Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Allosteric Regulation of Fibronectin/α5β1 Interaction by Fibronectin-Binding MSCRAMMs

  • Xiaowen Liang,

    Affiliation Center for Infectious and Inflammatory Diseases, Institute of Biosciences and Technology, Texas A&M Health Science Center, Houston, TX, 77030, United States of America

    ⨯
  • Brandon L. Garcia,

    Current address: Department of Biochemistry & Molecular Biophysics, Kansas State University, Manhattan, KS, 66506, United States of America

    Affiliation Center for Infectious and Inflammatory Diseases, Institute of Biosciences and Technology, Texas A&M Health Science Center, Houston, TX, 77030, United States of America

    ⨯
  • Livia Visai,

    Affiliations Dep. of Molecular Medicine, UdR INSTM, Center for Tissue Engineering (C.I.T.), University of Pavia, 27100, Pavia, Italy, Dep. of Occupational Medicine, Ergonomy and Disability, Salvatore Maugeri Foundation, IRCCS, Nanotechnology Laboratory, 27100, Pavia, Italy

    ⨯
  • Sabitha Prabhakaran,

    Current address: Houston Methodist Hospital Research Institute, Houston Methodist Hospital, Houston, TX, 77030, United States of America

    Affiliation Center for Infectious and Inflammatory Diseases, Institute of Biosciences and Technology, Texas A&M Health Science Center, Houston, TX, 77030, United States of America

    ⨯
  • Nicola A. G. Meenan,

    Affiliation Department of Biology, University of York, York, YO10 5DD, United Kingdom

    ⨯
  • Jennifer R. Potts,

    Affiliation Department of Biology, University of York, York, YO10 5DD, United Kingdom

    ⨯
  • Martin J. Humphries,

    Affiliation Wellcome Trust Centre for Cell-Matrix Research, Faculty of Life Sciences, University of Manchester, Manchester, M13 9PT, United Kingdom

    ⨯
  • Magnus Höök

    mhook@ibt.tamhsc.edu

    Affiliation Center for Infectious and Inflammatory Diseases, Institute of Biosciences and Technology, Texas A&M Health Science Center, Houston, TX, 77030, United States of America

    ⨯

Abstract

Adherence of microbes to host tissues is a hallmark of infectious disease and is often mediated by a class of adhesins termed MSCRAMMs (Microbial Surface Components Recognizing Adhesive Matrix Molecules). Numerous pathogens express MSCRAMMs that specifically bind the heterodimeric human glycoprotein fibronectin (Fn). In addition to roles in adhesion, Fn-binding MSCRAMMs exploit physiological Fn functions. For example, several pathogens can invade host cells by a mechanism whereby MSCRAMM-bound Fn bridges interaction with α5β1 integrin. Here, we investigate two Fn-binding MSCRAMMs, FnBPA (Staphylococcus aureus) and BBK32 (Borrelia burgdorferi) to probe structure-activity relationships of MSCRAMM-induced Fn/α5β1integrin activation. Circular dichroism, fluorescence resonance energy transfer, and dynamic light scattering techniques uncover a conformational rearrangement of Fn involving domains distant from the MSCRAMM binding site. Surface plasmon resonance experiments demonstrate a significant enhancement of Fn/α5β1 integrin affinity in the presence of FnBPA or BBK32. Detailed kinetic analysis of these interactions reveal that this change in affinity can be attributed solely to an increase in the initial Fn/α5β1 on-rate and that this rate-enhancement is dependent on high-affinity Fn-binding by MSCRAMMs. These data implicate MSCRAMM-induced perturbation of specific intramolecular contacts within the Fn heterodimer resulting in activation by exposing previously cryptic α5β1 interaction motifs. By correlating structural changes in Fn to a direct measurement of increased Fn/α5β1 affinity, this work significantly advances our understanding of the structural basis for the modulation of integrin function by Fn-binding MSCRAMMs.

Citation: Liang X, Garcia BL, Visai L, Prabhakaran S, Meenan NAG, Potts JR, et al. (2016) Allosteric Regulation of Fibronectin/α5β1 Interaction by Fibronectin-Binding MSCRAMMs. PLoS ONE 11(7): e0159118. https://doi.org/10.1371/journal.pone.0159118

Editor: Daniel Bouvard, Institut Albert Bonniot-INSERMU823, FRANCE

Received: May 7, 2016; Accepted: June 27, 2016; Published: July 19, 2016

Copyright: © 2016 Liang 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 relevant data are within the paper and its Supporting Information files.

Funding: This work was supported by NIH grants HL119648 and AI20624 to MH and AI113552 (to support BLG). This research was supported by the Wellcome Trust (072797) and the British Heart Foundation (FS/07/034). 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

Fibronectin (Fn) is a multidomain glycoprotein found in blood plasma, other bodily fluids and the extra-cellular matrix (ECM) and serves as a natural ligand for several integrins including α5β1, αvβ3, and α4β1 [1]. The functional roles of Fn are diverse and include ECM assembly, angiogenesis, wound-repair, and oncogenesis [2]. Fn is secreted as a C-terminally disulfide-linked heterodimer composed of three structurally defined repeating units termed Fn domains or modules (Fig 1A). Major forms of Fn are comprised of 12 type I modules (FnI) and two type II modules (FnII), while splice variation results in 15 to 17 type III modules (FnIII). Fn is found in two predominant forms; cellular Fn, which is tissue localized and assembled into a fibrillar matrix, and the hepatocyte expressed, soluble plasma Fn that is secreted and maintained in blood at ~0.3 mg ml-1 [3]. Despite having independent functional roles from cellular Fn, it is significant that plasma Fn accounts for a large fraction of Fn found in tissue ECM [3,4].

thumbnail
Download:
Fig 1. Domain architectures and sequences of the polypeptides.

(A) Fn is a large C-terminally disulfide linked heterodimer that is modular in nature. Alternative splicing results in Fn monomers which are similar but not identical [5]. For clarity, only the larger monomer is shown here. Fn consists of three domain types each with distinct folds: FnI, FnII and FnIII (Uniprot ID: P02751). Splice variation results in a variable region (V) in a single Fn monomer. Several functionally relevant protease-stable Fn fragments are known, including the N-terminal domain (Fn-NTD), the gelatin-binding domain (Fn-GBD), and the cell-binding domain (Fn-CBD). The Fn-CBD fragment harbors the canonical α5β1 “RGD” interaction motif within the 10th FnIII module and “synergy site” in the 9th FnIII module. Two free cysteines (SH) exist in each Fn monomer which were used to site-specifically label the FRET probes used in this study and are located in the 7th and 15th FnIII modules. Arrows indicate specific intramolecular interactions formed between domains of the Fn-NTD fragment and the distant Fn-CBD fragment. (B) Domain organizations of mature FnBPA from S. aureus. The extracellular region of FnBPA consists of an A domain that interacts with fibrinogen, followed by a sequential arrangement of 11 Fn binding repeats, the proline-rich (PR) and cell wall (W) spanning regions, and a C-terminal sortase A recognition motif (LPETG) (Uniprot ID: P14738). Sequence alignment of Fn-binding repeats used in this study are shown. Highlighted in cyan are the residues within each repeat that bind to or are presumed to engage Fn-NTD. (C) Domain organizations of mature BBK32 from B. burgdorferi (Uniprot ID: O50835). The residues involved in binding to Fn-NTD are shown in cyan. Unlike FnBPA BBK32 also encodes a binding site for Fn-GBD indicated by the underlined sequence [6]. The recently identified C1 binding domain is also indicated [7]. All domains are drawn to the scale using IBS (Illustrator for Biological Sequences, http://ibs.biocuckoo.org).

