https://orcid.org/0000-0002-9852-1722
This is an uncorrected proof.
Figures
Abstract
Leptospira interrogans is a causative agent of leptospirosis, a potentially life-threatening zoonotic disease. A major symptom of severe leptospirosis is acute kidney failure; this is characterized by the detachment of renal proximal tubule epithelial cells (RPTECs) from the basement membrane, which can lead to chronic kidney disease. The kidneys possess an intrinsic ability to self-repair following acute injury through collective cell migration; this process involves the movement of cells as cohesive units. Mechanical forces transmitted between cells via mechanosensors are important for this migration process. L. interrogans induces the displacement of mechanosensors from the plasma membrane, which can be prevented by inhibitors of eukaryotic proteolytic pathways. However, the effects of mechanosensors dysfunction on epithelial repair processes have not been investigated yet. Therefore, the aim of this study was to analyze proteasomal inhibition during epithelial barrier disruption and elucidate epithelial repair mechanisms by examining collective migration of infected RPTECs. We performed a proteome-wide analysis of protein ubiquitination in infected-RPTECs to understand the effect of proteasomal inhibitors. Moreover, we detected increased ubiquitination of ADP-ribosylation factor-like protein 2 (ARL2), a protein involved in microtubule dynamics. Immunofluorescence analysis revealed that L. interrogans induces proteasome-dependent disruption of the microtubule and F-actin cytoskeletal network. Scratch assays revealed aberrant collective cell migration and delayed wound healing in L. interrogans-infected RPTECs. Wound healing was improved by proteasomal inhibition followed by the addition of epidermal growth factor. In conclusion, our findings suggest that L. interrogans alters the circuit integrating mechanosensors at cell–cell contact junctions and the supracellular F-actin and microtubule cytoskeletal network, thereby disrupting epithelial repair responses.
Author summary
Leptospirosis is a bacterial zoonotic, waterborne disease mainly caused by Leptospira interrogans. One of the symptoms of severe leptospirosis is acute kidney failure, which can develop into chronic kidney disease. Leptospira–renal epithelial cell interactions induce epithelial barrier disruption, which can be prevented using eukaryotic proteolytic system inhibitors. The kidney possesses an intrinsic ability to self-repair after an acute injury, primarily through collective cell migration. However, the epithelial cell repair response to leptospiral infection has not yet been investigated. In this study, we demonstrate that Leptospira-induced cytoskeletal network disruption results in aberrant collective epithelial cell migration and delayed wound healing. Inhibition of cytoskeletal network disruption enhances collective cell migration. Overall, elucidation of tissue-scale repair mechanisms during infection enables the development of therapeutic strategies for preventing pathogen dissemination and chronic damage.
Citation: Barbee H, Sebastián I, Tokumon R, Wang Y, Yamashiro T, Toma C (2026) Proteasome-dependent cytoskeleton disruption during Leptospira interrogans infection induces aberrant collective cell migration. PLoS Pathog 22(8): e1014501. https://doi.org/10.1371/journal.ppat.1014501
Editor: Jon T. Skare, Texas A&M University, UNITED STATES OF AMERICA
Received: April 6, 2026; Accepted: July 26, 2026; Published: August 7, 2026
Copyright: © 2026 Barbee et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are in the manuscript and/or supporting information files. The proteomic analysis reported in this article have been deposited in ProteomeXchange and jPost under the dataset identifiers PXD073976 and JPST004377, respectively.
Funding: This work was supported by the Japan Society for the Promotion of Science (JSPS:https://www.jsps.go.jp/) KAKENHI 24K02281 and a Grant from the Chemo-Sero-Therapeutic Research Institute (https://www.kaketsuken.org/) to C.T. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The kidneys possess the intrinsic ability to self-repair after acute injury [1]. Surviving renal proximal tubule epithelial cells (RPTECs) that remain attached to the basement membrane contribute to tubular epithelium regeneration through collective cell migration and proliferation to mitigate the loss of neighboring cells [2,3]. Collective cell migration involves the movement of cells as a cohesive unit that maintains physical contact and influences each other’s behavior [4–6]. During migration, mechanical forces are transmitted between cells through mechanosensors located at the adherens junctions (AJs), cytoskeleton, and focal adhesions [7–9]. Furthermore, epidermal growth factor receptor (EGFR) activation induces the intercellular coupling of extracellular signal-regulated kinase (ERK) activation, which generates mechanochemical waves that guide the collective cell migration [10]. Following airway epithelium injury, repair is characterized by the spreading and migration of cells at the leading edge that surrounds the wound site. Abnormal wound repair has been observed in the airway epithelium of patients with chronic respiratory diseases [11]. Thus, understanding tissue repairing mechanisms is important for developing innovative therapeutic strategies for chronic complications.
Leptospirosis is a bacterial zoonotic, waterborne disease caused by pathogenic Leptospira spp., with L. interrogans being the major causative agent within the genus [12,13]. The clinical manifestations of leptospirosis range from asymptomatic to multiple organ dysfunction, including renal failure, jaundice, pulmonary hemorrhage, and meningitis [14,15]. Leptospiral kidney infection causes renal inflammation and damage; if the kidney does not heal, the disease might progress to chronic complications [16,17]. Inflammatory signaling pathways in infected kidneys have been explored in several studies; however, tissue repair mechanisms in Leptospira-infected renal epithelial cells have not yet been reported [18,19].
Several research groups, including ours, have reported that Leptospira–host cell interactions induce epithelial barrier disruption [20–22]. The following mechanisms have been reported: degradation of E-cadherin (E-cad) scaffold proteins (p0071 and p120-catenin [p120ctn]), and the AJ–F-actin linker protein (afadin); activation of the calmodulin/myosin light chain kinase cellular pathway [22–24]. However, the additional mechanisms remain unclear. Genomic comparisons of different strains and analyses of differentially expressed leptospiral proteins in animal models have been performed to identify the virulence factors responsible for F-actin disorganization [25,26]. A family of 12 proteins called virulence modifying proteins (VMPs), found only in pathogenic strains, show increased expression during infection; these proteins have emerged as the bacterial factors responsible for the Leptospira-induced F-actin disruption [27]. Furthermore, a detailed functional analysis of one of these VMPs (LA3490) was performed using HeLa cells; within 2 h of the addition of purified recombinant LA3490, this protein disrupted F-actin and induced cell detachment [28]. However, host cell responses to bacterial infection depend not only on the secreted virulence proteins, but also on several bacterial factors; these factors include heat-stable pathogen-associated molecular patterns (PAMPs), such as lipopolysaccharides (LPSs) and peptidoglycans, and the cell type used for analysis [29,30]. Immortal HeLa cells have contributed to many discoveries in biomedical research; however, HeLa cells exhibit abnormal tight junctions, which can impair the proper sensing of mechanical forces transmitted through cell–cell junctions. Thus, the mechanosensors in HeLa might not be able to induce normal cellular homeostatic responses [31].
Immunofluorescence analysis of leptospiral kidney colonization kinetics in a mouse infection model showed that L. interrogans attaches to the basal membrane of the RPTECs before translocation into the lumen [32]. Thus, we developed an in vitro infection model to analyze leptospiral–host cell interactions in RPTEC monolayers that can sense mechanical forces and respond collectively to external injuries [21]. We reported that L. interrogans and the non-pathogenic L. biflexa induce E-cad cleavage; however, only the pathogenic strain induces E-cad endocytosis and F-actin cytoskeletal disorganization. Consequently, transepithelial electrical resistance (TEER) of L. interrogans-infected RPTECs decreases [21,24]. Further detailed analysis of RPTECs infected with the highly pathogenic L. interrogans serovar Manilae UP-MMC-NIID strain (Mani) [33] revealed that a combination of inhibitors of eukaryotic proteolytic pathways (lysosomal: bafilomycin A, BF; proteasomal: MG132, MG) effectively prevented epithelial barrier disruption and p120-ctn degradation [23]. Protein degradation by the ubiquitin-proteasome system (UPS) requires the addition of one to several 76-amino acid-polypeptides (ubiquitin) to tag the protein for degradation [34]. However, we did not detect p120-ctn ubiquitination following Leptospira infection [23].
Antibiotic treatment can resolve leptospirosis; however, some complications such as the Jarisch-Herxheimer reaction, can occur within hours of treatment initiation, highlighting the need for additional therapeutic options [35]. Elucidation of the multifaceted strategy used by L. interrogans to disrupt host epithelial barrier regulation will guide the development of new approaches for treating leptospirosis and preventing leptospirosis-associated renal failure. To achieve this goal, in this study, we analyzed proteasomal inhibition during epithelial barrier disruption and elucidated epithelial repair mechanisms by examining collective cell migration in Mani-infected RPTECs. Overall, our findings show that L. interrogans induces proteasome-dependent disruption of the microtubule and F-actin cytoskeletal network which results in aberrant collective cell migration and delayed wound healing.
