New Results

Sortase-assembled pili promote extracellular electron transfer and iron acquisition in Enterococcus faecalis biofilm

Ling Ning Lam, Jun Jie Wong, Artur Matysik, Jason J. Paxman, Kelvin Kian Long Chong, Pui Man Low, Zhi Sheng Chua, Begoña Heras, Enrico Marsili, Kimberly A. Kline
doi: https://doi.org/10.1101/601666

Abstract

Enterococcus faecalis is an opportunistic human pathogen and the cause of biofilm-associated infections of the heart, catheterized urinary tract, wounds, and the dysbiotic gut where it can expand to high numbers upon microbiome perturbations. The E. faecalis sortase-assembled endocarditis and biofilm associated pilus (Ebp) is involved in adhesion and biofilm formation in vitro and in vivo. Extracellular electron transfer (EET) also promotes E. faecalis biofilm formation in iron-rich environments, however neither the mechanism underlying EET nor its role in virulence was previously known. Here we show that iron associated with Ebp serve as a terminal electron acceptor for EET, leading to extracellular iron reduction and intracellular iron accumulation. We found that a MIDAS motif within the EbpA tip adhesin is required for interaction with iron, EET, and FeoB-mediated iron uptake. We demonstrate that MenB and Ndh3, essential components of the aerobic respiratory chain and a specialized flavin-mediated electron transport chain, respectively, are required for iron-mediated EET. In addition, using a mouse gastrointestinal (GI) colonization model, we show that EET is essential for colonization of the GI tract, and Ebp is essential for augmented E. faecalis GI colonization when dietary iron is in excess. Taken together, our findings show that pilus mediated capture of iron within biofilms enables EET-mediated iron acquisition in E. faecalis, and that these processes plays an important role in E. faecalis expansion in the GI tract.

Significance Understanding enterococcal biofilm development is the first step towards improved therapeutics for the often antimicrobial resistant infections caused by these bacteria. Here we report a role for Enterococcus faecalis endocarditis and biofilm associated pili (Ebp) in mediating iron-dependent biofilm growth and contributing to extracellular electron transfer (EET) which in turn promotes iron acquisition. Furthermore, we characterize the mechanisms underlying electron transfer in the E. faecalis biofilm. Our findings support a model in which E. faecalis use EET to drive the reduction of pilus-associated ferric iron, leading to iron acquisition in E. faecalis biofilm, and contributing to enterococcal virulence in the GI tract.

Introduction

Enterococcus faecalis is an important human opportunistic pathogen that causes a variety of diseases including endocarditis, urinary tract infections (UTI), bacteremia, wound infection, and medical device-associated infections 1. Many of these infections are polymicrobial and biofilm-associated, rendering them more tolerant to antimicrobial and immune clearance, and contributing to their persistent nature 1. Therefore, a detailed understanding of enterococcal biofilm development is a critical step towards advancing new therapeutics for treatment of these often antimicrobial resistant infections 1, 2.

One important factor contributing to E. faecalis biofilm formation and virulence is the sortase-assembled endocarditis and biofilm associated pilus (Ebp) 3–7. The importance of Ebp to E. faecalis is supported by the presence of its coding sequence in the core E. faecalis genome 5, 8. The Ebp fiber is primarily composed of the major pilin subunit EbpC, along with two dispensable minor pilin subunits EbpB and EbpA 5, and these pilin subunits are covalently assembled by Sortase C on the cell membrane, prior to attachment to the cell wall by Sortase A 6, 9. EbpA is the tip adhesin of the pilus, and its N terminus encompasses a metal ion-dependent adhesion site (MIDAS) motif. In eukaryotes, the MIDAS motif is important for ligand binding 10 and coordinates divalent cations, most often Mg2+ 11, 12, for cell adhesion and interaction with extracellular matrix (ECM) proteins 12. The MIDAS motif within the N-terminal domain of E. faecalis EbpA contributes to in vitro biofilm formation, adherence to fibrinogen, and bladder colonization in a mouse catheter-associated urinary tract infection (CAUTI) model 13–15. Sortase-assembled pili are conserved in many Gram-positive bacteria, where they often contribute to adhesion, biofilm formation, modulation of host immune response, and virulence 16, 17. The adhesive tips of these pilins (PilA of Streptococcus agalactiae 18, RrgA of Streptococcus pneumoniae 19, and AP-1 of Streptococcus pyogenes 20) include a conserved von Willebrand factor A (vWFA) domain containing a MIDAS motif, which are essential for adhesion. However, the specificity and affinity for metal binding to the MIDAS motif in bacteria, including E. faecalis, has not been characterized.

Previously, we demonstrated that E. faecalis is electrogenic, and that iron promotes both extracellular electron transfer (EET) and biofilm growth in E. faecalis 21. Several mechanisms for EET have been described in both Gram-positive and Gram-negative bacteria 22, 23. In some bacteria, microbial nanowires, composed of extracellular filamentous protein fibers and/or of cytochrome-associated fibers, have been proposed to aid EET by directly connecting the bacterial cell surface to extracellular metal oxides, or alternatively, by interacting with extracellular soluble redox mediators 23, 24. Biochemical and structural analysis have recently demonstrated that the conductive filaments of Geobacter sulfurreducens consist of polymerized chains of the hexaheme cytochrome OmcS 25. Gram positive bacteria have evolved a highly conserved mechanism for EET, described in L. monocytogenes, which involves a specialized respiratory complex that channels electrons through discrete membrane-localized quinone pools to flavin intermediates, for subsequent delivery to terminal extracellular electron acceptors 26. In E. faecalis, electroactivity requires the ability to synthesize L-lactate dehydrogenase (LDH) 21 and demethylmenaquinone (DMK) 27, important for catalyzing redox reactions and electron transfer during respiration, respectively28.

In this study, we sought to understand the mechanism and physiological role of EET in E. faecalis. We show that EET in E. faecalis makes use of both DMK and the conserved flavin EET pathway described in L. monocytogenes for electron delivery to extracellular pilus-associated iron deposits. Iron induces the expression of Ebp pili, increasing the interaction sites for iron, which include the MIDAS motif of the EbpA tip adhesin, which is required for pilus association with iron as well as for iron-augmented biofilm formation. In addition, we show that EET drives the reduction of iron leading to FeoB-mediated uptake of iron into the E. faecalis biofilm cells. Finally, we demonstrate that a high iron diet promotes E. faecalis mouse gastrointestinal (GI) colonization in an ebp-dependent manner, and EET mutants are attenuated in the GI tract. Together, these findings suggest a model in which iron promotes pilus-mediated biofilm formation. We propose that the association of iron with sortase-assembled Ebp sequesters iron close to the cell surface within biofilms where it can serve as an extracellular electron acceptor for EET, resulting in the generation of ferrous iron for subsequent uptake into the cell. This new paradigm for iron acquisition in which pili and aggregates of piliated cells serve as iron sinks to promote biofilm formation and intracellular iron accumulation, significantly advances our mechanistic understanding of both biofilm formation and the pathogenesis of E. faecalis.

Results

Iron-augmented E. faecalis biofilm formation requires pilus expression

We previously demonstrated that E. faecalis biofilm growth in media supplemented with iron resulted in two-fold more biomass 21. To test the hypothesis that Ebp contributes to iron-augmented biofilm, we performed biofilm assays in growth media supplemented with 2 mM ferric chloride (FeCl3), and quantified total adherent biofilm biomass. In iron-supplemented media, deletion mutants of the entire Ebp pilus (ΔebpABC), pilus tip (ΔebpA) and pilus fiber (ΔebpC) displayed more than a 50% reduction in biofilm growth compared to the wild type control (Figure 1a) and mutant biofilm levels in iron-supplemented media were most similar to wild type biofilm in the absence of iron augmentation (Figure S1a). Biofilm formation by the pilus null ΔebpABC mutant was restored to wild type levels upon genetic complementation with the full locus (pebpABCsrtC). A ΔebpB deletion mutant also displayed significantly reduced biofilm growth as compared to the wild type control in iron-supplemented media; however, it was not as strongly attenuated compared to the other pilus mutants (Figure 1a). These findings are consistent with previous reports in which deletion of ebpA and ebpABC resulted in significantly reduced biofilm formation in normal media (Figure S1a, S2b) 6, 29. These results indicate that the fully assembled Ebp pilus fiber contributes substantially to iron-augmented biofilm formation with a key role for the EbpA pilus tip (Figure 1a, S1b). We next tested the contribution of the MIDAS motif within EbpA to iron-mediated biofilm growth. Using a MIDAS mutant (ebpAAWAGA), in which the native motif (Asp275–Trp-Ser277-Gly-Ser279) is mutated to (Ala275-Trp-Ala277-Gly-Ala279) 13, we observed that it was as attenuated as the ebpABC null mutant for biofilm formation in iron-supplemented media (Figure 1a, S1a). Collectively, these data demonstrate that both the MIDAS motif of the EbpA tip adhesin, as well as an intact EbpC pilus fiber, are essential for iron-mediated biofilm growth.