https://doi.org/10.1371/journal.pone.0159118.g001

To date, over one-hundred bacterial Fn-binding proteins (FnBPs) have been reported, and a majority of these belong to a protein family termed Microbial Surface Components Recognizing Adhesive Matrix Molecules (MSCRAMMs) [2]. Many Fn-binding MSCRAMMs contribute to virulence as in vivo studies demonstrate a critical role for these proteins in development of infectious diseases such as endocarditis, mastitis, and wound-infection [8–10]. A substantial body of evidence suggests that FnBPs can also manipulate physiological functions of Fn and thus contribute to pathogenesis in ways beyond mediating bacterial adhesion [6,11–16]. Indeed, several studies over the past decade involving the F1 and SfbI proteins from Streptococcus pyogenes [11,12,17], and more recently, BBK32 from Borrelia burgdorferi [6], have led to the development of a model whereby certain endogenous Fn activities are activated allosterically by Fn-binding MSCRAMMs. Specific intramolecular interactions exist in native plasma Fn which hold its solution conformation in a relatively compact state [18]. These interactions are mediated, in part, by FnI modules from the Fn-NTD fragment and FnIII modules originating from the distant Fn-CBD fragment (Fig 1A) [19–22]. Intrinsically disordered sequences from SfbI, F1, and BBK32 engage the Fn-NTD fragment via a tandem β-zipper model of binding [6,23–25] and, in doing so, compete for these intramolecular Fn contacts [6,12,24]. This competition results in a conformational expansion of Fn which has been directly measured using dynamic light scattering (DLS) upon binding of SfbI [12]. Interestingly, the SfbI-induced conformational change in Fn mirrors the transition of compact Fn to an elongated structure in solutions of increasing ionic strength [18].

A monoclonal antibody (mAbIII-10), which recognizes a conformationally-sensitive epitope within 10FnIII [26] has also been used to monitor structural changes induced in Fn by both streptococcal and borrelial FnBPs [6,12,17]. These studies reveal a conformational rearrangement in Fn that occurs in domains located far outside of the FnBP/Fn binding site. The conformational expansion of Fn induced by binding of SfbI, F1, or BBK32 has been linked to three primary effects; (i) the 10th FnIII module of Fn which harbors the “RGD” integrin recognition motif exposes a previously cryptic mAbIII-10 epitope, (ii) the motogenic “IGD” motifs of the 7th and 9th FnI modules of the Fn-GBD fragment become exposed, and (iii) in the case of SfbI, binding results in the blocking of Fn assembly into fibrils. In addition to binding to the Fn-NTD fragment, SfbI, F1 and BBK32 also harbor an Fn-GBD binding site. However, Fn-NTD interaction alone is sufficient to induce allosteric changes in Fn [12]. Consistent with this is the recent discovery of the Streptococcus equi FnBP termed SFS, which only binds Fn via Fn-GBD interaction, yet fails to cause conformational expansion of Fn [27].

Integrins are essential metazoan heterodimeric glycoproteins that mediate cell-adhesion, establish transmembrane connections to the cytoskeleton, and play an integral role in cell signaling pathways [28]. Interestingly, integrins are common targets of pathogens and often participate in bacterial and viral adhesion to host cells [29]. Two modes of microbial integrin recognition have been described and include direct binding by bacterial surface proteins [30–32] or indirect binding via physiological ligands such as Fn [29,33,34]. The latter mode is exemplified by an FnBP expressed by the gram-positive pathogen Staphylococcus aureus, termed FnBPA [35]. By acting as a molecular bridge, staphylococcal FnBPA exploits Fn’s role as a natural substrate for α5β1 integrin, causing activation of endocytic pathways, and ultimately resulting in the internalization of S. aureus by non-professional phagocytes [14–16,36–45]. Full-length FnBPA harbors eleven Fn-binding repeats that specifically interact with the 2-5FnI modules of the Fn-NTD fragment via a tandem beta-zipper [25,46] (Fig 1B). Unlike the streptococcal and borrelial FnBPs, FnBPA binds exclusively to Fn-NTD and makes no direct interactions with the Fn-GBD fragment. Of the 11 FnBPA Fn-binding repeats, six exhibit high-affinity binding of Fn [46] and a single high-affinity repeat is sufficient for α5β1 integrin-dependent in vitro invasion of endothelial cells by S. aureus [14].

While S. aureus is typically considered an extracellular pathogen, intracellular S. aureus is associated with several chronic and reoccurring infections [47–50] and can establish infection even in the presence of the vancomycin [51]. Thus, S. aureus internalized by non-professional phagocytes represents a bacterial reservoir protected against antibiotics and innate host defense systems. Although it is evident that FnBPA, Fn, and α5β1 integrin work in concert to promote an intracellular “life-style” for S. aureus [14–16,40], detailed knowledge of the underlying molecular events that lead to invasion have remained elusive. A mechanistic understanding of how FnBPA manipulates Fn function to enable entry of S. aureus into non-phagocytic cell types, will benefit renewed efforts [51] to target this form of the pathogen. With this in mind, the goals of this study were to determine if the allosteric model of Fn activation, which has been proposed for streptococcal and borrelial FnBPs, applies to staphylococcal FnBPA. Furthermore, we sought to better understand the nature of structural changes induced by Fn-binding MSCRAMMs and correlate these changes to quantitative and direct measurements of Fn/α5β1 integrin interaction. Therefore, in parallel with our studies of FnBPA, we have also evaluated the effect of borrelial BBK32 (Fig 1C) on the solution structure of Fn and its effect on Fn/α5β1 interaction. Herein, we apply a cross-disciplinary approach to dissect the structural basis for MSCRAMM-induced Fn activation using circular dichroism (CD), DLS, and fluorescence resonance energy transfer (FRET) to examine the conformational status of native Fn or MSCRAMM-bound Fn in solution. Surface plasmon resonance (SPR) is used to quantitatively measure the effects of FnBPA and BBK32 on direct Fn/α5β1 interaction. Together these experiments provide a detailed structure-activity relationship of MSCRAMM/host interaction and strongly suggest an allosteric mode of Fn/α5β1 affinity-enhancement for Fn-binding MSCRAMMs.

Materials and Methods

Materials

Human Fn was purified from freshly drawn citrated plasma (Gulf Coast Regional Blood Center, Houston, TX) using gelatin affinity chromatography combined with arginine affinity chromatography as described previously [52]. Fn was stored at 4°C in Tris-buffered saline (TBS: 50 mM Tris-HCl, pH 7.4, 150 mM NaCl), and used within one month. Fn dimer concentration was calculated from OD280 nm with EC280 (1%) = 12.8 and molecular weight of 500 kDa. The ecto-domain of integrin α5β1-Fc fusion protein, which contains the N-terminal 613 residues of α5-subunit and fragment 121–455 of β1-subunit, was generated and purified as previously reported [53]. Fn-NTD and Fn-CBD were purified as previously described [54]. Expression and purification of recombinant GST-FnBPAs, FnBPA-10 peptide and BBK32(21–205) was performed as previously described [46,55]. Rabbit anti-Fn pAb (polyclonal antibody) was purchased from ICN Pharmaceuticals, Inc. (Costa Mesa, California). BSA (bovine serum albumin), heparin sodium (from porcine intestinal mucosa, 17,000–19,000 Da) was from Sigma.

Dynamic light scattering

DLS measurements were performed using DynaPro Titan Ambient laser unit (Wyatt Technology Corporation) with wavelength at 828.5 nm and 10% to 12% laser power. Samples were dissolved in TBS and filtered through 0.22 μm filter to eliminate any particles. Fn concentration was maintained at 0.8 mg ml-1 (1.6 μM as dimer) in TBS (50 mM Tris-HCl pH 7.4, 150 mM NaCl) for the conditions tested. After equilibration at room temperature for 15 to 30 min, the samples were measured three times, with each measurement lasting 100 seconds (10 acquisitions, 10 seconds each). Data analysis was performed using Dynamics V6 software.

Circular Dichroism

CD measurements in the near UV region (350–250 nm) were carried out at ambient temperature on a Jasco J-720 spectropolarimeter (Easton, MD) with a 1 cm cell. Ten scans were collected and averaged at a scan speed of 200 nm min-1, with a time constant of 1 s and band width of 1 nm. A mean residue molecular weight of 110 was used in the calculation of mean residue ellipticity. The spectra of the Fn solution were background-corrected with the CD signal obtained from the buffer or molecules mixed with Fn.