Results
Cytoskeletal F-actin disorganization after leptospiral internalization into RPTECs
In our previous study, we showed that a combination of lysosomal (bafilomycin A, BF) and proteasomal (MG132, MG) inhibitors prevents Mani-induced epithelial barrier disruption [23]. Heat-stable pathogen-associated molecular patterns (PAMPs), such as lipopolysaccharides (LPSs), drive proinflammatory cytokines production via UPS-regulated pathways and induced epithelial barrier disruption [36]. Thus, we analyzed the epithelial integrity in RPTECs treated with live or heat-killed bacteria to determine whether heat-stable PAMPs are involved in Leptospira-induced epithelial barrier disruption. Incubation of RPTECs with heat-killed bacteria added at the basolateral side did not induce a transepithelial electrical resistance (TEER) decrease or F-actin disorganization (S1 Fig). As the non-motile heat-killed bacteria could not reach the RPTEC plasma membrane, we next performed a comparison of live and heat-killed bacteria added at the apical side, to rule out that the inability of heat-killed bacteria to induce TEER decrease is related to the loss of motility. Apical infection induced a TEER decrease and F-actin disorganization with live but not with heat-killed bacteria (S2 Fig). These findings suggest that live bacteria and/or heat-labile bacterial factors are essential to induce epithelial barrier disruption. The expression of secreted VMPs was upregulated in vivo; thus, these proteins were identified as key factors mediating L. interrogans-induced epithelial barrier disruption [22]. However, the proteins secreted during RPTEC infection did not induce TEER decrease or disassembly of AJs (S3 Fig). These results suggest that Leptospira–RPTEC interactions are important for epithelial barrier disruption in the RPTEC infection model.
A mechanism for disruption of F-actin dynamics via leptospiral attachment to endothelial cells surface has been reported previously [20]. Moreover, we have shown that leptospires maintain close contact with the RPTEC plasma membrane via extracellular adhesion or crawling; however, we did not observe a link between bacterial adherence and F-actin disruption in our previous study [21]. L. interrogans is considered an extracellular pathogen; however, bacteria have also been observed intracellularly at the early stage of infection in polarized kidney epithelial cells [21,37]. Because bacterial internalization into host cells can induce F-actin disruption [20], we analyzed early bacteria–RPTECs interactions in the presence or absence of BF + MG to determine whether the inhibitors affect bacterial cell adhesion and/or cell entry. Leptospires associated with dimethyl sulfoxide (DMSO)-treated and BF + MG-treated RPTECs were analyzed via immunofluorescence staining at 2 and 5 h post-infection (p.i.). Dual immunofluorescence of bacteria and lysosomal-associated membrane protein 1 (LAMP1) revealed that leptospires were located intracellularly within LAMP1-positive vacuoles in both DMSO-treated and BF + MG-treated RPTECs (Fig 1A). Next, we performed a gentamicin protection assay to quantify cell-associated (both extracellular and intracellular bacteria) and intracellular leptospires (gentamicin-protected bacteria) to understand the kinetics of RPTEC infection. In DMSO-treated RPTECs, intracellular leptospires were not detectable at 2 h p.i.; significantly more cell-associated than intracellular leptospires were detected at 5 h p.i. (p = 0.0135) (Fig 1B). In contrast, in BF + MG-treated RPTECs, live intracellular leptospires were observed at 2 h p.i.; however, the number of cell-associated leptospires significantly decreased at 5 h p.i. (p = 0.0177) to a level comparable to that of intracellular leptospires. These results suggest that a population of leptospires are internalized into RPTECs within 2 h p.i. and was rapidly killed, and this process was prevented by BF + MG treatment. Although the intracellular population survived in BF + MG-treated RPTECs, leptospiral adherence to the RPTECs was impaired at 5 h p.i. (Fig 1B).
RPTECs were pre-treated for 30 min with a combination of lysosomal and proteasomal inhibitors (bafilomycin A1: BF and MG132: MG) or dimethyl sulfoxide (DMSO; control) and then infected with Leptospira interrogans serovar Manilae UP-MMC-NIID strain (Mani). (A) Representative confocal images of infected cells at 2 and 5 h post-infection (p.i.). Bacteria were stained with Cy3-labeled anti-Mani antibodies (red) and lysosomal-associated membrane protein 1 (LAMP1) was stained with Alexa Fluor 488-labeled antibodies (green). Cell nuclei were stained with TO-PRO-3 (blue). White arrowheads: Mani-containing LAMP1 positive vacuoles. (B) Cell-associated and intracellular Mani were quantified using gentamicin protection assay. (C) Representative confocal images of non-infected and Mani-infected cells at 5 and 24 h p.i. Mani was stained with Alexa Fluor 488-labeled antibodies (green) and F-actin was stained with rhodamine-phalloidin (red). Cell nuclei were stained with TO-PRO-3 (blue). (D) Transepithelial electrical resistance (TEER) measurements at 5 and 24 h p.i. The values are the mean ± SD of at least three independent Transwells. *p < 0.05 and **p < 0.01. Scale bars: 10 μm.
Next, we analyzed F-actin structures and TEER at 5 and 24 h p.i. in non-infected and Mani-infected RPTECs. The cytoskeletal F-actin and epithelial barrier were not disrupted in non-infected RPTECs (Fig 1C and 1D). In Mani-infected RPTECs, F-actin belts were disorganized or disappeared from cell–cell junctions and TEER significantly decreased at 24 h p.i (p = 0.0053) (Fig 1C and 1D). Collectively, these results suggest that epithelial barrier disruption was not induced by bacterial adherence or cell entry, and sustained F-actin disorganization occurred after 5 h p.i.
Ubiquitinomic analysis detected ADP-ribosylation factor-like protein 2 (ARL2): A regulator of microtubule dynamics
Since F-actin disorganization occurred after 5 h p.i. (Fig 1C and 1D), we hypothesized that Mani-induced ubiquitination of host protein(s) leads to late-stage epithelial barrier disruption following RPTEC infection. The addition of BF + MG prior to infection prevented leptospiral induced TEER decrease (p = 0.00159), as previously reported [23]. Next, we tested the effect of BF + MG addition after infection. BF + MG added at 5 h p.i. prevented Mani-induced TEER decrease (p = 0.00769) (Fig 2A) and ubiquitinated protein degradation (Fig 2B). Therefore, further analyses were performed by adding BF + MG at 5 h p.i.
RPTECs were treated with a combination of lysosomal and proteasomal inhibitors (bafilomycin A1: BF and MG132: MG) or dimethyl sulfoxide (DM; control) 30 min before infection or at 5 h p.i. with Mani. (A) Transepithelial electrical resistance (TEER) measurements at 24 h p.i. Data are presented as the mean ± SD of at least three independent Transwell assay results. ** p < 0.01 (B) Whole cell lysates were subjected to western blotting to detect ubiquitinated proteins at 21 h p.i. (n.i.: non-infected). (C) Workflow of the experimental approach to identify ubiquitinated proteins. (D) Table listing the differentially ubiquitinated proteins at 21h p.i. (related to S1 Table). (E) RPTECs were infected with L. biflexa or Mani, and BF + MG was added at 5 h p.i. Whole-cell lysates were subjected to western blotting to detect ARL2 at 21 h p.i., using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a loading control, or (F and G) infected RPTECs were fixed with 2% paraformaldehyde (PFA) and prepared for immunofluorescence analysis. (F) Representative confocal images of infected RPTECs. Microtubules (β-tubulin) were stained with Alexa Fluor 488-labeled antibodies (green). Scale bar: 10 μm. (G) Relative fluorescence intensity of β-tubulin. ** p < 0.01.
To identify the proteins that showed increased ubiquitination following Mani infection, we performed a proteome-wide analysis of Mani- and L. biflexa-infected RPTECs using diglycine enrichment and mass spectrometry (Fig 2C). A comparison of the ubiquitinomes at 5 (just before adding the inhibitors) and 21 h p.i. revealed that 113 proteins showed more than a two-fold increase in ubiquitination in infected RPTECs at 21 h p.i. The ubiquitinome dataset was deposited in ProteomeXchange and jPost [38] under the dataset identifiers PXD073976 and JPST004377, respectively. Seventy proteins showed >100-fold higher ubiquitination rate in Mani- and L. biflexa-infected RPTECs compared with that in the control. Because the Proteome Discoverer could not quantify expression differences exceeding 100-fold, we could not determine whether ubiquitination of these 70 proteins increased during Mani infection (S1 Table). Of the remaining 43 proteins, six showed more than two-fold greater ubiquitination in Mani- compared with L. biflexa-infected RPTECs (Fig 2D; S1 Table). Two of these six proteins, namely, adenosine diphosphate (ADP)-ribosylation factor-like protein 2 (ARL2, P36404) and cytohesin-interacting protein (CYTIP, O60759), are involved in cytoskeletal dynamics [39,40].