Figure 1.
Figure 1. Contribution of pilus to iron mediated E. faecalis biofilm growth.

(a) E. faecalis biofilm biomass quantification grown for 120 hours using crystal violet staining. N=9 repeated on non-consecutive days. (b) Volume viewer of CSLM images of E. faecalis biofilm at 24 hr for control (40% TSBG) and iron-supplemented (40% TSBG, 0.2 mM FeCl3) at 63x magnification. E. faecalis cells are labeled with antibody against Group D antigen. (c) Overlap of cell aggregation density map with the piliated cell population density map using 3D reconstruction of CLSM z-stack images of iron-supplemented E. faecalis biofilm compared with control. N=3, repeated on non-consecutive days. (d) Quantification of immuno-fluorescent labelled anti-EbpC E. faecalis cells from 6-well microtiter plate 24 hour biofilm or planktonic culture grown at 37°C in control and metal supplemented media. N>3. Statistical significance was determined by two-way ANOVA. Error bars represent standard error of the mean (SEM).

Iron induces pilus expression and drives aggregation of pilus-expressing cells in E. faecalis biofilm

Iron-augmented E. faecalis biofilms have an altered 3-D structure 21. To determine whether pilus expression contributes to the biofilm ultrastructure, we used immunofluorescence microscopy to visualize pilus-expressing cells within biofilms. Augmented biofilm biomass after growth in iron-supplemented media, as previously reported, was not apparent for the ΔebpABC or the ebpAAWAGA mutants (Figure 1b) 21. In addition, pilus-expressing cells tended to aggregate in areas of higher cell density in biofilms grown in iron-supplemented media (Figure 1c). The abundance of piliated cell aggregates within the biofilm suggested that pilus expression may be induced in the presence of iron. We therefore quantified the number of biofilm cells expressing pili and observed a significantly larger pilus-expressing population in biofilms grown in ferric iron-supplemented media, but not in the presence of alternative iron sources or with other cationic metals (Figure 1d). By contrast, we observed no significant difference in planktonically grown pilus-expression in response to iron (Figure 1d). Together these results show that ferric iron induces pilus expression, which in turn increases the biofilm structure and mass.

Iron colocalizes with pili on the cell surface and is dependent on the EbpA MIDAS motif

While monitoring E. faecalis populations for piliation in iron-supplemented media, we observed dense deposits at the polar hemispheres of E. faecalis cells which co-localized with sites of EbpC deposition, suggesting that these dense deposits may be iron interacting with pili (Figure 2a, Figure S2a-b). We observed these dense deposits less frequently in the ebpAAWAGA mutant, even though this mutant strain expresses a similar proportion of piliated cells as WT (Figure S2c), and the dense deposits were completely absent in ΔebpABC deletion mutant (Figure 2a, Figure S2a-b), suggesting that pili may be interacting with ferric iron via the MIDAS motif of EbpA. To confirm the interaction between the EbpA MIDAS motif and iron, we extracted Ebp from E. faecalis wild type and ebpAAWAGA biofilms grown in normal or iron supplemented media, and in parallel performed control extractions from ΔebpABC biofilms. Pili were isolated under native conditions and analyzed for iron content by inductive-coupled plasma mass spectrometry (ICP-MS). Pilus extracts from wild type cells grown in the iron-supplemented media were associated with nearly three times more iron than wild type pilus extracts from biofilms grown in control media. Ebp association with iron was dependent on the EbpA MIDAS motif because pilus preps from ebpAAWAGA were similar to background iron levels in control ΔebpABC extracts (Figure 2b).

Figure 2.
Figure 2. Evidences for EbpA-iron interaction.

(a) Representative immunofluorescent images from 24 hour E. faecalis biofilm (wild type, ΔebpABC and ebpAAWAGA) labeled with EbpC antibody-conjugated fluorescent probe, grown from 6-well microtiter plate biofilm in the presence of iron, N=3. (b) ICP-MS analysis of iron (ppb) from pilus preps isolated under native condition. N=6. Statistical analysis is performed using two-way ANOVA Fisher LSD test. Error bars represent SEM. Black bar indicates 500nm. (c) Superimposition of EbpA structural model (N177-P620) (blue) with that of the S. agalactiae GBS104 (grey) and S. pneumoniae RrgA (pink). In agreement with other pilin adhesins, EbpA tip folds into a vWFA domain comprising of a 5 stranded β-sheet surrounded by 7 α-helices. The MIDAS motif is located at the N-terminus of the β-sheet. Only a partial model was obtained for the 2 extended arms of this domain. The EbpA tip is then followed by an IgG fold domain. (d) Closeup view of the superimposition of the MIDAS motif from EbpA (blue), GBS104 (grey) and RrgA (pink) showing conservation of D275, S277, S279, T350 and D378 residues (EbpA numbering). Bound Mg2+ ions (grey/magenta) in GBS104 and RrgA and modelled Fe3+(orange) in EbpA are shown as spheres. (e) The Fe3+ is coordinated by the EbpA MIDAS motif by ∼2Å direct interactions with S277, S279, D378 and a water mediated interaction with D275 along with interactions with two additional water molecules (water molecules shown as blue spheres). Upon a closed to open conformational change, bonding could shift from D378 to T350.

To better understand how the EbpA MIDAS motif could interact with ferric iron, we modelled the EbpA structure based upon the crystal structures of its Mg bound homologs, S. agalactiae PilA GBS104 (PDB: 3txa) 30 and S. pneumoniae RrgA (PDB: 2ww8) 19. The overall EbpA structural model (N177-P620) showed the expected vWFA structural fold encompassing the MIDAS motif, consistent with other pili tip adhesins (Figure 2c-d). The model supports an interaction of the MIDAS motif and associated metal binding residues with ferric iron (Figure 2e). D275 would make a water mediated contact with the iron, the hydroxyl oxygen atoms of S277 and S279 and the carboxylate oxygen of D378 would make direct ionic bonds with iron and the classic 6-coordination of iron would be completed by 2 water molecules. vWFA domains are known to undergo conformational changes from closed to open states, whereby iron bonding would be shifted from D378 to the nearby T350 31. The metal binding residues were confirmed by an alignment of the EbpA, GBS104 and RrgA sequences, which despite low sequence identity, showed strict conservation of the MIDAS motif D275, S277, S279, T350 and D378 residues (Figure S3).

Pili and the respiration-associated electron transport chain contribute to extracellular electron transfer (EET)

The association of Ebp with iron, coupled with the ability of E. faecalis to undergo iron-mediated EET 21, suggested that pili themselves may be involved in the EET. To address this, we measured the instantaneous current output by chronoamperometry and the cumulative charge production in biofilms grown in iron-supplemented media for 20 hours. We observed a reduction in cumulative charge production in both the ΔebpABC and ebpAAWAGA MIDAS mutants when grown in 0.5 mM iron-supplemented media (Figure 3a). At higher iron concentrations, we did not observe significant differences in cumulative charge production over 20 hours (Figure S4a). The current output in the first few hours was similar for all strains (Figure S4b), but both pilin mutants exhibited an inability to sustain current output as compared to the wild-type control at later time points (Figure S4a-b). We therefore conclude that pili contribute to sustained EET, but that other mechanisms also contribute to E. faecalis EET in the absence of the pili.

Figure 3.
Figure 3. Extracellular electron transfer in E. faecalis biofilm.

(a) Cumulative charge production (over 20 hours) in E. faecalis pilin mutants grown in the presence of 0.5 mM iron and (b) EET-associated mutants grown in the presence of 2 mM iron on a screen-printed carbon mini-electrode for 20 hours in a closed static system at 37°C. Experiments were repeated on non-consecutive days, N>3. Statistical analysis is performed using one-way ANOVA with Bonferroni multiple comparison test. Error bar represents standard error margin (SEM).

To identify additional pathways, in addition to pili, that contribute to iron-mediated EET, we examined the E. faecalis homolog of the L. monocytogenes EET-specific NADH dehydrogenase (ndh3) for flavin-mediated EET 26, as well as demethylmenaquinone (menB) for quinone-mediated EET 28. We hypothesized that these electron transfer pathways may work together to promote EET. Indeed, we observed attenuated charge production in both ndh3::Tn and menB::Tn mutants 32, 33 which was restored upon genetic complementation (Figure 3b), as well as in mutants predicted to work in concert with Ndh3 in the conserved EET pathway (Figure S5). Importantly, charge production was further reduced in the ndh3::Tn ΔmenB double mutant compared to either single mutant (Figure 3b), suggesting that DMK and Ndh3 work in non-redundant concert for efficient EET. Together, these findings demonstrate that both respiratory electron transport conduits and pili contribute to iron-mediated EET in E. faecalis biofilm.

