Mycobacterium tuberculosis Binding to Human Surfactant Proteins A and D, Fibronectin, and Small Airway Epithelial Cells under Shear Conditio
http://www.100md.com
感染与免疫杂志 2006年第6期
Center for Genomic Sciences, Allegheny-Singer Research Institute, Pittsburgh, Pennsylvania
Veterinary Molecular Biology, Montana State University, Bozeman, Montana
LigoCyte Pharmaceuticals, Inc., Bozeman, Montana
Division of Infectious Diseases, Department of Internal Medicine, Center for Microbial Interface Biology, The Ohio State University, Columbus, Ohio
Department of Mathematical Sciences, Montana State University, Bozeman, Montana
Department of Pathology and Immunology, Washington University, St. Louis, Missouri
ABSTRACT
A crucial step in infection is the initial attachment of a pathogen to host cells or tissue. Mycobacterium tuberculosis has evolved multiple strategies for establishing an infection within the host. The pulmonary microenvironment contains a complex milieu of pattern recognition molecules of the innate immune system that play a role in the primary response to inhaled pathogens. Encounters of M. tuberculosis with these recognition molecules likely influence the outcome of the bacillus-host interaction. Here we use a novel fluid shear assay to investigate the binding of M. tuberculosis to innate immune molecules that are produced by pulmonary epithelial cells and are thought to play a role in the lung innate immune response. Virulent and attenuated M. tuberculosis strains bound best to immobilized human fibronectin (FN) and surfactant protein A (SP-A) under this condition. Binding under fluid shear conditions was more consistent and significant compared to binding under static conditions. Soluble FN significantly increased the adherence of both virulent and attenuated M. tuberculosis strains to human primary small airway epithelial cells (SAEC) under fluid shear conditions. In contrast, SP-A and SP-D effects on bacterial adherence to SAEC differed between the two strains. The use of a fluid shear model to simulate physiological conditions within the lung and select for high-affinity binding interactions should prove useful for studies that investigate interactions between M. tuberculosis and host innate immune determinants.
INTRODUCTION
Tuberculosis (TB) remains a significant therapeutic challenge for the early 21st century. The global incidence and prevalence of infection and multidrug resistance and the impact of AIDS and poverty diminish the likelihood that TB will be controlled without a more effective vaccine (9, 15). Therefore, innovative approaches to discovering improved diagnostic methods, therapies, and vaccine candidates are very much needed.
Mycobacterium tuberculosis is a facultative intracellular pathogen of macrophages. Following inhalation of the tubercle bacilli in droplet nuclei into the lower respiratory tract, M. tuberculosis is phagocytosed by alveolar macrophages. These bacilli have acquired the ability to survive and multiply within these cells, and it is widely believed that this ability is critical for the establishment of infection. It has also been shown that M. tuberculosis can attach to and be taken up by alveolar epithelial cells, although the significance of this is not well understood (7, 8, 10, 21, 30).
Attachment of a pathogen to pattern recognition molecules is of central importance in the host innate immune response. Bacterial pathogens typically gain entry to host tissues using pathogen-associated molecular patterns that bind host broad-spectrum cognate ligands (16). Interactions between M. tuberculosis cell wall determinants and host molecules within the alveolar microenvironment are expected to play an important role in determining whether this pathogen successfully invades the host or is cleared by first-line host defenses. The binding interactions between M. tuberculosis and host recognition molecules, therefore, represent key opportunities to identify potential candidates for improved antimycobacterial therapy, since interrupting binding events that lead to tubercle bacilli invasion or facilitating binding events that lead to clearance would functionally disrupt bacilli entry into its host cell niche.
Surfactant protein A (SP-A), surfactant protein D (SP-D), and fibronectin (FN) are produced by pulmonary epithelial cells and play a role in the innate immune response in the lung. The collectins, SP-A and SP-D, are oligomeric carbohydrate binding proteins that regulate opsonization, phagocytosis, and agglutination of microorganisms within the respiratory tract. Their localization in the alveoli enables them to act as "first responders" to inhaled pathogens. SP-A acts both as an opsonin and regulator of cell receptor activity in enhancing phagocytosis by macrophages (4, 12, 49). SP-A-deficient mice are more susceptible to infection with respiratory pathogens, and the instillation of SP-A in the lungs of these mice enhances phagocytosis and clearance of pathogenic bacteria (17, 28). In contrast, SP-D binding to respiratory pathogens aggregates microorganisms and alters their interactions with host cells (12, 17). Previous work has shown that SP-D agglutination of virulent M. tuberculosis inhibits phagocytosis of the bacilli by alveolar macrophages (18, 19). Finally, FN is a glycoprotein that plays a role in the host response to tissue injury and opsonization (33). FN has multiple binding sites for collagen, heparin, and fibrin, and it has been shown to enhance the attachment of several pathogens, including mycobacteria, to host cells (32, 34, 46).
Binding interactions between pathogens and host molecules or cells are traditionally studied under static conditions. During inhalation, however, respiratory pathogens undergo shear forces as they travel to the terminal airways (29, 53). Once in the alveoli, expansion and contraction of the alveolar sac subject the cells lining the alveoli to deformations that change with lung tidal volume, i.e., bulging of the epithelium at low volumes and surface tensions (contraction) and stretching of the epithelium at higher volumes (expansion) (3). Therefore, the alveolar epithelium is subjected to micromechanical forces and shear from airflow—dynamic effects that are not well modeled by microorganisms and host cells in a static culture system.
The aim of this study was to evaluate binding of M. tuberculosis to purified human molecules using shear conditions to better simulate the dynamic lung microenvironment. To accomplish this aim, we modified a technique developed for investigating leukocyte interactions with other cells to assay for M. tuberculosis binding to putative host determinants under shear conditions. This method is sensitive and better simulates in vivo conditions by examining M. tuberculosis-host interactions when physiological shear is present. Importantly, shear conditions may reduce or abrogate nonspecific binding interactions that take place under static conditions and thus select for high-affinity binding interactions.
Using both a flow system and static binding assays, we investigated the binding patterns of both a virulent and an attenuated strain of M. tuberculosis to human SP-A, SP-D, and FN. Binding patterns of tubercle bacilli to SP-A and FN were significantly different compared with a nonspecific protein, bovine serum albumin (BSA), under shear conditions, suggesting high-affinity interactions. When M. tuberculosis was preincubated with soluble SP-A, SP-D, or FN and tested in a shear assay using primary human small airway epithelial cells (SAEC), these proteins affected the number of bacilli that adhered to these cells. FN significantly enhanced adherence to SAEC for both virulent and attenuated M. tuberculosis under shear conditions, whereas SP-A and SP-D showed strain-dependent adherence patterns. These results indicate that the shear model is useful in identifying robust binding interactions between M. tuberculosis and host molecules or cells. These interactions will enable further investigation of potential therapeutic and vaccine targets.
MATERIALS AND METHODS
Media and reagents. Middlebrook 7H9 broth and 7H11 agar were prepared with oleic acid albumin dextrose catalase supplement (Difco media; Becton Dickinson, Sparks, MD) and supplemented with 0.1% casein enzymatic hydrosylate and glycerol (Sigma) as described previously (11). Hanks balanced salt solution (HBSS) supplemented with HEPES, Ca2+, and Mg2+, and 2% fetal bovine serum (FBS) and RPMI (Gibco cell culture; Invitrogen Corporation, Carlsbad, CA) supplemented with 10% FBS (HyClone, Logan, UT), HEPES, penicillin/streptomycin, nonessential amino acids, sodium pyruvate, L-glutamine, and vitamins (Sigma) were used to rinse flow cells and in flow cell experiments. Human FN and BSA were purchased from Sigma.
Bacteria. M. tuberculosis H37Ra (ATCC 25177), H37Rv (ATCC 27294), and Erdman (ATCC 35801) were grown and harvested on Middlebrook 7H11 agar and used between 9 to 11 days as described previously (44). Immediately prior to assays, bacteria were scraped into polypropylene tubes containing 1 ml 7H9 broth and pulse vortexed five to six times (1 s per pulse) with two 3-mm-diameter glass beads. Clumped bacteria were allowed to settle, and the resulting suspension was counted using a Petroff-Hausser chamber, yielding >90% of a single cell bacterial suspension.
Surfactant protein purification. SP-A was purified as previously described (13). Following an approved Institutional Review Board protocol, surfactant lipid from human alveolar proteinosis bronchoalveolar lavage (BAL) fluid was isolated by centrifugation at 20,000 x g and washed repeatedly, and lipid-associated SP-A was eluted using 2 mM EDTA at 4°C. SP-A was purified from the EDTA elution supernatant by separation over a mannose-Sepharose affinity chromatography column, and EDTA was removed by dialysis against 10 mM HEPES plus 25 mM NaCl. SP-D was purified from human alveolar proteinosis bronchoalveolar lavage fluid as previously described (51), with minor modifications. Briefly, following removal of SP-A-containing surfactant lipid from BAL fluid by centrifugation, SP-D was purified from BAL fluid by loading BAL supernatant to a maltose-Sepharose affinity chromatography column. The column was then washed to remove unbound proteins, and SP-D was specifically eluted using MnCl2 and then dialyzed against 10 mM HEPES plus 150 mM NaCl to remove the MnCl2. Recombinant human SP-D dodecamers were also used in shear and static binding assays. There are no known functional differences between comparably oligomerized recombinant or native proteins. Recombinant SP-D was isolated by sequential maltosyl-agarose and gel filtration chromatography (24). Purified proteins were examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting using anti-SP-A and anti-SP-D specific antibodies.
Protein coating of flow cells and coverslips. To coat flow cells with FN, sterile glass flow cells (Friedrich and Dimmock, Inc., Millville, NJ) were filled with 100 μl of protein and incubated for 2 h at 37°C. The optimal concentration for attachment was determined empirically. Flow cells were then rinsed three times with RPMI. For covalent attachment of SP-A and SP-D, flow cells were filled with 100 μl of 0.1 mg/ml of poly-L-lysine diluted in phosphate-buffered saline (PBS) and allowed to incubate at room temperature (RT) for 60 min. Flow cells were washed three times with PBS and filled with 100 μl of 2.5% glutaraldehyde for 15 min and washed five times with sterile H2O and two times with PBS. Flow cells were incubated with 10 μg/ml SP-A, SP-D, or BSA at RT for 30 min, washed twice with PBS, and filled with 0.2 M glycine and incubated at RT overnight to quench untreated aldehyde groups. The flow cells were then washed twice with PBS and twice with RPMI. For static binding assays, glass coverslips in wells of a 24-well tissue culture plate were coated with proteins exactly as described above (44).
Flow cell and static binding assays. For flow cell assays, bacteria were stained with Syto 9 nucleic acid dye (Molecular Probes, Eugene, OR) for 30 min prior to the flow cell assay. Flow cells and loop tubing were filled with HBSS and placed in an Ismatec low-speed precision pump (Cole-Parmer, Vernon Hills, IL) to remove bubbles. The loops were clamped and a glass flow cell, coated with either the human protein of interest or BSA, was inserted into each loop tubing and secured to prevent leakage. A portable stage heater (Cole-Parmer) equilibrated to 37°C was placed beneath the flow cell, and flow was initiated at a flow rate of 0.75 dynes/cm2 (Reynold's number [Re] = 19) (calculated for fluid at 37°C). Stained bacilli, 1 x 108, were injected in each loop and allowed to circulate for 1 h at 37°C. Bacilli were then fixed in 3% glutaraldehyde (final concentration) for 10 min under flow to optimize mixing. Following fixation, flow was resumed and 30 volumes of HBSS were passed through the flow cell to remove nonadherent bacilli. For static binding assays, glass coverslips in tissue culture wells were incubated with 108 bacilli for 1 h at 37°C, rinsed to remove nonadherent bacteria, fixed, stained with auramine-rhodamine, and mounted on glass slides prior to enumeration. In control experiments, M. tuberculosis mannosyl lipoarabinomannan (ManLAM)-coated beads (108) (45) were incubated with coverslips under the same conditions, washed, fixed, and enumerated.
