., Tricoire-Leignel, H., Nicotra, A., Bourdoulous, S., Turowski, P., et al. (2005). Blood-brain barrier-specific properties of a human adult brain endothelial cell line. Faseb J 19 , 1872–1874. Nassif, X., Lowy, J., Stenberg, P., O’Gaora, P., Ganji, A., and So, M. (1993). Antigenic variation of pilin regulates adhesion of Neisseria meningitidis to human epithelial cells. Mol. Microbiol. 8 , 719–725. https://doi.org/10.1111/j.1365-2958.1993.tb01615.x. Morand, P.C., Tattevin, P., Eugene, E., Beretti, J.L., and Nassif, X. (2001). The adhesive property of the type IV pilus-associated component PilC1 of pathogenic Neisseria is supported by the conformational structure of the N-terminal part of the molecule. Mol Microbiol 40 , 846-56. Join-Lambert, O., Lecuyer, H., Miller, F., Lelievre, L., Jamet, A., Furio, L., Schmitt, A., Pelissier, P., Fraitag, S., Coureuil, M., et al. (2013). Meningococcal interaction to microvasculature triggers the tissular lesions of purpura fulminans. J. Infect. Dis. 208 , 1590–1597. https://doi.org/10.1093/infdis/jit301. Kellogg, D.S., Cohen, I.R., Norins, L.C., Schroeter, A.L., and Reising, G. (1968). Neisseria gonorrhoeae. II. Colonial variation and pathogenicity during 35 months in vitro. J. Bacteriol. 96 , 596–605. https://doi.org/10.1128/jb.96.3.596-605.1968. de Chaumont, F., Dallongeville, S., Chenouard, N., Hervé, N., Pop, S., Provoost, T., Meas-Yedid, V., Pankajakshan, P., Lecomte, T., Le Montagner, Y., et al. (2012). Icy: an open bioimage informatics platform for extended reproducible research. Nat. Methods 9 , 690–696. https://doi.org/10.1038/nmeth.2075. Paul-Gilloteaux, P., Heiligenstein, X., Belle, M., Domart, M.-C., Larijani, B., Collinson, L., Raposo, G., and Salamero, J. (2017). eC-CLEM: flexible multidimensional registration software for correlative microscopies. Nat. Methods 14 , 102–103. https://doi.org/10.1038/nmeth.4170. Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., et al. (2012). Fiji: an open-source platform for biological-image analysis. Nat. Methods 9 , 676–682. https://doi.org/10.1038/nmeth.2019. Berg, S., Kutra, D., Kroeger, T., Straehle, C.N., Kausler, B.X., Haubold, C., Schiegg, M., Ales, J., Beier, T., Rudy, M., et al. (2019). ilastik: interactive machine learning for (bio)image analysis. Nat. Methods 16 , 1226–1232. https://doi.org/10.1038/s41592-019-0582-9. Charrin, S., Le Naour, F., Oualid, M., Billard, M., Faure, G., Hanash, S.M., Boucheix, C., and Rubinstein, E. (2001). The major CD9 and CD81 molecular partner. Identification and characterization of the complexes. J. Biol. Chem. 276 , 14329–14337. https://doi.org/10.1074/jbc.M011297200. Sincock, P.M., Fitter, S., Parton, R.G., Berndt, M.C., Gamble, J.R., and Ashman, L.K. (1999). PETA-3/CD151, a member of the transmembrane 4 superfamily, is localised to the plasma membrane and endocytic system of endothelial cells, associates with multiple integrins and modulates cell function. J. Cell Sci. 112 ( Pt 6) , 833–844. https://doi.org/10.1242/jcs.112.6.833. Additional Declarations There is NO Competing Interest. Supplementary Files Supplementaryfigure.pdf Supplementary figure 1: (A) Complementary images from Figure 1A. Same representative confocal microscopy images with deconvolution of a wild type meningococcal colony adhering to human endothelial cell EA.hy926 than in Figure 1. CD9 (membrane marker, yellow) and ezrin (signaling marker, magenta) were immunostained and DAPI was used to reveal cell and bacterial DNA (cyan). Location of the orthogonal slicing (white dashed line) and different merge combinations for better appraisal of DAPI, CD9 and Ezrin staining relative to one-another. (B) Complementary images from Figure 1B. Representative correlative light and electron microscopy images obtained after repositioning confocal imaging with deconvolution with scanning electron microscopy (SEM). CD9 and ezrin were immunostained and DAPI was used to reveal cell and bacterial DNA. Black arrows indicate shifting of the bacteria between the immunofluorescence and SEM stages. CD9 and Ezrin immunofluorescence images superposition to the SEM image was done after the repositioning with the Icy software and EC-CLEM plugin. Full image and different merge combinations for better appraisal of DAPI, CD9 and Ezrin staining and SEM relative to one-another. Scale bars 5 µm. Supplementary figure 2: Complementary images from Figure 1. Representative immunofluorescence imaging of endothelial cells (A) EA.hy926, (B) hCMEC/D3 and (C) HDBEC, immunostained for endogenous tetraspanins CD9, CD81, CD151. The cells were infected with wild-type (2C4.3) and its derivative mutant for pilT (D T ) or the capsule mutant expressing OpaB adhesin ( SiaD - Opa + ) and its derivatives mutant for pilE (D E ) or for pilV (D V ). Z stack sum projection. Scale Bar 5µm (A) and 10 µm (B and C). Supplementary figure 3: (A-F) Infection with wild type 2C4.3 strain of EA.hy926 endothelial cells treated with siRNA targeting CD9 (A and D), CD81 (B and E) or CD151 (C and F). Immunofluorescence quantification of the accumulation of the targeted tetraspanin (A-C, respectively) or Ezrin (D-F) below the bacterial colonies, normalized with the volume of DAPI fluorescence of the corresponding colonies. Dashed line is the detection threshold. Three experiments were pooled. Data are medians ± interquartile range. (G, H, I) Phenotypes of triple CRISPR knockouts EA.hy926 cells. (G) Successive CRISPR knockouts were conducted in cells through lentivirus transfection to result in cell lines depleted of tetraspanins CD9, CD81, CD151 and all three. The purity of the KOs was assessed for each cell line through flow cytometry. Negative controls are stained with secondary antibody alone. The vertical black bar is a visual aid to separate assessed negative (left) or positive (right) staining. (H) Quantification of Ezrin accumulation in triple KO cells and below meningococcal colonies, normalized with DAPI fluorescence (Ezrin accumulation index). The dashed line represents the detection threshold. Three experiments were pooled. Data are presented as means ± 95% confidence intervals; Mann Whitney test. (I) Adhesion experiment of N. meningitidis strain 2C4.3 on wild type or triple KO cells. CFU mean percentage of the adhesion on wild type cells. Three experiments were pooled. Data are presented as means ± SD; Mann-Whitney test. Supplementary figure 4: First raw: representative confocal microscopy and DIC images of a wild type N. meningitidis (strain 2C4.3) colony adhering to EA.hy926 endothelial cells. The b2-adrenergic receptor tagged with YFP and Ezrin are both accumulated below a colony. Raw 2 to 4: representative microscopy images of EA.hy926 endothelial cells expressing CXCR4-YFP, CCD5, YFP, AT1R-YFP, CD4-YFP, Transferrin-receptor (TfR)-GFP, ACKR2-GFP (confocal microscopy), or TLR2-GFP, TLR4-GFP (apotome microscopy), or stained for endogenous CD44 (apotome microscopy). Scale bar 10µm (confocal microscopy) or 5 µm (apotome microscopy). Supplementary figure 5: (A) Complementary images from Figure 4K. Representative apotome microscopy images of meningococcal colonies mutant for pilT (D T ) adhering PM sheets of EA.hy926 endothelial cells. CD9 immunostaining reveal tubular membrane structures. DAPI was used to reveal bacterial DNA. Z-stack sum projection. White arrows indicate CD9 accumulation. Scale bar 5 µm. (B) Complementary images from Figure 5A, B. Quantification of ATP concentration in cells treated (2-DG) with 10mM 2-DG or not (control). Three experiments were pooled. Data are presented as means % of control ± SD. One sample t test. Supplementary figure 6: Schematic representation of our proposed model of N. meningitidis -cell interaction, integrating adhesion, tubular membrane structures and induced signalling. From left to right: (i) Meningococci interact with endothelial cells via the PilC1 adhesin. (ii) After interaction with an unknown PilC1 ligand, the cell membrane expands along T4P. This tubulation of the membrane is only possible during the few seconds of T4P immobility. This promotes the accumulation and clustering of receptors, allowing the T4P adhesins to interact with the signalling receptors CD147 and b2AR. Early tubular membrane structures can be revealed in fixed cells or using a pilT deletion mutant. (iii) Finally, T4P pulls on the plasma membrane, activating signalling via the CD147 and b2AR receptors, leading to the accumulation of ezrin and actin and the stabilization of large tubular membrane structures, which are the site of further accumulation of integral membrane proteins. This triggers an amplification loop leading to an increase in the membrane surface area to which bacteria can adhere and be protected from the shear stress of blood flow. Cite Share Download PDF Status: Published Journal Publication published 25 Nov, 2025 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-6302813","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":453383808,"identity":"ede586b6-565c-45a4-9314-99559b6c3a8c","order_by":0,"name":"Mathieu Coureuil","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAUlEQVRIie3OMUvDQBTA8VcO2uW1WV+o5BsIBweH4OBXSSno4iaIg0ghkCwBV/s5BCeHg4O4HGYVshQKnVNcMqh4SoqKubg63H96x92PewA+3/9sCPHXgfYC1o6jhUuwnwTDpB1R9ZDvId+9dJH9rCxWqwsNQZbLepseoHgY63p7fwc47TbSzEc8NhrIGLFcpoRST47Dm00FOIm7iZoPaZZq4HQq2PiToGSoKjjC7sVkubbkrSWvlogExfOLJegiTx+/LFoysIQz5FPoJWtJcXGCZIqzQf5ISHaxMFcVOkk524TN5WEUZMktNOdXUXBtRN2oKnKRXb+v/wA+n8/n6+sdhaxSmtq/vL0AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-7655-9685","institution":"Inserm","correspondingAuthor":true,"prefix":"","firstName":"Mathieu","middleName":"","lastName":"Coureuil","suffix":""},{"id":453383809,"identity":"a71d8296-c582-4ab7-976e-4ada4de4d2c1","order_by":1,"name":"Audrey LAURENT","email":"","orcid":"","institution":"INEM","correspondingAuthor":false,"prefix":"","firstName":"Audrey","middleName":"","lastName":"LAURENT","suffix":""},{"id":453383810,"identity":"6c8c294b-7135-4a1a-a05e-8770c7ba00f7","order_by":2,"name":"Kevin SOLLIER","email":"","orcid":"","institution":"Institut Cochin","correspondingAuthor":false,"prefix":"","firstName":"Kevin","middleName":"","lastName":"SOLLIER","suffix":""},{"id":453383811,"identity":"b3ed56f4-b6dd-49ca-8f52-efc15fa89377","order_by":3,"name":"Bruno Saubamea","email":"","orcid":"https://orcid.org/0000-0002-6218-0460","institution":"Cellular and Molecular Imaging facility, INSERM US25, UMS3612 CNRS, Faculté de Pharmacie de Paris, Université de Paris, Paris, France","correspondingAuthor":false,"prefix":"","firstName":"Bruno","middleName":"","lastName":"Saubamea","suffix":""},{"id":453383812,"identity":"fc274bc4-304a-4772-abba-61dccc967645","order_by":4,"name":"Virginie MIGNON","email":"","orcid":"","institution":"Faculté de pharmacie","correspondingAuthor":false,"prefix":"","firstName":"Virginie","middleName":"","lastName":"MIGNON","suffix":""},{"id":453383813,"identity":"eade552e-29cf-4248-a09e-f59d10780028","order_by":5,"name":"Nicolas GOUDIN","email":"","orcid":"https://orcid.org/0000-0002-3967-4016","institution":"SFR Necker","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"GOUDIN","suffix":""},{"id":453383814,"identity":"3f1861a7-c8e6-43f5-ad4a-9fcd07337536","order_by":6,"name":"Yaelle WORMSER","email":"","orcid":"https://orcid.org/0009-0003-7997-9425","institution":"INEM","correspondingAuthor":false,"prefix":"","firstName":"Yaelle","middleName":"","lastName":"WORMSER","suffix":""},{"id":453383815,"identity":"bfbda324-dd21-4188-bfcc-3201412409be","order_by":7,"name":"Morgane Wuckelt","email":"","orcid":"","institution":"Université Paris Cité, INSERM U1151, CNRS UMR8253, Institut Necker-Enfants Malades","correspondingAuthor":false,"prefix":"","firstName":"Morgane","middleName":"","lastName":"Wuckelt","suffix":""},{"id":453383816,"identity":"9c0f2571-b3ac-462f-b52b-dc776f4e908b","order_by":8,"name":"Mahmoud RIFAI","email":"","orcid":"https://orcid.org/0009-0008-6012-2890","institution":"INEM","correspondingAuthor":false,"prefix":"","firstName":"Mahmoud","middleName":"","lastName":"RIFAI","suffix":""},{"id":453383817,"identity":"0b29213f-f912-4006-9238-cc72506a7a37","order_by":9,"name":"Thomas HENG","email":"","orcid":"","institution":"INEM","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"HENG","suffix":""},{"id":453383818,"identity":"47bbd2d0-0903-413b-98df-0d0338e5b422","order_by":10,"name":"Lya L’HERMITTE","email":"","orcid":"","institution":"INEM","correspondingAuthor":false,"prefix":"","firstName":"Lya","middleName":"","lastName":"L’HERMITTE","suffix":""},{"id":453383819,"identity":"c82c7156-fc09-47a3-8aad-115a7ec39044","order_by":11,"name":"Marta CONFLITTI","email":"","orcid":"","institution":"INEM","correspondingAuthor":false,"prefix":"","firstName":"Marta","middleName":"","lastName":"CONFLITTI","suffix":""},{"id":453383820,"identity":"1cfb512f-03ba-4915-801a-88d2136f970f","order_by":12,"name":"Julie Meyer","email":"","orcid":"https://orcid.org/0000-0002-6144-927X","institution":"INSERM","correspondingAuthor":false,"prefix":"","firstName":"Julie","middleName":"","lastName":"Meyer","suffix":""},{"id":453383821,"identity":"3d4f3aae-cc61-463e-93f0-4153e7aca1a9","order_by":13,"name":"Hervé Lécuyer","email":"","orcid":"","institution":"Institut Necker Enfants Malades, Université Paris Descartes, INSERM U1151, CNRS UMR8253 \u0026 AP-HP, Hôpital Necker Enfants Malades, Service de Microbiologie Clinique","correspondingAuthor":false,"prefix":"","firstName":"Hervé","middleName":"","lastName":"Lécuyer","suffix":""},{"id":453383822,"identity":"1fac7c3c-7af0-4e15-ae14-1c06384608c1","order_by":14,"name":"Anne Jamet","email":"","orcid":"https://orcid.org/0000-0002-1902-6943","institution":"Université Paris Cité","correspondingAuthor":false,"prefix":"","firstName":"Anne","middleName":"","lastName":"Jamet","suffix":""},{"id":453383823,"identity":"372d5a1d-6ab1-44cc-981e-f1613cd7acfc","order_by":15,"name":"Nicolas BORGHI","email":"","orcid":"","institution":"Institut Jacques Monod","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"BORGHI","suffix":""},{"id":453383824,"identity":"44347319-4cf6-4552-b110-7af2e901ba83","order_by":16,"name":"Philippe GIRARD","email":"","orcid":"","institution":"Institut Jacques Monod","correspondingAuthor":false,"prefix":"","firstName":"Philippe","middleName":"","lastName":"GIRARD","suffix":""},{"id":453383825,"identity":"eab5650d-bfa1-453a-9a1b-d304758192b5","order_by":17,"name":"Emmanuelle Bille","email":"","orcid":"https://orcid.org/0000-0001-9172-0510","institution":"Université Paris Cité, INSERM U1151, CNRS UMR8253, Institut Necker-Enfants Malades","correspondingAuthor":false,"prefix":"","firstName":"Emmanuelle","middleName":"","lastName":"Bille","suffix":""},{"id":453383826,"identity":"e6149848-fc61-4599-92c6-dd1b30654079","order_by":18,"name":"Gregory LAVIEU","email":"","orcid":"https://orcid.org/0000-0001-9852-4977","institution":"Université Paris Cité","correspondingAuthor":false,"prefix":"","firstName":"Gregory","middleName":"","lastName":"LAVIEU","suffix":""},{"id":453383827,"identity":"88f3d9db-a55d-4dbe-8a94-5ebc5548300f","order_by":19,"name":"Eric RUBINSTEIN","email":"","orcid":"https://orcid.org/0000-0001-7623-9665","institution":"CimiParis","correspondingAuthor":false,"prefix":"","firstName":"Eric","middleName":"","lastName":"RUBINSTEIN","suffix":""},{"id":453383828,"identity":"747a60f9-2f19-4414-97cc-49a33af9391d","order_by":20,"name":"Stefano Marullo","email":"","orcid":"https://orcid.org/0000-0001-9604-9973","institution":"Université de Paris, Institut Cochin","correspondingAuthor":false,"prefix":"","firstName":"Stefano","middleName":"","lastName":"Marullo","suffix":""}],"badges":[],"createdAt":"2025-03-25 10:26:03","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6302813/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6302813/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-025-65436-1","type":"published","date":"2025-11-25T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":82352464,"identity":"dde2241f-168d-4a47-b6ee-9d7bf1a8900b","added_by":"auto","created_at":"2025-05-09 11:01:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1572104,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe membrane marker CD9 accumulates in the tubular plasma membrane structures induced by \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eN. meningitidis\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e, in contrast to the signaling marker ezrin.