Genes encoding Ebp and the FeoB ferrous iron transporter are upregulated in E. faecalis biofilm

To further understand how iron promotes E. faecalis biofilm, we performed RNA sequencing and found a total of 90 genes that were differentially regulated upon growth in iron-supplemented media (Table S1). As expected, the pilus genes ebpA (OG1RF_RS04555) and ebpB (OG1RF_RS04560) were induced 1.68-fold and 1.55-fold respectively, which we validated by qRT-PCR (Figure S6). Surprisingly, the only other iron-associated gene that showed both significant and differential expression was feoB (OG1RF_RS01950) encoding a predicted ferrous iron transporter, which was up-regulated 1.38-fold (Table S1). Importantly, this suggested that E. faecalis biofilm grown in iron-supplemented media may be importing ferrous iron. This finding was consistent with our earlier study that showed E. faecalis intracellular iron concentrations were increased in biofilm cells grown in iron-supplemented media 21. The simultaneous up-regulation of ebpABC and feoB as the sole iron-responsive transporter, along with our observations that Ebp co-purify with iron and are required for EET suggested that pilus-associated iron may be reduced during EET and subsequently taken up by the ferrous iron transporter FeoB.

Extracellular ferric iron reduction replenishes ferrous iron pool for uptake by E. faecalis biofilm cells

Since ferric iron reduction to ferrous iron requires EET and is facilitated by both Ebp pili and the respiration electron transport chain, we predicted that an inability to perform EET would coincide with attenuation in ferric iron reduction and subsequent diminished extracellular ferrous iron pools available for uptake via FeoB. We therefore performed ICP-MS to determine the intracellular iron content in pilus mutants and EET-associated mutants grown in iron-supplemented media. Since Ldh1 was previously shown to be important for E. faecalis EET 21, we predicted that the absence of ldh1 may also lead to reduced intracellular iron accumulation. Consistent with these predictions, we observed significantly reduced intracellular iron for all pilus deletion mutants which was restored upon complementation, as well as reduced intracellular iron in the ebpAAWAGAMIDAS and the ldh1 mutants.

Furthermore, we observed significantly reduced intracellular iron in the ndh3::Tn ΔmenB mutant (Figure 4a). As predicted, the absence of feoB also led to reduced intracellular iron, in a genetically complementable manner (Figure 4b). If intracellular ferrous iron accumulation is a direct consequence of EET catalyzed ferric iron reduction, EET mutants should display reduced ability for ferric iron reduction. We tested this using a ferrozine assay, which reacts with ferrous iron to form a stable colored product. With a functional EET mechanism, we predict the presence of increased ferrous iron pools. As expected, we observed a significant decrease in ferrous iron pools in ldh1, ndh3 and menB mutants (Figure 4c). Collectively, these data show that EET mutants are attenuated for ferric iron reduction and therefore intracellular iron accumulation.

Figure 4:
Figure 4: Intracellular iron accumulation is dependent on the ability for extracellular ferric iron reduction.

(a-b) ICP-MS analysis of intracellular iron (ppb) in E. faecalis biofilm cells grown in iron-supplemented media. N=9. (c) Quantification of ferric iron reduction activity by E. faecalis. Statistical significance was determined by one-way ANOVA. Error bars represent standard error margin (SEM).

Pili and respiration-associated electron transfer systems contribute to colonization of mouse GI tract

To understand how iron-augmented biofilms and EET impact E. faecalis pathogenesis, we examined the role of dietary iron in an antibiotic-treated mouse gastrointestinal (GI) colonization model. We fed mice a diet containing normal amounts of iron (control; 200 mg/kg ferric chloride) or excess iron (high iron; 2000 mg/kg ferric chloride), for 3 weeks prior to antibiotic exposure and subsequent ingestion of E. faecalis in the drinking water (Figure 5a). Twenty-four hours after inoculation, we recovered significantly more E. faecalis from the mouse colon in mice fed the high iron diet as compared to the control group and this increase was dependent on Ebp expression (Figure 5b). Mice fed with high iron diet had increased levels of colon tissue-associated iron, which was not apparent in the cecum or small intestine (Figure S7a). Accordingly, no iron-dependent colonization differences were observed in those compartments (Figure S7b). These data suggest that dietary iron promotes E. faecalis Ebp-dependent colonization in the mouse colon. We next asked whether EET also provides a colonization advantage for E. faecalis in the GI tract. Indeed, both Ndh3 and MenB are required for GI colonization, independent of dietary manipulation (Figure 5c, Figure S7c). Together, these findings demonstrate that both the pili and EET are important for GI colonization.

Figure 5.
Figure 5. E. faecalis colonization in lower gastrointestinal tract.

(a) Schematic of diet manipulation in GI colonization model. (b) Bacterial titers of pilus null mutant (N=3, n=5) and EET-associated mutants (for WT n=30, for mutants n=15) recovered from mouse colons, repeated on non-consecutive days. Statistical significance was determined by Mann-Whitney test. Black line indicates the median. Dotted line indicates limitation of detection at CFU <40.

Discussion

We previously showed that E. faecalis performs EET in the presence of extracellular iron, which was associated with ATP production, and that iron supplementation augments biofilm growth 21. In this study, we sought to understand the mechanism underlying EET and its physiological role in E. faecalis. We show that iron induces expression of the genes encoding the sortase-assembled Ebp, which in turn associate with iron via the MIDAS motif of the EbpA tip adhesin, and promote biofilm formation. We also show that EET is dependent on both DMK in the electron transport chain, and a specialized flavin-associated respiratory complex, leading to reduction of extracellular iron and FeoB-mediated uptake of iron (Figure 6). Furthermore, we demonstrate that pili are important for colonization in the iron-loaded mouse GI tract.

Figure 6.
Figure 6. Model for the role of iron in E. faecalis biofilm.

Biofilm growth in excess iron induces expression of the genes encoding the sortase-assembled Ebp, which in turn associate with iron via the MIDAS motif of the EbpA tip adhesin. EET is dependent on both DMK in the electron transport chain, as well as a specialized flavin-associated respiratory chain, leading to reduction of extracellular iron and FeoB-mediated uptake of reduced free ferrous iron.

Many microorganisms have the capacity to generate electrical current via dissimilatory metal reduction 22, 34, 35. In E. faecalis, it was previously known that EET relies upon DMK, part of its minimal electron transport chain 27, as well as Ldh1, involved in cellular redox homeostasis 36. Here we confirm that menB, required for DMK synthesis 37, contributes to EET, along with a recently reported flavin-associated alternative membrane-bound electron transport chain 26. Since mutation of either pathway results in decreased charge production and ferric iron reduction, whereas mutation of both pathways completely abrogates both activities, we suggest that electrons flow through both respiratory complexes for maximally efficient EET and ferric iron reduction. While the precise mechanism for electron transfer outside of the E. faecalis cell may vary depending on the environmental niche, like L. monocytogenes 26, E. faecalis is a flavin auxotroph 38, necessitating the presence of this soluble electron shuttle in the vicinity of the microbes. Indeed, addition of riboflavin to E. faecalis biofilms growing in microbial fuel cells augments current production 39. Our work also demonstrates that E. faecalis can use iron as a terminal acceptor for EET, which is likely relevant both in the outside environment where Enterococci can persist for long periods of time 40, as well as in the GI tract where less than 15% of dietary iron is absorbed at the duodenum, leaving the majority destined for excretion 41 and potentially available for EET. In addition, there are likely alternative terminal electron acceptors available for EET in the GI tract, such as host associated iron, host-derived nitrate, or microbially-derived molecules such as humic substances 42.

We previously demonstrated that E. faecalis biofilms, but not planktonic cells, grow better upon iron supplementation 21. We further showed that iron deposits accumulate in the E. faecalis biofilm matrix 21, which is composed of extracellular DNA 43, 44 as well as, most likely, sortase substrates including Ebp that are continuously shed from wild type cells 6, 9. Here we show that the same pili are required for the formation of biofilm matrix-associated iron deposits, and are both essential for iron-augmented biofilm formation and necessary for EET. We speculate that Ebp contribute to EET in E. faecalis by sequestering iron as terminal electron acceptors in close proximity to the cells, either as surface attached pili or within the tightly packed biofilm matrix, for efficient EET.