ARL2 inhibits tubulin specific chaperon D (TBCD)-dependent epithelial barrier disruption [41]. Therefore, we focused on ARL2, a 184 amino-acids small GTP-binding protein that showed diglycine at Lysine 53 as the ubiquitination site. We hypothesized that increased ubiquitination of ARL2 might induce its degradation via UPS. However, immunoblotting results showed that ARL2 protein levels did not decrease in Mani- infected RPTECs (Fig 2E). Functional ARL2 is also required for maintaining the microtubule network through a complex consisting of ARL2, TBCD, and β-tubulin [39]. Thus, we hypothesized that increased ARL2 ubiquitination might induce a defect in this trimer. Immunofluorescence analysis revealed a disrupted microtubule network in Mani-infected RPTECs compared to that in L. biflexa-infected RPTECs (p = 0.00053), this disruption was prevented by BF + MG treatment (Fig 2F and 2G). Collectively, these data suggest that UPS-mediated degradation of ARL2 is not responsible for the microtubule network defects induced by Mani. Treatment of RPTECs with BF + MG prevents microtubule destabilization and epithelial barrier disruption induced by the bacterial factor or Leptospira-activated signaling pathway after 5 h p.i.
Mani induces proteasome-dependent microtubule and F-actin cytoskeletal network disruption
To investigate the role of UPS in Mani-induced epithelial barrier disruption, we examined the effects of proteasomal inhibitors (MG or bortezomib [BZ]). Both inhibitors are peptide-base proteasome inhibitors which act primarily on the chymotrypsin-like site in the β-subunit of the 26S proteasome; however, they have different chemical structures [42]. We used two different inhibitors to ensure that the Mani-induced microtubule–F-actin network disruption is proteasome-dependent because chemical inhibitors can potentially have off-target effects by binding and altering unintended molecules. Mani-infected RPTECs showed a significant TEER decrease (p = 0.00159) which was prevented by proteasome inhibitors (Fig 3A). Furthermore, confocal images showed that Mani induces proteasome-dependent microtubule and F-actin cytoskeleton network disruption (p = 0.03583) (Fig 3B and 3C) at 24 h p.i.
RPTECs were infected with Mani from the basolateral side and treated with inhibitors (MG132: MG or Bortezomib: BZ) or DMSO (control) at 5 h p.i. (A) Transepithelial electrical resistance (TEER) measurements at 24 h p.i. Data are presented as the mean ± SD of at least three independent Transwell assay results. ** p < 0.01 (B) Representative confocal images of infected cells at 24 h p.i. Microtubules (β-tubulin) were stained with Alexa Fluor 488-labeled antibodies (green), F-actin with rhodamine-phalloidin (red), and Mani with Alexa 647-labeled antibodies (artificially recolored to white). Scale bar: 10 μm. (C) Relative fluorescence intensity of β-tubulin. * p < 0.05.
Recently, we analyzed clinical leptospiral strains and showed strain-dependent mechanisms of epithelial barrier disruption [24]. In one of these strains, epithelial barrier disruption and degradation of afadin, a protein that links the AJs to the actin cytoskeleton, were prevented by the single addition of MG or BZ [24]. Western blotting analyses revealed that afadin levels decreased in Mani-infected RPTECs compared with that in the control. However, only MG partially prevented afadin degradation (~ 50% of non-infected cells) (S4 Fig). Collectively, these results show that in Mani-infected RPTECs, proteasomal inhibitors maintained epithelial barrier integrity, despite afadin degradation, by preventing the supracellular F-actin and microtubule cytoskeletal network disruption.
Proteasomal inhibition prevents aberrant collective cell migration of Mani-infected RPTECs
The microtubule and F-actin cytoskeletal network and AJ proteins are important mechanosensors to induce epithelial cell responses and maintain epithelial barrier function in dynamic environments [43,44]. Thus, we hypothesized that proteasomal inhibitors could prevent Mani-induced cytoskeletal disruption and enhance epithelial cell responses by overcoming multiple pathogenic leptospiral mechanisms. Next, we evaluated tissue-level changes in subconfluent RPTECs infected with L. biflexa or Mani. Time-lapse imaging revealed that the cell-free areas in L. biflexa-infected RPTECs were covered with cells over time, resulting in the formation of monolayers. In contrast, in Mani-infected RPTECs, some cell-free areas became smaller; however, some areas became larger, and new cell-free areas appeared. Consequently, the epithelial monolayer was not formed (Fig 4; S1 and S2 Movies). The size reduction of some cell-free areas in Mani-infected RPTECs (yellow arrow in Fig 4) suggests that, at the tissue level, the collective cell migration of RPTECs can overcome the pathogenic mechanism.
Images of subconfluent RPTECs in collagen-coated dishes infected with L. biflexa or Mani (related to S1 and S2 Movies). The percentage of cell-free area calculated with ImageJ is denoted at the left bottom side of each image (cell-free area at 6 h p.i. was considered as 100%). The arrows indicate the changes from 6 h to 22 h p.i.: 1) cell-free area that merged with the monolayer (blue), 2) cell-free area that became smaller (yellow), and 3) new cell-free area that appeared during infection (red).
Since Mani-infected RPTECs could not form a monolayer, we hypothesized that collective cell migration might be affected following Leptospira-induced kidney injury, because it impairs the repair of epithelial barrier function. To test our hypothesis, we infected RPTECs with Mani from the apical side and evaluated the epithelial barrier integrity. Mani induced a significant TEER decrease which was prevented by proteasome inhibitors (S5A Fig). Proteasome-dependent cytoskeleton network disruption was also observed (S5B and S5C Fig) at 24 h p.i. Then, we set up a scratch assay and monitored wound healing using live imaging to examine directional collective cell migration of RPTECs (Fig 5A). The recovery rate of Mani-infected RPTECs was significantly lower than that of non-infected RPTECs at 24 h p.i. (p = 0.0097) (Fig 5B). The addition of MG did not improve the recovery rate at 24 h p.i., although destruction of the monolayer was prevented (Figs 5B; S6). EGFR activation has been reported to generate mechanochemical waves to guide the collective cell migration [10]. Moreover, EGF enhances RPTECs regeneration and repair, accelerating the recovery of renal function [45]. Thus, we activated EGFR signaling by replacing the culture medium with EGF-containing medium at 24 h p.i. and analyzed wound healing using time-lapse microscopy until 48 h p.i. Non-infected and L. biflexa-infected RPTECs showed supracellular polarity and leader cell formation; this resulted in collective cell migration toward the scratch area with a 90–100% recovery rate in non-inhibitor or inhibitor-treated RPTECs (Fig 5C, 5D, and 5E; S3– S6 Movies). The movement of Mani-infected RPTECs was less coordinated than that of L. biflexa-infected RPTECs, characterized by outward-facing leader-like protrusions, cell detachment from the monolayer, and a significantly decreased recovery rate by 36 h p.i. (p = 0.0037) (Fig 5C and 5D; S7 Movie). The addition of EGF to MG-treated, Mani-infected RPTECs improved collective cell migration with a recovery rate of approximately 90% by 48 h p.i. (Fig 5C and 5E; S8 Movie). Collectively, these results revealed aberrant collective cell migration of Mani-infected RPTECs. MG addition to Mani-infected RPTECs prevented monolayer destruction and following EGF addition enhanced collective cell migration compared with that in the control; the recovery rates were comparable to those in non-infected cells.
(A) Workflow of the scratch assay to analyze wound healing. (B) Recovery rate (%) of the scratched at 24 h p.i. in non-infected and infected-RPTECs. (C) Recovery rate (%) of the scratched area at 36 and 48 h p.i in non-infected and infected-RPTECs. (B and C) Single dot in the bar graphs represents the value for one determination (related to S1 Data). ** p < 0.01. (D and E) Representative images of wound healing in non-infected and infected RPTECs in the absence or presence of MG. Images are related to S3–S8 Movies.
Discussion
Epithelial cells exhibit sophisticated repair and defense responses to bacterial infections. These responses include activation of inflammatory signaling pathways, innate immune responses, collective infected cell extrusion, and coordinated tissue repair mechanisms [46–48]. Successful pathogens employ strategies to counteract these mechanisms to cause disease. Inflammatory signaling pathways induced by dead bacteria or proteins secreted during infection might induce epithelial barrier disruption; however, our results showed that live bacteria that can transmigrate through the paracellular route are essential to induce epithelial TEER decrease.
Bacterial invasion and intracellular persistence are tightly coupled with the disruption of the eukaryotic cytoskeletal machinery [49]. Classically, pathogenic bacteria are classified as extracellular, facultative intracellular, and obligate intracellular. L. interrogans is considered an extracellular pathogen; however, we showed that a bacterial population was internalized into RPTECs at an early stage of infection without disrupting F-actin (2 h p.i., Fig 1A and 1C). This population could not survive. However, the cell entry might be an important step in inducing host proteins ubiquitination after 5 h p.i., thereby driving epithelial barrier disruption and promoting bacterial dissemination within the host. Some extracellular bacterial pathogens have been shown to have an intracellular in vivo phase with pathogenic implications [50]. Our results emphasize the need to understand the intracellular phase of the L. interrogans infective cycle within the host. However, the intracellular state of L. interrogans is transient and no evidence has been achieved in vivo. In our model, BF + MG can prevent bacterial degradation. Therefore, this model will be useful in future studies to clarify if the intracellular state of L. interrogans is a pathogen-driven or host-driven event by comparing different leptospiral strains.