Fluorescence resonance energy transfer

Fluorescence labeling of human plasma Fn was performed at room temperature as described previously [56]. The fluorescence probes used in this study were purchased from Molecular Probes. Briefly, newly purified human plasma Fn at 1.2 mg ml-1 in TBS was partly denatured in 4 M GdnHCl (Sigma) for 15 min to expose the free cysteines in the Fn dimer. The denatured Fn was mixed with thiol-reactive Alexa Fluo® 546 C5-maleimide (AF546, ~15-fold molar excess over Fn monomer concentration) and allowed to incubate for 2 hr with gentle rocking. The acceptor fluorophore AF546-conjugated Fn (~0.9 mg ml-1) was labeled with a second probe, the amine-reactive Alexa Fluo® 488 carboxylic acid-succinimidyl ester-mixed-isomers (AF488, ~20-fold molar excess). After each labeling, the unbound dye was removed using a PD-10 desalting column (GE Healthcare). The labeling ratios were ~8 AF488 donors and ~4 AF546 acceptors in each Fn dimer.

FRET within donor and acceptor-labeled Fn was measured using spectrofluorimeter LS 50B (Perkin-Elmer) at room temperature with excitation wavelength at 493 nm. The sensitivity of FRET response to the Fn unfolding was evaluated under mild denaturant (up to 2 M guanidinium chloride (GdnHCl)). Measurement of MSCRAMM-induced FRET within Fn was performed by titrating 0.5, 2 and 5 μl of 50 μM of BBK32 or FnBPA-10 in to 500 μl of labeled Fn. Subsequently, the emission spectra collected for each titration was recorded and normalized to the donor emission peak so that changes in energy transfer were reflected only by changes in the acceptor peak. Control experiments were conducted using an equimolar solution of donor-labeled Fn with acceptor-labeled Fn in the presence and absence of 2M GdnHCl to confirm the absence of energy transfer between adjacent Fn molecules in solution.

SPR-based Biacore analysis of Fn and α5β1 interaction

Experiments were performed at 25°C on a Biacore 3000 (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) using HBS-T (10 mM HEPES, pH 7.3, 150 mM NaCl, 0.005% Tween-20) containing 1 mM MnCl2. To prepare sensor surface for kinetic experiments, ~1400 RU of α5β1 was immobilized to a CM5 chip by amine coupling. Briefly, 18 μL of protein (10 μg ml-1 in 10 mM sodium acetate, pH 4.8) was injected onto an activated (4 min activation) surface at a flow rate of 5 μl min-1. BSA was coupled to the adjacent flow cell and served as a reference surface (~5000 RU). A flow rate of 30 μL min-1 was used for all binding experiments. The sensor chip surface was regenerated to remove bound Fn by injection of 1 M NaCl for 1 min. All measurements were baseline corrected by subtracting the response from the reference surface. For kinetic analysis, signals from buffer blanks were subtracted.

Kinetic constants were obtained from curve fitting to the predefined bivalent analyte model using BIAevaluation software (Version 4.1). Heterodimeric Fn has two identical binding sites for integrin α5β1. The reaction between soluble Fn analyte (A) and immobilized integrin ligand (L) can be described by this equation:

Binding to the first ligand molecule is described by a single set of rate constants (ka1, kd1), so that the two sites on the analyte are equivalent in the first binding step. Binding of the second ligand molecule is described by a second set of rate constants (ka2, kd2), which is reported in an unconventional units of RU-1s-1. The first set of rate constants were used for calculating the dissociation constant KD (KD = kd1/ka1).

ELISA-type binding assays

Experiments were conducted at room temperature and each sample was assayed in triplicate. Immunoassay microplate Immulon 2HB (Thermo Scientific, Waltham, MA) was coated with 2 μg of integrins dissolved in TBS for 1 hour. The plate was then blocked for 1 h with 1% (w/v) BSA in a buffer “A” (TBS containing 1 mM MnCl2). Citrated plasma (stored at -20°C for less than 6 months) was fast-thawed at 37°C and spun to remove aggregates. The supernatant was diluted 10-fold in buffer “A” and then mixed with varying concentrations of BBK32 to a final concentration of 5% plasma. The sample mixtures were added onto the plate and incubated for 1 h. The plate was washed three times with buffer “B” (TBS with 1 mM, MnCl2, and 0.05% Tween-20) before addition of rabbit anti-Fn antibody. Following washing with buffer “B”, AP-conjugated goat-anti-rabbit IgG (BioRad, Hercules, CA) was added to the plate and incubated for 1 h. All antibodies were diluted 2000-fold in buffer “B” containing 1% BSA. Alkaline phosphatase substrate PNPP (Pierce, Rockford, IL) was added after washing, and the absorbance was measured at 405 nm using a Thermo Max plate reader (Molecular Devices, Sunnyvale, CA).

Results

The tertiary structure of plasma Fn is altered by FnBPA and BBK32

Although high resolution crystal structures have been solved for a number of smaller Fn fragments, a complete structure of full-length Fn remains to be determined. However, a wide range of biophysical studies using DLS [18,57–61], CD [59,62,63], steady-state fluorescence spectroscopy [59,63,64], small-angle X-ray scattering (SAXS) [65] and small-angle neutron scattering (SANS) [60,65,66] have yielded a wealth of information about the conformational states and structure of native Fn in solution. Conformational changes induced by FnBPs underlie the allosteric model of Fn activation proposed for streptococcal and borrelial FnBPs [6,12,17], and a previous study using DLS indicates an expansion of Fn structure upon SfbI binding which is similar to what is observed when Fn is incubated in a high salt medium [12]. To understand if analogous conformational changes are induced by staphylococcal FnBPA we used DLS to measure the hydrodynamic radii (Rh) of Fn in the presence or absence of the FnBPA Fn-binding repeat FnBPA-10 (Table 1). Fn adopted an extended conformation in the presence of high salt or heparin (Rh = 13.27 ± 0.15; 12.50 ± 0.29, respectively) compared to a more compact structure found in physiological buffer (Rh = 10.20 ± 0.12) which is consistent with previous measurements of Fn Rh by DLS [12,18]. Addition of staphylococcal FnBPA-10 (Rh = 12.73 ± 0.69) or the borrelial FnBP, BBK32 (Rh = 13.00 ± 0.38) resulted in a significantly increased Rh, suggesting that Fn indeed adopts an extended structure upon binding to these MSCRAMMs.

thumbnail
Download:
Table 1. Comparison of hydrodynamic radii of Fn under various conditions.

https://doi.org/10.1371/journal.pone.0159118.t001

Native Fn contains large numbers of aromatic residues in its type I and III domains which exist in common hydrophobic environments and give rise to a signature near-UV CD spectrum characterized by two strong negative peaks centered at 291 nm and 299 nm, and a weaker negative peak at 282 nm (Fig 2) [67,68]. Binding of Fn by the glycosaminoglycan heparin has been shown to alter Fn tertiary structure which can be monitored by shifts in this signature spectrum (Fig 2) [69]. When Fn was incubated in the presence of FnBPA-10 or BBK32, both complexes produce significantly different spectra relative to native Fn (Fig 2). Interestingly we noted the similarity in the near-UV spectra for heparin-bound Fn when compared to FnBPA-10/Fn (Fig 2), whereas, the BBK32 spectrum is reminiscent of the partial unfolding of Fn in 4M urea [63]. Importantly, the MSCRAMM fragments used in these experiments are intrinsically disordered polypeptides that do not contain tryptophan residues and therefore lack near-UV optical activity [70] (Fig 2). However, since the Fn-NTD is engaged by BBK32 and FnBPA-10, such interaction could potentially cause near-UV CD optical activity changes within NTD domain. As shown in S1 Fig, BBK32 and FnBPA-10 caused intrinsic Trp fluorescence quenching in Fn (S1A Fig) and Fn-NTD (S1B Fig), suggesting that Trp environments in the NTD and GBD (in the case for BBK32) are affected by the binding and may attribute partially to the near-UV CD changes in Fn.

thumbnail
Download:
Fig 2. Near-UV CD spectra of Fn.