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Representative confocal microscopy images with deconvolution of a wild type meningococcal colony adhering to human endothelial cell EA.hy926. CD9 (membrane marker, yellow) and ezrin (signaling marker, magenta) were immunostained and DAPI was used to reveal cell and bacterial DNA (cyan). On the top, split and merged channel. On the bottom, orthogonal XZ slice. (B) Representative correlative light and electron microscopy images obtained after repositioning confocal imaging with deconvolution with scanning electron microscopy (SEM) using Icy software and EC-CLEM plugin. CD9 and ezrin were immunostained and DAPI was used to reveal cell and bacterial DNA. White arrows and dashed black line indicate locations with a clear overlap between CD9 staining and tubular membrane structures where ezrin staining is absent. (C) Schematics of \u003cem\u003eN. meningitidis\u003c/em\u003e type IV pili (T4P), showing the main pilin PilE (blue), the adhesins PilC (green) and PilV (purple), and the retraction motor PilT (black). (D, E) Representative images of endothelial cells and colonies of adherent non-encapsulated meningococcal expressing OpaB (\u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u003c/sup\u003e \u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e) or the derivative deletion mutants for \u003cem\u003epilE \u003c/em\u003e(D\u003cem\u003eE\u003c/em\u003e), \u003cem\u003epilV \u003c/em\u003e(D\u003cem\u003eV\u003c/em\u003e) and \u003cem\u003epilC1 \u003c/em\u003e(D\u003cem\u003eC1\u003c/em\u003e). (D) Representative SEM images. (E) Representative immunofluorescence microscopy images of CD9. Bacterial colonies are indicated by white dashed circles. Z-stack sum projection. Scale bars 5 µm.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/aedbdb492334d97e3a710f0d.png"},{"id":82352466,"identity":"4b43d188-7cab-4a66-b793-aa0604a4e592","added_by":"auto","created_at":"2025-05-09 11:01:42","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":411790,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eN. meningitidis\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e promotes the sequestration of transmembrane proteins in host plasma-membrane protrusions.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A-D) FRAP experiments of fluorescent transmembrane proteins expressed in EA.hy926 cells infected with the wild type 2C4.3 strain. (A) Representative FRAP images sequence of plasma membrane (top) and tubular membrane structures (TMS) (bottom) of living EA.hy926 cells expressing YFP-CD9. A 1 µm² region of interest (ROI, red square) was photobleached with high laser power, and fluorescence recovery was monitored within the ROI over time. Scale bars 2 µm. (B) Normalized YFP-CD9, (C) β2AR-YFP and (D) CD4-YFP fluorescence intensity in plasma membranes (PM, grey) and tubular membrane structures (blue) over time were plotted. Thirty cells from 3 independent experiments were analyzed per condition. Data are presented as means ± SD. Unpaired \u003cem\u003et‑\u003c/em\u003etest (at 16s). (E-F) Quantification of transmembrane proteins accumulation in \u003cem\u003eN. meningitidis \u003c/em\u003einduced tubular membrane structures. (E) Example of representative living EA.hy926 cells accumulating YFP‑CD9 in tubular membrane structures. Maximum intensity values in tubular membrane structures (TMS, blue) and filipodia (F, red) were acquired on ImageJ from a trace line and normalized with plasma membranes values and quantified in (F). Filopodia and tubular membrane structures were studied on two different horizontal slices. YFP‑CD9 or b2AR‑YFP or CD4‑YFP were transfected in EA.hy926 cells. The membrane was stained using the Flipper probe that has been added just before infection. Thirty cells from 3 independent experiments were analyzed per condition. Data are presented as means ± SD. Kruskal-Wallis test with Dunn’s correction.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/f194200961cc1192321ad215.png"},{"id":82352473,"identity":"b31a6677-c041-4d86-b0ee-6fe7091c1445","added_by":"auto","created_at":"2025-05-09 11:01:42","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":461042,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eInitiation of tubular membrane structures in host cells is independent of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eN. meningitidis \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003einduced-signaling.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A-C) Impact of the inhibition of Ezrin Thr567 phosphorylation (3 µM; NSC668394 (NSC66)) in endothelial cells after infection with wild type meningococci (2C4.3). (A) Representative images of endothelial cells immunostained for Ezrin. Z-stack sum projection. Scale bar 5 µm. (B) Quantification of Ezrin accumulation below meningococcal colonies, normalized with DAPI fluorescence (Ezrin accumulation index) in DMSO- or NSC668394-treated endothelial cells. Dashed line represents the detection threshold. Three experiments were pooled. Data are presented as means ± 95% confidence intervals; Mann Whitney test. (C) Adhesion experiment of \u003cem\u003eN. meningitidis \u003c/em\u003estrain 2C4.3 on DMSO- or NSC668394-treated endothelial cells. CFU mean percentage of the cells treated with DMSO only. Data are presented as means ± SD; One way ANOVA with Bonferroni correction. (D) Quantification of CD9 accumulation below wild type (2C4.3) or its derivative mutant for \u003cem\u003epilT \u003c/em\u003e(D\u003cem\u003eT\u003c/em\u003e) meningococcal colonies in DMSO- or NSC668394-treated endothelial cells. Three experiments were pooled. Data are presented as means ± 95% confidence intervals; Kruskal-Wallis test with Dunn’s correction. (E-F) Representative SEM images of endothelial cells infected with 2C4.3 or its Δ\u003cem\u003eT\u003c/em\u003e derivative mutant after treatment with DMSO. Tubular membrane structures were digitally colored in orange. Scale bar 1µm. (G) FRAP experiments of YFP-CD9 expressed in EA.hy926 cells infected with 2C4.3 strain or with \u003cem\u003eΔT \u003c/em\u003estrain. FRAP was performed in tubular membrane structures (TMS) of cells treated with DMSO (grey). Thirty cells from 3 independent experiments were analyzed per condition. Data are presented as means ± SD. Unpaired \u003cem\u003et‑\u003c/em\u003etest (at 16s).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/4d20b3c3f3b82391bb91d074.png"},{"id":82354185,"identity":"ae457df4-ea62-4dbd-9432-d49aa4685d3a","added_by":"auto","created_at":"2025-05-09 11:09:42","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1443710,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePlasma membrane tubulation depends only on the plasma membrane dynamic.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A-G) Infection of living (grey) or prefixed with 4% PFA (blue) EA.hy926 endothelial cells with the 2C4.3 strain or its derivative deletion mutant for \u003cem\u003epilT \u003c/em\u003e(Δ\u003cem\u003eT\u003c/em\u003e). (A) Representative images of CD9 and ezrin immunostaining at the site meningococcal adhesion. CD9 (yellow) remains present in fixed cells in the absence of ezrin (magenta). DAPI was used to reveal bacterial DNA (cyan). On the left part: Z-stack sum projection. On the right part: Orthogonal slices. White arrows indicate puncta of CD9 accumulation. The red dashed lines indicate the position of the orthogonal slices. The yellow arrows and yellow dashed lines were added to represent the height of tubular membrane structures. Scale bar 5 µm. (B) Immunofluorescence quantification of CD9 accumulation below the meningococcal colonies, normalized with DAPI fluorescence of the corresponding colonies. Three experiments were pooled and analyzed with Kruskal-Wallis test with Dunn’s correction. Data are means ± 95% confidence intervals. (C) Maximum height of tubular membrane structures were estimated from CD9 immunofluorescence imaging. The orthogonal slices in A were only added to help visualizing the membrane structures. At least 90 colonies were studied from three independent experiments. One-way ANOVA with Bonferroni’s correction. Data are means ± SD. (D-G) Representative SEM images of infected endothelial cells. Tubular membrane structures were digitally colorized in orange. Scale bar 1µm. (H, I) FRAP Experiments of YFP-CD9 expressed in EA.hy926 cells. (H) FRAP experiments on tubular membrane structures (TMS) in living cells infected with the 2C4.3 strain (grey), prefixed cells infected with the 2C4.3 strain (blue) or its derivative deletion mutant for \u003cem\u003epilT \u003c/em\u003e(Δ\u003cem\u003eT\u003c/em\u003e, dark red). (I) FRAP experiments on plasma membrane (PM, green) or tubular membrane structures(TMS, blue) of prefixed cells infected with 2C4.3. Thirty cells from 3 independent experiments were analyzed per condition. Data are presented as means ± SD. Unpaired \u003cem\u003et‑\u003c/em\u003etest (at 16s). (J) Quantification of YFP-CD9 protein accumulation in \u003cem\u003eN. meningitidis \u003c/em\u003eD\u003cem\u003eT \u003c/em\u003einduced membrane structures, as in figure 2F. Maximum intensity values in tubular membrane structures (TMS, blue) and filipodia (F, red) were acquired on ImageJ from a trace line and normalized with plasma membranes values and quantified in prefixed EA.hy926 cells. Thirty cells from 3 independent experiments were analyzed per condition. Data are presented as means ± SD. Mann Whitney test. (K, L) Tubular membrane structures in PM sheets of EA.hy926 endothelial cells infected with the 2C4.3 strain (grey) or its derivative deletion mutant for \u003cem\u003epilT \u003c/em\u003e(Δ\u003cem\u003eT\u003c/em\u003e, dark red). (K) Representative images of CD9 immunostaining (yellow) at the site meningococcal adhesion. DAPI was used to reveal cell and bacterial DNA (cyan). Z-stack sum projection. White arrows indicate CD9 accumulation. Scale bar 5 µm. (L) Quantification of YFP-CD9 protein accumulation in \u003cem\u003eN. meningitidis \u003c/em\u003einduced tubular membrane structures, as above. Data are presented as means ± SD. Unpaired t test with Welch's correction.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/f93c5ae0273386a63b719262.png"},{"id":82352470,"identity":"3f8899b3-4f3a-4ec5-92c0-301a09355d14","added_by":"auto","created_at":"2025-05-09 11:01:42","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":588839,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eActin depolymerization frees plasma membrane tubulation upon infection with Δ\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003epilT \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003estrain.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A-C) Infection of EA.hy926 endothelial cells with the 2C4.3 strain or its derivative deletion mutant for \u003cem\u003epilT \u003c/em\u003e(Δ\u003cem\u003eT\u003c/em\u003e), with 2-DG (10mM, blue) or without (control, grey). (A) Representative images of CD9 and ezrin immunostaining at the site meningococcal adhesion. On the left part: Z-stack sum projection. On the right part: Orthogonal slices. The red dashed lines indicate the position of the orthogonal slices. The yellow arrows and yellow dashed lines were only added to represent the height of plasma membrane protrusions. Scale bar 5 µm. (B, C) Maximum height of tubular membrane structures were estimated from CD9 immunofluorescence imaging. Cells were infected after treatment with 2-DG (B) or cytochalasin D (CytD, light blue) and Latruncumin B (LatB, dark blue) (C) or without (control, grey). The orthogonal slices in A were only added to help visualize the height of plasma membrane protrusions. At least 90 colonies were studied from three independent experiments. Kruskal-Wallis test with Dunn’s correction. Data are means ± SD. Brown-Forsythe and Welch ANOVA with Games-Howell’s correction.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/222f6b330e36e0d68cf62881.png"},{"id":82354186,"identity":"a041b327-f2a3-44dd-91f5-88c284c5f311","added_by":"auto","created_at":"2025-05-09 11:09:42","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":918329,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eN. meningitidis-\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003einduced tubular membrane structures can be recapitulate in cells incompetent for meningococcal induced signaling.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Representative confocal and DIC images of HEK cells or MDCK cells expressing YFP-CD9 (yellow) infected with the non piliated 2C4.3 \u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u003c/sup\u003e \u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e Δ\u003cem\u003eE\u003c/em\u003e strain or the non-retracting 2C4.3 \u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u003c/sup\u003e \u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e Δ\u003cem\u003eT\u003c/em\u003e strain. Confocal images were acquired on different z positions: 1 (on the top of cells, blue) and 2 (on the bottom of cells, red). Bacterial colonies are indicated by white dashed shapes. Scale bar: 10µm. (B) FRAP Experiments of YFP-CD9 expressed in EA.hy926 cells on plasma membrane (PM, grey) or tubular membrane structures (TMS, dark red) in living cells infected with the \u003cem\u003epilT \u003c/em\u003emutant\u003cem\u003e \u003c/em\u003e(Δ\u003cem\u003eT\u003c/em\u003e). Thirty cells from 3 independent experiments were analyzed per condition. Data are presented as means ± SD. Unpaired \u003cem\u003et‑\u003c/em\u003etest (at 16s). (C) Representative confocal and DIC images of prefixed HEK cells expressing YFP-CD9 (yellow) infected with the 2C4.3 \u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u003c/sup\u003e \u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e Δ\u003cem\u003eT\u003c/em\u003e strain. One representative z position has been selected. Scale bar: 10µm.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/a64f0a8c7e82d60a1a22ae98.png"},{"id":82352467,"identity":"2a64204a-de9c-46dc-8bc9-f22581808a5a","added_by":"auto","created_at":"2025-05-09 11:01:42","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":72881,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePilC1-dependent adhesion is not rescued by PilT mutation, despite PilV expression.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Adhesion experiment of \u003cem\u003eN. meningitidis \u003c/em\u003estrain 2C4.3 or its derivate mutant for \u003cem\u003epilT \u003c/em\u003e(D\u003cem\u003eT\u003c/em\u003e) and \u003cem\u003epilC1 \u003c/em\u003e(D\u003cem\u003eC1\u003c/em\u003e) on living (grey) or prefixed (blue) EA.hy926 endothelial cells. Mean CFU. At least three independent experiments performed in duplicate were quantified per condition. Data are presented as Geometric means CFU ± 95% confidence interval; One way ANOVA with Bonferroni's correction. (B) Expression of PilV in pili preparation (T4P) and whole bacterial lysate (lysate) of the wild type strain 2C4.3 and its derivate mutant for \u003cem\u003epilT \u003c/em\u003eand \u003cem\u003epilC1 \u003c/em\u003e(D\u003cem\u003eTC1\u003c/em\u003e). On the left, western blotting using anti-PilV antibodies. On the right, total proteins were detected in the gel by Stain-Free technology.