Measuring bound M. tuberculosis. For flow assays, adherent bacteria were enumerated under a Leica DM-RXE microscope (Leica Microsystems, Exton, PA) (using the PL APO 63x/1.20 W lens or a 100x/1.4 oil lens) and counting two separate faces of the flow cell. By examining the top face of the flow cell, only bacteria that were adherent to the protein-coated surface, not bacteria that had detached and settled due to gravity, were enumerated. A minimum of 30 fields were counted per flow cell, and duplicate flow cells were run in parallel for each molecule. BSA control flow cells were run for each experiment to control for assay-to-assay variability. Average adherent bacilli per field were determined, and the area per microscopic field was calculated to yield a normalized number of bound bacilli per cm2. For static binding assays, a minimum of 20 high-power fields were counted per coverslip for either bacilli or ManLAM-coated beads and normalized as above to yield the number of bound bacilli or beads/cm2. Experiments were done in triplicate for each protein and particle tested (n = 6).
Cells and tissue culture. Cryopreserved SAEC were obtained from Clonetics (San Diego, CA) and subcultured in small airway basal medium (SABM; Clonetics), aliquoted, and frozen as stock suspensions containing 5 x 105 cells per ml. Frozen stock aliquots were removed from liquid N2 and cultured in Ham's F-12 media (Invitrogen, Calsbad, CA) supplemented with 10% FBS, HEPES, L-glutamine, nonessential amino acids, sodium pyruvate, and penicillin/streptomycin and incubated at 37°C and 5% CO2 for 48 h. Cells were harvested and centrifuged, and 5 x 106 SAEC were seeded onto MatTek glass-bottom culture dishes (MatTek Corp., Ashland, MA) and allowed to grow to confluence for 24 to 48 h.
Adherence to SAEC. SAEC monolayers were incubated with 5 x107 M. tuberculosis H37Rv or H37Ra cells (multiplicity of infection = 10:1) at 37°C for 1 h under gentle shear conditions (35 rpm) on a Lab-Line lab rotator (model no. 1304; Lab-Line, Melrose Park, IL). Cells were rinsed three times with tissue culture medium to remove nonadherent bacilli and bacteria were stained using a TB fluorescent stain kit M (Becton Dickenson, Sparks, MD) for 15 min and then rinsed three times to remove the auramine M. Following fixation with 3% glutaraldehyde for 10 min, the monolayers were rinsed three times to remove glutaraldehyde, counterstained briefly with potassium permanganate to enhance contrast, and rinsed twice more.
Confocal microscopy. Flow cell images were captured using a Leica DM-RXE microscope and TCS NT confocal scanning laser system (Leica Microsystems, Exton, PA) using the 488-nm laser line to visualize Syto 9-stained M. tuberculosis cells that were adherent to the flow cell after 1 h of shear. Tissue culture plates were analyzed with a Zeiss LSM 510 meta confocal microscope (Carl Zeiss Microimaging Inc., Thornwood, NY), using the 488-nm laser line to visualize fluorescent M. tuberculosis cells stained with auramine M and the transmission mode to visualize SAEC counterstained with potassium permanganate.
SEM. Monolayers were processed for scanning electron microscopy (SEM) by dehydration in a graded ethanol series following 3% glutaraldehyde fixation. Samples were sputter coated with 15 to 20 nm gold palladium using a Hummer VII sputtering system (Anatech Ltd., Alexandria, VA). Samples were visualized at an accelerator voltage of 12 kV using a JEOL JSM-6100 scanning electron microscope and digitized TIFF images were collected.
Statistical analysis. The data were analyzed using the R statistical environment (38) and the add-on package nlme (nonlinear and linear mixed effects) of Pinheiro and Bates (36). The analysis used a mixed effects model with random effects for the experiment and fixed effects for the strains and compounds (human molecules); the fixed model was a two-way analysis of variance model. Normality of residuals was examined using quantile-quantile plots, and transformations of the response were used to give a good fit to normality. The observation of many fields for each flow cell created pseudoreplications which were investigated for possible serial correlation (5). Several types of correlation were considered, and the best fitting, an autocorrelation of order 1, was used, thus avoiding the common assumption of independent observations by modeling the correlation structure.
RESULTS
The flow cell system. Shear caused by fluid flow is a function of the velocity, density and viscosity of the fluid (air) or air-liquid interface, and the diameter of the tube (blood vessel or airway). Laminar flow is predominant in the small airways (terminal bronchioles and alveoli) (29, 53). Physiologically relevant shear forces are measured in dynes/cm2 (shear forces occurring over square centimeters). Shear forces in the lung have been used primarily for purposes of calculating gas exchange where the fluid is air (29, 42, 53). Due to the complexity of the lung microenvironment with its three-dimensional branching structure, however, it is extremely difficult to determine these numbers, and so they are based largely on models using basic concepts of hydrodynamics or fluid transport (42).
To approximate physiological conditions present in the lower airways, purified proteins were adsorbed or covalently linked to the flow cell and M. tuberculosis was circulated under laminar flow conditions (Re = 19) determined using the following equations: Re = u(ave)(Dh)/ and u(ave) = Q/CSA, where u(ave) is the average flow velocity, Q is the flow rate, CSA is the cross-sectional area of the flow cell (1-mm2 flow chamber), Dh is the hydraulic diameter, and is the kinematic viscosity (taken for that of water at 37°C) (50). These conditions yielded good mixing and resulted in an operational shear equivalent to less than 1 dyne/cm2, simulating a very low laminar shear that would be expected in the terminal bronchioles and air sacs. The sterile flow system was completely disposable and precluded autoclaving and the potential buildup of bacterial toxins. Moreover, the modular system fit easily inside a biosafety cabinet in the biosafety level 3 facility and generated relatively small volumes of bacterial waste (flow cell volume, 3 ml). A similar system has been successfully used to monitor real-time binding and binding inhibition between larger cells such as epithelial cells and a fungal pathogen, Candida albicans (23).
M. tuberculosis binding to immobilized proteins. M. tuberculosis binding to human proteins was first tested under dynamic conditions by enumerating binding of the live attenuated M. tuberculosis strain H37Ra to immobilized proteins under shear conditions. FN was tested at two different flow rates (1.5 dyne/cm2 and 0.75 dyne/cm2) with no statistical difference in mycobacterial binding between rates (data not shown). Since we wished to simulate low shear conditions predicted in the alveolar space, we used the lower rate for all subsequent experiments. FN was initially tested at 10 and 100 μg/ml, and there was no significant difference in M. tuberculosis binding (compared to BSA control; data not shown), indicating that binding was saturable at 10 μg/ml FN (data not shown), consistent with other reports (34).
We next compared the binding of two strains of M. tuberculosis (a virulent strain, H37Rv, and an attenuated strain, H37Ra) to FN, SP-A, and SP-D under shear conditions. To adjust for possible changes in binding across experiments, a binding index for each molecule was calculated by dividing the average M. tuberculosis bacilli bound to the host molecule by the average number of bacilli bound to BSA (Table 1). Under shear conditions, virulent and attenuated M. tuberculosis bound to human FN approximately six- to sevenfold more than to the BSA control. M. tuberculosis binding to SP-A was approximately four to fivefold greater than to BSA. Binding of the virulent strain H37Rv to SP-D, however, was reduced compared with the nonspecific protein, BSA, showing a binding index of <1 (Table 1).
To allow for the assumption of constant variance of residuals in the mixed effects model, and to compare different binding assays, the binding indices were transformed to a log10 scale. Both strains bound similarly to FN and SP-A, but the confidence intervals of the attenuated and virulent strains did not overlap in SP-D binding conditions (data not shown), suggesting a difference in the binding patterns of virulent and attenuated M. tuberculosis to SP-D. Simple two-tailed t tests performed on the log10 data indicated significant differences between binding to BSA and to the collectins and FN, except for the binding of H37Ra to SP-D compared to BSA under shear conditions (Fig. 1).
Because SP-D was previously shown to bind to and agglutinate the virulent M. tuberculosis Erdman strain (18), our result was unexpected. Therefore, we further investigated the binding of M. tuberculosis H37Rv, H37Ra, and Erdman strains to covalently bound SP-D under static conditions. We also studied the binding of M. tuberculosis ManLAM-coated beads to covalently bound SP-D, since ManLAM also binds to SP-D (18). While the binding of the Erdman strain to both FN and SP-A was similar to that observed with the H37Ra and H37Rv strains under shear and static conditions, the binding to SP-D was variable and less than expected, indicating that the Erdman strain also did not bind well to SP-D covalently linked to the substratum. This suspicion was confirmed with binding assays using ManLAM beads, which failed to show specific binding to covalently linked SP-D (data not shown). Therefore we conclude that covalent attachment of SP-D to a platform disrupts the conformation of the SP-D carbohydrate recognition domains necessary for binding of M. tuberculosis.
Overall our data provide evidence that M. tuberculosis is able to bind rapidly (within 1 h) and significantly to immobilized human SP-A and FN under shear conditions. Binding under static conditions revealed similar trends, but there was a marked increase in variability. Binding under shear conditions reduced this variability and better enabled the identification of the specific receptor-ligand interactions.
M. tuberculosis adherence to primary human respiratory epithelial cells. To further simulate the dynamic physiological environment of the lung, we performed adherence assays of virulent and attenuated M. tuberculosis bacilli to human primary SAEC under shear conditions following preincubation with soluble human SP-A, SP-D, and FN. We chose to specifically examine tubercle bacilli adherence to SAEC since alveolar epithelial cells produce all of the molecules that showed significant binding interaction patterns in previous shear assays, and we wished to determine if soluble human collectin proteins or FN influence adherence of M. tuberculosis to primary human SAEC under shear.
Both M. tuberculosis H37Ra and H37Rv adhered to SAEC under rotational shear conditions at 37°C within 1 h (Fig. 2). These results are consistent with other reports that have shown that M. tuberculosis adheres to respiratory epithelial cells (7, 8, 30). When M. tuberculosis strains were preincubated with soluble human SP-A, SP-D, or FN and assessed for adherence to SAEC cell monolayers, FN significantly enhanced tubercle bacilli adherence to these cells for both virulent and attenuated M. tuberculosis strains, resulting in the greatest number of adherent bacilli to SAEC (Fig. 2) and the greatest difference compared with adherence to SAEC by untreated M. tuberculosis (P 0.0001) for both strains. Statistical analysis also showed that SP-D preincubation influenced attenuated M. tuberculosis H37Ra adherence to SAEC (P = 0.0258). The P value comparing SP-A-preincubated H37Ra to untreated M. tuberculosis was 0.0531. Interestingly, this pattern differed from the adherence pattern to SAEC by the virulent H37Rv strain, where adherence was unaffected by preincubation with the collectins compared with M. tuberculosis in the absence of host protein (SP-A, P = 0.8029; and SP-D, P = 0.9940). Adherence by M. tuberculosis in the presence of FN also differed significantly from bacilli preincubated with SP-A (P = 0.003) or SP-D (P = 0.002).