Ubiquitinomic approaches that rely on diglycine enrichment have been used to perform global analyses of host cell ubiquitinome changes after infection with Salmonella typhimurium, Mycobacterium tuberculosis, and Legionella pneumophila [51]. To the best of our knowledge, this is the first study to characterize the ubiquitinome of L. interrogans-infected host cells. Our data revealed that L. interrogans infection induces changes in the host ubiquitinome, affecting proteins involved in infection processes such as cytoskeleton dynamics (Fig 2; S1 Table). The increase in ubiquitination of the small GTPase ARL2 led us to hypothesize that ARL2 is degraded by UPS following leptospiral infection; however, we could not detect a decrease in ARL2 decrease by immunoblot analysis (Fig 2E). Nondegradative ubiquitination of small GTPases following L. pneumophila infection serves as a strategy to manipulate host subcellular trafficking and membrane dynamics [51]. Thus, ARL2 ubiquitination suggests the modulation of ARL2-dependent processes, including microtubule dynamics. However, the microtubule network was not affected in BF + MG-treated-infected-RPTECs (Fig 2F and 2G). Thus, direct evidence linking ubiquitination-dependent ARL2 dysfunction to disruption of the microtubule network was not established in this study.
Previous studies conducted using endothelial cells indicated that pathogenic Leptospira affects F-actin but does not affect microtubules [52–54]. In this study, we show for the first time that Leptospira disrupts microtubules in epithelial cells (Fig 2F and 2G). Cytoskeletal components have been studied independently for decades; however, the actin–microtubule crosstalk is important for many biological processes including the regulation of cell shape and polarity during epithelial cell migration [55]. Our results showed that during the leptospiral infection process, this crosstalk is targeted, leading to its interruption.
Gram negative pathogens such as Salmonella, Shigella, and Legionella deliver effector proteins through specialized secretion systems (Types III and IV) into the host cytoplasm to modulate cytoskeletal dynamics [56–58]. The mechanism by which L. interrogans induces host ubiquitination and microtubule disruption remains undefined; however, an increase in intracellular calcium, triggered by Leptospira-secreted VMPs, has recently been reported as a mediator of epithelial barrier disruption [22]. Alternatively, analysis of the tissue geometry that drives bacterial infection has shown that the mechanosensitive ion channel protein Piezo1 facilitates bacterial invasion not only in intracellular bacteria but also in many classic extracellular bacteria, such as Staphylococcus aureus, Escherichia coli and Bacillus cereus [59]. Activation of Piezo1 upon bacterial internalization triggers a cascade of mechano-transduction signals via Ca2+ influx. Our study shows that leptospiral internalization into RPTECs preceded epithelial barrier disruption, this finding suggests that Ca2+ influx through Piezo1 activation might contribute to the intracellular Ca2+ elevation that triggers the epithelial barrier disruption observed in a previous study [22].
L. interrogans is a slow-growing bacterium, and the TEER decrease in Mani-infected RPTECs is a late event during RPTEC infection (after 18 h p.i.) [23]. Our results suggest that bacteria require a lag period to induce virulence factor expression and achieved the protein levels needed for pathogenicity [22]. Epithelial repair and defense responses may overcome the mechanosensor dysfunction induced during the early stage of leptospiral infection. Interestingly, previous studies conducted using animal infection models showed that pathogenic Leptospira colonizes a limited number of proximal renal tubules, probably those renal tubules in which pathogenic strategies overcome host responses [32,60]. Leptospiral–host cell interactions have mainly been studied from the perspectives of bacterial and the host immune responses, without considering the host cells’ repair mechanisms [21,24,61]. Disruption of intestinal epithelial homeostasis and barrier function is a common strategy for intestinal pathogens; strengthening of homeostatic responses can counteract bacterial pathogenic strategies [57,62]. The live imaging of infected RPTECs in Fig 4 showed that some cell- free areas became smaller in Mani-infected cells, which means that a fine balance between “epithelial barrier destruction” and “epithelial barrier repair” might be related to the differences in the three possible final outcomes of the infection in different kinds of animals and in different human patients: acute disease with recovery, acute disease progressing to chronic infection, or chronic infection without acute disease.
L. interrogans-induced disruption of AJs and microtubule and F-actin organization may impede the propagation of cellular signaling (such as ERK activation waves) which is important for tissue repair [10,63]. One notable example is that E-cad at cell–cell junctions, which are displaced during infection, modulates collective cell migration by phosphorylating EGFR [64]. Delayed wound healing has been observed following infection with various bacterial pathogens. For example, Porphorymonas gingivalis compromises cell migration of oral epithelial cells [65]. Healing of RPTECs following leptospiral infection has not been previously studied. Thus, we conducted a scratch assay to examine the repair response of RPTECs. Aberrant collective cell migration and decreased recovery rates in Mani-infected RPTECs suggest a defect in cellular signaling (Fig 5D and S7 Movie). Surprisingly, a low recovery rate at 24 h p.i. was observed in RPTECs treated with MG, this is probably because the UPS-mediated degradation of Golgin45, which is required for Golgi dispersal and reorientation during collective cell migration, was inhibited [66]. Replacement of MG with EGF at 24 h p.i. enhanced cell migration, suggesting that the cellular signals were propagated through the RPTEC monolayer (Fig 5C).
Our study has several limitations: 1) for the ubiquitinome analysis, only Gly-Gly modifications on lysines were considered. However, the discovery of non-canonical pathways has shown that ubiquitination of serine, cysteine, and threonine can also induce UPS-mediated protein degradation [67,68]. Thus, we cannot rule out that we overlooked an important UPS-related pathway involved in leptospiral infection; 2) the renal repair process after acute injury involves a complex interplay of epithelial, endothelial, and immune cells, where inflammatory processes affect renal recovery [69]. We solely focused on the response of epithelial cells in our in vitro model. However, the model can be improved by co-culturing different cell types, and these interactions should be analyzed in future studies; 3) cellular signals affected in the scratch assay were not analyzed in this study. However, the collective cell migration assay optimized in this study provides a platform for conducting detailed mechanistic studies in the future.
Modulation of tissue mechanics is emerging as a new framework for combating infectious diseases [44]. In this study, we proposed a leptospiral pathogenic strategy in which proteasome hijacking is important for disassembling AJs and the cytoskeleton, thereby impairing host mechanoresponses. Although Leptospira were found within intracellular vacuoles, Leptospira did not evolve to replicate intracellularly in the same manner as other intracellular pathogens, such as Salmonella or Legionella [70]. Thus, although the intracellular leptospires were killed within RPTECs, this event might be important to efficient delivery of virulence factors into host cells for epithelial barrier disruption. In conclusion, our findings reveal a bacterial pathogenic strategy that disrupts a proteasome-dependent circuit integrating cell–cell adhesion and the cytoskeletal network. Proteasomal inhibition maintains this circuit to drive collective cell migration and restore the epithelial barrier after injury. Defects in mechanosensors associated with infectious diseases can alter tissue integrity and represent an intriguing field of study. Future studies focusing on the altered cellular signals during aberrant collective cell migration will open new opportunities for developing therapeutic strategies to prevent pathogen dissemination and chronic damage after an infectious disease.
Materials and methods
Bacterial cultures and strains
L. interrogans serovar Manilae strain UP-MMC-NIID [32], which was passaged no more than four times to maintain reproducible virulence [71], and L. biflexa serovar Patoc strain Patoc 1 were routinely stationary cultured in Ellinghausen–McCullough–Johnson–Harris (EMJH) broth at 30 °C. L. interrogans was diluted 20-fold, whereas L. biflexa was diluted 40-fold in fresh EMJH and cultured for three days with shaking at 30 °C for cell infection experiments.
Cell culture
RPTEC/TERT1 (American Type Culture Collection, ATCC CRL-4031) cells [72] were grown in Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F-12; Gibco, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 5 pM triiodothyronine, 10 ng/mL recombinant human epidermal growth factor (hEGF), 3.5 μg/mL ascorbic acid, 5 μg/mL transferrin, 5 μg/mL insulin, 8.65 ng/mL sodium selenite and 100 μg/mL G418; and stocked in Cell Banker 2 (Takara, Tokyo, Japan) at -80 °C at low number passage. To maintain experimental reproducibility and original properties of the cells, they were not passaged more than six times in our laboratory after purchasing from ATCC.
For infection experiments, cells were seeded at a density of 1 × 106 cells/well in polyethylene terephthalate hanging cell culture inserts with a pore size of 3 μm (Falcon, Corning, NY, USA) in Falcon Companion six-well tissue culture plates (Corning). Cells were maintained in a humidified incubator at 37 °C and 5% CO2 for 7–10 days after reaching confluence to allow monolayer maturation, and the medium was exchanged every two days. The TEER of the monolayers was measured using a Millicell-ERS cell resistance indicator (MilliporeSigma, Burlington, MA, USA). After subtracting the resistance value of a cell-free insert (blank), the mean TEER value was expressed as Ω·cm2. The TEERs of cells observed before infection were designated as baseline values. The percentage TEER relative to the baseline value was calculated using the following formula: (TEER of experimental wells/baseline TEER of experimental wells) × 100%.