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/22c596ca996a980d5301b983.png"},{"id":96798921,"identity":"216894e7-1f83-4b72-829e-fd3556ebc2cb","added_by":"auto","created_at":"2025-11-26 08:17:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6987432,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/36f13a7f-51c4-44fb-86ee-d62ddf7bd2a5.pdf"},{"id":82352465,"identity":"f90fa6af-3126-4bab-bb0c-2635a89d965a","added_by":"auto","created_at":"2025-05-09 11:01:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3996779,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary figure 1:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Complementary images from Figure 1A. Same representative confocal microscopy images with deconvolution of a wild type meningococcal colony adhering to human endothelial cell EA.hy926 than in Figure 1. CD9 (membrane marker, yellow) and ezrin (signaling marker, magenta) were immunostained and DAPI was used to reveal cell and bacterial DNA (cyan). Location of the orthogonal slicing (white dashed line) and different merge combinations for better appraisal of DAPI, CD9 and Ezrin staining relative to one-another. (B) Complementary images from Figure 1B. Representative correlative light and electron microscopy images obtained after repositioning confocal imaging with deconvolution with scanning electron microscopy (SEM). CD9 and ezrin were immunostained and DAPI was used to reveal cell and bacterial DNA. Black arrows indicate shifting of the bacteria between the immunofluorescence and SEM stages. CD9 and Ezrin immunofluorescence images superposition to the SEM image was done after the repositioning with the Icy software and EC-CLEM plugin. Full image and different merge combinations for better appraisal of DAPI, CD9 and Ezrin staining and SEM relative to one-another. Scale bars 5 µm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary figure 2:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eComplementary images from Figure 1. Representative immunofluorescence imaging of endothelial cells (A) EA.hy926, (B) hCMEC/D3 and (C) HDBEC, immunostained for endogenous tetraspanins CD9, CD81, CD151. The cells were infected with wild-type (2C4.3) and its derivative mutant for \u003cem\u003epilT \u003c/em\u003e(D\u003cem\u003eT\u003c/em\u003e) or the capsule mutant expressing OpaB adhesin (\u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u003c/sup\u003e \u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e) and its derivatives mutant for \u003cem\u003epilE \u003c/em\u003e(D\u003cem\u003eE\u003c/em\u003e) or for \u003cem\u003epilV \u003c/em\u003e(D\u003cem\u003eV\u003c/em\u003e). Z stack sum projection. Scale Bar 5µm (A) and 10 µm (B and C).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary figure 3:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A-F) Infection with wild type 2C4.3 strain of EA.hy926 endothelial cells treated with siRNA targeting CD9 (A and D), CD81 (B and E) or CD151 (C and F). Immunofluorescence quantification of the accumulation of the targeted tetraspanin (A-C, respectively) or Ezrin (D-F) below the bacterial colonies, normalized with the volume of DAPI fluorescence of the corresponding colonies. Dashed line is the detection threshold. Three experiments were pooled. Data are medians ± interquartile range. (G, H, I) Phenotypes of triple CRISPR knockouts EA.hy926 cells. (G) Successive CRISPR knockouts were conducted in cells through lentivirus transfection to result in cell lines depleted of tetraspanins CD9, CD81, CD151 and all three. The purity of the KOs was assessed for each cell line through flow cytometry. Negative controls are stained with secondary antibody alone.\u0026nbsp; The vertical black bar is a visual aid to separate assessed negative (left) or positive (right) staining. (H) Quantification of Ezrin accumulation in triple KO cells and below meningococcal colonies, normalized with DAPI fluorescence (Ezrin accumulation index). The dashed line represents the detection threshold. Three experiments were pooled. Data are presented as means ± 95% confidence intervals; Mann Whitney test. (I) Adhesion experiment of \u003cem\u003eN. meningitidis \u003c/em\u003estrain 2C4.3 on wild type or triple KO cells. CFU mean percentage of the adhesion on wild type cells. Three experiments were pooled. Data are presented as means ± SD; Mann-Whitney test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary figure 4:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFirst raw: representative confocal microscopy and DIC images of a wild type \u003cem\u003eN. meningitidis \u003c/em\u003e(strain 2C4.3) colony adhering to EA.hy926 endothelial cells. The b2-adrenergic receptor tagged with YFP and Ezrin are both accumulated below a colony. Raw 2 to 4: representative microscopy images of EA.hy926 endothelial cells expressing CXCR4-YFP, CCD5, YFP, AT1R-YFP, CD4-YFP, Transferrin-receptor (TfR)-GFP, ACKR2-GFP (confocal microscopy), or TLR2-GFP, TLR4-GFP (apotome microscopy), or stained for endogenous CD44 (apotome microscopy). Scale bar 10µm (confocal microscopy) or 5 µm (apotome microscopy).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary figure 5:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Complementary images from Figure 4K. Representative apotome microscopy images of meningococcal colonies mutant for \u003cem\u003epilT \u003c/em\u003e(D\u003cem\u003eT\u003c/em\u003e) adhering PM sheets of EA.hy926 endothelial cells. CD9 immunostaining reveal tubular membrane structures. DAPI was used to reveal bacterial DNA. Z-stack sum projection. White arrows indicate CD9 accumulation. Scale bar 5 µm. (B) Complementary images from Figure 5A, B. Quantification of ATP concentration in cells treated (2-DG) with 10mM 2-DG or not (control). Three experiments were pooled. Data are presented as means % of control ± SD. One sample \u003cem\u003et\u003c/em\u003e test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary figure 6:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSchematic representation of our proposed model of \u003cem\u003eN. meningitidis\u003c/em\u003e-cell interaction, integrating adhesion, tubular membrane structures and induced signalling. From left to right: (\u003cstrong\u003ei\u003c/strong\u003e) Meningococci interact with endothelial cells via the PilC1 adhesin. (\u003cstrong\u003eii\u003c/strong\u003e) After interaction with an unknown PilC1 ligand, the cell membrane expands along T4P. This tubulation of the membrane is only possible during the few seconds of T4P immobility. This promotes the accumulation and clustering of receptors, allowing the T4P adhesins to interact with the signalling receptors CD147 and b2AR. Early tubular membrane structures can be revealed in fixed cells or using a pilT deletion mutant. (\u003cstrong\u003eiii\u003c/strong\u003e) Finally, T4P pulls on the plasma membrane, activating signalling via the CD147 and b2AR receptors, leading to the accumulation of ezrin and actin and the stabilization of large tubular membrane structures, which are the site of further accumulation of integral membrane proteins. This triggers an amplification loop leading to an increase in the membrane surface area to which bacteria can adhere and be protected from the shear stress of blood flow.\u003c/p\u003e","description":"","filename":"Supplementaryfigure.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6302813/v1/58433f254be1cc338af05511.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Meningococci drive host membrane tubulation to recruit their signaling receptors","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eSignal activation in pathogen-infected cells, an essential event in the pathophysiology of infection, is currently viewed as a cascade of biochemical reactions elicited by the interaction of pathogen ligands with host cell receptors. In general, soluble ligands of biological fluids are in large excess over cognate plasma membrane receptors, granting a high probability of receptor binding and activation. However, many bacterial ligands are anchored to bacterial surfaces, such as the outer-membrane and the cell wall, or pathogen appendages, and their overall concentration is markedly lower than that of soluble ligands. Consequently, the probability of this type of ligand to eventually meet its receptor is theoretically much lower, especially if cognate receptors are expressed at low density. The challenge of \"finding the relevant receptor\" is particularly critical for bacteria, as they rely on interactions with host receptors for survival. One example is the model organism \u003cem\u003eListeria monocytogenes\u003c/em\u003e, for which the low probability of establishing a stable interaction with host tissues is balanced by the typically large number of infecting bacteria: 10 to 100\u0026nbsp;million bacteria are necessary for gut infection \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e or for neuro-invasion, while 10 thousand bacteria are required to infect hosts by intravenous injection \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. In addition, infectious bacteria present in the bloodstream generally have a limited time window to bind to and infect target tissues before being wiped out by host defense mechanisms.\u003c/p\u003e \u003cp\u003eIntriguingly, a poorly understood facilitating mechanism has been developed by \u003cem\u003eNeisseria meningitidis\u003c/em\u003e, for which only 5 diplococci are sufficient to cause septicaemia in mice \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. This extremely low number of bacteria is particularly remarkable since \u003cem\u003eN. meningitidis\u003c/em\u003e do not survive the resident immune cells checkpoints in the liver and the bladder. To infect endothelial cells lining blood vessels this pathogen requires stable adhesion to endothelial cells within a time window of a few seconds, despite the continuous shear forces exerted by the blood-flow \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. \u003cem\u003eN. meningitidis\u003c/em\u003e adhesion relies on their type IV pili (T4P) and on T4P-induced endothelial cell plasma membrane remodeling, consisting of long tubular villi-like structures emerging from the cell surface, designed from now on as \u0026ldquo;tubular membrane structures\u0026rdquo;. The mechanisms leading to the formation of these key structures, which are mandatory for tissue colonization \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e, and their role in early adhesion remain largely unknown.\u003c/p\u003e \u003cp\u003eT4P pili are common organelles made of filamentous polymers, belonging to the type IV filament family, which display a critical role in monoderm and diderm bacteria and in archaea \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. They are involved in aggregation, biofilm formation, DNA uptake, mechano-sensing, adhesion / interaction with biotic and abiotic surfaces, and are essential for bacterial virulence and pathogenesis \u003csup\u003e\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. T4P mediate the interaction with cell receptors and the activation of signaling pathways in host cells by many human pathogens, such as \u003cem\u003eP. aeruginosa\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eNeisseria meningitidis\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eNeisseria gonorrhoeae\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eClostridium difficile\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e, enteropathogenic or enterotoxigenic \u003cem\u003eE. coli\u003c/em\u003e (EPEC and ETEC) \u003csup\u003e\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e and \u003cem\u003eStreptococcus sanguinis\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Only a few studies were conducted to elucidate the mechanisms of this interaction and identify T4P molecular targets. Most of them pointed out the property of T4P to interact with glycans present on host cell receptors \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e or carried by mucins of the mucus layer protecting epithelial cells \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eT4P are dynamic structures, and their biogenesis relies on complex multi-protein machinery. Although the nomenclature differs between bacterial species \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, 16 widely conserved essential proteins contribute to T4P biogenesis in pathogenic meningococci (PilC1 and C2, D, E, F, G, H, I, J, K, M, N, O, P, Q, W). Among them, PilE corresponds to the main constitutive protein (also named pilin) of the pilus fiber, PilC1 and PilC2 to the pilus-tip adhesins \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Only PilC1 is an adhesin for endothelial cells, a PilC1 null mutant only expressing PilC2 being piliated and non-adhesive. T4P are continuously elongated and retracted thanks to PilF and PilT ATPases, respectively \u003csup\u003e\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. In addition, the minor pilin PilV is distributed along the fiber and necessary for adhesion and signaling in host cells \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. The cycles of elongation/retraction occur stochastically within few seconds and generate mechanical forces of 50\u0026ndash;100 piconewtons (pN) for each filament, reaching up to 1 nanonewton for a bundle containing 8 to 10 pili \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. These forces are larger than those required to break protein-protein or protein-glycan bonds, which are in the order of a few dozen to a hundred of pN \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Since T4P retraction forces are likely to rapidly disrupt the interaction between T4P and cognate receptors, the firm adhesion of bacteria to host cells must require some additional stabilizing mechanism(s) provided by host-cells. It was proposed, for example, that meningococci might exploit the \u0026ldquo;one dimensional\u0026rdquo; wetting property of the plasma membrane to form tubular-like structures along T4P \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Subsequent actin polymerization was proposed to stabilize these membrane structures to which meningococci are anchored \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. The stable interaction between meningococcal T4P and human endothelial cells relies on the accumulation and mechanical activation of signaling complexes, consisting of CD147 and the β2-adrenergic receptor (β2AR), cross-linked by the cytoskeletal protein α-actinin-4 \u003csup\u003e13,34\u0026ndash;36,19,3\u003c/sup\u003e. PilV and PilV are both needed to initiate bacterial adhesion: the first interaction of the pilus-tip PilV with host cells is thought to be reinforced by the interaction of the secondary adhesin PilV with plasma membrane receptors. A possible model including all mechanisms above would place T4P/PilC-induced wetting as instrumental in concentrating CD147/β2-AR/α-actinin-4 complexes close to bacterial pili, enabling PilV-receptor interaction and host cell response.\u003c/p\u003e \u003cp\u003eHere, we investigated the very early steps of T4P interaction with endothelial cells and demonstrated that the initial T4P-induced host cell tubular membrane structures only result from physical cues, independently of host cell cytoskeletal reorganization and signaling events. These tubular structures, promoted by membrane wetting, non-specifically accumulate multiple plasma membrane-associated proteins, including those specifically involved in meningococcal-induced signaling. Accumulation in these tubular structures allows their interaction with T4P ligands and subsequent activation of the signaling pathways, which ultimately drive the formation of the cellular protrusions - strengthened by cytoskeletal proteins - necessary for bacterial colonization. Wetting, by both increasing the interaction area and concentrating specific receptors, represents a general mechanism facilitating the encounter between membrane-associated ligands and cognate receptors in the context of cell-to-cell interactions.