CD spectra in the range of 250–350 nm are shown for individual molecules or mixtures: Fn dimer (0.82 mg ml-1, 1.6 μM), BBK32 (0.14 mg ml-1, 6.0 μM), FnBPA-10 (0.07 mg ml-1, 15 μM), and heparin (0.27 mg ml-1, 15 μM). All concentrations are final. TBS (50 mM Tris-HCl pH 7.4, 150 mM NaCl) was used for all experiments.

https://doi.org/10.1371/journal.pone.0159118.g002

Data obtained using an anti-Fn antibody (mAb10-III) that recognizes a conformational sensitive epitope indicates that structural changes distant from the streptococcal and borrelial FnBP binding site are induced in Fn [6,12,17] on MSCRAMM binding. To learn if FnBPA causes similar allosteric changes in Fn upon binding, and to confirm apparent conformational changes in solution Fn indicated by CD (Fig 2), we next employed a FRET-based approach. To this end, Fn was site-specifically labelled by conjugating an acceptor fluorophore to free cysteine residues using a technique previously described [56]. Native Fn contains only two free cysteines (4 per dimer molecule) and these are found in the C-terminally positioned 7FnIII and 15FnIII modules [71] (denoted in Fig 1A). When Fn is denatured by introducing increasing amounts of GdnHCl (Fig 3A and 3B), or incubated in the presence of increasing ionic strength (Fig 3C), the distance between the amine coupled donor fluorophores and the cysteine coupled acceptor fluorophores increased and the acceptor peak was diminished. There is not FRET observed between adjacent or denatured Fn molecules (Fig 3D). To determine if the MSCRAMM-induced structural changes measured by CD and DLS involve Fn domains outside of the MSCRAMM binding site, spectra were then collected in the presence of various concentrations of BBK32 (Fig 3E) and FnBPA-10 (Fig 3F) GST fusion proteins. These experiments showed a clear dose-dependent loss of acceptor peak signal for MSCRAMM-bound Fn. Importantly, an identical effect is observed when a 36 amino acid peptide lacking the GST fusion and corresponding to the high-affinity FnBPA repeat FnBPA-5 is used (Fig 3G). GST alone or a GST fusion to the low-affinity FnBPA repeat FnBPA-3 exhibits no effect in this assay system (Fig 3H). Taken together, the DLS, CD, and fluorescence based assays indicated that BBK32 and FnBPA profoundly modify the native solution structure of Fn by promoting an extended Fn conformation that results in large scale rearrangement of Fn domains at sites distant from the MSCRAMM binding site.

thumbnail
Download:
Fig 3. Fluorescent resonance energy transfer of Fn.

Fluorescence emission spectra of Fn-D/A (50 nM in 10 mM Tris, pH7.4, 150 mM NaCl) were collected at 493 nm excitation, (A) in the presence or absence of denaturant (GdnHCl) and (B) accompanying Acceptor/Donor peak ratio as a function of denaturant concentration. Fn-D/A fluorescence emission changes (C) under different ionic strengths, and (E-G) in the presence of different concentrations of BBK32, GST-FnBPA-10, or FnBPA-5 peptide. (H) Acceptor/Donor peak ratio vs. concentration (0, 1, 5, 20, 50 and 200 nM) of protein or peptide added. Spectra are normalized to the donor emission peak at 522 nm such that changes in energy transfer are reflected only by changes in the acceptor peak at 570 nm. (D) Control experiments were conducted by mixing 50 nM of donor-labeled Fn (Fndonor) with 50 nM of acceptor-labeled Fn (Fnacceptor) in TBS or 2 M GdnHCl.

https://doi.org/10.1371/journal.pone.0159118.g003

Fn recognition of the α5β1 integrin is enhanced by FnBPA and BBK32

Fn serves as a primary ECM substrate for several adhesion molecules including the major Fn cell surface receptor, α5β1 integrin [1]. We hypothesized that conformational changes induced upon binding by FnBPA or BBK32, as shown in Figs 2 and 3 and Table 1, may result in modified Fn/integrin interactions. Previous studies on streptococcal and borrelial Fn-binding MSCRAMMs have used the conformational Fn antibody (mAb10-III) as a surrogate for potential effects on Fn/α5β1 integrin interaction. To test this more directly, we developed an SPR-based system to study the effects of FnBPA and BBK32 on the recognition of α5β1 integrin by Fn. Briefly, a recombinant α5β1-Fc fusion protein representing the minimal functional unit of the integrin [53] was immobilized on the surface of a Biacore CM5 chip. Fn interacts with the α5β1 biosensor and binding is enhanced by Mn2+ as previously reported (S1A Fig) [53]. Further validation of this experimental approach was performed by using Fn fragments including the cell-binding domain (Fn-CBD) possessing the canonical α5β1-binding “RGD” (10FnIII) and PSHRN synergy sites (9FnIII) [72,73]. As expected, the Fn-CBD fragment exhibited dose-dependent interaction with α5β1, while the Fn-NTD fragment lacking the integrin interaction domains failed to bind the integrin (S1B and S1C Fig).

Using this approach we measured a moderate affinity of ~1 μM for Fn/α5β1 (Fig 4A–4D) that is in good agreement with affinities reported for fibroblast cell interaction with plasma Fn [74]. To understand if MSCRAMM binding to Fn affects the Fn/α5β1 interaction, we next injected mixtures of BBK32/Fn or FnBPA-10/Fn (fixed MSCRAMM + varied Fn concentrations) over the immobilized α5β1 integrin. A remarkable enhancement of the binding response was observed for MSCRAMM-bound Fn relative to native Fn (Fig 4B–4D). To obtain a quantitative measure of the increase in affinity, kinetic parameters were obtained by fitting SPR response curves to a bivalent analyte binding model (Fig 4B–4D). This model was selected on the basis of closeness of fit (χ2≈1, Table 1) and the expectation that dimeric Fn contributes one identical α5β1 binding site per subunit. Analysis for each interaction revealed that the approximately five-fold increase in apparent affinity (KD) (KDFn/α5β1 = 1000 nM vs. KDBBK32/Fn/α5β1 = 170 nM & KDFnBPA-10/Fn/α5β1 = 260 nM) was due to an increase in the initial association rate ([ka1 = (1.5, 6.1, & 7.5) x104 M-1s-1], respectively). Interestingly, all other rate constants (kd1, ka2 and kd2) (Table 2) were similar across experiments, indicating that the Fn/α5β1 binding site present in native Fn is the same for MSCRAMM-bound Fn.

thumbnail
Download:
Fig 4. SPR analysis of direct Fn/α5β1 interaction in the presence of Fn-binding MSCRAMMs.