Cell infection and treatment
Cells were infected with leptospires via the basolateral side at a multiplicity of infection (MOI) of 100 in supplement-free DMEM/F-12 medium. To examine the effects of inhibitor treatment after infection, RPTECs were infected in the absence of inhibitors; subsequently, at 5 h p.i., the culture inserts were transferred to new wells containing DMEM/F12 medium with DMSO (the solvent used to prepare inhibitor stock solutions) or inhibitor(s) to prevent further attachment of leptospires. The inhibitors used to treat RPTECs were 100 nM bafilomycin A1 (a lysosomal inhibitor; Sigma-Aldrich, St. Louis, MI, USA), 10 μM MG-132 (a proteasomal inhibitor; Sigma-Aldrich), and 260 nM bortezomib (a proteasomal inhibitor; Fujifilm Wako, Osaka, Japan). Subsequently, cells were incubated at 37 °C and 5% CO2 during infection or treatment, then fixed for immunostaining or lysed for ubiquitome or immunoblottings analysis.
To determine the relative numbers of gentamicin-protected and cell-associated bacteria, infected cells were transferred to a six-well plate containing pre-warmed DMEM/F-12 medium with and without gentamicin (100 μg/ml), respectively; infected RPTECs were incubated for 1 h at 37 °C and 5% CO2. RPTECs were lysed with 400 μL of distilled water, and 100 μL of each lysate were used for serial two-fold dilutions in EMJH broth using 96-well plates. The plates were incubated at 30 °C for 7–14 days, and the final dilution which showed leptospiral growth was determined [73].
To determine the effect of heat-stable PAMPs, RPTECs were treated with live and heat-killed bacteria (leptospires incubated at 56 °C for 10 min) from the basolateral or apical side. To determine if secreted proteins during RPTEC infection can induce a TEER decrease, at 5 h p.i., the DMEM/F12 medium from infected RPTECs was centrifuged for 5 min at 5,000 rpm to remove leptospires; the supernatant was filtered through a 0.2-μm pore-size membrane and transferred to an intact RPTEC monolayer grown in culture inserts, the RPTECs were further incubated for 24 h at 37 °C and 5% CO2.
Immunostaining
RPTECs were fixed with 2% paraformaldehyde in phosphate-buffered saline (PBS) for 2 h at 4 °C and washed twice with Tris-buffered saline (TBS; Nacalai Tesque, Kyoto, Japan). The filter membranes were detached from the culture inserts before immunostaining. For LAMP1 staining, the membranes were permeabilized and blocked for 1 h with TBS containing 0.2% saponin and 10% Blocking One (Nacalai). The antibodies were diluted in the same buffer. For β-tubulin immunostaining, the membranes were permeabilizated and blocked with buffer A (5% bovine serum albumin [BSA] and 1% Triton X-100 in TBS) for 15 min, and antibodies were diluted in TBS containing 1% BSA and 0.2% Triton X-100.
The primary antibodies were rabbit polyclonal anti-Mani (1:200; kindly provided by Sharon Y. A. M. Villanueva, University of the Philippines) [74], rabbit monoclonal anti-p120-ctn (1:300; cat. no.: 59854; Cell Signaling Technology, Danvers, MA, USA), and mouse monoclonal anti-LAMP1 (1:200; cat. no.: sc-20011; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) or anti- β-tubulin antibodies (1:200; cat.no.: 66240–1-Ig; Proteintech, Tokyo, Japan) antibodies. F-actin was stained with a 1:300 dilution of rhodamine-phalloidin (Abcam, Cambridge, UK). Cells were counterstained to label the DNA using a 1:50 dilution of TO-PRO-3 (Invitrogen, Waltham, MA, USA). Secondary antibodies used for the immunofluorescence analysis included anti-rabbit IgG Alexa Fluor 488 (1:100; cat. no.: 711-545-152; Jackson ImmunoResearch [JIR], West Grove, PA, USA), anti-rabbit IgG Alexa Fluor 647 (1:100; cat. no.: 711-605-152; JIR), and anti-mouse IgG Alexa Fluor 488 (1:100; cat. no.: 715-545-151; JIR). Immunostained samples were mounted using Slow Fade Diamond Antifade Mountant (Invitrogen). The compiled Z-stack images were acquired using a Leica TCS-SPE confocal laser-scanning microscope with LEICA LAS AF acquisition software (version 2.6.0.7266; Leica Microsystems CMS GmbH, Mannheim, Germany).
The fluorescence intensity of the microtubule network was quantified from maximum intensity projections of the Z-stack images using the ImageJ software (version 1.53). Relative fluorescence intensity was calculated in three independent microscopic fields per condition as the mean fluorescence intensity per unit area.
Affinity enrichment and nanoscale liquid chromatography-tandem mass spectrometry (nanoLC–MS/MS) analysis
Cell lysate samples were combined with 100 µL of reduction solution (100 mM NH4HCO3 containing 0.15% [w/v] dithiothreitol) and incubated at 57 °C for 30 min. After incubation, 100 µL of alkylation solution (100 mM NH4HCO3 containing 1% [w/v] iodoacetamide) was added, and the mixture was incubated at room temperature (20 °C –25 °C) for 30 min. After incubation, 100 µL of sequencing-grade modified trypsin (Promega, Madison, WI, USA) and 100 µL of 50 mM NH4HCO3 were added sequentially, followed by overnight incubation at 30 °C. Samples were centrifuged at 3,000 x g for 5 min, and the supernatants containing the digested peptides were desalted on MonoSpin C18 columns (GL Sciences, Tokyo, Japan). The eluted peptides were dried using a centrifugal concentrator (CC–105, Tomy Seiko, Tokyo, Japan). The digested peptides were resuspended in 1.4 mL immunoaffinity purification (IAP) buffer (50 mM 3-(N-morpholino) propanesulfonic acid [MOPS; pH 7.2], 10 mM Na2HPO4, and 50 mM NaCl); subsequently, the peptides were cleared at 20,000 x g for 5 min and incubated on a rotator for 2 h at 4 °C with 50 μL of PTMScan Ubiquitin Remnant Motif (K-ε-GG) Antibody Beads slurry (Cell Signaling Technology) equilibrated in IAP buffer. The K-ε-GG peptides were washed multiple times in IAP buffer and pure water, incubated with 55 μL of 0.15% trifluoroacetic acid for 10 min, and then eluted. The elution was repeated, and eluates were combined and desalted using MonoSpin C18 columns. The final peptide eluates were dried and dissolved in 30 µL of 0.1% formic acid.
Mass spectrometric analysis was performed using nano LC–MS/MS. The samples were loaded on an Aurora Series emitter column (25 cm x 75 µm ID, 1.6 µm C18; IonOpticks, Victoria, Australia) using nano high-performance liquid chromatography (nanoHPLC; Dionex UltiMate 3000, Thermo Fisher Scientific). The ionized peptides were analyzed using a Q Exactive Plus MS (Thermo Fisher Scientific). NanoLC and MS data were acquired using the Xcalibur software (Thermo Fisher Scientific).
Protein identification and quantification
NanoLC–MS/MS spectra were analyzed and quantified using Proteome Discoverer 2.5 with the Sequest HT (Thermo Fisher Scientific) and the Mascot (version 2.5.1) search engines. The spectral data were submitted to Swiss–Prot database for Homo sapiens (Taxonomy ID: 9606). The following Sequest HT and Mascot search parameters were used: ion score threshold, 10; peptide tolerance, 10 ppm; MS/MS tolerance, 0.1 Da; peptide charge, 2; trypsin as the enzyme, allowing up to two missed cleavages; carbamidomethylation of cysteines as a fixed modification; Gly-Gly modification on lysines and oxidation on methionine as variable modifications. The label-free quantification (LFQ) method was used for protein quantification. Peptide threshold was set to high, and false discovery rate was adjusted to < 1%.
Immunoblotting
Protein samples were separated using precast Mini-PROTEAN TGX 4–15% gels (Bio-Rad Laboratories, Hercules, CA, USA) and subsequently processed for immunoblotting. Primary antibodies used for immunoblotting included rabbit monoclonal antibodies: anti-ARL2 (1:1000; cat. no.: ab183510; Abcam) and anti-ubiquitin clone E6K4Y (1:2,000; cat. no.: 20316; Cell Signaling Technology); and mouse monoclonal antibodies: anti-multi-ubiquitin clone FK2 (1:2000, cat. no.: MBL#D058-3; Medical & Biological Laboratories, Tokyo, Japan), anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH; 1:2,500; cat. no.: sc-32233, Santa Cruz Biotechnology), and anti-afadin (1:500; cat. no.: 610732; BD Transduction Laboratories, Franklin Lakes, NJ, USA). The secondary antibodies were horseradish peroxidase (HRP)-conjugated anti-rabbit IgG-antibody (1:25,000; cat. no.: 111-035-144; JIR) and HRP-conjugated anti-mouse IgG-antibody (1:7,500; cat no.: 715-005-150; JIR). Afterward, an Amersham ImageQuant 800 Imaging System (GE Healthcare, Chicago, IL, USA) was used to visualize the protein bands.