Microscopic evaluation of adherent bacilli to SAEC. Confocal microscopy and SEM showed that tubercle bacilli were attached to the epithelial monolayer (Fig. 3 and 4). Figure 3 shows M. tuberculosis bacilli adhered to SEAC after 1 h of incubation under shear conditions. Bacilli were observed to be adherent to the cell monolayer even when not preincubated with human proteins. Large clumps of bacilli were noted more frequently with SP-A, SP-D, and FN preincubation (Fig. 3). In some confocal micrographs, bacilli appeared to be within the epithelial cell. SEM analysis demonstrated M. tuberculosis in the process of being surrounded by epithelial cell cytoplasm. Fine filaments of the monolayer appeared to envelop the edges of single bacteria or aggregates of cells (Fig. 4). These results suggest that bacilli are capable of invading human primary SAEC and that host proteins can influence the number of M. tuberculosis adherent to these primary airway cells under conditions that simulate physiological shear.
DISCUSSION
M. tuberculosis encounters a dynamic pulmonary microenvironment upon inhalation into the lung. The airway environment includes epithelial cells and host innate immune molecules, and previous studies provide evidence that interactions between M. tuberculosis and these recognition molecules influence uptake of the bacilli (4, 17-19, 22). Binding assays that examine interactions between microbial pathogens and recognition molecules are typically carried out under static conditions. These studies are useful in determining whether specific binding occurs between a ligand and its receptor; however, they overlook mechanical forces that arise during shear over a surface.
Shear forces can affect the binding of a microorganism as well as the effectiveness of a putative binding inhibitor (31, 41, 48). Bacterial cells can respond to hydrodynamic forces that affect bacterial adherence to substrata such as inert surfaces and extracellular matrix (31, 37, 41, 48). Specifically, the adherence of Staphylococcus aureus mutants to fibrinogen and FN differed depending on whether bacteria were incubated under static versus flow conditions (48). Moreover, shear may also select for higher-affinity adhesion interactions. A recombinant collagen adhesin fragment that abrogated S. aureus binding to collagen under static conditions failed to inhibit binding under shear conditions, suggesting that putative inhibitor candidates should have the ability to bind to a ligand and resist detachment under conditions that may be present physiologically (31). Shear also influences cellular responses such as mechanotransduction, membrane receptor regulation, cytokine expression, and repair mechanisms in many types of mammalian cells (20, 25, 43, 52).
A wide range of shear forces occur in the lung. Laminar flow is thought to be present in the small airways (terminal bronchioles and alveoli), whereas transitional or turbulent flows occur in branching and large airways, respectively (29, 53). Therefore, the lung epithelial surface is subjected to dynamic effects that are not well modeled in a static culture system. Given this dynamic pulmonary environment, we reason that an infectious particle might not simply settle out due to gravitational forces but is rather subjected to shear forces during its travel in the terminal airways and alveoli. Thus, we postulate that shear in the lung as studied using aqueous flow provides a better simulation of the dynamic conditions that exist physiologically in intact airways. To our knowledge, M. tuberculosis adherence under shear conditions has not been previously investigated.
Our results show that both virulent H37Rv and attenuated H37Ra M. tuberculosis bound to purified human FN and SP-A under shear conditions and that binding was significantly different than that for the BSA control (Fig. 1). Binding assays performed under static conditions showed similar trends; however, due to marked variability in the results, significant differences could not be seen between the host proteins and BSA control. Importantly, the significant differences exhibited under shear conditions were observed even with the very low shear used to model the alveolar space. This may be due to increased chances of bacilli binding to cognate receptors (better mixing) and/or the ability of shear to select for higher-affinity adhesion interactions. Data from static assays, on the other hand, may reflect a combination of specific binding and nonspecific binding, with the contribution of nonspecific binding creating background binding that is more difficult to distinguish from specific binding interactions.
Binding of M. tuberculosis strains to immobilized FN under shear conditions was increased greater than sixfold compared to BSA (Table 1). Binding to FN has been well characterized for several bacterial pathogens, including mycobacteria. Specifically, FN attachment proteins and binding proteins have been described for several mycobacterial species, including M. tuberculosis (1, 2, 32, 39, 40, 47, 54). Our results agree with previous studies that showed approximately a fourfold increase in binding M. bovis BCG to FN compared to a serum albumin protein control under static conditions; other assays using H37Rv demonstrated a similar level of binding (39). FN binding proteins have been used to immunize mice against challenge with Streptococcus pyogenes (26) and may induce some protective immunity to mycobacterial infection (6, 33).
SP-A and SP-D are important components of innate immunity and influence the ability of pathogens to be taken up by host cells and/or cleared by host defense mechanisms. Gaynor et al. showed that SP-A enhanced phagocytosis of the virulent M. tuberculosis Erdman strain by human alveolar macrophages within 1 h postinfection (22). Further studies revealed that the interaction of SP-A with macrophages was due to increased expression of the macrophage mannose receptor (4). Pasula et al. showed that attenuated M. tuberculosis H37Ra bound directly to SP-A with high affinity (Kd of 1.9 x 10–9 M) and that SP-A mediated increased adherence of bacilli to mouse alveolar macrophages (35). Direct binding of M. bovis BCG and nonpathogenic Mycobacterium smegmatis to human SP-A was also observed, suggesting that different mycobacterial species are capable of binding to SP-A (49). Our results show that both attenuated and virulent M. tuberculosis bound to immobilized human SP-A under shear conditions at approximately four to five times the level of binding to BSA.
We also investigated the binding of M. tuberculosis to immobilized SP-D under shear conditions because SP-D has previously been shown to bind and agglutinate virulent M. tuberculosis Erdman and reduce uptake of bacilli into human macrophages (18, 19). SP-D bound chiefly to the terminal mannosyl oligosaccharides of M. tuberculosis LAM, and agglutination of bacilli was mediated by bridging via the carbohydrate binding domains (CRDs) and facilitated by the oligomeric structure of intact SP-D. Based on these data, we had expected that the virulent M. tuberculosis would exhibit increased binding to SP-D under shear conditions compared to that for BSA. Our studies revealed that the covalent attachment of SP-D to the surface disrupted binding by virulent M. tuberculosis and M. tuberculosis ManLAM-coated beads. Covalent attachment may interfere with M. tuberculosis binding by preventing bacilli agglutination (and thereby potentially affecting the avidity of binding) and/or affecting the availability of the CRDs for binding, either by creating a less suitable conformation or decreasing the availability of the CRDs for binding. Under the condition where M. tuberculosis was preincubated with soluble SP-D, SEM images showed that bacilli were particularly clumped (Fig. 3d), indicating that soluble SP-D agglutinated M. tuberculosis. These results indicate that the conformation of SP-D is extremely important for binding.
In adherence assays using primary human SAEC monolayers, FN significantly enhanced M. tuberculosis adherence for both the virulent and attenuated M. tuberculosis strains (Fig. 2). SP-A and SP-D also slightly enhanced adherence of the H37Ra strain to SAEC. The ability of soluble FN, SP-A, and SP-D to act as bridging molecules that increase the level of bacilli-bacilli homotypic binding interactions may be important. Clumping of M. tuberculosis would presumably result in more bacilli, with corresponding ligands, engaging cognate receptors on host cells. Shear may enhance aggregation, or alternatively, shear may increase detachment of clumps from low-affinity receptors and promote clearance from the airway. Thus, differences among strains in binding to the collectins are of interest.
Confocal and SEM images indicated that M. tuberculosis bacilli were taken up by SAEC regardless of whether they were preincubated with soluble human proteins. However, our results indicate that FN enhances the M. tuberculosis-SAEC interaction. Other reports have shown that FN specifically enhanced mycobacterial adherence to both nonphagocytic and phagocytic cells. FN served as an opsonin to facilitate the adhesion and invasion of Mycobacterium leprae into both an epithelial and Schwann cell line (46), and FN enhanced the binding of M. tuberculosis H37Ra to mouse alveolar macrophages (34). Several other reports demonstrate M. tuberculosis adherence to and invasion of lung alveolar epithelial cells, suggesting that M. tuberculosis might use this mechanism as well as the entry into host monocytes to gain access to tissues (7, 8, 14, 30). M. tuberculosis H37Rv was internalized in A549 cells by macropinocytosis (21), a process that has been observed for another respiratory pathogen in airway epithelial cells (27). Microscopic evaluation revealed membrane ruffling surrounding adherent bacilli similar to that seen in our SEM micrographs.
Our results, in addition to those of other reports, suggest that FN contributes to a greater number of M. tuberculosis bacilli adhering to the alveolar epithelium and subsequently gaining access to epithelial cells by direct invasion. Our results further suggest that SP-A and SP-D enhance the adherence of the attenuated strain to human SAEC but that neither collectin affected adherence by virulent M. tuberculosis to SAEC compared with untreated bacilli under shear conditions. The relevance of these results to the pathogenesis of M. tuberculosis remains to be tested systematically.
We have shown that virulent and attenuated M. tuberculosis bacilli bind well to immobilized human FN and SP-A under conditions simulating shear in the lung microenvironment, suggesting high-affinity interactions. These proteins can also influence the interactions of bacilli with human pneumocytes. Our results extend previous observations to include binding interactions comparing virulent and attenuated M. tuberculosis and these innate immune molecules and primary SAEC under shear conditions.
The respiratory system has evolved a complex and multifaceted defense against inhaled pathogens, and M. tuberculosis has evolved impressive mechanisms for overcoming these defenses. A more comprehensive understanding of the cells and molecules that play a role in pathogenesis is important since vaccines and potential therapeutic regimes should protect against all routes of entry. By targeting early binding interactions between M. tuberculosis and various host molecules and cell types using shear to functionally define high-affinity interactions, attractive candidates for further study can be identified and further assessed.
ACKNOWLEDGMENTS
We thank Nancy Equall at Montana State University Image and Chemical Analysis Laboratory for help with the scanning electron microscopy. Additionally, we thank Pati M. Glee and Aiyappa Palecanda of LigoCyte Pharmaceuticals for initial advice on binding assays.
This work was supported by Public Health Service grant 1 R41 AI50368 from the National Institute of Allergy and Infectious Diseases (L.H.-S.), AI59639 (L.S.S.), and HL-44015 and HL-29594 (E.C.C.).
REFERENCES
1. Abou-Zeid, C., T. L. Ratliff, H. G. Wiker, M. Harboe, J. Bennedsen, and G. A. W. Rook. 1988. Characterization of fibronectin-binding antigens released by Mycobacterium tuberculosis and Mycobacterium bovis BCG. Infect. Immun. 56:3046-3051.
2. Abou-Zeid, C., T. Garbe, R. Lathigra, H. G. Wiker, M. Harboe, G. A. W. Rook, and D. B. Young. 1991. Genetic and immunological analysis of Mycobacterium tuberculosis fibronectin-binding proteins. Infect. Immun. 59:2712-2718.
3. Bachofen, H., and T. A. Wilson. 1997. Micromechanics of the acinus and alveolar walls, p. 1159-1167. In R. G. Crystal, J. B. West, et al. (ed.), The lung: scientific foundations, 2nd ed. Lippincott-Raven Publishers, Philadelphia, Pa.
4. Beharka, A. A., C. D Gaynor, B. K. Kang, D. R. Voelker, F. X. McCormack, and L. S. Schlesinger. 2002. Pulmonary surfactant protein A up-regulates activity of the mannose receptor, a pattern recognition receptor expressed on human macrophages. J. Immunol. 169:3565-3573.