(A) Two-fold linear dilution series of Fn in HBS-T containing 1 mM MnCl2, and in the presence of (B) 1 μM of FnBPA-10, or (C) 0.5 μM of BBK32 were injected over the α5β1 biosensor. (D) A higher Fn concentration series was used for measurement of Fn/α5β1 binding kinetics in the absence of MSCRAMMs. The SPR response curves generated from Fn binding to α5β1 are shown in black with lower curve corresponding to lower concentration of Fn injected. Dissociation constants (KD) for Fn/α5β1 the interaction (B-D) were obtained from fitting SPR response curves to a bivalent analyte binding model (fitted lines are shown in red) and the kinetic parameters are listed in Table 2. There was no detectable binding when FnBPA-10 (1 μM) or BBK32 (0.5 μM) were injected over the same α5β1 surface (data not shown).

https://doi.org/10.1371/journal.pone.0159118.g004

thumbnail
Download:
Table 2. Kinetic parameters for Fn::α5β1 interactions.

https://doi.org/10.1371/journal.pone.0159118.t002

To address the specificity of enhanced Fn/α5β1 activity and to further dissect the kinetic observations presented above, we next injected a mixture of varied MSCRAMM and fixed Fn concentrations (Fig 5A–5C). Again we observed an increase in response that behaved in a dose-dependent and saturable manner, similar to the results obtained in Fig 4B–4D. Likewise, only the initial association rate ka1 was changed, while the stability of the complex (kd1), the second rate step (ka2 and kd2), and maximum response when all the binding sites on active α5β1 are saturated (Rmax) were unaffected by the presence of MSCRAMMs (Fig 5B–5D). The initial association rate, ka1, is the only kinetic parameter directly affected by analyte concentration (units = M-1s-1). As the concentrations of Fn used across experiments were the same, the observed increase in ka1 for Fn bound by increasing concentrations of MSCRAMMs likely represents an increase in the apparent concentration of the active integrin binding site in Fn.

thumbnail
Download:
Fig 5. SPR analysis of kinetic components of the Fn/α5β1 interaction that are affected by MSCRAMMs.

Fn (50 nM) in the presence of indicated concentration of (A) FnBPA-10 or (C) BBK32 was injected over an α5β1 surface on a Biacore sensor chip. Kinetic components (ka1, kd1, ka2, kd2) and Rmax obtained from local fitting were normalized to those of Fn alone, and plotted as a function of (B) FnBPA-10 or (D) BBK32 concentrations.

https://doi.org/10.1371/journal.pone.0159118.g005

High-affinity FnBPA/Fn-NTD interaction is required for enhancement of Fn/α5β1 activity

Of the eleven Fn-binding repeats found in FnBPA, six have been characterized as high-affinity binders (FnBPA-1, -4, -5, -9, -10, -11). The higher affinity of these repeats has been attributed to their ability to interact with four sequential FnI modules of the Fn-NTD fragment [46]. For example, while FnBPA-10 contains correctly spaced motifs to interact with 2-5FnI in a β-tandem zipper model of binding, the low-affinity repeat FnBPA-3 appears to lack the 5FnI interaction motif. Intriguingly, these particular Fn modules have been implicated in mediating long-range intra-molecular contacts within native Fn dimers (Fig 1A) [19,21,22]. To understand if high-affinity interaction is required for an MSCRAMM-induced increase in Fn/α5β1 recognition, we next tested a panel of high and low-affinity FnBPA repeats (Fig 6). All high-affinity repeats studied (FnBPA-1, -5, -9, -10, and -11) increased the Fn/α5β1 binding response, while the low-affinity repeat FnBPA-3 exhibited control-level signal.

thumbnail
Download:
Fig 6. Relationship of high affinity Fn-NTD fragment interaction and α5β1 integrin activation.

SPR sensorgrams were generated by injecting 50 nM Fn in the presence of GST-fused FnBPA repeats (0.5 μM for high-affinity repeats -1, -5, -9, -10, -11 and 15 μM for low affinity repeat -3) over immobilized α5β1. The KD reported for direct binding of Fn-NTD by high affinity FnBPA repeats are provided for reference and were obtained from previously published isothermal titration calorimetry (ITC) experiments [46]. The weak affinity between FnBPA-3 and Fn-NTD was measured by SPR (this work, data not shown). None of the repeats bound directly to α5β1 (data not shown).

https://doi.org/10.1371/journal.pone.0159118.g006

MSCRAMMs enhance Fn/α5β1 interaction in human plasma

The SPR-based activity assay utilizing purified native and recombinant reagents allowed for a quantitative kinetic assessment of the MSCRAMM-induced enhancement of Fn/α5β1 binding. We next asked if this effect could be recapitulated using human plasma as a source of native Fn. Indeed, an MSCRAMM-dependent increase in plasma Fn/α5β1 binding was observed in the presence of FnBPA-10 or and to a greater extent for BBK32 when 5% human plasma (~15 μg ml-1 or equivalently ~30 nM Fn) was used in an ELISA-type binding assay (Fig 7).

thumbnail
Download:
Fig 7. ELISA detection of the effect of MSCRAMMs on intact Fn in plasma for α5β1 binding.

5% human blood plasma in TBS containing 1 mM MnCl2 was incubated with (A) FnBPA-10 or (B) BBK32, and added α5β1-coated wells or BSA (0.2 μg per well) coated surfaces. Bound Fn was probed with anti-Fn pAb. BBK32 did not bind directly to α5β1 when detected by anti-BBK32 pAb (data not shown). Data are one representative of three experiments.

https://doi.org/10.1371/journal.pone.0159118.g007

Discussion

Modeling the allosteric activation of Fn/α5β1 by Fn-binding MSCRAMMs

By manipulating the conformation of key Fn functional domains, Fn-binding MSCRAMMs enable pathogens to hijack normal host physiology. In the case of S. aureus, enhanced Fn/α5β1 interaction leads to recruitment of focal contact-associated proteins and subsequent integrin clustering at the bacterial attachment site initiates intracellular signaling through the focal adhesion kinase (FAK) and Src kinases [33,43,75]. This appears to confer an advantage to S. aureus by facilitating immune evasion and serves as a bacterial reservoir in chronic infections [33,51]. In light of the observations above, we propose a model for the allosteric enhancement of Fn/α5β1 binding by staphylococcal FnBPA (Fig 8), similar to that put forward for streptococcal and borrelial FnBPs [6,12,13,17]. We speculate that dimeric native Fn exists as an array of conformers in solution. At equilibrium, predominate forms of Fn are compact and sterically occlude α5β1 interaction domains (9FnIII-10FnIII) by long-range intramolecular interactions, thereby impeding Fn/α5β1 recognition. The equilibrium of Fn conformers shifts upon high-affinity binding of FnBPA and native intramolecular contacts are disrupted exposing previously cryptic α5β1 binding sites and promoting Fn/α5β1 binding. Domains involved in integrin binding by Fn are located at a site distant from the MSCRAMM binding site. It appears then that a critical feature of MSCRAMM-induced activation of Fn/α5β1 is allostery and allosteric modes of host protein modulation have been previously reported for S. aureus [76]. Our findings also suggest that the conformation of integrin ligands, like Fn, may fine-tune affinities and therefore contribute to integrin “activation” traditionally associated with conformational change of the integrin ectodomains [77].

thumbnail
Download:
Fig 8. An allosteric model of Fn activation by MSCRAMMs.

Based on the data presented here and the previously proposed allosteric model of Fn activation by streptococcal and borrelial FnBPs [6,12,17], we constructed the above model for allosteric affinity enhancement of the Fn/α5β1 interaction by staphylococcal FnBPA. Fn is represented in its native state (left) or bound by FnBPA (right). The native Fn model including intermolecular contacts are adapted from a previously proposed model based on multiple independent studies. In a native state, major solution conformations of Fn occlude the canonical α5β1 interaction motifs located in 9-10FnIII domains. Upon binding to Fn, FnBPA disrupts specific intermolecular contacts in the N-terminal domain of Fn (Fn-NTD) resulting in global structural rearrangements at sites distant from the FnBPA binding site. Previously cryptic integrin interaction sites are exposed, promoting interaction of Fn with α5β1. This initiates ‘outside-in’ signaling events leading to cytoskeletal rearrangement and eventual internalization of Fn-coated S. aureus.

https://doi.org/10.1371/journal.pone.0159118.g008

The data presented here provides a foundation to study the potential involvement of Fn-binding MSCRAMMs in several aspects of Fn physiology. In addition to Fn/integrin interaction, cryptic functional sites in Fn play a fundamental role in mechanoresponsive and motogenic pathways [19,78]. Interestingly, polymorphisms have been recently reported in two high-affinity FnBPA repeats (FnBPA-5 and -9) and S. aureus strains harboring these mutations are highly correlated with infection of cardiovascular devices [79,80]. How these FnBPA mutations impact bound Fn tertiary structure is currently unknown, however, we predict that sequence changes to residues in FnBPA repeats could result in different global Fn structure and thus function. Finally, as the clinical relevance of intracellular forms of S. aureus are becoming clearer [51], elucidation of the molecular mechanisms of Fn-binding MSCRAMMs, like FnBPA, stands to provide a major contribution towards improving the treatment of staphylococcal infectious diseases.