Cell scratch assay
Cells were seeded in 6-well plates (cat. no: 3810-006N; Iwaki, Shizuoka, Japan) at a density of 1 × 106 cells/well in supplemented DMEM/F-12; cells were cultured for a total of five days and reached confluence two days before the end of the culture period. Cells were infected at an MOI of 100 for 5 h in supplement-free DMEM/F-12 medium. At 5 h p.i., cells in each well were scratched with a gel-loading tip (cat. no.: 1034-R-204; Ina Optika, Osaka, Japan), the medium was replaced with DMEM/F-12 with or without the inhibitor. The scratch was photographed immediately to examine the initial scratch area and monolayer integrity; subsequently, the cells were incubated at 37 °C in a CO2 incubator. At 24 h p.i., the medium was replaced with DMEM/F-12 containing 10 ng/mL hEGF, images of ten regions along each scratch were taken at 20-minute intervals until 48 h p.i., using an all-in-one fluorescence microscope (BZ-X810 Keyence, Osaka, Japan) equipped with a time-lapse module with temperature and CO2 controller. Using ImageJ (version 1.54k), cell-free areas were measured in the middle of the scratch at 5 h and 24 h p.i. and within live-imaging regions at 26, 36, and 48 h p.i. Scratch closure was calculated using 5-h p.i. are as the baseline in Fig 5B, and the 26-h p.i. timepoint as the baseline in Fig 5C. Scratch closure was quantified for five regions in each experiment using the formula: 100 × (1 - [cell-free are at 36 or 48 h p.i./ cell-free are at 26 h p.i.]) (Fig 5C). The rate of closure for each experiment was calculated as the average of the five regions.
Statistical analyses
Statistical analyses were performed using OriginPro 2021 software. Analysis of a single variable across two groups was performed using a two-tailed Student’s t-test. Data from the scratch recovery assay, which involved multiple groups, were analyzed with one-way analysis of variance (ANOVA), followed by Tukey’s test for pairwise comparisons. Data are presented as the mean of at least three independent experiments. Differences were considered statistically significant at p-value < 0.05.
Supporting information
S1 Fig. Comparison of live and heat-killed bacteria after basolateral infection.
(A) Transepithelial electrical resistance (TEER) measurements at 24 h p.i. (B) Representative confocal images of renal proximal tubule epithelial cells (RPTECs) at 24 h p.i. Leptospira were stained with Alexa 488-labeled antibodies (green) and F-actin with rhodamine-phalloidin (red). The cell nuclei were stained with TO-PRO-3 (blue). Scale bar: 10 μm. ** p < 0.01.
https://doi.org/10.1371/journal.ppat.1014501.s001
(TIF)
S2 Fig. Comparison of live and heat-killed bacteria after apical infection.
(A) Transepithelial electrical resistance (TEER) measurements at 24 h p.i. (B) Representative confocal images of renal proximal tubule epithelial cells (RPTECs) at 24 h p.i. Leptospira were stained with Alexa 488-labeled antibodies (green) and F-actin with rhodamine-phalloidin (red). The cell nuclei were stained with TO-PRO-3 (blue). Scale bar: 10 μm. ** p < 0.01.
https://doi.org/10.1371/journal.ppat.1014501.s002
(TIF)
S3 Fig. Proteins secreted following RPTEC infection do not induce epithelial barrier disruption.
(A) Transepithelial electrical resistance (TEER) measurements at 24 h p.i. (B) Representative confocal images of RPTECs at 24 h p.i. The adherens junction (AJ) protein p120-catenin (p120-ctn) was stained with Alexa 488-labeled antibodies (green) and F-actin with rhodamine-phalloidin (red). The cell nuclei were stained with TO-PRO-3 (blue). Scale bar: 20 μm. ** p < 0.01.
https://doi.org/10.1371/journal.ppat.1014501.s003
(TIF)
S4 Fig. Proteasomal inhibition does not prevent AJ–F-actin linker protein (afadin) reduction in L. interrogans serovar Manilae UP-MMC-NIID strain (Mani)-infected RPTECs.
(A) Representative immunoblots of whole cell lysates for detection of afadin, ubiquitin, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH). (B) Normalized afadin levels calculated by considering the ratio of afadin/GAPDH in non-infected (n.i.) RPTECs as one. * p < 0.05.
https://doi.org/10.1371/journal.ppat.1014501.s004
(TIF)
S5 Fig. Mani induces proteasome-dependent disruption of the microtubule and F-actin cytoskeletal network.
RPTECs were infected with Mani from the apical side and treated with inhibitors (MG132: MG or Bortezomib: BZ) or DMSO (control) at 5 h p.i. (A) Transepithelial electrical resistance (TEER) measurements at 24 h p.i. Data are presented as the mean ± SD of at least three independent Transwell assay results. * p < 0.05 (B) Representative confocal images of infected cells at 24 h p.i. Microtubules (β-tubulin) were stained with Alexa Fluor 488-labeled antibodies (green) and F-actin with rhodamine-phalloidin (red). Scale bar: 20 μm. (C) Relative fluorescence intensity of β-tubulin. ** p < 0.01.
https://doi.org/10.1371/journal.ppat.1014501.s005
(TIF)
S6 Fig. Representative live images of Mani-infected RPTECs.
RPTEC monolayers were infected with Mani and images of live cells acquired at 24 h p.i. (A) RPTEC monolayer was destroyed. (B) The RPTEC monolayer was not destroyed in Mani-infected RPTECs when MG was added at 5 h pi. Scale bar: 25 μm.
https://doi.org/10.1371/journal.ppat.1014501.s006
(TIF)
S1 Table. Proteins showing increased ubiquitination at 21 h post-infection (p.i.).
https://doi.org/10.1371/journal.ppat.1014501.s007
(XLSX)
S1 Movie. Live imaging of subconfluent RPTECs grown in collagen-coated dishes and infected with Leptospira biflexa.
Images were acquired for 17 h (from 5 h 30 m to 22 h 30 m p.i.). Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s008
(AVI)
S2 Movie. Live imaging of sub-confluent RPTECs grown in collagen-coated dishes and infected with Mani.
Images were acquired for 17 h (from 5 h 30 m to 22 h 30 m p.i.). Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s009
(AVI)
S3 Movie. Representative live imaging of a scratched area in non-infected cells (non-inhibited) from 25 h 20 m to 48 h p.i.
Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s010
(MOV)
S4 Movie. Representative live imaging of a scratched area in non-infected cells (treated with MG132 [MG]) from 25 h 20 m to 48 h p.i.
Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s011
(MOV)
S5 Movie. Representative live imaging of a scratched area in L. biflexa-infected cells (non-inhibited) from 25 h 20 m to 48 h p.i.
Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s012
(MOV)
S6 Movie. Representative live imaging of a scratched area in L. biflexa-infected cells (MG-treated) from 25 h 20 m to 48 h p.i.
Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s013
(MOV)
S7 Movie. Representative live imaging of a scratched area in Mani-infected cells (non-inhibited) from 25 h 20 m to 48 h p.i.
Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s014
(MOV)
S8 Movie. Representative live imaging of a scratched area in Mani-infected cells (MG-treated) from 25 h 20 m to 48 h p.i.
Scale bar: 100 μm.
https://doi.org/10.1371/journal.ppat.1014501.s015
(MOV)
S1 Data. The datasheet includes all raw data for Figs 1–5, related to transepithelial electrical resistance (TEER) measurements, gentamicin protection assay, relative fluorescence intensity, and cell-free area analysis.
https://doi.org/10.1371/journal.ppat.1014501.s016
(XLSX)
S2 Data. The datasheet includes all raw data for S1–S5 Figs, related to transepithelial electrical resistance (TEER) measurements, protein quantification by western blotting, and relative fluorescence intensity.
https://doi.org/10.1371/journal.ppat.1014501.s017
(XLSX)
S1 Raw Gel. Original blot images for Figs 2 and S4.
https://doi.org/10.1371/journal.ppat.1014501.s018
(PDF)
Acknowledgments
We are grateful to the Research Laboratory Center at the University of the Ryukyus for providing access to the ImageQuant 800 Imaging System and Chemicals Evaluation and Research Institute for their support during proteome analysis. We also thank Jun Xu (University of the Ryukyus) for his assistance with data analyses.