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eN. meningitidis-\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003einduced tubular membrane structures are enriched in plasma membrane proteins.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eN. meningitidis\u0026nbsp;\u003c/em\u003einduces the formation of tubular membrane structures in human endothelial cells, which are necessary for efficient bacterial colonization \u003csup\u003e5,4,34\u003c/sup\u003e. These structures have been considered to be the result of signaling pathways elicited in host cells by the pathogen and involving ezrin phosphorylation and actin polymerization \u003csup\u003e37\u0026ndash;41,33\u003c/sup\u003e.Here, we investigated the key steps of formation of these structures by examining the distribution of a marker, the endogenous membrane curvature-sensitive CD9 tetraspanin \u003csup\u003e42\u003c/sup\u003e in human umbilical vein endothelial (EA.hy926) cells infected \u003cem\u003ein vitro\u003c/em\u003e with wild type meningococci (2C4.3 strain). CD9 labeling near bacterial colonies was reminiscent of that observed with the cell-signaling marker ezrin, which accumulates under \u003cem\u003eN. meningitidis\u0026nbsp;\u003c/em\u003ecolonies (Figure 1A). However, confocal imaging deconvolution revealed a very poor overlap between ezrin and CD9 labeling, the latter being concentrated in thin tubular plasma membrane structures surrounding adherent bacteria (Figure 1A and Figure S1A), which are distant or located above ezrin-positive areas (Figure 1A, bottom). We then compared CD9 and ezrin distribution using Correlated Light (confocal imaging) and (scanning) Electron Microscopy (CLEM) (Figure 1B and Figure S1B). While CD9 was accumulated into tubular membrane structures, ezrin appeared to be principally contained into distinct spotty areas with poor colocalization with CD9.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDense CD9 accumulation under meningococcal colonies has been considered as a facilitator of meningococcal adhesion to epithelial cell, reflecting the accumulation of receptors contributing to adhesion \u003csup\u003e43\u003c/sup\u003e. We therefore investigated whether CD9 and other tetraspanins would play an active role in the formation of the tubular membrane structures. We studied the localization of three plasma membrane tetraspanins CD9, CD81 and CD151 in different types of endothelial cells (EA.hy926, hCMEC/D3 and primary HDBEC) infected with wild type meningococci (Figure S2A-C). As for CD9 in EA.hy926 cells, all three tetraspanins accumulated under \u003cem\u003eN. meningitidis\u003c/em\u003e colonies, independently of the endothelial cell type. CD9 knockdown (KD) in EA.hy926 cells had no effect on \u003cem\u003eN. meningitidis\u0026nbsp;\u003c/em\u003einduced signaling, assessed by ezrin accumulation (Figure S3A,D) and the same observation was made after silencing CD81 (Figure S3B,E) and CD151 (Figure S3C,F). Consistently, \u003cem\u003eN. meningitidis\u003c/em\u003e-induced ezrin accumulation was not decreased by the CRISPR-mediated triple knockout (KO) \u0026nbsp;of CD9, CD81 and CD151 in EA.hy926 cells (Figure S3G, H), whereas the adhesion of wild-type bacteria on these triple KO cells was similar to that of wild-type cells (Figure S3I). Overall, these data indicate that the tetraspanins CD9, CD81 and CD151 were not required for \u003cem\u003eN. meningitidis\u003c/em\u003e adhesion, or for \u003cem\u003eN. meningitidis\u003c/em\u003e-induced signaling during the infection of endothelial cells. Several other plasma membrane proteins were recruited to bacterial colonies as CD9: the endogenous CD44, some exogenously expressed yellow fluorescent protein (YFP)-tagged G protein coupled receptors (CXCR4, CCR5, and AT1R), the YFP-tagged T lymphocyte cell marker CD4, and the YFP-tagged chemokine scavenger ACKR2 (Figure S4). Instead, other exogenously expressed plasma membrane receptors, such as the GFP-tagged transferrin receptor or the GFP-tagged Toll like receptors TLR2 and TLR4 were not recruited into these tubular membrane structures (Figure S4). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBacterial pili are indispensable appendages for adhesion and signaling in host cells. For meningococcal T4P, the main structural pilin PilE and the adhesive pilins PilV and PilC1 (Figure 1C) contribute to adhesion and signaling (including ezrin accumulation) in endothelial cells. The \u003cem\u003eN. meningitidis\u0026nbsp;\u003c/em\u003enon-encapsulated strain expressing the Opa outer membrane adhesins (\u003cem\u003eSiaD\u003csup\u003e-\u003c/sup\u003e Opa\u003csup\u003e+\u003c/sup\u003e\u003c/em\u003e) uses the secondary adhesins Opa proteins in addition to pilins to adhere to host cells, but still requires T4P to induce the formation of tubular membrane structures and signaling. The \u003cem\u003eSiaD\u003csup\u003e-\u003c/sup\u003e Opa\u003csup\u003e+\u003c/sup\u003e\u0026nbsp;\u003c/em\u003emeningococcal strains promoted CD9 accumulation into tubular plasma membrane structures surrounding adherent bacteria (Figure 1D,E top panels and S2) and ezrin accumulation under colonies \u003csup\u003e40\u003c/sup\u003e. In contrast, the non-piliated derivative \u0026Delta;\u003cem\u003ePilE\u003c/em\u003e mutant\u003cem\u003e\u0026nbsp;\u003c/em\u003eand the adhesion-defective derivatives \u0026Delta;\u003cem\u003ePilV\u003c/em\u003e and \u0026Delta;\u003cem\u003ePilC1\u003c/em\u003e mutant strains adhered to endothelial cells but failed to induce the formation of CD9-positive tubular membrane structures (Figure 1D,E and S2). These data indicate that adhesive T4P are necessary and sufficient to generate CD9-positive tubular membrane structures in endothelial cells.\u003c/p\u003e\n\u003cp\u003eWe then investigated the diffusion dynamics of plasma membrane proteins within the tubular membrane structures induced by meningococcal T4P. Fluorescence recovery after photobleaching (FRAP) experiments were conducted in endothelial cells expressing YFP-CD9, infected with the 2C4.3 strain (Figure 2A). YFP-CD9 mobility was significantly impaired in the tubular structures compared to control plasma membrane areas distant from bacterial colonies on the same coverslip (Figure 2B). Mobility of both the \u003cem\u003eN. meningitidis\u003c/em\u003e signaling receptor \u0026beta;2AR (\u0026beta;2AR-YFP) and of the T lymphocyte CD4 surface glycoprotein (CD4-YFP), which are not involved in signaling, was similarly reduced in tubular membrane structures, compared to control plasma membrane (Figure 2C, D). This suggests that these structures indiscriminately trap cell surface proteins and receptors, regardless of their potential role in \u003cem\u003eN. meningitidis\u003c/em\u003e-induced signaling in host cells. Such a phenomenon might drive the non-selective accumulation of multiple plasma-membrane-associated proteins in these structures, including the adhesion and signaling receptors for meningococci. To determine whether the \u003cem\u003eN. meningitidis\u003c/em\u003e signaling receptor \u0026beta;2AR might accumulate into these tubular structures, the fluorescent signal of \u0026beta;2AR-YFP and that of YFP-CD9 and CD4-YFP \u0026ndash; both used here as controls \u0026ndash; were quantified in \u003cem\u003eN. meningitidis\u003c/em\u003e-induced tubular membrane structures relative to that measured in physiological filopodia present in areas of the plasma membrane distant from meningococcal colonies (Figure 2E,F). We reasoned that if membrane shape was the only driver of protein accumulation, the same signal should be observed in physiological filopodia and in bacteria-induced tubular structures around the colonies. The \u0026beta;2AR-YFP appeared accumulated in the tubular membrane structures but not in filopodia (Figure 2F, 4.3 fold and 1.1 fold, respectively). YFP-CD9 and CD4-YFP were also significantly accumulated in \u003cem\u003eN. meningitidis\u003c/em\u003e-induced tubular membrane structures (Figure 2F, CD9: 5.65 fold in tubular structures (D) compared to 1.3 fold in filopodia (F); CD4: 4.3 fold in tubular structures compared to 1.1 fold in filopodia), ruling out the possibility that a simple change in membrane shape was the main driver of this enrichment. The signal of the FLIPPER-TR\u0026reg;, a fluorescent hydrophobic lipid membrane probe, was also significantly higher in tubular membrane structures compared to filopodia (2.18 fold and 0.95 fold, respectively), indicative of some plasma membrane accumulation in the focal acquisition plane of tubular membrane structures. It remained, however, significantly lower than that of YFP-CD9 (Figure 2F), suggesting that plasma membrane accumulation on its own accounted only in part for the observed local enrichment of membrane proteins. Overall, the data indicate that in the tubular membrane structures promoted by \u003cem\u003eN. meningitidis\u003c/em\u003e T4P, the diffusion of integral membrane proteins is decreased, possibly contributing to the accumulation of key receptors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eN. meningitidis\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e-induced initiation of tubular membrane structures in host cells is independent of signaling.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEzrin accumulation under bacteria is a hallmark of \u003cem\u003eN. meningitidis\u003c/em\u003e-promoted receptor signaling in host cells and plays an essential role in stabilizing bacterial adhesion and colony growth. We investigated its potential involvement in the formation of the tubular membrane structures promoted by T4P. Since ezrin activation and its plasma membrane relocation require its Thr567 phosphorylation \u003csup\u003e44\u003c/sup\u003e, EA.hy926 cells were treated prior to bacterial infection with NSC668394 \u003csup\u003e45\u003c/sup\u003e (from now on, NSC66), an ezrin Thr567 phosphorylation inhibitor, which selectively blocks ezrin accumulation (Figure 3A,B) and its downstream effects. In particular, NSC66 treatment inhibits sustained bacterial adhesion after 2h of infection, which depends on the \u0026beta;2AR signaling pathway and on ezrin accumulation under colonies \u003csup\u003e33,36,19\u003c/sup\u003e (Figure 3C). Then, using CD9-associated fluorescence as a quantitative marker, we investigated the potential contribution of ezrin accumulation to plasma membrane tubulation. To take into account the decrease of bacterial adhesion 2h after infection, we quantified CD9 labeling intensity below bacterial colonies and normalized each value with the respective intensity of the DAPI staining which was used as a proxy of the colony size. Interestingly, the inhibition of ezrin phosphorylation (Figure 3D) did not affect the level of CD9 accumulation around wild type colonies when normalized on their size, indicating that initial \u003cem\u003eN. meningitidis\u003c/em\u003e-promoted tubular membrane structures develop independently of ezrin activation and accumulation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eT4P retraction, which depends on the PilT ATPase activity and is turned off in the \u003cem\u003eN. meningitidis\u003c/em\u003e \u0026Delta;\u003cem\u003epilT\u0026nbsp;\u003c/em\u003estrain \u003csup\u003e46\u003c/sup\u003e, contributes to generating the mechanical forces that activate the T4P-induced b2AR signaling leading to ezrin recruitment in host cells \u003csup\u003e19\u003c/sup\u003e. CD9 accumulation (Figure 3D) and the formation of tubular membrane structures (Figure 3E,F) persisted, although reduced, in cells infected with \u0026Delta;\u003cem\u003epilT\u003c/em\u003e mutant bacteria compared to wild type 2C4.3 meningococci. Consistent with the observation above, NSC66 did not further decrease CD9 accumulation to \u0026Delta;\u003cem\u003epilT\u003c/em\u003e mutant bacteria (Figure 3D). Moreover, CD9 mobility in tubular deformations, whether promoted by wild type or by \u0026Delta;\u003cem\u003epilT\u0026nbsp;\u003c/em\u003ebacteria, was comparable in FRAP experiments (Figure 3G). Based on these results, the cellular process leading to the formation of early tubular membrane structures appeared to be independent on T4P-dependent receptor signaling or ezrin phosphorylation and activation. Of note, the persistence of some tubular membrane structures in the absence of any T4P retraction and ezrin accumulation raises the hypothesis that some purely physical, signaling-independent and T4P retraction-independent mechanism might be involved in this phenomenon.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTubulation of the plasma membrane along T4P is maintained when biochemical activity in host cells is inhibited.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe shape of the apical membrane in endothelial cells is dynamically controlled by the cytoskeleton and by receptor-driven signaling. To impede any biochemical reaction that would interfere with plasma membrane dynamics, we examined the formation of \u003cem\u003eN. meningitidis\u003c/em\u003e-induced tubular membrane structures in cells prefixed 15 min with 4% paraformaldehyde (PFA). Upon infection of PFA-treated cells with wild type meningococci, some tubular membrane structures containing endogenous CD9 still grew at the interface with bacteria (Figure 4A), although much less numerous (Figure 4D,E) and shorter (Figure 4C,E) than in living cells. As expected, no ezrin accumulation was observed in prefixed cells (Figure 4A-D).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSurprisingly, upon infection with the \u0026Delta;\u003cem\u003epilT\u003c/em\u003e mutant strain, the accumulation of CD9 around bacterial colonies was much more abundant in PFA-fixed cells than in living cells, and comparable to that observed in living cells infected with the 2C4.3 strain (Figure 4A,B). \u0026nbsp;CD9 diffusion was similarly impaired by PFA fixation, whatever was the infecting strain, both in tubular structures and plasma membrane filopodia (Figure 4H,I). Thus, the larger CD9 accumulation in \u0026Delta;\u003cem\u003epilT\u003c/em\u003e-induced tubular membrane structures of fixed cells (Figure 4J,\u0026nbsp;about two-fold higher CD9 median intensity in tubular membrane structures, than in filopodia) is likely due to membrane accumulation and not to decreased diffusion as in non-fixed cells. Moreover, in fixed cells infected with \u0026Delta;\u003cem\u003epilT\u003c/em\u003e bacteria, not only the maximal length of tubular membrane structures was greater than that measured in fixed or living cells infected with the wild type strain (Figure 4C), their number was also higher than in living cells, or fixed cells infected with the wild type strain (Figure 4 compare G with E and F). Overall, these findings indicate that the tubular membrane structures growing along T4P do not require any cell signaling but are limited/interrupted by pilus retraction. Moreover, in living cells this membrane tubulation along immobile T4P is restrained by some fixation sensitive host cell factor(s).