5. Benjamini, Y., and Y. Hochberg. 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57:289-300.
6. Bentley-Hibbert, S. I., X. Quan, T. Newman, K. Huygen, and H. P. Godfrey. 1999. Pathophysiology of antigen 85 in patients with active tuberculosis: antigen 85 circulates as complexes with fibronectin and immunoglobulin G. Infect. Immun. 67:581-588.
7. Bermudez, L. E., and J. Goodman. 1996. Mycobacterium tuberculosis invades and replicates within type II alveolar cells. Infect. Immun. 64:1400-1406.
8. Bermudez, L. E., F. J. Sangari, P. Kolonoski, M. Petrofsky, and J. Goodman. 2002. The efficiency of the translocation of Mycobacterium tuberculosis across a bilayer of epithelial cells and endothelial cells as a model of the alveolar wall is a consequence of transport within mononuclear phagocytes and invasion of alveolar epithelial cells. Infect. Immun. 70:140-146.
9. Brennan, M. J. 2005. The tuberculosis vaccine challenge. Tuberculosis 85:7-12.
10. Castro-Garza, J., C. H. King, W. E. Swords, and F. D. Quinn. 2002. Demonstration of spread by Mycobacterium tuberculosis bacilli in A549 epithelial cell monolayers. FEMS Microbiol. Lett. 212:145-149.
11. Cohn, M. L., R. F. Waggoner, and J. K. McClatchy. 1968. The 7H11 medium for the cultivation of mycobacteria. Am. Rev. Respir. Dis. 98:295-296.
12. Crouch, E. C. 1998. Collectins and pulmonary host defense. Am. J. Respir. Cell Mol. Biol. 19:177-201.
13. Crowther, J. E., V. K. Kutala, P. Kuppusamy, J. S. Ferguson, A. A. Beharka, J. L. Zweier, F. X. McCormack, and L. S. Schlesinger. 2004. Pulmonary surfactant protein A inhibits macrophage reactive oxygen intermediate production in response to stimuli by reducing NADPH oxidase activity. J. Immunol. 172:6866-6874.
14. Dobos, K. M., E. A. Spotts, F. D. Quinn, and C. H. King. 2000. Necrosis of lung epithelial cells during infection with Mycobacterium tuberculosis is preceded by cell permeation. Infect. Immun. 68:6300-6310.
15. Dye, C., S. Sheele, P. Dolin, V. Pathania, and M. C. Raviglione. 1999. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. JAMA 282:677686.
16. Fearon, D., and R. M. Locksley. 1996. The instructive role of innate immunity in the acquired immune response. Science 272:50-54.
17. Ferguson, J. S., and L. S. Schlesinger. 2000. Pulmonary surfactant in innate immunity and the pathogenesis of tuberculosis. Tuber. Lung Dis. 80:173-184.
18. Ferguson, J. S., D. R. Voelker, F. X. McCormack, and L. S. Schlesinger. 1999. Surfactant protein D binds to Mycobacterium tuberculosis bacilli and lipoarabinomannan via a carbohydrate-lectin interactions resulting in reduced phagocytosis of the bacteria by macrophages. J. Immunol. 163:312-321.
19. Ferguson, J. S., D. R. Voelker, J. A. Ufnar, A. J. Dawson, and L. S. Schlesinger. 2002. Surfactant protein D inhibition of human macrophage uptake of Mycobacterium tuberculosis is independent of bacterial agglutination. J. Immunol. 168:1309-1314.
20. Frangos, J. A., L. V. McIntire, and S. G. Eskin. 1988. Shear stress induced stimulation of mammalian cell metabolism. Biotech. Bioeng. 32:1053-1060.
21. Garcia-Perez, B. E., R. Mondragon-Flores, and J. Luna-Herrera. 2003. Internalization of Mycobacterium tuberculosis by macropinocytosis in non-phagocytic cells. Microb. Pathog. 35:49-55.
22. Gaynor, C. D., F. X. McCormack, D. R Voelker, S. E. McGowan, and L. S. Schlesinger. 1995. Pulmonary surfactant protein A mediates enhanced phagocytosis of Mycobacterium tuberculosis by a direct interaction with human macrophages. J. Immunol. 155:5343-5351.
23. Glee, P. M., J. E. Cutler, E. E. Benson, R. F. Bargatze, and K. C. Hazen. 2001. Inhibition of hydrophobic protein-mediated Candida albicans attachment to endothelial cells during physiologic shear flow. Infect. Immun. 69:2815-2820.
24. Hartshorn, K. L., M. R. White, and E. C. Crouch. 2002. Contributions of the N- and C-terminal domains of surfactant protein D to the binding, aggregation, and phagocytic uptake of bacteria. Infect. Immun. 70:6129-6139.
25. Jalali, S., M. A. del Pozo, K.-D. Chen, H. Miao, Y.-S. Li, et al. 2001. Integrin-mediated mechanotransduction requires its dynamic interaction with specific extracellular matrix ligands. Proc. Natl. Acad. Sci. USA 98:1042-1046.
26. Kawabata, S., E. Kunitomo, Y. Terao, I. Nakagawa, K. Kikuchi, K. Totsuka, and S. Hamada. 2001. Systemic and mucosal immunizations with fibronectin binding protein FBP54 induce protective immune responses against Streptococcus pyogenes challenge in mice. Infect. Immun. 69:924-930.
27. Ketterer, M., J. Q. Shao, D. B. Hornick, B. Buscher, V. K. Bandi, and M. A. Apicella. 1999. Infection of primary human bronchial epithelial cells by Haemophilus influenzae: macropinocytosis as a mechanism of airway epithelial cell entry. Infect. Immun. 67:4161-4170.
28. LeVine, A. M., J. A. Whitsett, J. A. Gwozdz, et al. 2000. Distinct effects of surfactant protein A or D deficiency during bacterial infection on the lung. J. Immunol. 165:3934-3940.
29. Levitsky, M. G. 1982. Pulmonary physiology, p. 32-48. McGraw-Hill, New York, N.Y.
30. McDonough, K. A., and Y. Kress. 1995. Cytotoxicity for lung epithelial cells is a virulence-associated phenotype of Mycobacterium tuberculosis. Infect. Immun. 63:4802-4811.
31. Mohamed, N., M. A. Teeters, J. M. Patti, M. Hook, and J. M. Ross. 1999. Inhibition of Staphylococcus aureus adherence to collagen under dynamic conditions. Infect. Immun. 67:589-594.
32. Naito, M., N. Ohara, S. Matsumoto, and T. Yamada. 1998. The novel fibronectin-binding motif and key residues of Mycobacteria. J. Biol. Chem. 273:2905-2909.
33. Naito, M., T. Fukuda, K. Sekiguchi, and T. Yamada. 2000. The domains of human fibronectin mediating the binding of antigen, the most immunopotent antigen of mycobacteria that induces protective immunity against mycobacterial infection. Biochem. J. 347:725-731.
34. Pasula, R., P. Wisniowski, and W. J. Martin II. 2002. Fibronectin facilitates Mycobacterium tuberculosis attachment to murine alveolar macrophages. Infect. Immun. 70:1287-1292.
35. Pasula, R., J. F. Downing, J. R. Wright, D. L. Kachel, T. E. Davis, Jr., and W. J. Martin II. 1997. Surfactant protein A (SP-A) mediates attachment of Mycobacterium tuberculosis to murine alveolar macrophages. Am. J. Respir. Cell Mol. Biol. 17:209-217.
36. Pinheiro, J., and D. Bates. 2000. Mixed effects models in S and SPLUS. Springer, New York, N.Y.
37. Purevdorj, B., J. W. Costerton, and P. Stoodley. 2002. Influence of hydrodynamics and cell signaling on the structure and behavior of Pseudomonas aeruginosa biofilms. Appl. Environ. Microbiol. 68:4457-4464.
38. R Development Core Team. 2005. A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. [Online.] http://www.R-project.org.
39. Ratliff, T. L., J. A. McGarr, and C. Abou-Zeid. 1988. Attachment of mycobacteria to fibronectin-coated surfaces. J. Gen. Microbiol. 134:1307-1313.
40. Ratliff, T. L., R. McCarthy, W. B. Telle, and E. J. Brown. 1993. Purification of a mycobacterial adhesin for fibronectin. Infect. Immun. 61:1889-1894.
41. Reddy, K., and J. M. Ross. 2001. Shear stress prevents fibronectin binding protein-mediated Staphylococcus aureus adhesion to resting endothelial cells. Infect. Immun. 69:3472-3475.
42. Roselli, R. J., G. Tack, and T. R. Harris. 1997. A model of fluid, erythrocyte, and solute transport in the lung. Ann. Biomed. Eng. 25:46-61.
43. Savla, U., and C. M. Waters. 1998. Mechanical strain inhibits repair of airway epithelium in vitro. Am. J. Physiol. 274:L883-L892.
44. Schlesinger, L. S., C. G. Bellinger-Kawahara, N. R. Payne, and M. A. Horwitz. 1990. Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte complement receptors and complement component C3. J. Immunol. 144:2771-2780.
45. Schlesinger, L. S., S. R. Hull, and T. M. Kaufman. 1994. Binding of the terminal mannosyl units of lipoarabinomannan from a virulent strain of Mycobacterium tuberculosis to human macrophages. J. Immunol. 52:4070-4079.
46. Schorey, J. S., Q. Li, D. W. McCourt, M. Bong-Mastek, J. E. Clark-Curtiss, T. L. Ratliff, and E. J. Brown. 1995. A Mycobacterium leprae gene encoding a fibronectin binding protein is used for efficient invasion of epithelial cells and Schwann cells. Infect. Immun. 63:2652-2657.
47. Schorey, J. S., M. A. Holsti, T. L. Ratliff, P. M. Allen, and E. J. Brown. 1996. Characterization of the fibronectin-attachment protein of Mycobacterim avium reveals a fibronectin-binding motif conserved among mycobacteria. Mol. Micro. 21:321-329.
48. Shenkman, B., E. Rubinstein, A. L. Cheung, G. E. Brill, R. Dardik, I. Tamarin, N. Savion, and D. Varon. 2001. Adherence properties of Staphylococcus aureus under static and flow conditions: roles of agr and sar loci, platelets, and plasma ligands. Infect. Immun. 69:4473-4478.
49. Sidobre, S., J. Nigou, G. Puzo, and M. Riviere. 2000. Lipoglycans are putative ligands for the human pulmonary surfactant protein A attachment to mycobacteria. J. Biol. Chem. 275:2415-2422.
50. Stoodley, P., L. Hall-Stoodley, and H. M. Lappin-Scott. 2001. Detachment, surface migration, and other dynamic behavior in bacterial biofilms revealed by digital time-lapse imaging. Methods Enzymol. 337:306-319.
51. Strong, P., U. Kishore, C. Morgan, A. Lopez Bernal, M. Singh, and K. B. Reid. 1998. A novel method of purifying lung surfactant proteins A and D from the lung lavage of alveolar proteinosis patients and from pooled amniotic fluid. J. Immunol. Methods 220:139-149.
52. Tzima, E., M. A. del Pozo, S. J. Shattil, S. Chien, and M. A. Schwartz. 2001. Activation of integrins in endothelial cells by fluid shear stress mediates Rho-dependent cytoskeletal alignment. EMBO J. 3:4639-4647.