References
- 1. Li Z-L, Li X-Y, Zhou Y, Wang B, Lv L-L, Liu B-C. Renal tubular epithelial cells response to injury in acute kidney injury. EBioMedicine. 2024;107:105294. pmid:39178744
- 2. Palmyre A, Lee J, Ryklin G, Camarata T, Selig MK, Duchemin A-L, et al. Collective epithelial migration drives kidney repair after acute injury. PLoS One. 2014;9(7):e101304. pmid:25010471
- 3. Zhuang S, Dang Y, Schnellmann RG. Requirement of the epidermal growth factor receptor in renal epithelial cell proliferation and migration. Am J Physiol Renal Physiol. 2004;287(3):F365-72. pmid:15213065
- 4. Stehbens SJ, Scarpa E, White MD. Perspectives in collective cell migration - moving forward. J Cell Sci. 2024;137(12):jcs261549. pmid:38904172
- 5. Cheung KJ, Horne-Badovinac S. Collective migration modes in development, tissue repair and cancer. Nat Rev Mol Cell Biol. 2025;26(10):741–58. pmid:40473829
- 6. Venhuizen J-H, Zegers MM. Making Heads or Tails of It: Cell-Cell Adhesion in Cellular and Supracellular Polarity in Collective Migration. Cold Spring Harb Perspect Biol. 2017;9(11):a027854. pmid:28246177
- 7. Capuana L, Boström A, Etienne-Manneville S. Multicellular scale front-to-rear polarity in collective migration. Curr Opin Cell Biol. 2020;62:114–22. pmid:31756576
- 8. Jetta D, Shireen T, Hua SZ. Epithelial cells sense local stiffness via Piezo1 mediated cytoskeletal reorganization. Front Cell Dev Biol. 2023;11:1198109. pmid:37293127
- 9. Xiao B. Mechanisms of mechanotransduction and physiological roles of PIEZO channels. Nat Rev Mol Cell Biol. 2024;25(11):886–903. pmid:39251883
- 10. Hino N, Rossetti L, Marín-Llauradó A, Aoki K, Trepat X, Matsuda M, et al. ERK-Mediated Mechanochemical Waves Direct Collective Cell Polarization. Dev Cell. 2020;53(6):646-660.e8. pmid:32497487
- 11. Iosifidis T, Sutanto EN, Buckley AG, Coleman L, Gill EE, Lee AH, et al. Aberrant cell migration contributes to defective airway epithelial repair in childhood wheeze. JCI Insight. 2020;5(7):e133125. pmid:32208383
- 12. Rajapakse S, Fernando N, Dreyfus A, Smith C, Rodrigo C. Leptospirosis. Nat Rev Dis Primers. 2025;11:32.
- 13. Fouts DE, Matthias MA, Adhikarla H, Adler B, Amorim-Santos L, Berg DE, et al. What Makes a Bacterial Species Pathogenic?:Comparative Genomic Analysis of the Genus Leptospira. PLoS Negl Trop Dis. 2016;10(2):e0004403. pmid:26890609
- 14. Picardeau M. Virulence of the zoonotic agent of leptospirosis: still terra incognita?. Nat Rev Microbiol. 2017;15(5):297–307. pmid:28260786
- 15. Karpagam KB, Ganesh B. Leptospirosis: a neglected tropical zoonotic infection of public health importance-an updated review. Eur J Clin Microbiol Infect Dis. 2020;39(5):835–46. pmid:31898795
- 16. Chou L-F, Yang H-Y, Hung C-C, Tian Y-C, Hsu S-H, Yang C-W. Leptospirosis kidney disease: Evolution from acute to chronic kidney disease. Biomed J. 2023;46(4):100595. pmid:37142093
- 17. Prado LG, Barbosa AS. Understanding the Renal Fibrotic Process in Leptospirosis. Int J Mol Sci. 2021;22(19):10779. pmid:34639117
- 18. Chou L-F, Chen T-W, Yang H-Y, Chang M-Y, Hsu S-H, Tsai C-Y, et al. Murine Renal Transcriptome Profiles Upon Leptospiral Infection: Implications for Chronic Kidney Diseases. J Infect Dis. 2018;218(9):1411–23. pmid:29868892
- 19. Fanton d’Andon M, Quellard N, Fernandez B, Ratet G, Lacroix-Lamandé S, Vandewalle A. Leptospira interrogans induces fibrosis in the mouse kidney through Inos-dependent, TLR- and NLR-independent signaling pathways. PLOS Negl Trop Dis. 2014;8:e2664. pmid:24498450
- 20. Martinez-Lopez DG, Fahey M, Coburn J. Responses of human endothelial cells to pathogenic and non-pathogenic Leptospira species. PLoS Negl Trop Dis. 2010;4(12):e918. pmid:21179504
- 21. Sebastián I, Okura N, Humbel BM, Xu J, Hermawan I, Matsuura C, et al. Disassembly of the apical junctional complex during the transmigration of Leptospira interrogans across polarized renal proximal tubule epithelial cells. Cell Microbiol. 2021;23(9):e13343. pmid:33864347
- 22. Giraud-Gatineau A, Haustant G, Monot M, Picardeau M, Benaroudj N. In vivo dual RNA-Seq uncovers key effectors of epithelial barrier disruption by an extracellular pathogen. Nat Commun. 2026;17(1):2274. pmid:41634022
- 23. Tokumon R, Sebastián I, Humbel BM, Okura N, Yamanaka H, Yamashiro T, et al. Degradation of p0071 and p120-catenin during adherens junction disassembly by Leptospira interrogans. Front Cell Infect Microbiol. 2023;13:1228051. pmid:37795382
- 24. Kakita T, Toma C, Sebastián I, Barbee H, Yamashiro T. Distinct strategies of epithelial cell barrier disruption by Leptospira interrogans isolated from human patients in Okinawa, Japan. PLoS Negl Trop Dis. 2025;19(11):e0013693. pmid:41187201
- 25. Lehmann JS, Fouts DE, Haft DH, Cannella AP, Ricaldi JN, Brinkac L, et al. Pathogenomic inference of virulence-associated genes in Leptospira interrogans. PLoS Negl Trop Dis. 2013;7(10):e2468. pmid:24098822
- 26. Giraud-Gatineau A, Nieves C, Harrison LB, Benaroudj N, Veyrier FJ, Picardeau M. Evolutionary insights into the emergence of virulent Leptospira spirochetes. PLoS Pathog. 2024;20(7):e1012161. pmid:39018329
- 27. Chaurasia R, Vinetz JM. In silico prediction of molecular mechanisms of toxicity mediated by the leptospiral PF07598 gene family-encoded virulence-modifying proteins. Front Mol Biosci. 2023;9:1092197. pmid:36756251
- 28. Chaurasia R, Marroquin AS, Vinetz JM, Matthias MA. Pathogenic Leptospira Evolved a Unique Gene Family Comprised of Ricin B-Like Lectin Domain-Containing Cytotoxins. Front Microbiol. 2022;13:859680. pmid:35422779
- 29. Hsu S-H, Chang M-Y, Lin S-M, Ko Y-C, Chou L-F, Tian Y-C, et al. Peptidoglycan mediates Leptospira outer membrane protein Loa22 to toll-like receptor 2 for inflammatory interaction: a novel innate immune recognition. Sci Rep. 2021;11(1):1064. pmid:33441663
- 30. Li D, Wu M. Pattern recognition receptors in health and diseases. Signal Transduct Target Ther. 2021;6(1):291. pmid:34344870
- 31. Shi Y, Li R, Yang J, Li X. No tight junctions in tight junction protein-1 expressing HeLa and fibroblast cells. Int J Physiol Pathophysiol Pharmacol. 2020;12(2):70–8. pmid:32419902
- 32. Yamaguchi T, Higa N, Okura N, Matsumoto A, Hermawan I, Yamashiro T, et al. Characterizing interactions of Leptospira interrogans with proximal renal tubule epithelial cells. BMC Microbiol. 2018;18(1):64. pmid:29973159
- 33. Koizumi N, Watanabe H. Leptospiral immunoglobulin-like proteins elicit protective immunity. Vaccine. 2004;22(11–12):1545–52. pmid:15063580
- 34. Cai J, Culley MK, Zhao Y, Zhao J. The role of ubiquitination and deubiquitination in the regulation of cell junctions. Protein Cell. 2018;9(9):754–69. pmid:29080116
- 35. Cagliero J, Loarec A, Lebon J, Baur F, Lefevre P, Follenfant E, et al. Anti-inflammatory cytokine profile and Jarisch-Herxheimer reaction in Leptospirosis patients: A prospective case-series study in New Caledonia. PLoS Negl Trop Dis. 2025;19(9):e0013189. pmid:40986630
- 36. Cagliero J, Villanueva SYAM, Matsui M. Leptospirosis Pathophysiology: Into the Storm of Cytokines. Front Cell Infect Microbiol. 2018;8:204. pmid:29974037
- 37. Barocchi MA, Ko AI, Reis MG, McDonald KL, Riley LW. Rapid translocation of polarized MDCK cell monolayers by Leptospira interrogans, an invasive but nonintracellular pathogen. Infect Immun. 2002;70(12):6926–32. pmid:12438371
- 38. Okuda S, Yoshizawa AC, Kobayashi D, Takahashi Y, Watanabe Y, Moriya Y, et al. jPOST environment accelerates the reuse and reanalysis of public proteome mass spectrometry data. Nucleic Acids Res. 2025;53(D1):D462–7. pmid:39526391