Both pilus associated EET and biofilm formation depend on an intact MIDAS motif within the EbpA pilus tip adhesin of a fully polymerized pilus. Ebp also co-purifies with iron in a MIDAS-dependent manner, demonstrating for the first time that the MIDAS motif contributes to the association of a protein with iron. Indeed, our structural model of EbpA aligns well with other pilus adhesin tips such as S. agalactiae GBS104 and S. pneumoniae RrgA, with a vWFA fold and MIDAS motif configured for binding cationic metals 76, 77. MIDAS motifs have been shown to bind a range of cationic metals, but in the case of EbpA this can include Fe3+ due to the local available concentrations in iron-rich environments. Apart from binding metals, MIDAS motifs also form part of substrate binding sites, whereby iron binding to this site could influence the interaction of EbpA with biofilm factors to promote an increase in biofilm mass 78. Collectively, our data confirm that iron binding at the MIDAS motif has a dual role as both a terminal electron acceptor for EET and to promote biofilm formation. Moreover, vWFA domains can undergo conformational changes to open and closed states 78, so it is tempting to speculate that such a conformational change may have a role in the interaction with biofilm promoting factors and/or the release of iron after its reduction to Fe2+.

Consistent with an important role for Ebp pili in an iron-rich environment, we show that the ebp operon is transcriptionally induced when iron is supplemented. Ebp are displayed on the surface of a subset of cells in any given E. faecalis population 5, 6, 9, 29, 45, and this population can be increased in the presence of serum 5 and bicarbonate 46. Here we add another environmental signal, iron abundance, that can influence Ebp expression. Several intrinsic factors can impact ebp gene expression including transcriptional regulators, EbpR 47, AhrC (also annotated as ArgR3) and ArgR2 48, the FsrB quorum sensing peptide 49 and the RNA processing enzyme RNase J2 50, 51. While none of these factors were differentially regulated upon iron supplementation which may suggest uncharacterized regulatory pathways are responsible for iron-regulated ebp transcription, we cannot rule out the possibility that excess iron impacts the function of these known regulators of ebp expression.

In addition to ebp, the only iron-regulated E. faecalis gene annotated as iron-associated was feoB, involved in ferrous iron uptake. Our discoveries that 1) Ebp expressing cells aggregate within the biofilm, 2) E. faecalis accumulate intracellular iron in an Ebp-, EbpA MIDAS-, and FeoB-dependent manner, and that 3) intracellular iron accumulation is abrogated in the absence of EET and ability for ferric iron reduction, together suggest the presence of spatially segregated areas within the biofilm that may be especially enriched for ferric iron sequestration and reduction, subsequent ferrous iron acquisition, leading to biofilm growth and restructuring. Our observation that E. faecalis can combine dissimilatory metal reduction with iron acquisition is similar to early reports for Shewanella putrefaciens 52. However, the precise mechanism by which iron acquisition contributes to E. faecalis biofilm formation is currently under investigation.

Enterococci are common but minor members of the GI microbiome. However, antibiotic mediated dysbiosis favors overgrowth of Enterococci, harbouring both intrinsic and acquired antibiotic resistance, in the lower GI tract 53–56. In addition, their inherent tolerance to oxidative stress 57 and high concentration of metals 21, 58, 59 can provide a selective advantage for E. faecalis to colonize and bloom in niches unfavorable to other microbes. Our observation that E. faecalis can both use EET and take advantage of iron-overload in the GI tract provides yet another mechanism by which E. faecalis can thrive in this niche.

Collectively, the findings of this study demonstrate that Ebp and EET facilitate iron reduction for its subsequent uptake, biofilm augmentation, and virulence in the antibiotic treated gut. Given the strong conservation of sortase-assembled pili containing MIDAS motifs in their tip adhesins among Gram positive pathogens, along with the conservation of genes encoding EET components in the same organisms, we propose that pilus-mediated metal capture and acquisition may be a common mechanism used by microbes to acquire limiting nutrients in a diversity of niches. This work therefore raises the possibility for new therapeutic strategies for E. faecalis, and potentially many other pathogens, aimed inhibiting extracellular electron flow, iron reduction and acquisition, and EET-associated ATP production.

Materials and Methods

Bacterial Strains and Growth Conditions

E. faecalis was grown in Brain Heart Infusion broth (BHI; Becton, Dickinson and Company, Franklin Lakes, NJ) and cultured at 37°C under static or shaking (200rpm) conditions, as indicated. Preparation of inocula for biofilm and planktonic growth assays was performed as previously described following growth in Tryptic Soy Broth (TSB) or agar, supplemented with 0.175% glucose (TSBG) (Oxoid Inc., Ontario, Canada) 21. Bacterial strains used in this study are listed in Table S2. Where appropriate, strains harbouring pGCP123 plasmids were grown in 1000 µg/mL kanamycin (Sigma Aldrich, USA) and strains harbouring pMSP3535 plasmids were grown in 100 µg/mL erythromycin unless stated otherwise. Metals were filtered sterilized and supplemented during medium preparation in autoclaved TSBG media. For experiments using ferric chloride only, metal is supplemented in TSBG media and autoclaved together. Ferric citrate hydrate ≥98%, magnesium chloride anhydrous ≥98%, copper chloride dihydrate ≥99%, ferrous sulphate heptahydrate ≥99%, ferric sulphate hydrate ≥97%, ferric chloride anhydrous ≥99%, heme ≥90%, and the chelator 2,2’dipyridyl ≥99% were all supplied by Sigma Aldrich, St Louis, MO, USA. Manganese chloride tetrahydrate and zinc chloride were supplied by Merk Millipore, Singapore.

General cloning techniques

Both menB and ndh3 nucleotide sequences are based on the E. faecalis OG1RF genome obtained from BioCyc 60. The Wizard genome DNA purification kit (Promega Corp., Madison, WI) was used for isolation of bacterial genomic DNA (gDNA), and Monarch® Plasmid miniprep Kit (New England BioLabs, Ipswitch, MA) was used for purification of plasmid for gene expression and construction of deletion mutant. The Monarch® DNA Gel Extraction Kit (New England BioLabs, Ipswitch, MA) was used to isolate PCR products during extension overlay PCR. In-Fusion HD Cloning Kit (TaKara Bio, USA) was used for fast, directional cloning of DNA fragments into vector for both expression vector and in-frame deletion vector. All plasmids used in the study are listed in Table S3. T4 DNA ligase and restriction endonucleases were purchased from New England BioLabs (Ipswitch, MA). Colony PCR was performed using Taq DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) and PCR of gene of interest for plasmid construction was performed using Phusion DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA). Ligations were transformed into E. coli Dh5α cells. Plasmids derived in this study were confirmed by sequencing of purified plasmid.

Strain construction

To construct menB and ndh3 complementation plasmids, primers (menB_F’ and menB_R’ for menB, or ndh3_F’ and ndh3_R’ for ndh3; Table S3) were designed with BamHI restriction site or SpeI restriction sites flanking the gene of interest, to generate DNA fragments as templates for In-Fusion cloning (Takara Bio USA Inc.) using primers (menB_F’_Infusion and menB_R’_Infusion for menB, or ndh3_F’_Infusion and ndh3_R’_Infusion for ndh3) with at least 15 bp complementary sequence for ligation into the nisin-inducible vector pMSP3535 digested with the same restriction enzymes. Both pMSP3535::menB and pMSP3535::ndh3 plasmids were generated in E. coli Dh5α, verified by sequencing, and transformed into E. faecalis as described previously 13. Deletion of the menB coding sequence from the OG1RF chromosome was accomplished by allelic replacement using pGCP213 temperature sensitive shuttle vector as described previously 13. The deletion allele was constructed by extension overlap PCR, consisting of the upstream (IFD_menB_Frag2_F’ and IFD_menB_Frag2_R’; Table S3) and downstream (IFD_menB_Frag1_F’ and IFD_menB_Frag1_R’; Table S3) of the menB coding sequence, and introduced into pGCP213 using in-fusion cloning with IFD_menB_Frag2_F’ and IFD_menB_Frag1_R’ primers which has at least 15 bp complementary. Deletion of menB was performed in ndh3::Tn parent strain to generate a double mutant. Transformants were PCR screened for in-frame deletion as previously described 13.

Immunofluorescence Imaging and 3D Reconstruction of Biofilm

Bacterial cultures were normalized to OD600 0.7 and diluted 1000x prior inoculating into each well of µ-Slide 8 well glass bottom slides (ibiTreat coated) (ibidi Inc., USA) containing 40% v/v TSBG for incubation at 37°C under static conditions for 24 hours. After incubation, adherent biofilm was fixed in 4% paraformaldehyde (Sigma Aldrich, USA)-PBS for 30 minutes, blocked with 3% bovine serum album (Sigma Aldrich, USA) (BSA)-PBS for 30 minutes, and incubated with primary antibody (Rabbit Anti-Group D antigen, Thermo Scientific Singapore) or Guinea Pig Anti-EbpC 45 at 1:500 dilution for 30 minutes. The biofilm was then washed with 3% BSA-PBS and incubated with secondary antibody (IgG horseradish peroxidase-conjugated anti-rabbit or anti guinea-pig, Thermo Scientific Singapore) at 1:500 dilution for 30 minutes and washed with 3% BSA-PBS. Biofilms were hydrated with PBS prior to imaging. Biofilm morphology, biomass, and cell distribution were analysed by confocal laser scanning microscopy (CLSM). Images were acquired using LSM780 confocal microscope (Zeiss, Germany) equipped with 63x/1.4 Oil DIC M27 and controlled by ZEN software.