53. West, J. B. 1974. Respiratory physiology: the essentials, p. 1-13. Williams and Wilkins, Baltimore, Md.
54. Zhao, W., J. S. Schorey, R. Groger, P. A. Allen, E. J. Brown, and T. L. Ratliff. 1999. Characterization of the fibronectin binding motif for a unique mycobacterial fibronectin attachment protein, FAP. J. Biol. Chem. 274:4521-4526.(Luanne Hall-Stoodley, Gay)
Veterinary Molecular Biology, Montana State University, Bozeman, Montana
LigoCyte Pharmaceuticals, Inc., Bozeman, Montana
Division of Infectious Diseases, Department of Internal Medicine, Center for Microbial Interface Biology, The Ohio State University, Columbus, Ohio
Department of Mathematical Sciences, Montana State University, Bozeman, Montana
Department of Pathology and Immunology, Washington University, St. Louis, Missouri
ABSTRACT
A crucial step in infection is the initial attachment of a pathogen to host cells or tissue. Mycobacterium tuberculosis has evolved multiple strategies for establishing an infection within the host. The pulmonary microenvironment contains a complex milieu of pattern recognition molecules of the innate immune system that play a role in the primary response to inhaled pathogens. Encounters of M. tuberculosis with these recognition molecules likely influence the outcome of the bacillus-host interaction. Here we use a novel fluid shear assay to investigate the binding of M. tuberculosis to innate immune molecules that are produced by pulmonary epithelial cells and are thought to play a role in the lung innate immune response. Virulent and attenuated M. tuberculosis strains bound best to immobilized human fibronectin (FN) and surfactant protein A (SP-A) under this condition. Binding under fluid shear conditions was more consistent and significant compared to binding under static conditions. Soluble FN significantly increased the adherence of both virulent and attenuated M. tuberculosis strains to human primary small airway epithelial cells (SAEC) under fluid shear conditions. In contrast, SP-A and SP-D effects on bacterial adherence to SAEC differed between the two strains. The use of a fluid shear model to simulate physiological conditions within the lung and select for high-affinity binding interactions should prove useful for studies that investigate interactions between M. tuberculosis and host innate immune determinants.
INTRODUCTION
Tuberculosis (TB) remains a significant therapeutic challenge for the early 21st century. The global incidence and prevalence of infection and multidrug resistance and the impact of AIDS and poverty diminish the likelihood that TB will be controlled without a more effective vaccine (9, 15). Therefore, innovative approaches to discovering improved diagnostic methods, therapies, and vaccine candidates are very much needed.
Mycobacterium tuberculosis is a facultative intracellular pathogen of macrophages. Following inhalation of the tubercle bacilli in droplet nuclei into the lower respiratory tract, M. tuberculosis is phagocytosed by alveolar macrophages. These bacilli have acquired the ability to survive and multiply within these cells, and it is widely believed that this ability is critical for the establishment of infection. It has also been shown that M. tuberculosis can attach to and be taken up by alveolar epithelial cells, although the significance of this is not well understood (7, 8, 10, 21, 30).
Attachment of a pathogen to pattern recognition molecules is of central importance in the host innate immune response. Bacterial pathogens typically gain entry to host tissues using pathogen-associated molecular patterns that bind host broad-spectrum cognate ligands (16). Interactions between M. tuberculosis cell wall determinants and host molecules within the alveolar microenvironment are expected to play an important role in determining whether this pathogen successfully invades the host or is cleared by first-line host defenses. The binding interactions between M. tuberculosis and host recognition molecules, therefore, represent key opportunities to identify potential candidates for improved antimycobacterial therapy, since interrupting binding events that lead to tubercle bacilli invasion or facilitating binding events that lead to clearance would functionally disrupt bacilli entry into its host cell niche.
Surfactant protein A (SP-A), surfactant protein D (SP-D), and fibronectin (FN) are produced by pulmonary epithelial cells and play a role in the innate immune response in the lung. The collectins, SP-A and SP-D, are oligomeric carbohydrate binding proteins that regulate opsonization, phagocytosis, and agglutination of microorganisms within the respiratory tract. Their localization in the alveoli enables them to act as "first responders" to inhaled pathogens. SP-A acts both as an opsonin and regulator of cell receptor activity in enhancing phagocytosis by macrophages (4, 12, 49). SP-A-deficient mice are more susceptible to infection with respiratory pathogens, and the instillation of SP-A in the lungs of these mice enhances phagocytosis and clearance of pathogenic bacteria (17, 28). In contrast, SP-D binding to respiratory pathogens aggregates microorganisms and alters their interactions with host cells (12, 17). Previous work has shown that SP-D agglutination of virulent M. tuberculosis inhibits phagocytosis of the bacilli by alveolar macrophages (18, 19). Finally, FN is a glycoprotein that plays a role in the host response to tissue injury and opsonization (33). FN has multiple binding sites for collagen, heparin, and fibrin, and it has been shown to enhance the attachment of several pathogens, including mycobacteria, to host cells (32, 34, 46).
Binding interactions between pathogens and host molecules or cells are traditionally studied under static conditions. During inhalation, however, respiratory pathogens undergo shear forces as they travel to the terminal airways (29, 53). Once in the alveoli, expansion and contraction of the alveolar sac subject the cells lining the alveoli to deformations that change with lung tidal volume, i.e., bulging of the epithelium at low volumes and surface tensions (contraction) and stretching of the epithelium at higher volumes (expansion) (3). Therefore, the alveolar epithelium is subjected to micromechanical forces and shear from airflow—dynamic effects that are not well modeled by microorganisms and host cells in a static culture system.
The aim of this study was to evaluate binding of M. tuberculosis to purified human molecules using shear conditions to better simulate the dynamic lung microenvironment. To accomplish this aim, we modified a technique developed for investigating leukocyte interactions with other cells to assay for M. tuberculosis binding to putative host determinants under shear conditions. This method is sensitive and better simulates in vivo conditions by examining M. tuberculosis-host interactions when physiological shear is present. Importantly, shear conditions may reduce or abrogate nonspecific binding interactions that take place under static conditions and thus select for high-affinity binding interactions.
Using both a flow system and static binding assays, we investigated the binding patterns of both a virulent and an attenuated strain of M. tuberculosis to human SP-A, SP-D, and FN. Binding patterns of tubercle bacilli to SP-A and FN were significantly different compared with a nonspecific protein, bovine serum albumin (BSA), under shear conditions, suggesting high-affinity interactions. When M. tuberculosis was preincubated with soluble SP-A, SP-D, or FN and tested in a shear assay using primary human small airway epithelial cells (SAEC), these proteins affected the number of bacilli that adhered to these cells. FN significantly enhanced adherence to SAEC for both virulent and attenuated M. tuberculosis under shear conditions, whereas SP-A and SP-D showed strain-dependent adherence patterns. These results indicate that the shear model is useful in identifying robust binding interactions between M. tuberculosis and host molecules or cells. These interactions will enable further investigation of potential therapeutic and vaccine targets.
MATERIALS AND METHODS
Media and reagents. Middlebrook 7H9 broth and 7H11 agar were prepared with oleic acid albumin dextrose catalase supplement (Difco media; Becton Dickinson, Sparks, MD) and supplemented with 0.1% casein enzymatic hydrosylate and glycerol (Sigma) as described previously (11). Hanks balanced salt solution (HBSS) supplemented with HEPES, Ca2+, and Mg2+, and 2% fetal bovine serum (FBS) and RPMI (Gibco cell culture; Invitrogen Corporation, Carlsbad, CA) supplemented with 10% FBS (HyClone, Logan, UT), HEPES, penicillin/streptomycin, nonessential amino acids, sodium pyruvate, L-glutamine, and vitamins (Sigma) were used to rinse flow cells and in flow cell experiments. Human FN and BSA were purchased from Sigma.
Bacteria. M. tuberculosis H37Ra (ATCC 25177), H37Rv (ATCC 27294), and Erdman (ATCC 35801) were grown and harvested on Middlebrook 7H11 agar and used between 9 to 11 days as described previously (44). Immediately prior to assays, bacteria were scraped into polypropylene tubes containing 1 ml 7H9 broth and pulse vortexed five to six times (1 s per pulse) with two 3-mm-diameter glass beads. Clumped bacteria were allowed to settle, and the resulting suspension was counted using a Petroff-Hausser chamber, yielding >90% of a single cell bacterial suspension.
Surfactant protein purification. SP-A was purified as previously described (13). Following an approved Institutional Review Board protocol, surfactant lipid from human alveolar proteinosis bronchoalveolar lavage (BAL) fluid was isolated by centrifugation at 20,000 x g and washed repeatedly, and lipid-associated SP-A was eluted using 2 mM EDTA at 4°C. SP-A was purified from the EDTA elution supernatant by separation over a mannose-Sepharose affinity chromatography column, and EDTA was removed by dialysis against 10 mM HEPES plus 25 mM NaCl. SP-D was purified from human alveolar proteinosis bronchoalveolar lavage fluid as previously described (51), with minor modifications. Briefly, following removal of SP-A-containing surfactant lipid from BAL fluid by centrifugation, SP-D was purified from BAL fluid by loading BAL supernatant to a maltose-Sepharose affinity chromatography column. The column was then washed to remove unbound proteins, and SP-D was specifically eluted using MnCl2 and then dialyzed against 10 mM HEPES plus 150 mM NaCl to remove the MnCl2. Recombinant human SP-D dodecamers were also used in shear and static binding assays. There are no known functional differences between comparably oligomerized recombinant or native proteins. Recombinant SP-D was isolated by sequential maltosyl-agarose and gel filtration chromatography (24). Purified proteins were examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting using anti-SP-A and anti-SP-D specific antibodies.
Protein coating of flow cells and coverslips. To coat flow cells with FN, sterile glass flow cells (Friedrich and Dimmock, Inc., Millville, NJ) were filled with 100 μl of protein and incubated for 2 h at 37°C. The optimal concentration for attachment was determined empirically. Flow cells were then rinsed three times with RPMI. For covalent attachment of SP-A and SP-D, flow cells were filled with 100 μl of 0.1 mg/ml of poly-L-lysine diluted in phosphate-buffered saline (PBS) and allowed to incubate at room temperature (RT) for 60 min. Flow cells were washed three times with PBS and filled with 100 μl of 2.5% glutaraldehyde for 15 min and washed five times with sterile H2O and two times with PBS. Flow cells were incubated with 10 μg/ml SP-A, SP-D, or BSA at RT for 30 min, washed twice with PBS, and filled with 0.2 M glycine and incubated at RT overnight to quench untreated aldehyde groups. The flow cells were then washed twice with PBS and twice with RPMI. For static binding assays, glass coverslips in wells of a 24-well tissue culture plate were coated with proteins exactly as described above (44).
Flow cell and static binding assays. For flow cell assays, bacteria were stained with Syto 9 nucleic acid dye (Molecular Probes, Eugene, OR) for 30 min prior to the flow cell assay. Flow cells and loop tubing were filled with HBSS and placed in an Ismatec low-speed precision pump (Cole-Parmer, Vernon Hills, IL) to remove bubbles. The loops were clamped and a glass flow cell, coated with either the human protein of interest or BSA, was inserted into each loop tubing and secured to prevent leakage. A portable stage heater (Cole-Parmer) equilibrated to 37°C was placed beneath the flow cell, and flow was initiated at a flow rate of 0.75 dynes/cm2 (Reynold's number [Re] = 19) (calculated for fluid at 37°C). Stained bacilli, 1 x 108, were injected in each loop and allowed to circulate for 1 h at 37°C. Bacilli were then fixed in 3% glutaraldehyde (final concentration) for 10 min under flow to optimize mixing. Following fixation, flow was resumed and 30 volumes of HBSS were passed through the flow cell to remove nonadherent bacilli. For static binding assays, glass coverslips in tissue culture wells were incubated with 108 bacilli for 1 h at 37°C, rinsed to remove nonadherent bacteria, fixed, stained with auramine-rhodamine, and mounted on glass slides prior to enumeration. In control experiments, M. tuberculosis mannosyl lipoarabinomannan (ManLAM)-coated beads (108) (45) were incubated with coverslips under the same conditions, washed, fixed, and enumerated.