\u003c/p\u003e\n\u003cp\u003eTo extend our observations in a more representative experimental setup, purified plasma membrane (PM) sheets \u003csup\u003e47,48\u003c/sup\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003eof EA.hy926 cells were prepared from a monolayer of living endothelial cells and then attached to the bottom of Ibidi \u0026micro;-slides coated with anti-clathrin antibodies. The topology of clathrin-coated zones (flat lattices and coated pits) at the inner side of the membrane facilitates the attachment of PM sheets with the extracellular side oriented upwards. This attachment, which is limited to a few anchoring points, is thought to allow the lipid fraction of the plasma membrane to move freely, while biochemical reactions such as ezrin phosphorylation and actin polymerization are abolished \u003csup\u003e47,48\u003c/sup\u003e. PM sheets were then incubated with wild type or \u0026Delta;\u003cem\u003epilT\u0026nbsp;\u003c/em\u003ebacteria and the maximal length of tubular structures was determined (Figure 4K, L and S5A). Consistent with the findings in fixed cells, CD9 accumulation was hardly detected below wild type bacteria but clearly visible below \u0026Delta;\u003cem\u003epilT\u0026nbsp;\u003c/em\u003ecolonies, with the maximal length of tubular membrane structures induced by \u0026Delta;\u003cem\u003epilT\u0026nbsp;\u003c/em\u003ebacteria being similar to that observed on fixed cells (2.59 \u0026micro;m and 3.09 \u0026micro;m, respectively).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eActin polymerization inhibits tubulation of the plasma membrane induced by \u003cem\u003eN.\u0026nbsp;meningitidis\u003c/em\u003e T4P.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFrom the data above, it appears that some host cell factors limit the formation of plasma tubular membrane structures promoted by immobile T4P. To address the possibility that this might be due to the reaction of the cortical cytoskeleton to a local increase in membrane tension, we examined whether the inhibition of membrane tubulation along T4P would require ATP as energy source. Endothelial cells were incubated with 10mM 2-deoxy-d-glucose (2-DG) to deplete cellular ATP (Figure S5B), before meningococcal infection. The length of CD9-labelled tubular membrane structures was then estimated using microscopy orthogonal sections (Figure 5A,B). ATP depletion, at levels sufficient to inhibit ezrin accumulation (Figure 5A), significantly reduced the length of protrusions in cells infected with the wild type 2C4.3 strain (Figure 5B), confirming the requirement of host cell signaling in the formation of mature protrusions in and around the\u0026nbsp;colony. However, reminiscent of what was observed on PM sheets or upon PFA fixation of endothelial cells (see Figure 4), ATP depletion markedly enhanced the length of tubular protrusions induced by the \u0026Delta;\u003cem\u003epilT\u0026nbsp;\u003c/em\u003estrain infection (Figure 5A, B). Actin polymerization, typically observed under bacterial colonies in infected endothelial cells \u003csup\u003e36,40\u003c/sup\u003e, requires ATP. Pre-treatment of endothelial cells with the actin polymerization inhibitors Cytochalasin D (CytD) and Latrunculin B (LatB) enhanced the maximum length of tubular membrane structures promoted by \u0026Delta;\u003cem\u003epilT\u003c/em\u003e bacteria to the same level of low ATP conditions (Figure 5B,C). Altogether, our findings indicate that the signaling-independent growth of tubular membrane structures is counteracted by actin cytoskeleton remodeling in host cells.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eN. meningitidis\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003ecan\u003cem\u003e\u0026nbsp;\u003c/em\u003einduce tubular membrane structures\u0026nbsp;in cells incompetent for meningococcal-promoted signaling.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further validate the hypothesis that the tubular membrane structures induced by T4P involve a purely physical mechanism, additional studies were conducted in cell types, which are not competent for meningococcal infection and meningococcal-induced signaling or ezrin recruitment. Kidney epithelial HEK cells cannot be infected by wild type \u003cem\u003eN. meningitidis\u003c/em\u003e because bacteria cannot adhere to their surface and, in case of surrogate adhesion, bacteria do not induce signaling in the absence of exogenous b2AR and b-arrestins \u003csup\u003e34\u003c/sup\u003e. Surrogate adhesion of a \u003cem\u003eN. meningitidis\u003c/em\u003e \u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e strain can be reconstituted in HEK cells by expressing hCEACAM1, the adhesion receptor of opacity proteins (Opa) \u003csup\u003e49,40\u003c/sup\u003e. In hCEACAM-expressing HEK cells, CD9 - used as a marker of the tubular structures induced by T4P - accumulated in and around adherent \u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u0026nbsp;\u003c/sup\u003e\u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u0026nbsp;\u003c/sup\u003e\u0026Delta;\u003cem\u003epilT\u003c/em\u003e colonies (Figure 6A, left panels). Control non-piliated \u0026Delta;\u003cem\u003epilE\u003c/em\u003e bacteria failed to induce any CD9 accumulation (Figure 6A, middle panels), consistent with the absolute requirement of T4P in this process. \u003cem\u003eSiaD\u003c/em\u003e\u003csup\u003e-\u0026nbsp;\u003c/sup\u003e\u003cem\u003eOpa\u003c/em\u003e\u003csup\u003e+\u0026nbsp;\u003c/sup\u003e\u0026Delta;\u003cem\u003epilT\u003c/em\u003e colonies also induced the accumulation of CD9 in canine MDCK hCEACAM cells, in which no meningococcal-promoted signaling can be observed (Figure 6A, right panels). As shown by FRAP experiments in endothelial cells, YFP-CD9 mobility was decreased in infected HEK cell membrane structures surrounding bacteria, compared to control distant plasma membrane areas (Figure 6B, compare with Figure 2B). Furthermore, CD9 accumulation around bacteria was maintained in PFA-fixed HEK cells (Figure 6C), recapitulating what was observed in endothelial cells. These findings support the hypothesis that adhesive \u003cem\u003eN. meningitidis\u003c/em\u003e T4P generate a purely physical process, independent of b2AR signaling, driving plasma membrane tubulation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePilC1-dependent adhesion is the initiating event of membrane tubulation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data accumulated so far did not determine what component of T4P could be the initial driver of the tubulation process in host cells. The PilV pilin and the tip-located PilC1 adhesin are both required for adhesion, CD9 recruitment and membrane tubulation (Figure 1 D,E). PilV interacts with CD147 and b2AR receptors and activates T4P retraction-induced signaling in host cells, suggesting that PilV is mainly involved in meningococcal adhesion after receptors accumulation in tubular membrane structures. Although also involved in adhesion, PilC1 has no known signaling properties. PilC1 thus appears as a strong candidate for the initiation of membrane tubulations in host cells. The genetic deletion of PilC1 in wild type or D\u003cem\u003epilT\u003c/em\u003e strains (D\u003cem\u003eC1\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;\u003c/em\u003eD\u003cem\u003eTC1\u003c/em\u003e mutants, respectively) significantly inhibited meningococcal adhesion in both intact and PFA-fixed cells (Figure 7A). The preserved expression of PilV in T4P of D\u003cem\u003eTC1\u003c/em\u003e bacteria was therefore not sufficient for maintaining normal adhesion (Figure 7A) and could not rescue membrane tubulation in \u003cem\u003eSiaD\u003csup\u003e-\u003c/sup\u003e Opa\u003csup\u003e+\u003c/sup\u003e\u003c/em\u003e \u0026Delta;\u003cem\u003eC1\u003c/em\u003e bacteria (Figure 1D,E). Also, the enhanced adhesion observed in fixed cells infected by the D\u003cem\u003epilT\u0026nbsp;\u003c/em\u003estrain was abolished by the absence of PilC1 (Figure 7 A), despite the advantage provided by PilT suppression in terms of prolonged T4P contact with host cells plasma membrane and T4P length (see Figure 4). All together, these data demonstrate a predominant role of PilC1 as the inducer of tubular membrane structures in host cells at the initial steps of adhesion.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eWe uncovered the mechanism adopted by \u003cem\u003eN. meningitidis\u003c/em\u003e to facilitate the encounter with cognate host-cells receptors despite the initial low multiplicity of infection usually observed with this pathogen. Bacterial T4P interact through the pilus tip adhesin PilC1 with host cells, inducing an early signaling-independent host plasma membrane tubulation (Figure S6, step 1). These tubular membrane structures extend the interaction surface between meningococcal ligands, such as PilV molecules with potential receptors. The plasma membrane-associated receptors found in these tubular structures exhibit reduced mobility, compared to areas of the plasma membrane distant from infecting bacteria. The increased local concentration of potential adhesion and signaling receptors in the vicinity of bacterial ligands eventually enable PilV to bind to CD147 / β2AR oligomers during the short time frame (few seconds) of pili immobility, before PilT-dependent retraction. PilV interaction with CD147 facilitates additional PilV molecules distributed along the pilus to bind to glycan chains of β2AR signalling receptors (Figure S6, step 2). The mechanical forces resulting from both T4P retraction and blood flow, applied on the β2AR via PilV, trigger signaling in host cells. The subsequent signaling- and ATP-dependent enrichment of the tubular membrane structures with ezrin and actin polymers allows their enlargement and reinforcement. The cellular protrusions resulting from this process stabilize meningococci at the cell surface, permitting the interaction of additional T4P ligands with additional receptors, and then the progressive growth of a bacterial colony despite the forces exerted by the blood flow (Figure S6, step 3).\u003c/p\u003e \u003cp\u003eThe ability of the plasma membrane to form early tubular structures along meningococcal T4P was previously observed, but its functional relevance in the context of \u003cem\u003eN. meningitidis\u003c/em\u003e pathophysiology had not been addressed so far. The formation of these tubular structures was described as a physical process of one-dimensional \u0026ldquo;wetting\u0026rdquo; occurring when membranes interact with nanofibers \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. The physical nature of this phenomenon is supported here by the demonstration that it is independent of meningococcus-induced signaling, it does not require ATP as energy source and can be reproduced in fixed cells or with purified plasma membrane sheets. Our data also point out the central role of the PilC1 pilus-tip adhesin docking to the plasma membrane, in the initiation of the one-dimensional wetting.\u003c/p\u003e \u003cp\u003eThese early tubular structures, promoted by T4P, display several properties that differentiate them from common plasma membrane filopodia. They accumulate multiple integral membrane proteins, including the tetraspanins CD9, CD81 and CD151, G protein-coupled and chemokine scavenger receptors, and the single transmembrane domain proteins CD4, CD44 and CD147. Using CD9 as a marker of these tubular structures, we could determine that its local enrichment around bacteria might not only depend on membrane accumulation - the tubular formation itself - but also on its markedly decreased mobility. Indeed, once in the tubular structure, membrane proteins can move along the walls of the tube, with a lower probability of exiting from its basis, keeping them trapped. Our results (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF) also suggest an active receptor accumulation phenomenon in living cells that has yet to be determined. Moreover, it is not clear how other integral membrane proteins, such as the transferrin receptor or Toll like receptors are excluded from these structures.\u003c/p\u003e \u003cp\u003eSeveral factors regulate the size and the lifetime of these T4P-promoted tubular structures in living cells. As mentioned above, PilT-dependent pilus retraction limit their duration to a few seconds \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, unless signaling events promoted in host cells stabilize and turn them into larger cellular protrusions. In this context, ezrin recruitment and phosphorylation play an essential role for adhesion and protrusion stabilization 2h after infection. The role of cortical actin polymerization is more complex, since it has a variable effect depending on the duration of T4P interaction with the host cell plasma membrane. On one hand, its inhibition by ATP depletion in cells infected by the wild type strain (which retracts its T4P) reduces the maximal size of tubular structures. On the other hand, in cells infected with the Δ\u003cem\u003epilT\u003c/em\u003e strain, the inhibition of actin polymerization, either indirectly through ATP depletion or directly using cytochalasin D or latrunculin B, markedly enhances the length of tubular structures.\u003c/p\u003e \u003cp\u003eIn addition to \u003cem\u003eNeisseria\u003c/em\u003e family bacteria, many other species express fimbriae or pili for adhesion to and infection of host cells. Interestingly, pili-tip adhesin proteins, such as \u003cem\u003ePseudomonas\u003c/em\u003e PilY1, uropathogenic \u003cem\u003eE. coli\u003c/em\u003e FimH and \u003cem\u003eStreptococcus pneumoniae\u003c/em\u003e RrgA are similarly exposed and directly engaged in the interaction with host cells by targeting specific receptors or extracellular matrix (ECM) components. Since the pili-dependent formation of tubular membrane structures appears to be a physical process of one-dimensional wetting resulting from the interaction of membranes with nanofibers (i.e. pili), it is plausible that the mechanism of cognate receptor recruitment in these tubular structures for \u003cem\u003eN. meningitidis\u003c/em\u003e extends to other bacterial species. Mammalian cells can also produce nanofibers, including actin-based structures (filipodia, microvilli) and ECM components (collagen, fibronectin), which play a role in cell signaling and adhesion. It is not excluded that these structures might induce equivalent membrane tubular processes in adjacent/interacting cells.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eCell lines and culture conditions\u003c/h2\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003eCell lines and media\u003c/h2\u003e \u003cp\u003eThe EA.hy926 cell line (ATCC #CRL-2922) was obtained from the American Type Culture Collection (ATCC). EA.hy926 cells were grown in Dulbecco's Modified Eagle Medium (DMEM; Gibco) supplemented with 10% decomplemented fetal calf serum (FCS; Gibco) and 1% penicillin\u0026ndash;streptomycin. Human Cerebral Microvascular Endothelial Cells hCMEC/D3s \u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e are fully differentiated brain endothelial cell derived from human brain capillaries. hCMEC/D3 were grown onto cultrex rat collagen type I-coated dishes (Bio-techne) in Endothelial Cell Basal Medium-2 (Lonza) supplemented with 5% of FCS, 1.4 \u0026micro;M hydrocortisone (Lonza), 5 \u0026micro;g/mL ascorbic acid (Lonza), 1 ng/mL b-FGF (Lonza), at 37\u0026deg;C in 5% CO2. Primary Human Dermal Blood (Microvascular) Endothelial Cells (HDBEC) (Promocell) are isolated from the dermis of juvenile foreskin and adult skin (different locations). HDBEC were grown onto cultrex rat collagen type I-coated dishes (Bio-techne) in their specific endothelial cell growth medium (PromoCell). HEK-293 (ATCC CRL-1573) cells were grown in flasks coated with 0.01% poly-l-lysine (Sigma) in supplemented with 10% FCS and 1% penicillin\u0026ndash;streptomycin. Cells were grown at 37˚C in a humidified incubator under 5% CO2.