Samples were illuminated with 488 nm and 561nm Argon laser line; GFP emitted fluorescence was collected in the 493-580 nm range and RFP emitted fluorescence was collected in the 568-712 nm range. Optical sections (134.95×134.95 µm) were collected every 0.637 µm through the entire biofilm thickness and signal from each section was averaged 4 times. Fiji software was used for further processing (levels adjustment, stack resliced). To visualize the biofilm spatial organization coordinates representing cocci position were plotted as spheres with cell-size diameter as described previously21. For each cell, local density was calculated as number of neighbour cells within 4µm radius, normalized and visualized using colour gradient.

Biofilm Assay

Bacterial cultures were normalized as previously described 21 and inoculated in TSBG in a 96-well flat bottom transparent microtiter plate (Thermo Scientific, Waltman, MA, USA), and incubated at 37°C under static conditions for 5 days unless specified otherwise. Strains harboring pGCP123 complementation plasmids were grown in the presence of kanamycin prior to the biofilm assay, but kanamycin was not added to the biofilm assay itself because we found that kanamycin precipitates in the presence of excess iron leading to aberrant biofilm formation (data not shown). Adherent biofilm biomass was stained using 0.1% w/v crystal violet (Sigma-Aldrich, St Louis, MO, USA) at 4°C for 30 minutes. The microtiter plate was washed twice with PBS followed by crystal violet solubilization with ethanol:acetone (4:1) for 45 minutes at room temperature. Quantification of adherent biofilm biomass was measured by absorbance at OD595nm using a Tecan Infinite 200 PRO spectrophotometer (Tecan Group Ltd., Männedorf, Switzerland).

Planktonic Growth Assay

Bacterial cultures were normalized as previously described 21 and serially diluted by a dilution factor of 200. Diluted cultures were inoculated into the media at a ratio of 1:25, which is 8 µL of the inoculum in 200 µL of media, incubated at 37°C for 18 hours, and absorbance at OD600nm was measured using a Tecan Infinite 200 PRO spectrophotometer (Tecan Group Ltd., Männedorf, Switzerland) at 15 minute intervals.

Population Analysis of Pilus Expression

Biofilms were grown in 6 well tissue culture plates at 37°C under static conditions for 24 hours in TSBG or TSBG supplemented with metals where appropriate, and planktonic cultures were grown in 50-ml falcon tube under shaking conditions (200 rpm). Immunoblotting was performed as described previously 61 with some modification. Spent media was removed, and the adherent biofilm was suspended in PBS. A cell scraper was used to dislodge the biofilm, and the dislodged cells were centrifuged at 14,000 rpm for 2 minutes at room temperature to remove the supernatant.

Biofilm cells were suspended in PBS and normalized to OD600nm 1.0. Normalized suspensions were fixed in 4% paraformaldehyde (Sigma Aldrich, USA)-PBS for 20 minutes at 4°C, centrifuged at 14,000 rpm for 2 minutes. and the supernatant then discarded. The pellets were washed with PBS, centrifuged at 14,000 rpm for 2 minutes, and the supernatant discarded again. Immunofluorescence microscopy was performed as described previously 61 with the following modifications: 1:500 dilution of guinea pig anti-EbpC (Thermo Scientific, Singapore) in PBS-3% bovine serum albumin, and 1:500 dilution of goat Anti-guinea pig AlexaFluor-568 (Invitrogen Inc., USA) incubation for 30 minutes. After washing with PBS twice, resuspended cells were applied to poly-L-lysine slides (Polysciences Inc., USA), dried for 5 minutes, mounted with Vectashield®, and sealed with a cover slip. Images were acquired using an inverted Epi-fluorescence microscope (Zeiss Axio observer Z1, Germany) equipped with an EC Plan-Neofluar 100x/1.3 Oil objective and controlled by ZEN software. For each biological sample, 5 images were acquired and the percentage of the cell population expressing pili was quantified. At least 300 bacterial cells per strain per experiment were scored for pilus expression. The metal concentrations used during the pilus expression experiments were based on the highest concentration in which E. faecalis cells did not exhibit growth inhibition (data not shown).

EbpA Structural Modelling

The EbpA amino acid sequence was submitted to the structural homology-modelling server Swiss-Model (https://swissmodel.expasy.org) 62. Template structures for structural modelling were automatically selected based upon sequence identity. The initial template search derived 723 templates that were filtered down to 50 templates, whereby 14 models were constructed and ranked based on sequence similarity and coverage. The top solution was based upon the PDB 3txa (GBS104 from Streptococcus agalactiae) that showed 34% sequence similarity and 38% sequence coverage. The final EbpA model included N177-P620. The Fe3+ was created with Sketcher 63 and placed into position using the MIDAS bound magnesium of 3txa and 2ww8 as a guide with the program COOT 64. The two conserved water molecules from both 3txa and 2ww8 were fitted into EbpA by a similar means. Final alignments, analysis and structural figures were created with PyMOL 65.

Quantitative Real time PCR (qRT-PCR) and RNA sequencing

Biofilm cells were prepared as described above for TEM. After washing, the biofilm cell pellet was incubated with lysozyme from chicken egg white (10mg/ml) (Sigma Aldrich, USA) for 30 minutes at 37°C to remove the cell wall and centrifuged at 14,000 rpm for 2 minutes at room temperature to remove supernatant prior to cell lysis. RNA extraction was performed in a Purifier® filtered PCR enclosure using the PureLink™ RNA mini kit (Invitrogen, USA) according to the manufacturer’s instructions. RNA purification and removal of DNA was performed using TURBO DNA-free™ kit (Thermo Fisher, USA) and Agencourt® RNAClean® XP Kit (Beckman Coulter, USA). Measurement of RNA yield and quality was performed using Qubit® RNA HS assay kit (Thermo Fisher, USA) and RNA ScreenTape System and 2200 TapeStation (Agilent, USA). Synthesis of cDNA was performed using SuperScript III First-strand (Invitrogen, USA). Quantitative real-time PCR using cDNA was performed using KAPA SYBR fast qPCR master mix kit (Sigma Aldrich, USA) and Applied Biosystems StepOne Plus Real-Time PCR system. The expression of ebpC, ebpR, srtA, srtC, argR3 and gyrA were analysed using primer pairs listed in Table S2. For each primer set, a standard curve was established using genomic DNA from E. faecalis OG1RF. Normalized concentrations of cDNA were used to determine the relative fold change in gene expression as compared to E. faecalis OG1RF biofilm grown in TSBG. For RNA sequencing, ribosomal RNA depletion was performed after RNA purification using Ribo-Zero™ rRNA removal kit (Illumina, USA). cDNA library synthesis was performed using NEBNext RNA First-strand and NEBNext Ultra directional RNA Second-strand synthesis module (New England BioLab, US). Transcriptome library preparation was performed using 300bp paired end Illumina sequencing.

Pilus Extraction and Purification

Cell surface protein extracts from biofilms were prepared as described previously 5 with minor modifications. Biofilms were grown cells dislodged as described above for TEM. Dislodged biofilm cells were centrifuge at 3300 g for 4 minutes at 15°C, washed with Tris buffer (10 mM Tris-HCl pH 8, 1 mM EDTA, 1 mM DTT) and suspended in protoplast buffer with Pierce™ protease inhibitors (Thermo Scientific, USA). Mutanolysin from Streptomyces globisporus ATCC 21553 (Sigma Aldrich, USA) and benzonase nuclease (Sigma Aldrich, USA) were added and incubated at 37°C for 6 hours with shaking (200 rpm). Following incubation, biofilm cells were centrifuged at 3300 g for 4 minutes, 15°C to obtain the supernatant. Supernatants were filtered through a 0.45µm Supor® membrane (Sigma Aldrich, USA) and loaded into Corning Spin-X UF concentrators (MWCO 100kDa) (Sigma Aldrich, USA), centrifuged for 20 minutes, 15°C at 6000 g, and dialyzed using Tris buffer using Spectra/Por® Float-A-Lyzer® G2 dialysis device (MWCO 300kDa) (Spectrum Labs, USA). Pilus extracts were collected, quantified using Nanodrop 2000 Spectrophotometer (Thermofisher, USA), and normalized to 5 µg per sample. Samples were mixed with an equal volume of NuPAGE® Tris-glycine native sample buffer (Novex®) and loaded on to NuPAGE™ 3-8% Tris-acetate gel (Thermofisher, USA). Gel blots were run at 150 V for 3 hours. Western blotting was performed as described previously 6 with minor modifications. Membranes were blocked in 0.05% v/v Tween-3% v/v bovine serum albumin-PBS (Sigma Aldrich, USA) overnight at 4°C and washed twice with 0.05% v/v Tween (Sigma Aldrich, USA)-PBS prior to incubation with primary and secondary antibodies, and prior to incubation with detection substrate. Primary antibody (guinea pig Anti-EbpC 45or rabbit Anti-SecA6) diluted 1:1500, and secondary antibody (IgG guinea pig or rabbit conjugated horseradish peroxidase) diluted 1:4000 were used. Protein was detected using SuperSignal® west femto maximum sensitivity substrate kit (Thermo Scientific, USA).