Measuring bound M. tuberculosis. For flow assays, adherent bacteria were enumerated under a Leica DM-RXE microscope (Leica Microsystems, Exton, PA) (using the PL APO 63x/1.20 W lens or a 100x/1.4 oil lens) and counting two separate faces of the flow cell. By examining the top face of the flow cell, only bacteria that were adherent to the protein-coated surface, not bacteria that had detached and settled due to gravity, were enumerated. A minimum of 30 fields were counted per flow cell, and duplicate flow cells were run in parallel for each molecule. BSA control flow cells were run for each experiment to control for assay-to-assay variability. Average adherent bacilli per field were determined, and the area per microscopic field was calculated to yield a normalized number of bound bacilli per cm2. For static binding assays, a minimum of 20 high-power fields were counted per coverslip for either bacilli or ManLAM-coated beads and normalized as above to yield the number of bound bacilli or beads/cm2. Experiments were done in triplicate for each protein and particle tested (n = 6).
Cells and tissue culture. Cryopreserved SAEC were obtained from Clonetics (San Diego, CA) and subcultured in small airway basal medium (SABM; Clonetics), aliquoted, and frozen as stock suspensions containing 5 x 105 cells per ml. Frozen stock aliquots were removed from liquid N2 and cultured in Ham's F-12 media (Invitrogen, Calsbad, CA) supplemented with 10% FBS, HEPES, L-glutamine, nonessential amino acids, sodium pyruvate, and penicillin/streptomycin and incubated at 37°C and 5% CO2 for 48 h. Cells were harvested and centrifuged, and 5 x 106 SAEC were seeded onto MatTek glass-bottom culture dishes (MatTek Corp., Ashland, MA) and allowed to grow to confluence for 24 to 48 h.
Adherence to SAEC. SAEC monolayers were incubated with 5 x107 M. tuberculosis H37Rv or H37Ra cells (multiplicity of infection = 10:1) at 37°C for 1 h under gentle shear conditions (35 rpm) on a Lab-Line lab rotator (model no. 1304; Lab-Line, Melrose Park, IL). Cells were rinsed three times with tissue culture medium to remove nonadherent bacilli and bacteria were stained using a TB fluorescent stain kit M (Becton Dickenson, Sparks, MD) for 15 min and then rinsed three times to remove the auramine M. Following fixation with 3% glutaraldehyde for 10 min, the monolayers were rinsed three times to remove glutaraldehyde, counterstained briefly with potassium permanganate to enhance contrast, and rinsed twice more.
Confocal microscopy. Flow cell images were captured using a Leica DM-RXE microscope and TCS NT confocal scanning laser system (Leica Microsystems, Exton, PA) using the 488-nm laser line to visualize Syto 9-stained M. tuberculosis cells that were adherent to the flow cell after 1 h of shear. Tissue culture plates were analyzed with a Zeiss LSM 510 meta confocal microscope (Carl Zeiss Microimaging Inc., Thornwood, NY), using the 488-nm laser line to visualize fluorescent M. tuberculosis cells stained with auramine M and the transmission mode to visualize SAEC counterstained with potassium permanganate.
SEM. Monolayers were processed for scanning electron microscopy (SEM) by dehydration in a graded ethanol series following 3% glutaraldehyde fixation. Samples were sputter coated with 15 to 20 nm gold palladium using a Hummer VII sputtering system (Anatech Ltd., Alexandria, VA). Samples were visualized at an accelerator voltage of 12 kV using a JEOL JSM-6100 scanning electron microscope and digitized TIFF images were collected.
Statistical analysis. The data were analyzed using the R statistical environment (38) and the add-on package nlme (nonlinear and linear mixed effects) of Pinheiro and Bates (36). The analysis used a mixed effects model with random effects for the experiment and fixed effects for the strains and compounds (human molecules); the fixed model was a two-way analysis of variance model. Normality of residuals was examined using quantile-quantile plots, and transformations of the response were used to give a good fit to normality. The observation of many fields for each flow cell created pseudoreplications which were investigated for possible serial correlation (5). Several types of correlation were considered, and the best fitting, an autocorrelation of order 1, was used, thus avoiding the common assumption of independent observations by modeling the correlation structure.
RESULTS
The flow cell system. Shear caused by fluid flow is a function of the velocity, density and viscosity of the fluid (air) or air-liquid interface, and the diameter of the tube (blood vessel or airway). Laminar flow is predominant in the small airways (terminal bronchioles and alveoli) (29, 53). Physiologically relevant shear forces are measured in dynes/cm2 (shear forces occurring over square centimeters). Shear forces in the lung have been used primarily for purposes of calculating gas exchange where the fluid is air (29, 42, 53). Due to the complexity of the lung microenvironment with its three-dimensional branching structure, however, it is extremely difficult to determine these numbers, and so they are based largely on models using basic concepts of hydrodynamics or fluid transport (42).
To approximate physiological conditions present in the lower airways, purified proteins were adsorbed or covalently linked to the flow cell and M. tuberculosis was circulated under laminar flow conditions (Re = 19) determined using the following equations: Re = u(ave)(Dh)/ and u(ave) = Q/CSA, where u(ave) is the average flow velocity, Q is the flow rate, CSA is the cross-sectional area of the flow cell (1-mm2 flow chamber), Dh is the hydraulic diameter, and is the kinematic viscosity (taken for that of water at 37°C) (50). These conditions yielded good mixing and resulted in an operational shear equivalent to less than 1 dyne/cm2, simulating a very low laminar shear that would be expected in the terminal bronchioles and air sacs. The sterile flow system was completely disposable and precluded autoclaving and the potential buildup of bacterial toxins. Moreover, the modular system fit easily inside a biosafety cabinet in the biosafety level 3 facility and generated relatively small volumes of bacterial waste (flow cell volume, 3 ml). A similar system has been successfully used to monitor real-time binding and binding inhibition between larger cells such as epithelial cells and a fungal pathogen, Candida albicans (23).
M. tuberculosis binding to immobilized proteins. M. tuberculosis binding to human proteins was first tested under dynamic conditions by enumerating binding of the live attenuated M. tuberculosis strain H37Ra to immobilized proteins under shear conditions. FN was tested at two different flow rates (1.5 dyne/cm2 and 0.75 dyne/cm2) with no statistical difference in mycobacterial binding between rates (data not shown). Since we wished to simulate low shear conditions predicted in the alveolar space, we used the lower rate for all subsequent experiments. FN was initially tested at 10 and 100 μg/ml, and there was no significant difference in M. tuberculosis binding (compared to BSA control; data not shown), indicating that binding was saturable at 10 μg/ml FN (data not shown), consistent with other reports (34).
We next compared the binding of two strains of M. tuberculosis (a virulent strain, H37Rv, and an attenuated strain, H37Ra) to FN, SP-A, and SP-D under shear conditions. To adjust for possible changes in binding across experiments, a binding index for each molecule was calculated by dividing the average M. tuberculosis bacilli bound to the host molecule by the average number of bacilli bound to BSA (Table 1). Under shear conditions, virulent and attenuated M. tuberculosis bound to human FN approximately six- to sevenfold more than to the BSA control. M. tuberculosis binding to SP-A was approximately four to fivefold greater than to BSA. Binding of the virulent strain H37Rv to SP-D, however, was reduced compared with the nonspecific protein, BSA, showing a binding index of <1 (Table 1).
To allow for the assumption of constant variance of residuals in the mixed effects model, and to compare different binding assays, the binding indices were transformed to a log10 scale. Both strains bound similarly to FN and SP-A, but the confidence intervals of the attenuated and virulent strains did not overlap in SP-D binding conditions (data not shown), suggesting a difference in the binding patterns of virulent and attenuated M. tuberculosis to SP-D. Simple two-tailed t tests performed on the log10 data indicated significant differences between binding to BSA and to the collectins and FN, except for the binding of H37Ra to SP-D compared to BSA under shear conditions (Fig. 1).
Because SP-D was previously shown to bind to and agglutinate the virulent M. tuberculosis Erdman strain (18), our result was unexpected. Therefore, we further investigated the binding of M. tuberculosis H37Rv, H37Ra, and Erdman strains to covalently bound SP-D under static conditions. We also studied the binding of M. tuberculosis ManLAM-coated beads to covalently bound SP-D, since ManLAM also binds to SP-D (18). While the binding of the Erdman strain to both FN and SP-A was similar to that observed with the H37Ra and H37Rv strains under shear and static conditions, the binding to SP-D was variable and less than expected, indicating that the Erdman strain also did not bind well to SP-D covalently linked to the substratum. This suspicion was confirmed with binding assays using ManLAM beads, which failed to show specific binding to covalently linked SP-D (data not shown). Therefore we conclude that covalent attachment of SP-D to a platform disrupts the conformation of the SP-D carbohydrate recognition domains necessary for binding of M. tuberculosis.
Overall our data provide evidence that M. tuberculosis is able to bind rapidly (within 1 h) and significantly to immobilized human SP-A and FN under shear conditions. Binding under static conditions revealed similar trends, but there was a marked increase in variability. Binding under shear conditions reduced this variability and better enabled the identification of the specific receptor-ligand interactions.
M. tuberculosis adherence to primary human respiratory epithelial cells. To further simulate the dynamic physiological environment of the lung, we performed adherence assays of virulent and attenuated M. tuberculosis bacilli to human primary SAEC under shear conditions following preincubation with soluble human SP-A, SP-D, and FN. We chose to specifically examine tubercle bacilli adherence to SAEC since alveolar epithelial cells produce all of the molecules that showed significant binding interaction patterns in previous shear assays, and we wished to determine if soluble human collectin proteins or FN influence adherence of M. tuberculosis to primary human SAEC under shear.
Both M. tuberculosis H37Ra and H37Rv adhered to SAEC under rotational shear conditions at 37°C within 1 h (Fig. 2). These results are consistent with other reports that have shown that M. tuberculosis adheres to respiratory epithelial cells (7, 8, 30). When M. tuberculosis strains were preincubated with soluble human SP-A, SP-D, or FN and assessed for adherence to SAEC cell monolayers, FN significantly enhanced tubercle bacilli adherence to these cells for both virulent and attenuated M. tuberculosis strains, resulting in the greatest number of adherent bacilli to SAEC (Fig. 2) and the greatest difference compared with adherence to SAEC by untreated M. tuberculosis (P 0.0001) for both strains. Statistical analysis also showed that SP-D preincubation influenced attenuated M. tuberculosis H37Ra adherence to SAEC (P = 0.0258). The P value comparing SP-A-preincubated H37Ra to untreated M. tuberculosis was 0.0531. Interestingly, this pattern differed from the adherence pattern to SAEC by the virulent H37Rv strain, where adherence was unaffected by preincubation with the collectins compared with M. tuberculosis in the absence of host protein (SP-A, P = 0.8029; and SP-D, P = 0.9940). Adherence by M. tuberculosis in the presence of FN also differed significantly from bacilli preincubated with SP-A (P = 0.003) or SP-D (P = 0.002).