Supporting Information

S1 Fig. Steady-state fluorescence spectroscopy.

Intrinsic tryptophan fluorescence spectra of Fn solutions were obtained on a Spectrofluorimeter LS 50B (Perkin-Elmer) at ambient temperature. (A) Fn (0.1 μM) was incubated in TBS in the presence of 1 M GdnHCl, 0.2 μM BBK32 or 2 μM of FnBPA-10. (B) FnNTD-30K (1.6 μM) were incubated in TBS in the presence of 16 μM of FnBPA-10. Samples were excited at 295 nm with and excitation slit of 5 nm, and emission spectra were collected with an emission slit of 5 nm. All spectra were corrected for background fluorescence by subtraction of the buffer blanks.

https://doi.org/10.1371/journal.pone.0159118.s001

(TIF)

S2 Fig. SPR analysis of Fn/α5β1 interaction.

The α5β1 biosensor was validated by demonstrating metal ion dependence and domain specificity: (A) Fn (250 nM) was injected over immobilized α5β1 surface in the presence of 1 mM MnCl2 (solid line) or 3 mM EDTA (dashed line). (B) Response curves for a two-fold linear dilution series of Fn-CBD over immobilized α5β1 are shown. (C) Comparison of Fn (250 nM, solid line) and the Fn-NTD fragment lacking the canonical integrin binding RGD-motif (500 nM, dashed line).

https://doi.org/10.1371/journal.pone.0159118.s002

(TIFF)

Acknowledgments

The authors are grateful for Susan E. Craig from the Wellcome Trust Centre for Cell­Matrix Research at University of Manchester for the recombinant integrin α5β1.

Author Contributions

Conceived and designed the experiments: MH XL. Performed the experiments: XL. Analyzed the data: MH XL BLG. Contributed reagents/materials/analysis tools: SP LV NAGM MJH JRP. Wrote the paper: MH XL BLG.