Inductive-Coupled Plasma Mass Spectrometry (ICP-MS)

Pilus extracts, normalized to 5µg per sample, were run under native condition as described above. Using the western blot prepared gel as a reference, a separate identical NuPAGE™ Tris-acetate gel (Thermofisher, USA) was cut to isolate pilus extract ladder above a molecular weight of 100 kDa. At a ratio of 2:1, 70% nitric acid (Sigma Aldrich, USA) and 30% hydrogen peroxide (Sigma Aldrich, USA) was added to the excised gel slice containing the pilus protein, and left under room temperature for 3 days to allow complete digestion. The digested samples were diluted with LC-MS grade water and filtered using 0.2 µm membrane, prior to analysis using ICP-MS. Analysis of trace metals in samples was performed using ICP-MS model Elan-DRCe, Meinhard Nebulizer model TR-30-C3 (Perkin Elmer; Model: N8122006 (Elan Standard Torch)).

Electrochemical Assay and Analysis

Screen printed electrodes (SPE) (model DRP-C110; DropSens, Spain) consisting of a 4 mm diameter carbon working electrode, carbon counter electrode, and a Ag pseudo-reference electrode were controlled by a multichannel potentiostat (VSP, Bio-Logic, France) in an electrochemical cell of 9 mL working volume sealed with a Teflon cap. Chronoamperometry was performed as previously described with minor modifications 21. All electrochemical experiments were conducted at 37°C in TSBG medium supplemented with 1 mM FeCl3 unless otherwise stated.

Ferrozine assay of ferric iron reductase activity

E. faecalis cells were normalized to OD600nm 0.7, diluted 1:200 in PBS and resuspended in TSBG supplemented with 2 mM iron (III) chloride as previously described 21 and then supplemented with 0.5 mM ferrozine. Experiments were performed as previously described with minor modification 26. OD592nm measurements were made every 2 mins for up to 7 hours on E. faecalis biofilms grown statically. Corresponding change in optical density measurements correlates with color change due to ferrous iron binding to ferrozine after ferric iron is reduced to ferrous iron.

Mouse Gastrointestinal Tract (GI) Infection Model

Three week old male C57BL/6NTac mice were administered ampicillin (VWR, USA) in their drinking water (1 g/L) for 5 days as previously described 53, 55. Mice were then given one day of recovery from antibiotic treatment prior to administration of approximately 1-5 x 108 CFU/ml E. faecalis (OD600nm 0.5) in the drinking water for 3 days as previously described 66. Before and after infection, mice were monitored for signs of disease and weight loss. All animal experiments were approved and performed in compliance with the Nanyang Technological University Institutional Animal Care and Use Committee (IACUC). For dietary manipulation, customized synthesized diets (C1038 iron deficient diet with 200 mg/kg ferric chloride as control diet, C1038 iron deficient diet with 2000 mg/kg ferric chloride as high iron diet) (Altromin, Germany) were given to mice prior to infection and throughout infection. Mice were kept on customized diets for 3 weeks before administrating ampicillin as previously described with minor modification in the synthesized diet 67, 68. At the indicated timepoints, the small intestine, colon, and cecum were harvested. Tissue samples were homogenised in PBS, serial diluted in PBS, and spot-plated on BHI agar with 10 mg/L colistin, 10 mg/L nalidixic acid, 100 mg/L rifampicin, 25 mg/L fusidic acid for CFU enumeration. All antibiotics were obtained from Sigma Aldrich, USA.

Authors Contributions

L.N.L and K.A.K. conceptualized the study. L.N.L, E.M and K.A.K designed the experiments, analyzed the data and prepared the manuscript. L.N.L performed biofilm experiments, immuno-fluorescence assays, ICP-MS, and analyzed the data. A.M analyzed the confocal data and generated 3D reconstruction models. LN.L, P.M.L., Z.S.C, and E.M performed the electrochemistry experiments and analyzed the data. L.N.L and J.J.W performed mouse GI experiments and analyzed the data. L.N.L prepared the RNA samples for RNA sequencing and K.K.L.C analyzed the data. J.P. and B.H. performed modeling experiments. All authors reviewed the manuscript.

Acknowledgments

This work was supported by the National Research Foundation and Ministry of Education Singapore under its Research Centre of Excellence Programme and by the Ministry of Education Singapore under its Tier 2 programme (MOE2014-T2-2-124). Artur Matysik was supported by the National Medical Research Council under its Clinical Basic Research Grant (NMRC/CBRG/0086/2015). We thank Wandy Betty (Washington University, St Louis, USA) for performing TEM, Hailyn V. Nielsen and Scott Hultgren (Washington University, St Louis, USA) for providing the pilus mutants, and Gary Dunny (University of Minnesota) for providing transposon mutants. Finally, we thank Sam Light and Dan Portnoy (University of California, Berkeley) for sharing unpublished data and for critical discussions of the manuscript.