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eTransfection and siRNA inhibition\u003c/h2\u003e \u003cp\u003eEA.hy926 cells were transfected with 1\u0026micro;g of plasmids using Lonza 4D Nucleofector\u0026reg; System (Lonza) according to manufacturer\u0026rsquo;s instructions and 5.10\u003csup\u003e5\u003c/sup\u003e transfected cells were seeded into \u0026micro;-Slide 4 Well (Ibidi) 24h prior to infection. For HEK cell, 1.5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e cells were seeded in \u0026micro;-Slide 4 Well and were transfected 24h prior to infection with 200ng of plasmid encoding hCEACAM and YFP-CD9, using EcoTransfect\u0026trade; (Oz Biosciences) according to manufacturer\u0026rsquo;s instructions. For relative intensity and accumulation quantification experiments, infections were performed in Opti-MEM (Gibco). The siRNA inhibitions of CD9, CD81 and CD151 were obtained from siRNA pools (ON-TARGETplus Human CD9 (928) siRNA - SMARTpool (Horizon); ON-TARGETplus Human CD81 (975) siRNA - SMARTpool (Horizon); MISSION Pre-designed siRNA \u0026minus;\u0026thinsp;2 OD, Human CD151, SASI_Hs01_00188401 (Sigma)) and controlled with MISSION\u0026reg; siRNA Universal Negative Control #1 (Sigma). The siRNAs were transfected with Lipofectamine RNAiMAX (Thermofisher) on non-confluent cells. For CD9 siRNA, the transfection was repeated a second time after 3 days. Experiments were performed 3 days after the last transfection. Because tetraspanins are accumulated below meningococcal colonies, the whole amount of protein in cells did not reflect the true level of protein accumulation. Therefore, the inhibitions of tetraspanins expression were controlled by immunofluorescence quantification under bacterial colonies (see the image analysis section below).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eProduction of the CRISPR/Cas9 knockout cell lines\u003c/h3\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eCell line mutation\u003c/h2\u003e \u003cp\u003eThe triple knockout (KO) EA.hy926 cell line was obtained by serial knockout of CD9, then CD81 and then CD151. CD9 and CD81 KO cell lines were obtained using the lentiCRISPR v2 plasmid (gifts from Dr Eric Rubinstein; Addgene plasmid #52961). The CD151 KO cell lines were obtained using the Sigma-Aldrich CRIPRD HSPD0000006331 and HSPD0000006332 plasmids. For each of the three separate tetraspanins we produced two lentiviruses with different guiding sequence. After obtaining each KO cell lines, tetraspanin depletion, adhesion of bacteria and induced signalling was assessed. One cell line was kept for further transduction. Guiding sequences are as follow: CD9sg95 TTGGACTATGGCTCCGATTC \u0026amp; CD9sg314 ATTCGCCATTGAAATAGCTG; CD81sg166 ACACCTTCTATGTAGGTGAG \u0026amp; CD81sg253 AGGAATCCCAGTGCCTGCTG; CD151 n\u0026deg;6331 CTGGTAGTAGGCGTAGGCG \u0026amp; CD151 n\u0026deg;6332 CCAAGCGCTACCACCAGCC. EA.hy926 cells were transduced with lentivirus (MOI 1:1000). 48 h after transduction, cells were treated with 1 \u0026micro;g/ml puromycin to kill non-transduced cells. FACS sorting of tetraspanin-depleted cells was performed 2 to 3 times over the scope of 1 or 2 months to ensure the purity of each CRISPR KO cell lines. Note that we did not select a clonal population but the whole population of the CD9, CD81 and CD151 depleted cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCytometry\u003c/h2\u003e \u003cp\u003eFor surface protein expression analysis, cells were washed twice in Phosphate-Buffered-Saline (PBS), trypsinized, fixed in 4% paraformaldehyde (PFA) for 15 minutes, washed twice in PBS and kept at 4\u0026deg;C until staining. For cell sorting, cells were washed twice in PBS, trypsinized, and immediately washed twice in PBS before staining. In both experiments, cells were stained for 30 minutes with the appropriate primary antibody at 4\u0026deg;C in PBS/Bovine Serum Albumin (BSA) 0.1% and washed three times in PBS-BSA before Alexa Fluor-coupled secondary antibody staining for 30 minutes. Cells were then washed twice in PBS-BSA before FACS analysis. Negative controls were assessed with secondary antibody-only staining on the same cells. Data were acquired using a BD LSR Fortessa instrument (BD Biosciences) for fixed samples, and with BD FACSAria II for live cell sorting. Data was analyzed using the FlowJo Software. A minimum of 20 000 cells were acquired for each experiment.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eTreatments of cells\u003c/h2\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003eActin polymerization and ezrin inhibition drugs\u003c/h2\u003e \u003cp\u003eTreatments with drugs were performed in DMEM\u0026thinsp;+\u0026thinsp;10% FBS for NSC668394 (Sigma-Aldrich 341216; 30 \u0026micro;M 2h prior to infection, and 3\u0026micro;M throughout infection), wheat germ agglutinin (WGA; Vector Laboratories, Burlingame, CA, USA; 20 \u0026micro;M, 1h prior to infection and maintained throughout infection), Cytochalasin D (Sigma C8273; 2\u0026ndash;4 \u0026micro;M, 2h prior to infection and maintained throughout infection), Latrunculin B (Abcam ab141409; 0,1\u0026ndash;1 \u0026micro;g/ml, 2h prior to infection and maintained throughout). Controls were assessed with DMSO when relevant.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eATP depletion\u003c/h2\u003e \u003cp\u003eFor ATP depletion, treatment was performed in DMEM without glucose (Invitrogen)\u0026thinsp;+\u0026thinsp;1 mM lactate (as carbon source for bacteria) and 10 mM 2-deoxy-d-glucose (2-DG, Sigma-Aldrich). Control cells were incubated in DMEM without glucose supplemented with 10% FBS. Treatment was performed overnight prior to infection and maintained through the course of the experiment. ATP depletion efficiency was assessed using the commercially available kit Adenosine 5\u0026prime;-triphosphate Bioluminescent Assay (Sigma-Aldrich).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eChemical pre-fixation assay\u003c/h2\u003e \u003cp\u003eCells were chemically fixed on the day of the infection with 4% paraformaldehyde (PFA, Thermofisher 28908) for 15 minutes and quenched with 50 mM NH\u003csub\u003e4\u003c/sub\u003eCl for 5 minutes. Cells were then washed several times in PBS and DMEM\u0026thinsp;+\u0026thinsp;10% FBS prior to infection.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eBacterial strains and infection\u003c/h2\u003e \u003cp\u003e \u003cem\u003eN. meningitidis\u003c/em\u003e strains used in this study are derived from serogroup C meningococcal strain 8013, designated as 2C4.3 \u003csup\u003e51\u003c/sup\u003e. 2C4.3 is a piliated encapsulated Opa\u0026thinsp;\u0026minus;\u0026thinsp;Opc\u0026thinsp;\u0026minus;\u0026thinsp;variant. Mutant strains Δ\u003cem\u003epilC1\u003c/em\u003e \u003csup\u003e52\u003c/sup\u003e, Δ\u003cem\u003epilE\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e, Δ\u003cem\u003epilT\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e, Δ\u003cem\u003epilV\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, Δ\u003cem\u003esiaD\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e as well as double mutants were designed in the laboratory. \u003cem\u003eN. meningitidis\u003c/em\u003e were grown at 37\u0026deg;C in 5% CO\u003csub\u003e2\u003c/sub\u003e on gonococcal base (GCB) agar (Difco) plates containing 12 \u0026micro;M FeSO4 and Kellogg\u0026rsquo;s supplements \u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e or in Dulbecco's Modified Eagle Medium (DMEM; Gibco) supplemented with 10% decomplemented fetal calf serum (FCS; Gibco). For antibiotic selection of \u003cem\u003eN. meningitidis\u003c/em\u003e strains, kanamycin was used at a concentration of 100 \u0026micro;g/ml, chloramphenicol at 6 \u0026micro;g/ml and erythromycin at 2 \u0026micro;g/ml. Before cell infection, bacteria were first sub-cultured to OD600\u0026thinsp;=\u0026thinsp;0.1 in pre-warmed cell culture medium and incubated for 2 h at 37\u0026deg;C with agitation and 5% CO\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eImaging assay\u003c/h2\u003e \u003cp\u003eCells were infected with bacteria at a multiplicity of infection of 30 bacteria per cell for 30 min and washed with media to remove non-adherent bacteria. Infection then proceeded for 2 h and was concluded with two PBS washes and 4% PFA fixation for 15 minutes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eBacterial adhesion assay\u003c/h2\u003e \u003cp\u003eAt least two days before infection, cells were seeded on 2 cm\u0026sup2; wells. The day of infection, confluent adherent cells were infected with an estimated number of 10\u003csup\u003e7\u003c/sup\u003e bacteria each for 40 min. The exact number of colony forming units (CFU) in the inoculum was determined by serial dilution and counting of CFU. After 40 min, the cells were washed six times with PBS to remove non-adherent bacteria and adherent bacteria were detached in 500 \u0026micro;L DMEM with 10% FBS by mechanical scratching. To determine the adhesion frequency, adherent bacteria were diluted and spread on agar plates and the CFU were counted the next day. Adhesion frequency was obtained by dividing the adherent bacteria CFU count with the inoculum CFU count. For longer adhesion assay, infection was allowed to proceed for 2 h after the first washes at the 40-minute mark and was concluded similarly to the early adhesion assay.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eImaging\u003c/h2\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003ch2\u003eImmunofluorescence microscopy\u003c/h2\u003e \u003cp\u003eAll incubations were performed at room temperature. Cells were grown and infected on glass coverslips. They were fixed in 4% PFA for 15 minutes, quenched with ammonium chloride 50 mM solution for 5 min and rinsed with PBS with ions. Immunostaining steps were performed with 0.1% BSA blocking and 0.1% Saponin permeabilization at all times. Coverslips were incubated in BSA and Saponin buffer for 20 minutes before staining. Proteins of interest were stained with primary antibodies for 45 min to 1 h. After three washes in PBS, the coverslips were incubated with Alexa Fluor-conjugated secondary antibodies for 45 min. Nuclear DNA and actin were stained with 4\u0026prime;,6\u0026prime;-diamidino-2-phenylindole (DAPI) at 1 \u0026micro;g/ml and Alexa Fluor-conjugated phalloidin (Thermofisher), respectively. After three PBS washes, the coverslips were rinsed in water and mounted in Mowiol for observation. Immunofluorescence imaging was performed either on a Zeiss Spinning Disk microscope (63X, NA 1.4), a Zeiss Apotome fluorescence microscope or a SP8 laser scanning confocal microscope (Leica Microsystems) using a oil immersion objective. When mentioned deconvolution was performed using Huygens software.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eScanning Electron Microscopy\u003c/h2\u003e \u003cp\u003eAll incubations were performed at room temperature except when otherwise indicated. Cells were grown and infected on glass coverslips. They were fixed in 0.1M sodium cacodylate (Euromedex) buffer, pH 7.4 containing 2.5% glutaraldehyde (GA, Euromedex) and 1% PFA for 60 min, washed in cacodylate buffer (2x10 min), and then fixed in 1% OsO4 diluted in cacodylate buffer for 45 min at 4\u0026deg;C. After washing in cacodylate buffer (2x10 min), samples were dehydrated in an ascending series of ethanol (30%, 50%, 70%, 95%, 100%, 100%, 100% \u0026minus;\u0026thinsp;10 min each), followed by Hexamethyldisilazane (HMDS, Sigma-Aldrich)/ethanol (1/1:v/v) for 10 min and HMDS for 10 min. After overnight air drying, each coverslip was placed on a double-sided sticky tape on the top of an aluminum stub and sputter coated with Au/Pd. Images were acquired using a Jeol LV6510 (Jeol, Croissy-sur-Seine, France).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eCorrelative Light Electron Microscopy (CLEM)\u003c/h2\u003e \u003cp\u003eCells were grown and infected in 35mm \u0026micro;\u0026ndash;Dishes with a removable gridded glass coverslip bottom (Ibidi 81158). They were first processed for fluorescence imaging as follows. After fixation in 4% PFA\u0026thinsp;+\u0026thinsp;0.2% GA, they were immunostained similarly to a classic immunofluorescence microscopy. They were then imaged in a SP8 laser scanning confocal microscope (Leica Microsystems) using a X63/1.40 oil immersion objective and 405/410\u0026ndash;450, 488/495\u0026ndash;530, and 552/560\u0026ndash;600 excitation/emission wavelengths (in nm) for the detection of DAPI, AF488, and AF555 respectively. A mosaic image of the whole grid, including the lettered and numbered squares, was acquired in the DAPI channel and used as a map to locate the cells that were then observed at high resolution. For each selected region of interest (ROI), a z-stack with a step of 0.3 \u0026micro;m was acquired and deconvoluted using Huyghens software. Following confocal imaging, cells were fixed and processed for SEM as indicated above. Each ROI previously acquired in the confocal microscope was imaged in the SEM. For correlative image registration between electron microscopy (EM) and fluorescence the Icy (v2.5.2.0) \u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e EC-CLEM plugin \u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e has been used with the DAPI signal as the referent channel for the registration.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eFRAP\u003c/h2\u003e \u003cp\u003eLive experiment data were acquired with a TCS SP8 X confocal microscope (Leica) equipped with a \u0026times;100, 1.40 oil HC PL APO objective and a heated stage maintained at 37\u0026deg;C. For FRAP analysis, series of 40 fluorescence measures were captured from 30 images for all experiments from three independent experiments, with time intervals between scans ranging from 100 to 1000 ms. To perform FRAP, 1\u0026micro;m\u0026sup2;-ROI were bleached with 10 WLL2 laser pulse iterations at 100% power. Image analysis and fluorescence recovery was made by LAS X software (Leica). Data were normalized with average value of pre-FRAP values and were plotted using Prism9 software (GraphPad), a one-phase decay fit curve was applied to the plotted results.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003ePlasma membrane sheets (PM sheets) preparation\u003c/h2\u003e \u003cp\u003eThe protocol was adapted from \u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. EA.hy926 cells were separated with 5mM EDTA (ThermoFisher 15575020) on the day of cell passage and seeded on a 6-well plaque pre-coated with Poly-L-Lysine (Sigma P4707). One day after seeding, cells were rinsed twice with PBS and were submitted to sonication in 10 mL cold PBS using a probe sonicator (six pulses of 0.5 s, 20% duty cycle, output control level 2). Cytoplasm released and detached cells in the supernatant were discarded by rinsing twice with HBSS (ThermoFisher 14170088; 5 ml, 5 min). Remaining attached membranes were scrapped and resuspended in 150 \u0026micro;L HBSS with protease cocktail inhibitor (Sigma 4693159001) with cut pipette tips to avoid destroying the collected PM-sheets. PM-sheets were then deposited and stayed overnight into a 8 well \u0026micro;-Slide (ibidi 80826) previously treated with Poly-D-Lysine (ThermoFisher A3890401) for 1h and then coated with anti-clathrin antibodies (ThermoFisher PA5-143896) overnight. Wells were then washed twice in PBS and adherent PM-sheets were incubated in cell culture media and infected with bacteria and fixed for immunostaining.\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eImage analysis\u003c/h2\u003e \u003cdiv id=\"Sec26\" class=\"Section4\"\u003e \u003ch2\u003eRelative intensity accumulation quantification\u003c/h2\u003e \u003cp\u003eEA.hy926 cells were transfected and infected as described above. For each condition, 30 confocal images from three independent experiments of filopodia and tubular membrane structures were acquired along plasma membrane. Maximum intensity value were determined from a trace line using ImageJ software. For each image, filopodia and tubular membrane structures values were normalized with the corresponding plasma membrane value and plotted using Prism9 software.