References

  1. 1. Leiss M, Beckmann K, Giros A, Costell M, Fassler R. The role of integrin binding sites in fibronectin matrix assembly in vivo. Curr Opin Cell Biol. 2008;20: 502–507. pmid:18586094
  2. 2. Henderson B, Nair S, Pallas J, Williams MA. Fibronectin: a multidomain host adhesin targeted by bacterial fibronectin-binding proteins. FEMS Microbiol Rev. 2011;35: 147–200. pmid:20695902
  3. 3. Moretti FA, Chauhan AK, Iaconcig A, Porro F, Baralle FE, Muro AF. A major fraction of fibronectin present in the extracellular matrix of tissues is plasma-derived. J Biol Chem. 2007;282: 28057–28062. M611315200 [pii]. pmid:17644525
  4. 4. To WS, Midwood KS. Plasma and cellular fibronectin: distinct and independent functions during tissue repair. Fibrogenesis Tissue Repair. 2011;4: 21-1536-4-21.
  5. 5. Peters JH, Loredo GA, Chen G, Maunder R, Hahn TJ, Willits NH, et al. Plasma levels of fibronectin bearing the alternatively spliced EIIIB segment are increased after major trauma. J Lab Clin Med. 2003;141: 401–410. pmid:12819638
  6. 6. Harris G, Ma W, Maurer LM, Potts JR, Mosher DF. Borrelia burgdorferi protein BBK32 binds to soluble fibronectin via the N-terminal 70-kDa region, causing fibronectin to undergo conformational extension. J Biol Chem. 2014;289: 22490–22499. pmid:24962582
  7. 7. Garcia BL, Zhi H, Wager B, Hook M, Skare JT. Borrelia burgdorferi BBK32 Inhibits the Classical Pathway by Blocking Activation of the C1 Complement Complex. PLoS Pathog. 2016;12: e1005404. pmid:26808924
  8. 8. Heying R, van de Gevel J, Que YA, Moreillon P, Beekhuizen H. Fibronectin-binding proteins and clumping factor A in Staphylococcus aureus experimental endocarditis: FnBPA is sufficient to activate human endothelial cells. Thromb Haemost. 2007;97: 617–626. 07040617 [pii]. pmid:17393025
  9. 9. Fisher M, Huang YS, Li X, McIver KS, Toukoki C, Eichenbaum Z. Shr is a broad-spectrum surface receptor that contributes to adherence and virulence in group A streptococcus. Infect Immun. 2008;76: 5006–5015. pmid:18710861
  10. 10. Brouillette E, Talbot BG, Malouin F. The fibronectin-binding proteins of Staphylococcus aureus may promote mammary gland colonization in a lactating mouse model of mastitis. Infect Immun. 2003;71: 2292–2295. pmid:12654860
  11. 11. Marjenberg ZR, Ellis IR, Hagan RM, Prabhakaran S, Hook M, Talay SR, et al. Cooperative binding and activation of fibronectin by a bacterial surface protein. J Biol Chem. 2011;286: 1884–1894. pmid:21059652
  12. 12. Maurer LM, Ma W, Eickstaedt NL, Johnson IA, Tomasini-Johansson BR, Annis DS, et al. Ligation of the fibrin-binding domain by beta-strand addition is sufficient for expansion of soluble fibronectin. J Biol Chem. 2012;287: 13303–13312. pmid:22351755
  13. 13. Maurer LM, Tomasini-Johansson BR, Ma W, Annis DS, Eickstaedt NL, Ensenberger MG, et al. Extended binding site on fibronectin for the functional upstream domain of protein F1 of Streptococcus pyogenes. J Biol Chem. 2010;285: 41087–41099. pmid:20947497
  14. 14. Edwards AM, Potts JR, Josefsson E, Massey RC. Staphylococcus aureus host cell invasion and virulence in sepsis is facilitated by the multiple repeats within FnBPA. PLoS Pathog. 2010;6: e1000964. pmid:20585570
  15. 15. Edwards AM, Potter U, Meenan NA, Potts JR, Massey RC. Staphylococcus aureus keratinocyte invasion is dependent upon multiple high-affinity fibronectin-binding repeats within FnBPA. PLoS One. 2011;6: e18899. pmid:21526122
  16. 16. Massey RC, Kantzanou MN, Fowler T, Day NP, Schofield K, Wann ER, et al. Fibronectin-binding protein A of Staphylococcus aureus has multiple, substituting, binding regions that mediate adherence to fibronectin and invasion of endothelial cells. Cell Microbiol. 2001;3: 839–851. 157 [pii]. pmid:11736995
  17. 17. Ensenberger MG, Annis DS, Mosher DF. Actions of the functional upstream domain of protein F1 of Streptococcus pyogenes on the conformation of fibronectin. Biophys Chem. 2004;112: 201–207. S0301-4622(04)00188-7 [pii]. pmid:15572249
  18. 18. Nelea V, Nakano Y, Kaartinen MT. Size distribution and molecular associations of plasma fibronectin and fibronectin crosslinked by transglutaminase 2. Protein J. 2008;27: 223–233. pmid:18330684
  19. 19. Vakonakis I, Staunton D, Ellis IR, Sarkies P, Flanagan A, Schor AM, et al. Motogenic sites in human fibronectin are masked by long range interactions. J Biol Chem. 2009;284: 15668–15675. pmid:19366708
  20. 20. Johnson KJ, Sage H, Briscoe G, Erickson HP. The compact conformation of fibronectin is determined by intramolecular ionic interactions. J Biol Chem. 1999;274: 15473–15479. pmid:10336438
  21. 21. Bultmann H, Santas AJ, Peters DM. Fibronectin fibrillogenesis involves the heparin II binding domain of fibronectin. J Biol Chem. 1998;273: 2601–2609. pmid:9446562
  22. 22. Homandberg GA, Erickson JW. Model of fibronectin tertiary structure based on studies of interactions between fragments. Biochemistry. 1986;25: 6917–6925. pmid:3801402
  23. 23. Raibaud S, Schwarz-Linek U, Kim JH, Jenkins HT, Baines ER, Gurusiddappa S, et al. Borrelia burgdorferi binds fibronectin through a tandem beta-zipper, a common mechanism of fibronectin binding in staphylococci, streptococci, and spirochetes. J Biol Chem. 2005;280: 18803–18809. M501731200 [pii]. pmid:15737988
  24. 24. Norris NC, Bingham RJ, Harris G, Speakman A, Jones RP, Leech A, et al. Structural and functional analysis of the tandem beta-zipper interaction of a Streptococcal protein with human fibronectin. J Biol Chem. 2011;286: 38311–38320. pmid:21840989
  25. 25. Schwarz-Linek U, Werner JM, Pickford AR, Gurusiddappa S, Kim JH, Pilka ES, et al. Pathogenic bacteria attach to human fibronectin through a tandem beta-zipper. Nature. 2003;423: 177–181. pmid:12736686
  26. 26. Ugarova TP, Zamarron C, Veklich Y, Bowditch RD, Ginsberg MH, Weisel JW, et al. Conformational transitions in the cell binding domain of fibronectin. Biochemistry. 1995;34: 4457–4466. pmid:7535564
  27. 27. Ma W, Ma H, Fogerty FJ, Mosher DF. Bivalent ligation of the collagen-binding modules of fibronectin by SFS, a non-anchored bacterial protein of Streptococcus equi. J Biol Chem. 2015;290: 4866–4876. pmid:25525266
  28. 28. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell. 2002;110: 673–687. S0092867402009716 [pii]. pmid:12297042
  29. 29. Hauck CR, Borisova M, Muenzner P. Exploitation of integrin function by pathogenic microbes. Curr Opin Cell Biol. 2012;24: 637–644. pmid:22884865
  30. 30. Coburn J, Chege W, Magoun L, Bodary SC, Leong JM. Characterization of a candidate Borrelia burgdorferi beta3-chain integrin ligand identified using a phage display library. Mol Microbiol. 1999;34: 926–940. mmi1654 [pii]. pmid:10594819
  31. 31. Kwok T, Zabler D, Urman S, Rohde M, Hartig R, Wessler S, et al. Helicobacter exploits integrin for type IV secretion and kinase activation. Nature. 2007;449: 862–866. nature06187 [pii]. pmid:17943123
  32. 32. Isberg RR, Leong JM. Multiple beta 1 chain integrins are receptors for invasin, a protein that promotes bacterial penetration into mammalian cells. Cell. 1990;60: 861–871. 0092-8674(90)90099-Z [pii]. pmid:2311122
  33. 33. Hauck CR, Ohlsen K. Sticky connections: extracellular matrix protein recognition and integrin-mediated cellular invasion by Staphylococcus aureus. Curr Opin Microbiol. 2006;9: 5–11. S1369-5274(05)00196-7 [pii]. pmid:16406780
  34. 34. Hoffmann C, Ohlsen K, Hauck CR. Integrin-mediated uptake of fibronectin-binding bacteria. Eur J Cell Biol. 2011;90: 891–896. pmid:21561684
  35. 35. Foster TJ, Hook M. Surface protein adhesins of Staphylococcus aureus. Trends Microbiol. 1998;6: 484–488. S0966-842X(98)01400-0 [pii]. pmid:10036727
  36. 36. Sinha B, Francois PP, Nusse O, Foti M, Hartford OM, Vaudaux P, et al. Fibronectin-binding protein acts as Staphylococcus aureus invasin via fibronectin bridging to integrin alpha5beta1. Cell Microbiol. 1999;1: 101–117. cmi11 [pii]. pmid:11207545
  37. 37. Dziewanowska K, Patti JM, Deobald CF, Bayles KW, Trumble WR, Bohach GA. Fibronectin binding protein and host cell tyrosine kinase are required for internalization of Staphylococcus aureus by epithelial cells. Infect Immun. 1999;67: 4673–4678. pmid:10456915
  38. 38. Lammers A, Nuijten PJ, Smith HE. The fibronectin binding proteins of Staphylococcus aureus are required for adhesion to and invasion of bovine mammary gland cells. FEMS Microbiol Lett. 1999;180: 103–109. S0378-1097(99)00470-X [pii]. pmid:10547450
  39. 39. Fowler T, Wann ER, Joh D, Johansson S, Foster TJ, Hook M. Cellular invasion by Staphylococcus aureus involves a fibronectin bridge between the bacterial fibronectin-binding MSCRAMMs and host cell beta1 integrins. Eur J Cell Biol. 2000;79: 672–679. pmid:11089915
  40. 40. Schroder A, Schroder B, Roppenser B, Linder S, Sinha B, Fassler R, et al. Staphylococcus aureus fibronectin binding protein-A induces motile attachment sites and complex actin remodeling in living endothelial cells. Mol Biol Cell. 2006;17: 5198–5210. E06-05-0463 [pii]. pmid:17021255