References

  1. ↵
    Ch’ng, J. H., Chong, K. K. L., Lam, L. N., Wong, J. J. & Kline, K. A. Biofilm-associated infection by enterococci. Nature reviews. Microbiology 17, 82–94, doi:10.1038/s41579-018-0107-z (2019).
    OpenUrlCrossRef
  2. ↵
    O’Driscoll, T. & Crank, C. W. Vancomycin-resistant enterococcal infections: epidemiology, clinical manifestations, and optimal management. Infection and drug resistance 8, 217–230, doi:10.2147/idr.s54125 (2015).
    OpenUrlCrossRef
  3. ↵
    Nallapareddy, S. R. et al. Conservation of Ebp-type pilus genes among Enterococci and demonstration of their role in adherence of Enterococcus faecalis to human platelets. Infection and immunity 79, 2911–2920, doi:10.1128/iai.00039-11 (2011).
    OpenUrlAbstract/FREE Full Text
  4. Nallapareddy, S. R., Singh, K. V., Sillanpaa, J., Zhao, M. & Murray, B. E. Relative contributions of Ebp Pili and the collagen adhesin ace to host extracellular matrix protein adherence and experimental urinary tract infection by Enterococcus faecalis OG1RF. Infection and immunity 79, 2901–2910, doi:10.1128/iai.00038-11 (2011).
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Nallapareddy, S. R. et al. Endocarditis and biofilm-associated pili of Enterococcus faecalis. The Journal of clinical investigation 116, 2799–2807, doi:10.1172/jci29021 (2006).
    OpenUrlCrossRefPubMedWeb of Science
  6. ↵
    Nielsen, H. V. et al. Pilin and sortase residues critical for endocarditis-and biofilm-associated pilus biogenesis in Enterococcus faecalis. Journal of bacteriology 195, 4484–4495, doi:10.1128/jb.00451-13 (2013).
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Singh, K. V., Nallapareddy, S. R. & Murray, B. E. Importance of the ebp (endocarditis-and biofilm-associated pilus) locus in the pathogenesis of Enterococcus faecalis ascending urinary tract infection. J Infect Dis 195, 1671–1677 (2007).
    OpenUrlCrossRefPubMedWeb of Science
  8. ↵
    Sillanpää, J., Xu, Y., Nallapareddy, S. R., Murray, B. E. & Höök, M. A family of putative MSCRAMMs from Enterococcus faecalis. Microbiology 150, 2069–2078, doi:doi:10.1099/mic.0.27074-0 (2004).
    OpenUrlCrossRefPubMedWeb of Science
  9. ↵
    Kline, K. A. et al. Mechanism for sortase localization and the role of sortase localization in efficient pilus assembly in Enterococcus faecalis. Journal of bacteriology 191, 3237–3247, doi:10.1128/JB.01837-08 (2009).
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Plow, E. F., Haas, T. A., Zhang, L., Loftus, J. & Smith, J. W. Ligand Binding to Integrins. Journal of Biological Chemistry 275, 21785–21788, doi:10.1074/jbc.R000003200 (2000).
    OpenUrlFREE Full Text
  11. ↵
    Bergelson, J. M. & Hemler, M. E. Integrin-ligand binding. Do integrins use a ‘MIDAS touch’ to grasp an Asp? Current biology : CB 5, 615–617 (1995).
    OpenUrl
  12. ↵
    Whittaker, C. A. & Hynes, R. O. Distribution and evolution of von Willebrand/integrin A domains: widely dispersed domains with roles in cell adhesion and elsewhere. Molecular biology of the cell 13, 3369–3387, doi:10.1091/mbc.e02-05-0259 (2002).
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Nielsen, H. V. et al. The Metal Ion-Dependent Adhesion Site Motif of the Enterococcus faecalis EbpA Pilin Mediates Pilus Function in Catheter-Associated Urinary Tract Infection. mBio 3, e00177–00112, doi:10.1128/mBio.00177-12 (2012).
    OpenUrlCrossRefPubMed
  14. Flores-Mireles, A. L., Pinkner, J. S., Caparon, M. G. & Hultgren, S. J. EbpA vaccine antibodies block binding of Enterococcus faecalis to fibrinogen to prevent catheter-associated bladder infection in mice. Science translational medicine 6, 254ra127–254ra127, doi:10.1126/scitranslmed.3009384 (2014).
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Schembri, M. A., Lo, A. W. & Ulett, G. C. Blocking the ‘MIDAS’ touch of Enterococcus faecalis. Annals of translational medicine 3, S3–S3, doi:10.3978/j.issn.2305-5839.2015.03.32 (2015).
    OpenUrlCrossRef
  16. ↵
    Proft, T. & Baker, E. N. Pili in Gram-negative and Gram-positive bacteria -structure, assembly and their role in disease. Cellular and molecular life sciences : CMLS 66, 613–635, doi:10.1007/s00018-008-8477-4 (2009).
    OpenUrlCrossRefPubMedWeb of Science
  17. ↵
    Danne, C. & Dramsi, S. Pili of Gram-positive bacteria: roles in host colonization. Research in Microbiology 163, 645–658, doi:https://doi.org/10.1016/j.resmic.2012.10.012 (2012).
    OpenUrlCrossRefWeb of Science
  18. ↵
    Konto-Ghiorghi, Y. et al. Dual Role for Pilus in Adherence to Epithelial Cells and Biofilm Formation in Streptococcus agalactiae. PLOS Pathogens 5, e1000422, doi:10.1371/journal.ppat.1000422 (2009).
    OpenUrlCrossRefPubMed
  19. ↵
    Izore, T. et al. Structural basis of host cell recognition by the pilus adhesin from Streptococcus pneumoniae. Structure (London, England : 1993) 18, 106–115, doi:10.1016/j.str.2009.10.019 (2010).
    OpenUrlCrossRefPubMed
  20. ↵
    Becherelli, M. et al. The ancillary protein 1 of Streptococcus pyogenes FCT-1 pili mediates cell adhesion and biofilm formation through heterophilic as well as homophilic interactions. Molecular microbiology 83, 1035–1047, doi:10.1111/j.1365-2958.2012.07987.x (2012).
    OpenUrlCrossRefPubMed
  21. ↵
    Keogh, D. et al. Extracellular Electron Transfer Powers Enterococcus faecalis Biofilm Metabolism. mBio 9, doi:10.1128/mBio.00626-17 (2018).
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Shi, L. et al. Extracellular electron transfer mechanisms between microorganisms and minerals. Nature reviews. Microbiology 14, 651–662, doi:10.1038/nrmicro.2016.93 (2016).
    OpenUrlCrossRef
  23. ↵
    Weber, K. A., Achenbach, L. A. & Coates, J. D. Microorganisms pumping iron: anaerobic microbial iron oxidation and reduction. Nature reviews. Microbiology 4, 752–764, doi:10.1038/nrmicro1490 (2006).
    OpenUrlCrossRefPubMedWeb of Science
  24. ↵
    Sure, S. K., Ackland, L. M., Torriero, A. A., Adholeya, A. & Kochar, M. Microbial Nanowires: An Electrifying Tale. Microbiology, doi:10.1099/mic.0.000382 (2016).
    OpenUrlCrossRef
  25. ↵
    Wang, F. et al. Structure of Microbial Nanowires Reveals Stacked Hemes that Transport Electrons over Micrometers. Cell 177, 361–369.e310, doi:10.1016/j.cell.2019.03.029 (2019).
    OpenUrlCrossRef
  26. ↵
    Light, S. H. et al. A flavin-based extracellular electron transfer mechanism in diverse Gram-positive bacteria. Nature 562, 140–144, doi:10.1038/s41586-018-0498-z (2018).
    OpenUrlCrossRef
  27. ↵
    Pankratova, G., Leech, D., Gorton, L. & Hederstedt, L. Extracellular Electron Transfer by the Gram-Positive Bacterium Enterococcus faecalis. Biochemistry 57, 4597–4603, doi:10.1021/acs.biochem.8b00600 (2018).
    OpenUrlCrossRef
  28. ↵
    1. M. S. Gilmore,
    2. D. B. Clewell,
    3. Y. Ike, &
    4. N. Shankar
    Ramsey, M., Hartke, A. & Huycke, M. in Enterococci: From Commensals to Leading Causes of Drug Resistant Infection (eds M. S. Gilmore, D. B. Clewell, Y. Ike, & N. Shankar) (Massachusetts Eye and Ear Infirmary, 2014).
  29. ↵
    Sillanpaa, J. et al. Contribution of individual Ebp Pilus subunits of Enterococcus faecalis OG1RF to pilus biogenesis, biofilm formation and urinary tract infection. PloS one 8, e68813, doi:10.1371/journal.pone.0068813 (2013).
    OpenUrlCrossRefPubMed
  30. ↵
    Krishnan, V. et al. Structure of Streptococcus agalactiae tip pilin GBS104: a model for GBS pili assembly and host interactions. Acta crystallographica. Section D, Biological crystallography 69, 1073–1089, doi:10.1107/S0907444913004642 (2013).
    OpenUrlCrossRefPubMed
  31. ↵
    Shimaoka, M., Takagi, J. & Springer, T. A. Conformational regulation of integrin structure and function. Annual review of biophysics and biomolecular structure 31, 485–516, doi:10.1146/annurev.biophys.31.101101.140922 (2002).
    OpenUrlCrossRefPubMedWeb of Science
  32. ↵
    Kristich, C. J. et al. Development and use of an efficient system for random mariner transposon mutagenesis to identify novel genetic determinants of biofilm formation in the core Enterococcus faecalis genome. Applied and environmental microbiology 74, 3377–3386, doi:10.1128/aem.02665-07 (2008).
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Dale, J. L. et al. Comprehensive Functional Analysis of the Enterococcus faecalis Core Genome Using an Ordered, Sequence-Defined Collection of Insertional Mutations in Strain OG1RF. mSystems 3, doi:10.1128/mSystems.00062-18 (2018).
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Logan, B. E., Rossi, R., Ragab, A. a. & Saikaly, P. E. Electroactive microorganisms in bioelectrochemical systems. Nature Reviews Microbiology, doi:10.1038/s41579-019-0173-x (2019).
    OpenUrlCrossRef
  35. ↵