- 39. Francis JW, Newman LE, Cunningham LA, Kahn RA. A Trimer Consisting of the Tubulin-specific Chaperone D (TBCD), Regulatory GTPase ARL2, and β-Tubulin Is Required for Maintaining the Microtubule Network. J Biol Chem. 2017;292(10):4336–49. pmid:28126905
- 40. Zhai Y, Leung KT, Lau K-F. Regulation of cytohesins by their interactors in the nervous system. Neural Regen Res. 2023;18(10):2186–7. pmid:37056129
- 41. Shultz T, Shmuel M, Hyman T, Altschuler Y. Beta-tubulin cofactor D and ARL2 take part in apical junctional complex disassembly and abrogate epithelial structure. FASEB J. 2008;22(1):168–82. pmid:17704193
- 42. Goldberg AL. Development of proteasome inhibitors as research tools and cancer drugs. J Cell Biol. 2012;199(4):583–8. pmid:23148232
- 43. Sala S, Caillier A, Oakes PW. Principles and regulation of mechanosensing. J Cell Sci. 2024;137(18):jcs261338. pmid:39297391
- 44. Di X, Gao X, Peng L, Ai J, Jin X, Qi S, et al. Cellular mechanotransduction in health and diseases: from molecular mechanism to therapeutic targets. Signal Transduct Target Ther. 2023;8(1):282. pmid:37518181
- 45. Humes HD, Cieslinski DA, Coimbra TM, Messana JM, Galvao C. Epidermal growth factor enhances renal tubule cell regeneration and repair and accelerates the recovery of renal function in postischemic acute renal failure. J Clin Invest. 1989;84(6):1757–61. pmid:2592559
- 46. Kim M, Ashida H, Ogawa M, Yoshikawa Y, Mimuro H, Sasakawa C. Bacterial interactions with the host epithelium. Cell Host Microbe. 2010;8(1):20–35. pmid:20638639
- 47. Hundsdorfer L, Muenkel M, Aparicio-Yuste R, Sanchez-Rendon JC, Gomez-Benito MJ, Balmes A, et al. ERK activation waves coordinate mechanical cell competition leading to collective elimination via extrusion of bacterially infected cells. Cell Rep. 2025;44(1):115193. pmid:39817903
- 48. Robinson KS, Boucher D. Inflammasomes in epithelial innate immunity: front line warriors. FEBS Lett. 2024;598(11):1335–53. pmid:38485451
- 49. de Souza Santos M, Orth K. Subversion of the cytoskeleton by intracellular bacteria: lessons from Listeria, Salmonella and Vibrio. Cell Microbiol. 2015;17(2):164–73. pmid:25440316
- 50. Silva MT. Classical labeling of bacterial pathogens according to their lifestyle in the host: inconsistencies and alternatives. Front Microbiol. 2012;3:71. pmid:22393329
- 51. Steinbach A, Bhadkamkar V, Jimenez-Morales D, Stevenson E, Jang GM, Krogan NJ, et al. Cross-family small GTPase ubiquitination by the intracellular pathogen Legionella pneumophila. Mol Biol Cell. 2024;35(3):ar27. pmid:38117589
- 52. Sato H, Coburn J. Leptospira interrogans causes quantitative and morphological disturbances in adherens junctions and other biological groups of proteins in human endothelial cells. PLoS Negl Trop Dis. 2017;11(7):e0005830. pmid:28750011
- 53. Zhao X, Guo J, Jia X, Yang Y, Liu L, Nie W, et al. Internalization of Leptospira interrogans via diverse endocytosis mechanisms in human macrophages and vascular endothelial cells. PLoS Negl Trop Dis. 2022;16(9):e0010778. pmid:36137081
- 54. Daroz BB, Fernandes LGV, Cavenague MF, Kochi LT, Passalia FJ, Takahashi MB, et al. A Review on Host-Leptospira Interactions: What We Know and Future Expectations. Front Cell Infect Microbiol. 2021;11:777709. pmid:34900757
- 55. Dogterom M, Koenderink GH. Actin-microtubule crosstalk in cell biology. Nat Rev Mol Cell Biol. 2019;20(1):38–54. pmid:30323238
- 56. Radhakrishnan GK, Splitter GA. Modulation of host microtubule dynamics by pathogenic bacteria. Biomol Concepts. 2012;3(6):571–80. pmid:23585820
- 57. Citalán-Madrid AF, García-Ponce A, Vargas-Robles H, Betanzos A, Schnoor M. Small GTPases of the Ras superfamily regulate intestinal epithelial homeostasis and barrier function via common and unique mechanisms. Tissue Barriers. 2013;1(5):e26938. pmid:24868497
- 58. Berglund J, Gjondrekaj R, Verney E, Maupin-Furlow JA, Edelmann MJ. Modification of the host ubiquitome by bacterial enzymes. Microbiol Res. 2020;235:126429. pmid:32109687
- 59. Han Y, Liu X, Qu S, Duan X, Xiang Y, Jiang N, et al. Tissue geometry spatiotemporally drives bacterial infections. Cell. 2025;188(13):3459-3476.e21. pmid:40262607
- 60. Ratet G, Veyrier FJ, Fanton d’Andon M, Kammerscheit X, Nicola M-A, Picardeau M, et al. Live imaging of bioluminescent leptospira interrogans in mice reveals renal colonization as a stealth escape from the blood defenses and antibiotics. PLoS Negl Trop Dis. 2014;8(12):e3359. pmid:25474719
- 61. Anderson PN, Hahn BL, Thao G, Johnson MS, Giraud-Gatineau A, Gao Y, et al. In vitro and in vivo endothelial interactions of Leptospira species are markers of virulence. PLoS Negl Trop Dis. 2026;20(1):e0013939. pmid:41592120
- 62. Martens K, Pugin B, De Boeck I, Spacova I, Steelant B, Seys SF, et al. Probiotics for the airways: Potential to improve epithelial and immune homeostasis. Allergy. 2018;73(10):1954–63. pmid:29869783
- 63. Miyata M, Ogita H, Komura H, Nakata S, Okamoto R, Ozaki M, et al. Localization of nectin-free afadin at the leading edge and its involvement in directional cell movement induced by platelet-derived growth factor. J Cell Sci. 2009;122(Pt 23):4319–29. pmid:19887591
- 64. Fu C, Dilasser F, Lin S-Z, Karnat M, Arora A, Rajendiran H, et al. Regulation of intercellular viscosity by E-cadherin-dependent phosphorylation of EGFR in collective cell migration. Proc Natl Acad Sci U S A. 2024;121(37):e2405560121. pmid:39231206
- 65. Bhattacharya R, Xu F, Dong G, Li S, Tian C, Ponugoti B, et al. Effect of bacteria on the wound healing behavior of oral epithelial cells. PLoS One. 2014;9(2):e89475. pmid:24586806
- 66. Sharma R, Dey Das K, Srinivasula SM. EGF-mediated Golgi dynamics and cell migration require CARP2. Cell Rep. 2024;43(11):114896. pmid:39441718
- 67. Akizuki Y, Kaypee S, Ohtake F, Ikeda F. The emerging roles of non-canonical ubiquitination in proteostasis and beyond. J Cell Biol. 2024;223(5):e202311171. pmid:38517379
- 68. Kelsall IR. Non-lysine ubiquitylation: Doing things differently. Front Mol Biosci. 2022;9:1008175. pmid:36200073
- 69. Kumar S. Cellular and molecular pathways of renal repair after acute kidney injury. Kidney Int. 2018;93(1):27–40. pmid:29291820
- 70. Vaughn B, Abu Kwaik Y. Idiosyncratic Biogenesis of Intracellular Pathogens-Containing Vacuoles. Front Cell Infect Microbiol. 2021;11:722433. pmid:34858868
- 71. Toma C, Murray GL, Nohara T, Mizuyama M, Koizumi N, Adler B, et al. Leptospiral outer membrane protein LMB216 is involved in enhancement of phagocytic uptake by macrophages. Cell Microbiol. 2014;16(9):1366–77. pmid:24655538
- 72. Wieser M, Stadler G, Jennings P, Streubel B, Pfaller W, Ambros P, et al. hTERT alone immortalizes epithelial cells of renal proximal tubules without changing their functional characteristics. Am J Physiol Renal Physiol. 2008;295(5):F1365-75. pmid:18715936
- 73. Nikaido Y, Ogawa M, Fukuda K, Yokoyama M, Kanemaru T, Nakayama T, et al. Transbronchial Invasion and Proliferation of Leptospira interrogans in Lung without Inflammatory Cell Infiltration in a Hamster Model. Infect Immun. 2019;87(12):e00727-19. pmid:31548321
- 74. Villanueva SYAM, Saito M, Tsutsumi Y, Segawa T, Baterna RA, Chakraborty A, et al. High virulence in hamsters of four dominant Leptospira serovars isolated from rats in the Philippines. Microbiology (Reading). 2014;160(Pt 2):418–28. pmid:24257815