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003eProtein accumulation index\u003c/h2\u003e \u003cp\u003eThe accumulation index of CD9 and Ezrin was determined by immunofluorescence volume analysis. Stack images of bacterial colony were acquired with a Zeiss Spinning Disk microscope (63X, NA 1.4). We manually recognized bacterial colonies using Fiji ROI \u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e (v2.14.0). Then with a macro designed for this work, each ROI for each image were extracted as a cropped tiff stack image. Then each cropped images channel were isolated with a Fiji macro in order to do a 3D shallow learning pixel classification with ilastik \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e (v1.4.0post1). DAPI was used to recognize bacteria. Bacteria and protein-of-interest volume (Bv and POIv, respectively) were obtained with a last Fiji macro and the accumulation index (AI) was calculated, AI\u0026thinsp;=\u0026thinsp;POIv/Bv.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eMaximum tubular membrane length quantification\u003c/h2\u003e \u003cp\u003eThe maximum tubular membrane length was assessed for n\u0026thinsp;\u0026ge;\u0026thinsp;30 colonies among two technical replicates for a total of three experimental replicates for every experimental condition. For each colony, chosen at random, CD9 immunofluorescence was used to spot tubular membrane structure. The distance between the apical plasma membrane focal plane and the tip of tubular structures was estimated with the z-stack tool on a Zeiss Apotome fluorescence microscope at x63 by multiplying the steps amount and step size of 0,24 \u0026micro;m.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eT4P purification\u003c/h2\u003e \u003cp\u003e \u003cem\u003eAmmonium sulfate precipitation. N. meningitidis\u003c/em\u003e strains were grown overnight on GCB agar plates, scraped off of the plates and resuspended in 2 mL of 20 mM ethanolamine, pH 10.5 supplemented with 1 mM dithiothreitol (DTT) at 4\u0026deg;C. To shear T4P off, the bacteria were vortexed vigorously 3 times for 1 min bursts, returning them to ice for 1 min between each burst. Bacterial cells were removed from the pilus suspension by two successive centrifugations at 10 000 x \u003cem\u003eg\u003c/em\u003e for 20 min at 4\u0026deg;C. Cell pellets were resuspended in lysis buffer (50 mM Tris pH 7.5, 25 mM HEPES, 2 mM EDTA, 1% (w/v) SDS - bacterial fraction). The supernatant containing the T4P was collected and supplemented with saturated ammonium sulfate in 20 mM ethanolamine, pH 10.5 at a final concentration of 0.15 M, before overnight agitation at 4\u0026deg;C. Aggregated T4P were pelleted by centrifugation at 17 000 x g for 20 min at 4\u0026deg;C and resuspended overnight at 4\u0026deg;C, in 400 \u0026micro;l of 20 mM ethanolamine, pH 10.5. The T4P solution was centrifuged at 10 000 x g for 20 min at 4\u0026deg;C to remove residual cell debris and the supernatant was concentrated 10-fold using an Amicon 10 kDa MWCO membrane (Merck Millipore).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eImmunoblotting\u003c/h3\u003e\n\u003cp\u003eFor analysis of cell lysates, one bacterial loop of each strain of interest \u0026ndash; grown overnight in GCB agar\u0026thinsp;+\u0026thinsp;corresponding antibiotics \u0026ndash; was resuspended in 500 \u0026micro;l RIPA buffer (50 mM Tris pH 7,5, 150 nM NaCl, 25 nm HEPES, 2 mM EDTA, 1% w/vol SDS). This suspension was heated at 95\u0026deg;C for 5 min. For analysis of T4P, pili were purified as described above. Whole lysate or T4P preparation was deposited on a 12% acrylamide SDS-PAGE gel. After transfer to nitrocellulose membrane, the membrane was incubated for 20 min in a blocking solution (PBS\u0026thinsp;+\u0026thinsp;0.1% tween-20\u0026thinsp;+\u0026thinsp;4% milk) and washed twice with PBS\u0026thinsp;+\u0026thinsp;0.1% tween-20 (PBST). The membrane was incubated for 1 h in the presence of anti-PilV primary antibody \u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e diluted 1:1,000 in PBST, washed three times for 5 min with PBST and then incubated for 45 min with a rabbit anti-IgG secondary antibody coupled to horseradish peroxidase diluted at 1:10,000 in PBST. Proteins were detected by chemiluminescence using the Clarity Western ECL substrate detection kit (Bio-Rad) and a Chemidoc imaging system (Bio-rad). Whole proteins were detected on gel using the stain-free technology from Bio-Rad.\u003c/p\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003eAntibodies and staining reagents used in this work\u003c/h2\u003e \u003cp\u003eFLIPPER-TR\u0026reg; was used at 1\u0026micro;M in OptiMEM (Spirochrome AG, Switzerland) for 30mn before imaging of living infected cells. 4\u0026prime;,6\u0026prime;-diamidino-2-phenylindole (DAPI) was used at 1 \u0026micro;g/ml to stain eukaryotic and prokaryotic DNA. Alexa Fluor-conjugated phalloidin (Thermofisher, A22287) was used at the recommended concentration to stain filamentous actin. TS9 and TS81 \u003csup\u003e59\u003c/sup\u003e were used at 1:200 dilution to stain for CD9 and CD81 in cytometry and microscopy assays. TS151 \u003csup\u003e59\u003c/sup\u003e and 11B1G4 (generously provided by Leonie Ashman) \u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e were used at 1:200 dilution for CD151 staining in cytometry and microscopy respectively. Ezrin was stained with an anti-ezrin antibody at 1:500 dilution (generously provided by Dr. P. Mangeat (CNRS, UMR5539. Montpellier, France). CD44 was stained with proteintech 15675-1-AP antibody at 1:50 dilution. Alexa Fluor-conjugated goat IgG secondary antibodies (Thermofisher, A11029 and A11010) were used at 1:200 dilution in fluorescent microscopy. Clathrin polyclonal antibody (Thermofisher, PA5-143896) was used at 0.5 \u0026micro;g/mL in PM sheets preparation. Anti-PilV antibody \u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e and HRP-conjugated goat IgG secondary antibody (Thermofisher, 656120) were used in immunoblotting at 1:10 000 dilution.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec32\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses were performed with GraphPad Prism 8 or 9. When needed the data were log transformed. Normality of the distribution of whole data sets were assessed using QQ plot. The variance was assessed with a Brown-Forsythe test. When variance was different and the distribution was normal a Brown-Forsythe and Welch ANOVA or \u003cem\u003et\u003c/em\u003e-test was performed. For non-normal distribution, a non-parametric test was performed (Mann Whitney test or Kruskal-Wallis test). Corrections were indicated in the legend of figures. The H0 hypothesis was rejected for a significance level of p\u0026thinsp;\u0026le;\u0026thinsp;0.05. Figures have been created using GraphPad Prism and Adobe Illustrator. Raw data and statistical analysis are listed in the data_and_statistics and supplementary_data_and_statistics files.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the research grant ANR-19-CE14-0045-002 (to NB, PG, SM and MC) and funding from Inserm. AL and MW were supported by a scholarship from Université Paris Cité. KS fellowship was supported by the research grant ANR-19-CE14-0045-002.\u0026nbsp;We warmly thank: Nicolas Biais, Sandrine Bourdoulous, Etienne Morel, Cédric Delevoye and Philippe Chavrier for in-depth discussions; Béatrice Durel, Julie Lesieur, Thomas Guilbert and Pierre Bourdoncle for their expertise in microscopy (conventional, FRAP and FLIM); Jean-Baptiste Manneville for his expertise in biophysics. Image acquisition and image analysis were performed at the Imaging Facility of Structure Fédérative de Recherche (SFR) Necker, INSERM US24/CNRS UAR3633 and\u0026nbsp;the IMAG’IC Facility of the National Infrastructure France BioImaging (ANR-10-INBS-04). Flow cytometry and analysis was performed at the Cytometry facility of SFR Necker, INSERM US24/CNRS UAR3633. Lentiviruses were obtained from the Viral Vectors and Genes Transfer facility (VVTG) platform of SFR Necker, INSERM US24/CNRS UAR3633.\u0026nbsp;Illustrations were created with Biorender.com.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003edeclaration of interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, AL, KS, SM, MC; Methodology, AL, KS, BS, NG, GL, ER, SM, MC; Investigation, AL, KS, BS, VM, YW, MW, MR, TH, LL, MC, JM, PG; Validation, HL, EB, SM, MC ; Writing – Original Draft, AL, KS, SM, MC; Writing – Review \u0026amp; Editing, AL, KS, BS, NG, AJ, NB, PG, EB, GL, ER, SM, MC; Visualization, AL, KS, SM, MC; Funding Acquisition, \u0026nbsp;NB, SM, MC; Resources, NG, ER; Supervision, SM, MC.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eFarber, J.M., Ross, W.H., and Harwig, J. (1996). Health risk assessment of Listeria monocytogenes in Canada. Int. J. Food Microbiol. \u003cem\u003e30\u003c/em\u003e, 145\u0026ndash;156. https://doi.org/10.1016/0168-1605(96)01107-5.\u003c/li\u003e\n \u003cli\u003eMaudet, C., Kheloufi, M., Levallois, S., Gaillard, J., Huang, L., Gaultier, C., Tsai, Y.-H., Disson, O., and Lecuit, M. (2022). Bacterial inhibition of Fas-mediated killing promotes neuroinvasion and persistence. Nature \u003cem\u003e603\u003c/em\u003e, 900\u0026ndash;906. https://doi.org/10.1038/s41586-022-04505-7.\u003c/li\u003e\n \u003cli\u003eBarnier, J.-P., Euphrasie, D., Join-Lambert, O., Audry, M., Schonherr-Hellec, S., Schmitt, T., Bourdoulous, S., Coureuil, M., Nassif, X., and El Behi, M. (2021). Type IV pilus retraction enables sustained bacteremia and plays a key role in the outcome of meningococcal sepsis in a humanized mouse model. PLoS Pathog. \u003cem\u003e17\u003c/em\u003e, e1009299. https://doi.org/10.1371/journal.ppat.1009299.\u003c/li\u003e\n \u003cli\u003eMairey, E., Genovesio, A., Donnadieu, E., Bernard, C., Jaubert, F., Pinard, E., Seylaz, J., Olivo-Marin, J.-C., Nassif, X., and Dum\u0026eacute;nil, G. (2006). Cerebral microcirculation shear stress levels determine Neisseria meningitidis attachment sites along the blood-brain barrier. J. Exp. Med. \u003cem\u003e203\u003c/em\u003e, 1939\u0026ndash;1950. https://doi.org/10.1084/jem.20060482.\u003c/li\u003e\n \u003cli\u003eMikaty, G., Soyer, M., Mairey, E., Henry, N., Dyer, D., Forest, K.T., Morand, P., Guadagnini, S., Pr\u0026eacute;vost, M.C., Nassif, X., et al. (2009). Extracellular bacterial pathogen induces host cell surface reorganization to resist shear stress. PLoS Pathog. \u003cem\u003e5\u003c/em\u003e, e1000314. https://doi.org/10.1371/journal.ppat.1000314.\u003c/li\u003e\n \u003cli\u003eBerry, J.L., and Pelicic, V. (2015). Exceptionally widespread nanomachines composed of type IV pilins: the prokaryotic Swiss Army knives. FEMS Microbiol Rev \u003cem\u003e39\u003c/em\u003e, 134\u0026ndash;154. https://doi.org/10.1093/femsre/fuu001.\u003c/li\u003e\n \u003cli\u003eCraig, L., Forest, K.T., and Maier, B. (2019). Type IV pili: dynamics, biophysics and functional consequences. Nat. Rev. Microbiol. \u003cem\u003e17\u003c/em\u003e, 429\u0026ndash;440. https://doi.org/10.1038/s41579-019-0195-4.\u003c/li\u003e\n \u003cli\u003eLigthart, K., Belzer, C., de Vos, W.M., and Tytgat, H.L.P. (2020). Bridging Bacteria and the Gut: Functional Aspects of Type IV Pili. Trends Microbiol. \u003cem\u003e28\u003c/em\u003e, 340\u0026ndash;348. https://doi.org/10.1016/j.tim.2020.02.003.\u003c/li\u003e\n \u003cli\u003ePelicic, V. (2023). Mechanism of assembly of type 4 filaments: everything you always wanted to know (but were afraid to ask). Microbiol. Read. Engl. \u003cem\u003e169\u003c/em\u003e, 001311. https://doi.org/10.1099/mic.0.001311.\u003c/li\u003e\n \u003cli\u003eBucior, I., Pielage, J.F., and Engel, J.N. (2012). Pseudomonas aeruginosa pili and flagella mediate distinct binding and signaling events at the apical and basolateral surface of airway epithelium. PLoS Pathog \u003cem\u003e8\u003c/em\u003e, e1002616. https://doi.org/10.1371/journal.ppat.1002616.\u003c/li\u003e\n \u003cli\u003eCoureuil, M., Jamet, A., Bille, E., L\u0026eacute;cuyer, H., Bourdoulous, S., and Nassif, X. (2019). Molecular interactions between Neisseria meningitidis and its human host. Cell. Microbiol. \u003cem\u003e21\u003c/em\u003e, e13063. https://doi.org/10.1111/cmi.13063.\u003c/li\u003e\n \u003cli\u003eMarullo, S., Scott, M.G.H., Enslen, H., and Coureuil, M. (2022). Mechanical Activation of the \u0026beta;2-Adrenergic Receptor by Meningococcus: A Historical and Future Perspective Analysis of How a Bacterial Probe Can Reveal Signalling Pathways in Endothelial Cells, and a Unique Mode of Receptor Activation Involving Its N-Terminal Glycan Chains. Front. Endocrinol. \u003cem\u003e13\u003c/em\u003e, 883568. https://doi.org/10.3389/fendo.2022.883568.\u003c/li\u003e\n \u003cli\u003eMerz, A.J., Enns, C.A., and So, M. (1999). Type IV pili of pathogenic Neisseriae elicit cortical plaque formation in epithelial cells. Mol. Microbiol. \u003cem\u003e32\u003c/em\u003e, 1316\u0026ndash;1332. https://doi.org/10.1046/j.1365-2958.1999.01459.x.\u003c/li\u003e\n \u003cli\u003eMcKee, R.W., Aleksanyan, N., Garrett, E.M., and Tamayo, R. (2018). Type IV Pili Promote Clostridium difficile Adherence and Persistence in a Mouse Model of Infection. Infect. Immun. \u003cem\u003e86\u003c/em\u003e, e00943-17. https://doi.org/10.1128/IAI.00943-17.\u003c/li\u003e\n \u003cli\u003eMazariego-Espinosa, K., Cruz, A., Ledesma, M.A., Ochoa, S.A., and Xicohtencatl-Cortes, J. (2010). Longus, a type IV pilus of enterotoxigenic Escherichia coli, is involved in adherence to intestinal epithelial cells. J. Bacteriol. \u003cem\u003e192\u003c/em\u003e, 2791\u0026ndash;2800. https://doi.org/10.1128/JB.01595-09.\u003c/li\u003e\n \u003cli\u003eAroeti, B., Friedman, G., Zlotkin-Rivkin, E., and Donnenberg, M.S. (2012). Retraction of enteropathogenic E. coli type IV pili promotes efficient host cell colonization, effector translocation and tight junction disruption. Gut Microbes \u003cem\u003e3\u003c/em\u003e, 267\u0026ndash;271. https://doi.org/10.4161/gmic.19814.\u003c/li\u003e\n \u003cli\u003eOki, H., Kawahara, K., Maruno, T., Imai, T., Muroga, Y., Fukakusa, S., Iwashita, T., Kobayashi, Y., Matsuda, S., Kodama, T., et al. (2018). Interplay of a secreted protein with type IVb pilus for efficient enterotoxigenic Escherichia coli colonization. Proc. Natl. Acad. Sci. U. S. A. \u003cem\u003e115\u003c/em\u003e, 7422\u0026ndash;7427. https://doi.org/10.1073/pnas.1805671115.\u003c/li\u003e\n \u003cli\u003eRaynaud, C., Sheppard, D., Berry, J.-L., Gurung, I., and Pelicic, V. (2021). PilB from Streptococcus sanguinis is a bimodular type IV pilin with a direct role in adhesion. Proc. Natl. Acad. Sci. U. S. A. \u003cem\u003e118\u003c/em\u003e, e2102092118. https://doi.org/10.1073/pnas.2102092118.\u003c/li\u003e\n \u003cli\u003eVirion, Z., Doly, S., Saha, K., Lambert, M., Guillonneau, F., Bied, C., Duke, R.M., Rudd, P.M., Robbe-Masselot, C., Nassif, X., et al. (2019). Sialic acid mediated mechanical activation of \u0026beta;2 adrenergic receptors by bacterial pili. Nat. Commun. \u003cem\u003e10\u003c/em\u003e, 4752. https://doi.org/10.1038/s41467-019-12685-6.\u003c/li\u003e\n \u003cli\u003eLe Guennec, L., Virion, Z., Bouzinba-S\u0026eacute;gard, H., Robbe-Masselot, C., L\u0026eacute;onard, R., Nassif, X., Bourdoulous, S., and Coureuil, M. (2020). Receptor recognition by meningococcal type IV pili relies on a specific complex N-glycan. Proc. Natl. Acad. Sci. U. S. A. \u003cem\u003e117\u003c/em\u003e, 2606\u0026ndash;2612. https://doi.org/10.1073/pnas.1919567117.\u003c/li\u003e\n \u003cli\u003eShahin, M., Sheppard, D., Raynaud, C., Berry, J.-L., Gurung, I., Silva, L.M., Feizi, T., Liu, Y., and Pelicic, V. (2023). Characterization of a glycan-binding complex of minor pilins completes the analysis of Streptococcus sanguinis type 4 pili subunits. Proc. Natl. Acad. Sci. U. S. A. \u003cem\u003e120\u003c/em\u003e, e2216237120. https://doi.org/10.1073/pnas.2216237120.\u003c/li\u003e\n \u003cli\u003eRossy, T., Distler, T., Meirelles, L.A., Pezoldt, J., Kim, J., Tal\u0026agrave;, L., Bouklas, N., Deplancke, B., and Persat, A. (2023). Pseudomonas aeruginosa type IV pili actively induce mucus contraction to form biofilms in tissue-engineered human airways. PLoS Biol. \u003cem\u003e21\u003c/em\u003e, e3002209. https://doi.org/10.1371/journal.pbio.3002209.