  41. 41. Sinha B, Francois P, Que YA, Hussain M, Heilmann C, Moreillon P, et al. Heterologously expressed Staphylococcus aureus fibronectin-binding proteins are sufficient for invasion of host cells. Infect Immun. 2000;68: 6871–6878. pmid:11083807
  42. 42. Sinha B, Herrmann M. Mechanism and consequences of invasion of endothelial cells by Staphylococcus aureus. Thromb Haemost. 2005;94: 266–277. 05080266 [pii]. pmid:16113815
  43. 43. Agerer F, Lux S, Michel A, Rohde M, Ohlsen K, Hauck CR. Cellular invasion by Staphylococcus aureus reveals a functional link between focal adhesion kinase and cortactin in integrin-mediated internalisation. J Cell Sci. 2005;118: 2189–2200. jcs.02328 [pii]. pmid:15855238
  44. 44. Shinji H, Yosizawa Y, Tajima A, Iwase T, Sugimoto S, Seki K, et al. Role of fibronectin-binding proteins A and B in in vitro cellular infections and in vivo septic infections by Staphylococcus aureus. Infect Immun. 2011;79: 2215–2223. pmid:21422173
  45. 45. Kerdudou S, Laschke MW, Sinha B, Preissner KT, Menger MD, Herrmann M. Fibronectin binding proteins contribute to the adherence of Staphylococcus aureus to intact endothelium in vivo. Thromb Haemost. 2006;96: 183–189. 06080183 [pii]. pmid:16894462
  46. 46. Meenan NA, Visai L, Valtulina V, Schwarz-Linek U, Norris NC, Gurusiddappa S, et al. The tandem beta-zipper model defines high affinity fibronectin-binding repeats within Staphylococcus aureus FnBPA. J Biol Chem. 2007;282: 25893–25902. M703063200 [pii]. pmid:17606607
  47. 47. Bosse MJ, Gruber HE, Ramp WK. Internalization of bacteria by osteoblasts in a patient with recurrent, long-term osteomyelitis. A case report. J Bone Joint Surg Am. 2005;87: 1343–1347. 87/6/1343 [pii]. pmid:15930546
  48. 48. Clement S, Vaudaux P, Francois P, Schrenzel J, Huggler E, Kampf S, et al. Evidence of an intracellular reservoir in the nasal mucosa of patients with recurrent Staphylococcus aureus rhinosinusitis. J Infect Dis. 2005;192: 1023–1028. JID34531 [pii]. pmid:16107955
  49. 49. Jarry TM, Memmi G, Cheung AL. The expression of alpha-haemolysin is required for Staphylococcus aureus phagosomal escape after internalization in CFT-1 cells. Cell Microbiol. 2008;10: 1801–1814. pmid:18466345
  50. 50. Que YA, Haefliger JA, Piroth L, Francois P, Widmer E, Entenza JM, et al. Fibrinogen and fibronectin binding cooperate for valve infection and invasion in Staphylococcus aureus experimental endocarditis. J Exp Med. 2005;201: 1627–1635. jem.20050125 [pii]. pmid:15897276
  51. 51. Lehar SM, Pillow T, Xu M, Staben L, Kajihara KK, Vandlen R, et al. Novel antibody-antibiotic conjugate eliminates intracellular S. aureus. Nature. 2015;527: 323–328. pmid:26536114
  52. 52. Speziale P, Visai L, Rindi S, Di Poto A. Purification of human plasma fibronectin using immobilized gelatin and Arg affinity chromatography. Nat Protoc. 2008;3: 525–533. pmid:18323821
  53. 53. Coe AP, Askari JA, Kline AD, Robinson MK, Kirby H, Stephens PE, et al. Generation of a minimal alpha5beta1 integrin-Fc fragment. J Biol Chem. 2001;276: 35854–35866. pmid:11389148
  54. 54. Zardi L, Carnemolla B, Balza E, Borsi L, Castellani P, Rocco M, et al. Elution of fibronectin proteolytic fragments from a hydroxyapatite chromatography column. A simple procedure for the purification of fibronectin domains. Eur J Biochem. 1985;146: 571–579. pmid:2982601
  55. 55. Prabhakaran S, Liang X, Skare JT, Potts JR, Hook M. A novel fibronectin binding motif in MSCRAMMs targets F3 modules. PLoS One. 2009;4: e5412. pmid:19404402
  56. 56. Baneyx G, Baugh L, Vogel V. Coexisting conformations of fibronectin in cell culture imaged using fluorescence resonance energy transfer. Proc Natl Acad Sci U S A. 2001;98: 14464–14468. pmid:11717404
  57. 57. Williams EC, Janmey PA, Ferry JD, Mosher DF. Conformational states of fibronectin. Effects of pH, ionic strength, and collagen binding. J Biol Chem. 1982;257: 14973–14978. pmid:7174679
  58. 58. Rocco M, Carson M, Hantgan R, McDonagh J, Hermans J. Dependence of the shape of the plasma fibronectin molecule on solvent composition. Ionic strength and glycerol content. J Biol Chem. 1983;258: 14545–14549. pmid:6643501
  59. 59. Lai CS, Wolff CE, Novello D, Griffone L, Cuniberti C, Molina F, et al. Solution structure of human plasma fibronectin under different solvent conditions. Fluorescence energy transfer, circular dichroism and light-scattering studies. J Mol Biol. 1993;230: 625–640. S0022-2836(83)71174-5 [pii]. pmid:8464068
  60. 60. Pelta J, Berry H, Fadda GC, Pauthe E, Lairez D. Statistical conformation of human plasma fibronectin. Biochemistry. 2000;39: 5146–5154. bi992770x [pii]. pmid:10819982
  61. 61. Pauthe E, Pelta J, Patel S, Lairez D, Goubard F. Temperature-induced beta-aggregation of fibronectin in aqueous solution. Biochim Biophys Acta. 2002;1597: 12–21. S0167483802002716 [pii]. pmid:12009397
  62. 62. Lai CS, Tooney NM, Ankel EG. Structure and flexibility of plasma fibronectin in solution: electron spin resonance spin-label, circular dichroism, and sedimentation studies. Biochemistry. 1984;23: 6393–6397. pmid:6099139
  63. 63. Patel S, Chaffotte AF, Goubard F, Pauthe E. Urea-induced sequential unfolding of fibronectin: a fluorescence spectroscopy and circular dichroism study. Biochemistry. 2004;43: 1724–1735. pmid:14769050
  64. 64. Benecky MJ, Kolvenbach CG, Wine RW, DiOrio JP, Mosesson MW. Human plasma fibronectin structure probed by steady-state fluorescence polarization: evidence for a rigid oblate structure. Biochemistry. 1990;29: 3082–3091. pmid:2337580
  65. 65. Sjoberg B, Pap S, Osterlund E, Osterlund K, Vuento M, Kjems J. Solution structure of human plasma fibronectin using small-angle X-ray and neutron scattering at physiological pH and ionic strength. Arch Biochem Biophys. 1987;255: 347–353. pmid:3592678
  66. 66. Lairez D, Pauthe E, Pelta J. Refolding of a high molecular weight protein: salt effect on collapse. Biophys J. 2003;84: 3904–3916. S0006-3495(03)75118-2 [pii]. pmid:12770896
  67. 67. Alexander SS Jr, Colonna G, Edelhoch H. The structure and stability of human plasma cold-insoluble globulin. J Biol Chem. 1979;254: 1501–1505. pmid:762148
  68. 68. Osterlund E, Eronen I, Osterlund K, Vuento M. Secondary structure of human plasma fibronectin: conformational change induced by calf alveolar heparan sulfates. Biochemistry. 1985;24: 2661–2667. pmid:3161537
  69. 69. Khan MY, Medow MS, Newman SA. Unfolding transitions of fibronectin and its domains. Stabilization and structural alteration of the N-terminal domain by heparin. Biochem J. 1990;270: 33–38. pmid:2396990
  70. 70. Brumfeld V, Werber MM. Studies on fibronectin and its domains. II. Secondary structure and spatial configuration of fibronectin and of its domains. Arch Biochem Biophys. 1993;302: 134–143. S0003-9861(83)71191-4 [pii]. pmid:8470891
  71. 71. Wolff CE, Lai CS. Inter-sulfhydryl distances in plasma fibronectin determined by fluorescence energy transfer: effect of environmental factors. Biochemistry. 1990;29: 3354–3361. pmid:2334697
  72. 72. Takagi J, Strokovich K, Springer TA, Walz T. Structure of integrin alpha5beta1 in complex with fibronectin. EMBO J. 2003;22: 4607–4615. pmid:12970173
  73. 73. Campbell ID, Humphries MJ. Integrin structure, activation, and interactions. Cold Spring Harb Perspect Biol. 2011;3:
  74. 74. Akiyama SK, Hasegawa E, Hasegawa T, Yamada KM. The interaction of fibronectin fragments with fibroblastic cells. J Biol Chem. 1985;260: 13256–13260. pmid:2932436
  75. 75. Fraunholz M, Sinha B. Intracellular Staphylococcus aureus: live-in and let die. Front Cell Infect Microbiol. 2012;2: 43. pmid:22919634
  76. 76. Chen H, Ricklin D, Hammel M, Garcia BL, McWhorter WJ, Sfyroera G, et al. Allosteric inhibition of complement function by a staphylococcal immune evasion protein. Proc Natl Acad Sci U S A. 2010;107: 17621–17626. pmid:20876141
  77. 77. Shattil SJ, Kim C, Ginsberg MH. The final steps of integrin activation: the end game. Nat Rev Mol Cell Biol. 2010;11: 288–300. pmid:20308986
  78. 78. Vogel V. Mechanotransduction involving multimodular proteins: converting force into biochemical signals. Annu Rev Biophys Biomol Struct. 2006;35: 459–488. pmid:16689645
  79. 79. Lower SK, Lamlertthon S, Casillas-Ituarte NN, Lins RD, Yongsunthon R, Taylor ES, et al. Polymorphisms in fibronectin binding protein A of Staphylococcus aureus are associated with infection of cardiovascular devices. Proc Natl Acad Sci U S A. 2011;108: 18372–18377. pmid:22025727
  80. 80. Xiong YQ, Sharma-Kuinkel BK, Casillas-Ituarte NN, Fowler VG Jr, Rude T, DiBartola AC, et al. Endovascular infections caused by methicillin-resistant Staphylococcus aureus are linked to clonal complex-specific alterations in binding and invasion domains of fibronectin-binding protein A as well as the occurrence of fnbB. Infect Immun. 2015;83: 4772–4780. pmid:26416903