Microscopic evaluation of adherent bacilli to SAEC. Confocal microscopy and SEM showed that tubercle bacilli were attached to the epithelial monolayer (Fig. 3 and 4). Figure 3 shows M. tuberculosis bacilli adhered to SEAC after 1 h of incubation under shear conditions. Bacilli were observed to be adherent to the cell monolayer even when not preincubated with human proteins. Large clumps of bacilli were noted more frequently with SP-A, SP-D, and FN preincubation (Fig. 3). In some confocal micrographs, bacilli appeared to be within the epithelial cell. SEM analysis demonstrated M. tuberculosis in the process of being surrounded by epithelial cell cytoplasm. Fine filaments of the monolayer appeared to envelop the edges of single bacteria or aggregates of cells (Fig. 4). These results suggest that bacilli are capable of invading human primary SAEC and that host proteins can influence the number of M. tuberculosis adherent to these primary airway cells under conditions that simulate physiological shear.
DISCUSSION
M. tuberculosis encounters a dynamic pulmonary microenvironment upon inhalation into the lung. The airway environment includes epithelial cells and host innate immune molecules, and previous studies provide evidence that interactions between M. tuberculosis and these recognition molecules influence uptake of the bacilli (4, 17-19, 22). Binding assays that examine interactions between microbial pathogens and recognition molecules are typically carried out under static conditions. These studies are useful in determining whether specific binding occurs between a ligand and its receptor; however, they overlook mechanical forces that arise during shear over a surface.
Shear forces can affect the binding of a microorganism as well as the effectiveness of a putative binding inhibitor (31, 41, 48). Bacterial cells can respond to hydrodynamic forces that affect bacterial adherence to substrata such as inert surfaces and extracellular matrix (31, 37, 41, 48). Specifically, the adherence of Staphylococcus aureus mutants to fibrinogen and FN differed depending on whether bacteria were incubated under static versus flow conditions (48). Moreover, shear may also select for higher-affinity adhesion interactions. A recombinant collagen adhesin fragment that abrogated S. aureus binding to collagen under static conditions failed to inhibit binding under shear conditions, suggesting that putative inhibitor candidates should have the ability to bind to a ligand and resist detachment under conditions that may be present physiologically (31). Shear also influences cellular responses such as mechanotransduction, membrane receptor regulation, cytokine expression, and repair mechanisms in many types of mammalian cells (20, 25, 43, 52).
A wide range of shear forces occur in the lung. Laminar flow is thought to be present in the small airways (terminal bronchioles and alveoli), whereas transitional or turbulent flows occur in branching and large airways, respectively (29, 53). Therefore, the lung epithelial surface is subjected to dynamic effects that are not well modeled in a static culture system. Given this dynamic pulmonary environment, we reason that an infectious particle might not simply settle out due to gravitational forces but is rather subjected to shear forces during its travel in the terminal airways and alveoli. Thus, we postulate that shear in the lung as studied using aqueous flow provides a better simulation of the dynamic conditions that exist physiologically in intact airways. To our knowledge, M. tuberculosis adherence under shear conditions has not been previously investigated.
Our results show that both virulent H37Rv and attenuated H37Ra M. tuberculosis bound to purified human FN and SP-A under shear conditions and that binding was significantly different than that for the BSA control (Fig. 1). Binding assays performed under static conditions showed similar trends; however, due to marked variability in the results, significant differences could not be seen between the host proteins and BSA control. Importantly, the significant differences exhibited under shear conditions were observed even with the very low shear used to model the alveolar space. This may be due to increased chances of bacilli binding to cognate receptors (better mixing) and/or the ability of shear to select for higher-affinity adhesion interactions. Data from static assays, on the other hand, may reflect a combination of specific binding and nonspecific binding, with the contribution of nonspecific binding creating background binding that is more difficult to distinguish from specific binding interactions.
Binding of M. tuberculosis strains to immobilized FN under shear conditions was increased greater than sixfold compared to BSA (Table 1). Binding to FN has been well characterized for several bacterial pathogens, including mycobacteria. Specifically, FN attachment proteins and binding proteins have been described for several mycobacterial species, including M. tuberculosis (1, 2, 32, 39, 40, 47, 54). Our results agree with previous studies that showed approximately a fourfold increase in binding M. bovis BCG to FN compared to a serum albumin protein control under static conditions; other assays using H37Rv demonstrated a similar level of binding (39). FN binding proteins have been used to immunize mice against challenge with Streptococcus pyogenes (26) and may induce some protective immunity to mycobacterial infection (6, 33).
SP-A and SP-D are important components of innate immunity and influence the ability of pathogens to be taken up by host cells and/or cleared by host defense mechanisms. Gaynor et al. showed that SP-A enhanced phagocytosis of the virulent M. tuberculosis Erdman strain by human alveolar macrophages within 1 h postinfection (22). Further studies revealed that the interaction of SP-A with macrophages was due to increased expression of the macrophage mannose receptor (4). Pasula et al. showed that attenuated M. tuberculosis H37Ra bound directly to SP-A with high affinity (Kd of 1.9 x 10–9 M) and that SP-A mediated increased adherence of bacilli to mouse alveolar macrophages (35). Direct binding of M. bovis BCG and nonpathogenic Mycobacterium smegmatis to human SP-A was also observed, suggesting that different mycobacterial species are capable of binding to SP-A (49). Our results show that both attenuated and virulent M. tuberculosis bound to immobilized human SP-A under shear conditions at approximately four to five times the level of binding to BSA.
We also investigated the binding of M. tuberculosis to immobilized SP-D under shear conditions because SP-D has previously been shown to bind and agglutinate virulent M. tuberculosis Erdman and reduce uptake of bacilli into human macrophages (18, 19). SP-D bound chiefly to the terminal mannosyl oligosaccharides of M. tuberculosis LAM, and agglutination of bacilli was mediated by bridging via the carbohydrate binding domains (CRDs) and facilitated by the oligomeric structure of intact SP-D. Based on these data, we had expected that the virulent M. tuberculosis would exhibit increased binding to SP-D under shear conditions compared to that for BSA. Our studies revealed that the covalent attachment of SP-D to the surface disrupted binding by virulent M. tuberculosis and M. tuberculosis ManLAM-coated beads. Covalent attachment may interfere with M. tuberculosis binding by preventing bacilli agglutination (and thereby potentially affecting the avidity of binding) and/or affecting the availability of the CRDs for binding, either by creating a less suitable conformation or decreasing the availability of the CRDs for binding. Under the condition where M. tuberculosis was preincubated with soluble SP-D, SEM images showed that bacilli were particularly clumped (Fig. 3d), indicating that soluble SP-D agglutinated M. tuberculosis. These results indicate that the conformation of SP-D is extremely important for binding.
In adherence assays using primary human SAEC monolayers, FN significantly enhanced M. tuberculosis adherence for both the virulent and attenuated M. tuberculosis strains (Fig. 2). SP-A and SP-D also slightly enhanced adherence of the H37Ra strain to SAEC. The ability of soluble FN, SP-A, and SP-D to act as bridging molecules that increase the level of bacilli-bacilli homotypic binding interactions may be important. Clumping of M. tuberculosis would presumably result in more bacilli, with corresponding ligands, engaging cognate receptors on host cells. Shear may enhance aggregation, or alternatively, shear may increase detachment of clumps from low-affinity receptors and promote clearance from the airway. Thus, differences among strains in binding to the collectins are of interest.
Confocal and SEM images indicated that M. tuberculosis bacilli were taken up by SAEC regardless of whether they were preincubated with soluble human proteins. However, our results indicate that FN enhances the M. tuberculosis-SAEC interaction. Other reports have shown that FN specifically enhanced mycobacterial adherence to both nonphagocytic and phagocytic cells. FN served as an opsonin to facilitate the adhesion and invasion of Mycobacterium leprae into both an epithelial and Schwann cell line (46), and FN enhanced the binding of M. tuberculosis H37Ra to mouse alveolar macrophages (34). Several other reports demonstrate M. tuberculosis adherence to and invasion of lung alveolar epithelial cells, suggesting that M. tuberculosis might use this mechanism as well as the entry into host monocytes to gain access to tissues (7, 8, 14, 30). M. tuberculosis H37Rv was internalized in A549 cells by macropinocytosis (21), a process that has been observed for another respiratory pathogen in airway epithelial cells (27). Microscopic evaluation revealed membrane ruffling surrounding adherent bacilli similar to that seen in our SEM micrographs.
Our results, in addition to those of other reports, suggest that FN contributes to a greater number of M. tuberculosis bacilli adhering to the alveolar epithelium and subsequently gaining access to epithelial cells by direct invasion. Our results further suggest that SP-A and SP-D enhance the adherence of the attenuated strain to human SAEC but that neither collectin affected adherence by virulent M. tuberculosis to SAEC compared with untreated bacilli under shear conditions. The relevance of these results to the pathogenesis of M. tuberculosis remains to be tested systematically.
We have shown that virulent and attenuated M. tuberculosis bacilli bind well to immobilized human FN and SP-A under conditions simulating shear in the lung microenvironment, suggesting high-affinity interactions. These proteins can also influence the interactions of bacilli with human pneumocytes. Our results extend previous observations to include binding interactions comparing virulent and attenuated M. tuberculosis and these innate immune molecules and primary SAEC under shear conditions.
The respiratory system has evolved a complex and multifaceted defense against inhaled pathogens, and M. tuberculosis has evolved impressive mechanisms for overcoming these defenses. A more comprehensive understanding of the cells and molecules that play a role in pathogenesis is important since vaccines and potential therapeutic regimes should protect against all routes of entry. By targeting early binding interactions between M. tuberculosis and various host molecules and cell types using shear to functionally define high-affinity interactions, attractive candidates for further study can be identified and further assessed.
ACKNOWLEDGMENTS
We thank Nancy Equall at Montana State University Image and Chemical Analysis Laboratory for help with the scanning electron microscopy. Additionally, we thank Pati M. Glee and Aiyappa Palecanda of LigoCyte Pharmaceuticals for initial advice on binding assays.
This work was supported by Public Health Service grant 1 R41 AI50368 from the National Institute of Allergy and Infectious Diseases (L.H.-S.), AI59639 (L.S.S.), and HL-44015 and HL-29594 (E.C.C.).
REFERENCES
1. Abou-Zeid, C., T. L. Ratliff, H. G. Wiker, M. Harboe, J. Bennedsen, and G. A. W. Rook. 1988. Characterization of fibronectin-binding antigens released by Mycobacterium tuberculosis and Mycobacterium bovis BCG. Infect. Immun. 56:3046-3051.
2. Abou-Zeid, C., T. Garbe, R. Lathigra, H. G. Wiker, M. Harboe, G. A. W. Rook, and D. B. Young. 1991. Genetic and immunological analysis of Mycobacterium tuberculosis fibronectin-binding proteins. Infect. Immun. 59:2712-2718.
3. Bachofen, H., and T. A. Wilson. 1997. Micromechanics of the acinus and alveolar walls, p. 1159-1167. In R. G. Crystal, J. B. West, et al. (ed.), The lung: scientific foundations, 2nd ed. Lippincott-Raven Publishers, Philadelphia, Pa.
4. Beharka, A. A., C. D Gaynor, B. K. Kang, D. R. Voelker, F. X. McCormack, and L. S. Schlesinger. 2002. Pulmonary surfactant protein A up-regulates activity of the mannose receptor, a pattern recognition receptor expressed on human macrophages. J. Immunol. 169:3565-3573.
5. Benjamini, Y., and Y. Hochberg. 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57:289-300.
6. Bentley-Hibbert, S. I., X. Quan, T. Newman, K. Huygen, and H. P. Godfrey. 1999. Pathophysiology of antigen 85 in patients with active tuberculosis: antigen 85 circulates as complexes with fibronectin and immunoglobulin G. Infect. Immun. 67:581-588.