\u003c/li\u003e\n \u003cli\u003eMuir, A., Gurung, I., Cehovin, A., Bazin, A., Vallenet, D., and Pelicic, V. (2020). Construction of a complete set of Neisseria meningitidis mutants and its use for the phenotypic profiling of this human pathogen. Nat. Commun. \u003cem\u003e11\u003c/em\u003e, 5541. https://doi.org/10.1038/s41467-020-19347-y.\u003c/li\u003e\n \u003cli\u003eNolan, L.M., Whitchurch, C.B., Barquist, L., Katrib, M., Boinett, C.J., Mayho, M., Goulding, D., Charles, I.G., Filloux, A., Parkhill, J., et al. (2018). A global genomic approach uncovers novel components for twitching motility-mediated biofilm expansion in Pseudomonas aeruginosa. Microb. Genomics \u003cem\u003e4\u003c/em\u003e, e000229. https://doi.org/10.1099/mgen.0.000229.\u003c/li\u003e\n \u003cli\u003eRudel, T., Scheurerpflug, I., and Meyer, T.F. (1995). Neisseria PilC protein identified as type-4 pilus tip-located adhesin. Nature \u003cem\u003e373\u003c/em\u003e, 357\u0026ndash;359.\u003c/li\u003e\n \u003cli\u003eTreuner-Lange, A., Chang, Y.-W., Glatter, T., Herfurth, M., Lindow, S., Chreifi, G., Jensen, G.J., and S\u0026oslash;gaard-Andersen, L. (2020). PilY1 and minor pilins form a complex priming the type IVa pilus in Myxococcus xanthus. Nat. Commun. \u003cem\u003e11\u003c/em\u003e, 5054. https://doi.org/10.1038/s41467-020-18803-z.\u003c/li\u003e\n \u003cli\u003eBarnier, J.-P., Meyer, J., Kolappan, S., Bouzinba-S\u0026eacute;gard, H., Gesbert, G., Jamet, A., Frapy, E., Sch\u0026ouml;nherr-Hellec, S., Capel, E., Virion, Z., et al. (2021). The minor pilin PilV provides a conserved adhesion site throughout the antigenically variable meningococcal type IV pilus. Proc. Natl. Acad. Sci. U. S. A. \u003cem\u003e118\u003c/em\u003e, e2109364118. https://doi.org/10.1073/pnas.2109364118.\u003c/li\u003e\n \u003cli\u003eBiais, N., Ladoux, B., Higashi, D., So, M., and Sheetz, M. (2008). Cooperative retraction of bundled type IV pili enables nanonewton force generation. PLoS Biol. \u003cem\u003e6\u003c/em\u003e, e87. https://doi.org/10.1371/journal.pbio.0060087.\u003c/li\u003e\n \u003cli\u003eKoch, M.D., Fei, C., Wingreen, N.S., Shaevitz, J.W., and Gitai, Z. (2021). Competitive binding of independent extension and retraction motors explains the quantitative dynamics of type IV pili. Proc. Natl. Acad. Sci. \u003cem\u003e118\u003c/em\u003e, e2014926118. https://doi.org/10.1073/pnas.2014926118.\u003c/li\u003e\n \u003cli\u003eFerrer, J.M., Lee, H., Chen, J., Pelz, B., Nakamura, F., Kamm, R.D., and Lang, M.J. (2008). Measuring molecular rupture forces between single actin filaments and actin-binding proteins. Proc. Natl. Acad. Sci. \u003cem\u003e105\u003c/em\u003e, 9221\u0026ndash;9226. https://doi.org/10.1073/pnas.0706124105.\u003c/li\u003e\n \u003cli\u003eSchiavone, M., Sieczkowski, N., Castex, M., Trevisiol, E., Dague, E., and Fran\u0026ccedil;ois, J.M. (2019). AFM dendritips functionalized with molecular probes specific to cell wall polysaccharides as a tool to investigate cell surface structure and organization. Cell Surf. \u003cem\u003e5\u003c/em\u003e, 100027. https://doi.org/10.1016/j.tcsw.2019.100027.\u003c/li\u003e\n \u003cli\u003eCharles-Orszag, A., Tsai, F.C., Bonazzi, D., Manriquez, V., Sachse, M., Mallet, A., Salles, A., Melican, K., Staneva, R., Bertin, A., et al. (2018). Adhesion to nanofibers drives cell membrane remodeling through one-dimensional wetting. Nat Commun \u003cem\u003e9\u003c/em\u003e, 4450. https://doi.org/10.1038/s41467-018-06948-x.\u003c/li\u003e\n \u003cli\u003eSoyer, M., Charles-Orszag, A., Lagache, T., Machata, S., Imhaus, A.F., Dumont, A., Millien, C., Olivo-Marin, J.C., and Dumenil, G. (2014). Early sequence of events triggered by the interaction of Neisseria meningitidis with endothelial cells. Cell Microbiol \u003cem\u003e16\u003c/em\u003e, 878\u0026ndash;895. https://doi.org/10.1111/cmi.12248.\u003c/li\u003e\n \u003cli\u003eCoureuil, M., L\u0026eacute;cuyer, H., Scott, M.G.H., Boularan, C., Enslen, H., Soyer, M., Mikaty, G., Bourdoulous, S., Nassif, X., and Marullo, S. (2010). Meningococcus Hijacks a \u0026beta;2-adrenoceptor/\u0026beta;-Arrestin pathway to cross brain microvasculature endothelium. Cell \u003cem\u003e143\u003c/em\u003e, 1149\u0026ndash;1160. https://doi.org/10.1016/j.cell.2010.11.035.\u003c/li\u003e\n \u003cli\u003eBernard, S.C., Simpson, N., Join-Lambert, O., Federici, C., Laran-Chich, M.-P., Ma\u0026iuml;ssa, N., Bouzinba-S\u0026eacute;gard, H., Morand, P.C., Chretien, F., Taouji, S., et al. (2014). Pathogenic Neisseria meningitidis utilizes CD147 for vascular colonization. Nat. Med. \u003cem\u003e20\u003c/em\u003e, 725\u0026ndash;731. https://doi.org/10.1038/nm.3563.\u003c/li\u003e\n \u003cli\u003eMa\u0026iuml;ssa, N., Covarelli, V., Janel, S., Durel, B., Simpson, N., Bernard, S.C., Pardo-Lopez, L., Bouzinba-S\u0026eacute;gard, H., Faure, C., Scott, M.G.H., et al. (2017). Strength of Neisseria meningitidis binding to endothelial cells requires highly-ordered CD147/\u0026beta;2-adrenoceptor clusters assembled by alpha-actinin-4. Nat. Commun. \u003cem\u003e8\u003c/em\u003e, 15764. https://doi.org/10.1038/ncomms15764.\u003c/li\u003e\n \u003cli\u003eMerz, A.J., and So, M. (1997). Attachment of piliated, Opa- and Opc- gonococci and meningococci to epithelial cells elicits cortical actin rearrangements and clustering of tyrosine-phosphorylated proteins. Infect Immun \u003cem\u003e65\u003c/em\u003e, 4341\u0026ndash;4349.\u003c/li\u003e\n \u003cli\u003eHoffmann, I., Eug\u0026egrave;ne, E., Nassif, X., Couraud, P.O., and Bourdoulous, S. (2001). Activation of ErbB2 receptor tyrosine kinase supports invasion of endothelial cells by Neisseria meningitidis. J. Cell Biol. \u003cem\u003e155\u003c/em\u003e, 133\u0026ndash;143. https://doi.org/10.1083/jcb.200106148.\u003c/li\u003e\n \u003cli\u003eLambotin, M., Hoffmann, I., Laran-chich, M.P., Nassif, X., Couraud, P.O., and Bourdoulous, S. (2005). Invasion of endothelial cells by Neisseria meningitidis requires cortactin recruitment by a PI3-Kinase/Rac1 signalling pathway triggered by the lipo-oligosaccharide. J Cell Sci \u003cem\u003e118\u003c/em\u003e, 3805\u0026ndash;3816.\u003c/li\u003e\n \u003cli\u003eCoureuil, M., Mikaty, G., Miller, F., L\u0026eacute;cuyer, H., Bernard, C., Bourdoulous, S., Dum\u0026eacute;nil, G., M\u0026egrave;ge, R.-M., Weksler, B.B., Romero, I.A., et al. (2009). Meningococcal type IV pili recruit the polarity complex to cross the brain endothelium. Science \u003cem\u003e325\u003c/em\u003e, 83\u0026ndash;87. https://doi.org/10.1126/science.1173196.\u003c/li\u003e\n \u003cli\u003eSlanina, H., Hebling, S., Hauck, C.R., and Schubert-Unkmeir, A. (2012). Cell invasion by Neisseria meningitidis requires a functional interplay between the focal adhesion kinase, Src and cortactin. PLoS One \u003cem\u003e7\u003c/em\u003e, e39613. https://doi.org/10.1371/journal.pone.0039613.\u003c/li\u003e\n \u003cli\u003eDharan, R., Goren, S., Cheppali, S.K., Shendrik, P., Brand, G., Vaknin, A., Yu, L., Kozlov, M.M., and Sorkin, R. (2022). Transmembrane proteins tetraspanin 4 and CD9 sense membrane curvature. Proc. Natl. Acad. Sci. U. S. A. \u003cem\u003e119\u003c/em\u003e, e2208993119. https://doi.org/10.1073/pnas.2208993119.\u003c/li\u003e\n \u003cli\u003eWolverson, P.A., Parreira, I.F., Collins, M.O., Shaw, J.G., and Green, L.R. (2024). Dynamics of the CD9 interactome during bacterial infection of epithelial cells by proximity labelling proteomics. Preprint at bioRxiv, https://doi.org/10.1101/2024.12.13.628358 https://doi.org/10.1101/2024.12.13.628358.\u003c/li\u003e\n \u003cli\u003eFehon, R.G., McClatchey, A.I., and Bretscher, A. (2010). Organizing the cell cortex: the role of ERM proteins. Nat. Rev. Mol. Cell Biol. \u003cem\u003e11\u003c/em\u003e, 276\u0026ndash;287. https://doi.org/10.1038/nrm2866.\u003c/li\u003e\n \u003cli\u003eBulut, G., Hong, S.-H., Chen, K., Beauchamp, E.M., Rahim, S., Kosturko, G.W., Glasgow, E., Dakshanamurthy, S., Lee, H.-S., Daar, I., et al. (2012). Small molecule inhibitors of ezrin inhibit the invasive phenotype of osteosarcoma cells. Oncogene \u003cem\u003e31\u003c/em\u003e, 269\u0026ndash;281. https://doi.org/10.1038/onc.2011.245.\u003c/li\u003e\n \u003cli\u003eMaier, B., Koomey, M., and Sheetz, M.P. (2004). A force-dependent switch reverses type IV pilus retraction. Proc Natl Acad Sci U A \u003cem\u003e101\u003c/em\u003e, 10961\u0026ndash;10966. https://doi.org/10.1073/pnas.0402305101.\u003c/li\u003e\n \u003cli\u003eBonsergent, E., Grisard, E., Buchrieser, J., Schwartz, O., Th\u0026eacute;ry, C., and Lavieu, G. (2021). Quantitative characterization of extracellular vesicle uptake and content delivery within mammalian cells. Nat. Commun. \u003cem\u003e12\u003c/em\u003e, 1864. https://doi.org/10.1038/s41467-021-22126-y.\u003c/li\u003e\n \u003cli\u003eBonsergent, E., and Lavieu, G. (2019). Content release of extracellular vesicles in a cell-free extract. FEBS Lett. \u003cem\u003e593\u003c/em\u003e, 1983\u0026ndash;1992. https://doi.org/10.1002/1873-3468.13472.\u003c/li\u003e\n \u003cli\u003eVirji, M., Makepeace, K., Ferguson, D.J., Achtman, M., and Moxon, E.R. (1993). Meningococcal Opa and Opc proteins: their role in colonization and invasion of human epithelial and endothelial cells. Mol Microbiol \u003cem\u003e10\u003c/em\u003e, 499\u0026ndash;510.\u003c/li\u003e\n \u003cli\u003eWeksler, B.B., Subileau, E.A., Perriere, N., Charneau, P., Holloway, K., Leveque, M., Tricoire-Leignel, H., Nicotra, A., Bourdoulous, S., Turowski, P., et al. (2005). Blood-brain barrier-specific properties of a human adult brain endothelial cell line. Faseb J \u003cem\u003e19\u003c/em\u003e, 1872\u0026ndash;1874.\u003c/li\u003e\n \u003cli\u003eNassif, X., Lowy, J., Stenberg, P., O\u0026rsquo;Gaora, P., Ganji, A., and So, M. (1993). Antigenic variation of pilin regulates adhesion of Neisseria meningitidis to human epithelial cells. Mol. Microbiol. \u003cem\u003e8\u003c/em\u003e, 719\u0026ndash;725. https://doi.org/10.1111/j.1365-2958.1993.tb01615.x.\u003c/li\u003e\n \u003cli\u003eMorand, P.C., Tattevin, P., Eugene, E., Beretti, J.L., and Nassif, X. (2001). The adhesive property of the type IV pilus-associated component PilC1 of pathogenic Neisseria is supported by the conformational structure of the N-terminal part of the molecule. Mol Microbiol \u003cem\u003e40\u003c/em\u003e, 846-56.\u003c/li\u003e\n \u003cli\u003eJoin-Lambert, O., Lecuyer, H., Miller, F., Lelievre, L., Jamet, A., Furio, L., Schmitt, A., Pelissier, P., Fraitag, S., Coureuil, M., et al. (2013). Meningococcal interaction to microvasculature triggers the tissular lesions of purpura fulminans. J. Infect. Dis. \u003cem\u003e208\u003c/em\u003e, 1590\u0026ndash;1597. https://doi.org/10.1093/infdis/jit301.\u003c/li\u003e\n \u003cli\u003eKellogg, D.S., Cohen, I.R., Norins, L.C., Schroeter, A.L., and Reising, G. (1968). Neisseria gonorrhoeae. II. Colonial variation and pathogenicity during 35 months in vitro. J. Bacteriol. \u003cem\u003e96\u003c/em\u003e, 596\u0026ndash;605. https://doi.org/10.1128/jb.96.3.596-605.1968.\u003c/li\u003e\n \u003cli\u003ede Chaumont, F., Dallongeville, S., Chenouard, N., Herv\u0026eacute;, N., Pop, S., Provoost, T., Meas-Yedid, V., Pankajakshan, P., Lecomte, T., Le Montagner, Y., et al. (2012). Icy: an open bioimage informatics platform for extended reproducible research. Nat. Methods \u003cem\u003e9\u003c/em\u003e, 690\u0026ndash;696. https://doi.org/10.1038/nmeth.2075.\u003c/li\u003e\n \u003cli\u003ePaul-Gilloteaux, P., Heiligenstein, X., Belle, M., Domart, M.-C., Larijani, B., Collinson, L., Raposo, G., and Salamero, J. (2017). eC-CLEM: flexible multidimensional registration software for correlative microscopies. Nat. Methods \u003cem\u003e14\u003c/em\u003e, 102\u0026ndash;103. https://doi.org/10.1038/nmeth.4170.\u003c/li\u003e\n \u003cli\u003eSchindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., et al. (2012). Fiji: an open-source platform for biological-image analysis. Nat. Methods \u003cem\u003e9\u003c/em\u003e, 676\u0026ndash;682. https://doi.org/10.1038/nmeth.2019.\u003c/li\u003e\n \u003cli\u003eBerg, S., Kutra, D., Kroeger, T., Straehle, C.N., Kausler, B.X., Haubold, C., Schiegg, M., Ales, J., Beier, T., Rudy, M., et al. (2019). ilastik: interactive machine learning for (bio)image analysis. Nat. Methods \u003cem\u003e16\u003c/em\u003e, 1226\u0026ndash;1232. https://doi.org/10.1038/s41592-019-0582-9.\u003c/li\u003e\n \u003cli\u003eCharrin, S., Le Naour, F., Oualid, M., Billard, M., Faure, G., Hanash, S.M., Boucheix, C., and Rubinstein, E. (2001). The major CD9 and CD81 molecular partner. Identification and characterization of the complexes. J. Biol. Chem. \u003cem\u003e276\u003c/em\u003e, 14329\u0026ndash;14337. https://doi.org/10.1074/jbc.M011297200.\u003c/li\u003e\n \u003cli\u003eSincock, P.M., Fitter, S., Parton, R.G., Berndt, M.C., Gamble, J.R., and Ashman, L.K. (1999). PETA-3/CD151, a member of the transmembrane 4 superfamily, is localised to the plasma membrane and endocytic system of endothelial cells, associates with multiple integrins and modulates cell function. J. Cell Sci. \u003cem\u003e112 ( Pt 6)\u003c/em\u003e, 833\u0026ndash;844. https://doi.org/10.1242/jcs.112.6.833.\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":"[email protected]","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":"","lastPublishedDoi":"10.21203/rs.3.rs-6302813/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6302813/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eOnce passed into the bloodstream, bacterial pathogens have a limited time to interact with permissive receptors at the surface of host cells. \u003cem\u003eNeisseria meningitidis\u003c/em\u003e has developed an extremely effective strategy allowing it to find its receptors in a few seconds. Here, we report that \u003cem\u003eN. meningitidis \u003c/em\u003etype IV pili exploit the physical properties of host cells plasma membranes to promote the formation of early tubular membrane structures essential for initial bacterial adhesion. These tubular structures, which form before any signaling events in host cells, concentrate and trap multiple plasma membrane-associated proteins in the vicinity of bacteria, thereby facilitating the selection, interaction and activation of specific adhesion and signaling receptors by bacterial ligands bound to type IV pili. Our results define an additional paradigm for the recruitment of specific receptors by pathogenic bacteria, which depends on the physical property of bacterial pili to induce the formation of tubular plasma membrane structures enriched in integral membrane receptor.\u003c/p\u003e","manuscriptTitle":"Meningococci drive host membrane tubulation to recruit their signaling receptors","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-09 11:01:37","doi":"10.21203/rs.3.rs-6302813/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","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":"84677284-b094-4b87-9128-333f0cdd333f","owner":[],"postedDate":"May 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":48215966,"name":"Biological sciences/Microbiology/Cellular microbiology"},{"id":48215967,"name":"Biological sciences/Cell biology/Cell signalling"}],"tags":[],"updatedAt":"2025-11-26T08:17:16+00:00","versionOfRecord":{"articleIdentity":"rs-6302813","link":"https://doi.org/10.1038/s41467-025-65436-1","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2025-11-25 05:00:00","publishedOnDateReadable":"November 25th, 2025"},"versionCreatedAt":"2025-05-09 11:01:37","video":"","vorDoi":"10.1038/s41467-025-65436-1","vorDoiUrl":"https://doi.org/10.1038/s41467-025-65436-1","workflowStages":[]},"version":"v1","identity":"rs-6302813","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6302813","identity":"rs-6302813","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}