    Logan, B. E. Exoelectrogenic bacteria that power microbial fuel cells. Nature Reviews Microbiology 7, 375, doi:10.1038/nrmicro2113 (2009).
    OpenUrlCrossRefPubMedWeb of Science
  36. ↵
    Rana, N. F. et al. Redox balance via lactate dehydrogenase is important for multiple stress resistance and virulence in Enterococcus faecalis. Infection and immunity 81, 2662–2668, doi:10.1128/IAI.01299-12 (2013).
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Huycke, M. M. et al. Extracellular superoxide production by Enterococcus faecalis requires demethylmenaquinone and is attenuated by functional terminal quinol oxidases. Molecular microbiology 42, 729–740 (2001).
    OpenUrlCrossRefPubMedWeb of Science
  38. ↵
    Vitreschak, A. G., Rodionov, D. A., Mironov, A. A. & Gelfand, M. S. Regulation of riboflavin biosynthesis and transport genes in bacteria by transcriptional and translational attenuation. Nucleic acids research 30, 3141–3151 (2002).
    OpenUrlCrossRefPubMedWeb of Science
  39. ↵
    Zhang, E., Cai, Y., Luo, Y. & Piao, Z. Riboflavin-shuttled extracellular electron transfer from Enterococcus faecalis to electrodes in microbial fuel cells. Canadian journal of microbiology 60, 753–759, doi:10.1139/cjm-2014-0389 (2014).
    OpenUrlCrossRef
  40. ↵
    Byappanahalli, M. N., Nevers, M. B., Korajkic, A., Staley, Z. R. & Harwood, V. J. Enterococci in the environment. Microbiology and molecular biology reviews : MMBR 76, 685–706, doi:10.1128/MMBR.00023-12 (2012).
    OpenUrlAbstract/FREE Full Text
  41. ↵
    Benito, P. & Miller, D. Iron absorption and bioavailability: An updated review. Nutrition Research 18, 581–603, doi:https://doi.org/10.1016/S0271-5317(98)00044-X (1998).
    OpenUrlCrossRef
  42. ↵
    Saunders, S. H. & Newman, D. K. Extracellular Electron Transfer Transcends Microbe-Mineral Interactions. Cell Host & Microbe 24, 611–613, doi:https://doi.org/10.1016/j.chom.2018.10.018 (2018).
    OpenUrl
  43. ↵
    Guiton, P. S. et al. Contribution of autolysin and Sortase a during Enterococcus faecalis DNA-dependent biofilm development. Infection and immunity 77, 3626–3638, doi:10.1128/iai.00219-09 (2009).
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Thomas, V. C., Thurlow, L. R., Boyle, D. & Hancock, L. E. Regulation of autolysis-dependent extracellular DNA release by Enterococcus faecalis extracellular proteases influences biofilm development. Journal of bacteriology 190, 5690–5698, doi:10.1128/jb.00314-08 (2008).
    OpenUrlAbstract/FREE Full Text
  45. ↵
    Afonina, I., Lim, X. N., Tan, R. & Kline, K. A. Planktonic Interference and Biofilm Alliance between Aggregation Substance and Endocarditis-and Biofilm-Associated Pili in Enterococcus faecalis. Journal of bacteriology 200, doi:10.1128/jb.00361-18 (2018).
    OpenUrlCrossRef
  46. ↵
    Bourgogne, A., Thomson, L. C. & Murray, B. E. Bicarbonate enhances expression of the endocarditis and biofilm associated pilus locus, ebpR-ebpABC, in Enterococcus faecalis. BMC Microbiol 10, 17, doi:10.1186/1471-2180-10-17 (2010).
    OpenUrlCrossRefPubMed
  47. ↵
    Bourgogne, A. et al. EbpR is important for biofilm formation by activating expression of the endocarditis and biofilm-associated pilus operon (ebpABC) of Enterococcus faecalis OG1RF. Journal of bacteriology 189, 6490–6493, doi:10.1128/JB.00594-07 (2007).
    OpenUrlAbstract/FREE Full Text
  48. ↵
    Manias, D. A. & Dunny, G. M. Expression of Adhesive Pili and the Collagen-Binding Adhesin Ace Is Activated by ArgR Family Transcription Factors in Enterococcus faecalis. Journal of bacteriology 200, doi:10.1128/jb.00269-18 (2018).
    OpenUrlCrossRef
  49. ↵
    Bourgogne, A., Hilsenbeck, S. G., Dunny, G. M. & Murray, B. E. Comparison of OG1RF and an Isogenic fsrB Deletion Mutant by Transcriptional Analysis: the Fsr System of Enterococcus faecalis Is More than the Activator of Gelatinase and Serine Protease. Journal of bacteriology 188, 2875–2884, doi:10.1128/JB.188.8.2875-2884.2006 (2006).
    OpenUrlAbstract/FREE Full Text
  50. ↵
    Gao, P. et al. Enterococcus faecalis rnjB is required for pilin gene expression and biofilm formation. Journal of bacteriology 192, 5489–5498, doi:10.1128/jb.00725-10 (2010).
    OpenUrlAbstract/FREE Full Text
  51. ↵
    Gao, P. et al. Functional studies of E. faecalis RNase J2 and its role in virulence and fitness. PloS one 12, e0175212, doi:10.1371/journal.pone.0175212 (2017).
    OpenUrlCrossRef
  52. ↵
    Glasauer, S., Langley, S. & Beveridge, T. J. Intracellular iron minerals in a dissimilatory iron-reducing bacterium. Science (New York, N.Y.) 295, 117–119, doi:10.1126/science.1066577 (2002).
    OpenUrlAbstract/FREE Full Text
  53. ↵
    Ubeda, C. et al. Vancomycin-resistant Enterococcus domination of intestinal microbiota is enabled by antibiotic treatment in mice and precedes bloodstream invasion in humans. The Journal of clinical investigation 120, 4332–4341, doi:10.1172/JCI43918 (2010).
    OpenUrlCrossRefPubMedWeb of Science
  54. Caballero, S. et al. Distinct but Spatially Overlapping Intestinal Niches for Vancomycin-Resistant Enterococcus faecium and Carbapenem-Resistant Klebsiella pneumoniae. PLOS Pathogens 11, e1005132, doi:10.1371/journal.ppat.1005132 (2015).
    OpenUrlCrossRefPubMed
  55. ↵
    Caballero, S. et al. Cooperating Commensals Restore Colonization Resistance to Vancomycin-Resistant Enterococcus faecium. Cell Host Microbe 21, 592–602.e594, doi:10.1016/j.chom.2017.04.002 (2017).
    OpenUrlCrossRefPubMed
  56. ↵
    La Rosa, S. L. et al. In vivo assessment of growth and virulence gene expression during commensal and pathogenic lifestyles of luxABCDE-tagged Enterococcus faecalis strains in murine gastrointestinal and intravenous infection models. Applied and environmental microbiology 79, 3986–3997, doi:10.1128/aem.00831-13 (2013).
    OpenUrlAbstract/FREE Full Text
  57. ↵
    Riboulet, E. et al. Relationships between oxidative stress response and virulence in Enterococcus faecalis. Journal of molecular microbiology and biotechnology 13, 140–146, doi:10.1159/000103605 (2007).
    OpenUrlCrossRefPubMedWeb of Science
  58. ↵
    Aktan, Y., Tan, S. & Icgen, B. Characterization of lead-resistant river isolate Enterococcus faecalis and assessment of its multiple metal and antibiotic resistance. Vol. 185 (2012).
  59. ↵
    Colomer-Winter, C., Gaca, A. O. & Lemos, J. A. Association of Metal Homeostasis and (p)ppGpp Regulation in the Pathophysiology of Enterococcus faecalis. Infection and immunity 85, doi:10.1128/iai.00260-17 (2017).
    OpenUrlCrossRef
  60. ↵
    Karp, P. D. et al. The BioCyc collection of microbial genomes and metabolic pathways. Briefings in bioinformatics, doi:10.1093/bib/bbx085 (2017).
    OpenUrlCrossRef
  61. ↵
    Kline, K. A. et al. Mechanism for Sortase Localization and the Role of Sortase Localization in Efficient Pilus Assembly in Enterococcus faecalis. Journal of bacteriology 191, 3237–3247, doi:10.1128/jb.01837-08 (2009).
    OpenUrlAbstract/FREE Full Text
  62. ↵
    Waterhouse, A. et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res 46, W296–w303, doi:10.1093/nar/gky427 (2018).
    OpenUrlCrossRefPubMed
  63. ↵
    Winn, M. D. et al. Overview of the CCP4 suite and current developments. Acta Crystallogr D Biol Crystallogr 67, 235–242, doi:10.1107/s0907444910045749 (2011).
    OpenUrlCrossRefPubMedWeb of Science
  64. ↵
    Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 60, 2126–2132, doi:10.1107/s0907444904019158 (2004).
    OpenUrlCrossRefPubMedWeb of Science
  65. ↵
    L DeLano, W. The PyMOL Molecular Graphics System (2002) DeLano Scientific, Palo Alto, CA, USA. http://www.pymol.org. (2002).
  66. ↵
    Banla, I. L. et al. Modulators of Enterococcus faecalis Cell Envelope Integrity and Antimicrobial Resistance Influence Stable Colonization of the Mammalian Gastrointestinal Tract. Infection and immunity 86, e00381–00317, doi:10.1128/IAI.00381-17 (2017).
    OpenUrlAbstract/FREE Full Text
  67. ↵
    Zackular, J. P. et al. Dietary zinc alters the microbiota and decreases resistance to Clostridium difficile infection. Nat Med 22, 1330–1334, doi:10.1038/nm.4174 (2016).
    OpenUrlCrossRef
  68. ↵
    Juttukonda, L. J. et al. Dietary Manganese Promotes Staphylococcal Infection of the Heart. Cell Host Microbe 22, 531–542.e538, doi:10.1016/j.chom.2017.08.009 (2017).
    OpenUrlCrossRef
Back to top
Posted April 07, 2019.
Download PDF
Email
Sortase-assembled pili promote extracellular electron transfer and iron acquisition in Enterococcus faecalis biofilm
Ling Ning Lam, Jun Jie Wong, Artur Matysik, Jason J. Paxman, Kelvin Kian Long Chong, Pui Man Low, Zhi Sheng Chua, Begoña Heras, Enrico Marsili, Kimberly A. Kline
bioRxiv 601666; doi: https://doi.org/10.1101/601666
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area