7. Bermudez, L. E., and J. Goodman. 1996. Mycobacterium tuberculosis invades and replicates within type II alveolar cells. Infect. Immun. 64:1400-1406.
8. Bermudez, L. E., F. J. Sangari, P. Kolonoski, M. Petrofsky, and J. Goodman. 2002. The efficiency of the translocation of Mycobacterium tuberculosis across a bilayer of epithelial cells and endothelial cells as a model of the alveolar wall is a consequence of transport within mononuclear phagocytes and invasion of alveolar epithelial cells. Infect. Immun. 70:140-146.
9. Brennan, M. J. 2005. The tuberculosis vaccine challenge. Tuberculosis 85:7-12.
10. Castro-Garza, J., C. H. King, W. E. Swords, and F. D. Quinn. 2002. Demonstration of spread by Mycobacterium tuberculosis bacilli in A549 epithelial cell monolayers. FEMS Microbiol. Lett. 212:145-149.
11. Cohn, M. L., R. F. Waggoner, and J. K. McClatchy. 1968. The 7H11 medium for the cultivation of mycobacteria. Am. Rev. Respir. Dis. 98:295-296.
12. Crouch, E. C. 1998. Collectins and pulmonary host defense. Am. J. Respir. Cell Mol. Biol. 19:177-201.
13. Crowther, J. E., V. K. Kutala, P. Kuppusamy, J. S. Ferguson, A. A. Beharka, J. L. Zweier, F. X. McCormack, and L. S. Schlesinger. 2004. Pulmonary surfactant protein A inhibits macrophage reactive oxygen intermediate production in response to stimuli by reducing NADPH oxidase activity. J. Immunol. 172:6866-6874.
14. Dobos, K. M., E. A. Spotts, F. D. Quinn, and C. H. King. 2000. Necrosis of lung epithelial cells during infection with Mycobacterium tuberculosis is preceded by cell permeation. Infect. Immun. 68:6300-6310.
15. Dye, C., S. Sheele, P. Dolin, V. Pathania, and M. C. Raviglione. 1999. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. JAMA 282:677686.
16. Fearon, D., and R. M. Locksley. 1996. The instructive role of innate immunity in the acquired immune response. Science 272:50-54.
17. Ferguson, J. S., and L. S. Schlesinger. 2000. Pulmonary surfactant in innate immunity and the pathogenesis of tuberculosis. Tuber. Lung Dis. 80:173-184.
18. Ferguson, J. S., D. R. Voelker, F. X. McCormack, and L. S. Schlesinger. 1999. Surfactant protein D binds to Mycobacterium tuberculosis bacilli and lipoarabinomannan via a carbohydrate-lectin interactions resulting in reduced phagocytosis of the bacteria by macrophages. J. Immunol. 163:312-321.
19. Ferguson, J. S., D. R. Voelker, J. A. Ufnar, A. J. Dawson, and L. S. Schlesinger. 2002. Surfactant protein D inhibition of human macrophage uptake of Mycobacterium tuberculosis is independent of bacterial agglutination. J. Immunol. 168:1309-1314.
20. Frangos, J. A., L. V. McIntire, and S. G. Eskin. 1988. Shear stress induced stimulation of mammalian cell metabolism. Biotech. Bioeng. 32:1053-1060.
21. Garcia-Perez, B. E., R. Mondragon-Flores, and J. Luna-Herrera. 2003. Internalization of Mycobacterium tuberculosis by macropinocytosis in non-phagocytic cells. Microb. Pathog. 35:49-55.
22. Gaynor, C. D., F. X. McCormack, D. R Voelker, S. E. McGowan, and L. S. Schlesinger. 1995. Pulmonary surfactant protein A mediates enhanced phagocytosis of Mycobacterium tuberculosis by a direct interaction with human macrophages. J. Immunol. 155:5343-5351.
23. Glee, P. M., J. E. Cutler, E. E. Benson, R. F. Bargatze, and K. C. Hazen. 2001. Inhibition of hydrophobic protein-mediated Candida albicans attachment to endothelial cells during physiologic shear flow. Infect. Immun. 69:2815-2820.
24. Hartshorn, K. L., M. R. White, and E. C. Crouch. 2002. Contributions of the N- and C-terminal domains of surfactant protein D to the binding, aggregation, and phagocytic uptake of bacteria. Infect. Immun. 70:6129-6139.
25. Jalali, S., M. A. del Pozo, K.-D. Chen, H. Miao, Y.-S. Li, et al. 2001. Integrin-mediated mechanotransduction requires its dynamic interaction with specific extracellular matrix ligands. Proc. Natl. Acad. Sci. USA 98:1042-1046.
26. Kawabata, S., E. Kunitomo, Y. Terao, I. Nakagawa, K. Kikuchi, K. Totsuka, and S. Hamada. 2001. Systemic and mucosal immunizations with fibronectin binding protein FBP54 induce protective immune responses against Streptococcus pyogenes challenge in mice. Infect. Immun. 69:924-930.
27. Ketterer, M., J. Q. Shao, D. B. Hornick, B. Buscher, V. K. Bandi, and M. A. Apicella. 1999. Infection of primary human bronchial epithelial cells by Haemophilus influenzae: macropinocytosis as a mechanism of airway epithelial cell entry. Infect. Immun. 67:4161-4170.
28. LeVine, A. M., J. A. Whitsett, J. A. Gwozdz, et al. 2000. Distinct effects of surfactant protein A or D deficiency during bacterial infection on the lung. J. Immunol. 165:3934-3940.
29. Levitsky, M. G. 1982. Pulmonary physiology, p. 32-48. McGraw-Hill, New York, N.Y.
30. McDonough, K. A., and Y. Kress. 1995. Cytotoxicity for lung epithelial cells is a virulence-associated phenotype of Mycobacterium tuberculosis. Infect. Immun. 63:4802-4811.
31. Mohamed, N., M. A. Teeters, J. M. Patti, M. Hook, and J. M. Ross. 1999. Inhibition of Staphylococcus aureus adherence to collagen under dynamic conditions. Infect. Immun. 67:589-594.
32. Naito, M., N. Ohara, S. Matsumoto, and T. Yamada. 1998. The novel fibronectin-binding motif and key residues of Mycobacteria. J. Biol. Chem. 273:2905-2909.
33. Naito, M., T. Fukuda, K. Sekiguchi, and T. Yamada. 2000. The domains of human fibronectin mediating the binding of antigen, the most immunopotent antigen of mycobacteria that induces protective immunity against mycobacterial infection. Biochem. J. 347:725-731.
34. Pasula, R., P. Wisniowski, and W. J. Martin II. 2002. Fibronectin facilitates Mycobacterium tuberculosis attachment to murine alveolar macrophages. Infect. Immun. 70:1287-1292.
35. Pasula, R., J. F. Downing, J. R. Wright, D. L. Kachel, T. E. Davis, Jr., and W. J. Martin II. 1997. Surfactant protein A (SP-A) mediates attachment of Mycobacterium tuberculosis to murine alveolar macrophages. Am. J. Respir. Cell Mol. Biol. 17:209-217.
36. Pinheiro, J., and D. Bates. 2000. Mixed effects models in S and SPLUS. Springer, New York, N.Y.
37. Purevdorj, B., J. W. Costerton, and P. Stoodley. 2002. Influence of hydrodynamics and cell signaling on the structure and behavior of Pseudomonas aeruginosa biofilms. Appl. Environ. Microbiol. 68:4457-4464.
38. R Development Core Team. 2005. A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. [Online.] http://www.R-project.org.
39. Ratliff, T. L., J. A. McGarr, and C. Abou-Zeid. 1988. Attachment of mycobacteria to fibronectin-coated surfaces. J. Gen. Microbiol. 134:1307-1313.
40. Ratliff, T. L., R. McCarthy, W. B. Telle, and E. J. Brown. 1993. Purification of a mycobacterial adhesin for fibronectin. Infect. Immun. 61:1889-1894.
41. Reddy, K., and J. M. Ross. 2001. Shear stress prevents fibronectin binding protein-mediated Staphylococcus aureus adhesion to resting endothelial cells. Infect. Immun. 69:3472-3475.
42. Roselli, R. J., G. Tack, and T. R. Harris. 1997. A model of fluid, erythrocyte, and solute transport in the lung. Ann. Biomed. Eng. 25:46-61.
43. Savla, U., and C. M. Waters. 1998. Mechanical strain inhibits repair of airway epithelium in vitro. Am. J. Physiol. 274:L883-L892.
44. Schlesinger, L. S., C. G. Bellinger-Kawahara, N. R. Payne, and M. A. Horwitz. 1990. Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte complement receptors and complement component C3. J. Immunol. 144:2771-2780.
45. Schlesinger, L. S., S. R. Hull, and T. M. Kaufman. 1994. Binding of the terminal mannosyl units of lipoarabinomannan from a virulent strain of Mycobacterium tuberculosis to human macrophages. J. Immunol. 52:4070-4079.
46. Schorey, J. S., Q. Li, D. W. McCourt, M. Bong-Mastek, J. E. Clark-Curtiss, T. L. Ratliff, and E. J. Brown. 1995. A Mycobacterium leprae gene encoding a fibronectin binding protein is used for efficient invasion of epithelial cells and Schwann cells. Infect. Immun. 63:2652-2657.
47. Schorey, J. S., M. A. Holsti, T. L. Ratliff, P. M. Allen, and E. J. Brown. 1996. Characterization of the fibronectin-attachment protein of Mycobacterim avium reveals a fibronectin-binding motif conserved among mycobacteria. Mol. Micro. 21:321-329.
48. Shenkman, B., E. Rubinstein, A. L. Cheung, G. E. Brill, R. Dardik, I. Tamarin, N. Savion, and D. Varon. 2001. Adherence properties of Staphylococcus aureus under static and flow conditions: roles of agr and sar loci, platelets, and plasma ligands. Infect. Immun. 69:4473-4478.
49. Sidobre, S., J. Nigou, G. Puzo, and M. Riviere. 2000. Lipoglycans are putative ligands for the human pulmonary surfactant protein A attachment to mycobacteria. J. Biol. Chem. 275:2415-2422.
50. Stoodley, P., L. Hall-Stoodley, and H. M. Lappin-Scott. 2001. Detachment, surface migration, and other dynamic behavior in bacterial biofilms revealed by digital time-lapse imaging. Methods Enzymol. 337:306-319.
51. Strong, P., U. Kishore, C. Morgan, A. Lopez Bernal, M. Singh, and K. B. Reid. 1998. A novel method of purifying lung surfactant proteins A and D from the lung lavage of alveolar proteinosis patients and from pooled amniotic fluid. J. Immunol. Methods 220:139-149.
52. Tzima, E., M. A. del Pozo, S. J. Shattil, S. Chien, and M. A. Schwartz. 2001. Activation of integrins in endothelial cells by fluid shear stress mediates Rho-dependent cytoskeletal alignment. EMBO J. 3:4639-4647.
53. West, J. B. 1974. Respiratory physiology: the essentials, p. 1-13. Williams and Wilkins, Baltimore, Md.
54. Zhao, W., J. S. Schorey, R. Groger, P. A. Allen, E. J. Brown, and T. L. Ratliff. 1999. Characterization of the fibronectin binding motif for a unique mycobacterial fibronectin attachment protein, FAP. J. Biol. Chem. 274:4521-4526.(Luanne Hall-Stoodley, Gay)