Phosphatidylinositol Mannoside from Mycobacterium tuberculosis Binds α5β1 Integrin (VLA-5) on CD4+ T Cells and Induces Adhesion to Fibronectin

Roxana E. Rojas, Jeremy J. Thomas, Adam J. Gehring, Preston J. Hill, John T. Belisle, Clifford V. Harding and W. Henry Boom
J Immunol September 1, 2006, 177 (5) 2959-2968; DOI: https://doi.org/10.4049/jimmunol.177.5.2959

Abstract

The pathological hallmark of the host response to Mycobacterium tuberculosis is the granuloma where T cells and macrophages interact with the extracellular matrix (ECM) to control the infection. Recruitment and retention of T cells within inflamed tissues depend on adhesion to the ECM. T cells use integrins to adhere to the ECM, and fibronectin (FN) is one of its major components. We have found that the major M. tuberculosis cell wall glycolipid, phosphatidylinositol mannoside (PIM), induces homotypic adhesion of human CD4+ T cells and T cell adhesion to immobilized FN. Treatment with EDTA and cytochalasin D prevented PIM-induced T cell adhesion. PIM-induced T cell adhesion to FN was blocked with mAbs against α5 integrin chain and with RGD-containing peptides. α5β1 (VLA-5) is one of two major FN receptors on T cells. PIM was found to bind directly to purified human VLA-5. Thus, PIM interacts directly with VLA-5 on CD4+ T lymphocytes, inducing activation of the integrin, and promoting adhesion to the ECM glycoprotein, FN. This is the first report of direct binding of a M. tuberculosis molecule to a receptor on human T cells resulting in a change in CD4+ T cell function.

The pathological hallmark of the host response to Mycobacterium tuberculosis is the granuloma. The granuloma is the site where the host contains M. tuberculosis and where the organism can evade the host and persist (1, 2). The tuberculous granuloma is an organized collection of differentiated macrophages surrounded by T lymphocytes, some B lymphocytes, dendritic cells, fibroblasts, and extracellular matrix (ECM)3 components (3). Formation of granulomas is poorly understood, and little is known about how lymphocytes migrate and are retained in the outermost region of the granuloma. The role of ECM proteins in this process is also undefined. While cytokines and chemokines, induced by mycobacterial infection of macrophages, promote lymphocyte recruitment (4, 5), there may be a role for mycobacterial molecules as direct signals for lymphocyte adhesion and retention at sites of infection.

Activation of T lymphocytes by M. tuberculosis is generally thought to be mediated primarily by APC. Although the majority of M. tuberculosis bacilli reside within macrophages, mycobacterial components readily can gain access to the extracellular environment (6, 7). Vesicles containing mycobacterial cell wall lipids can be exocytosed by macrophages, and mycobacterial proteins have been found in serum, circulating immune complexes, and cerebrospinal fluid (8, 9, 10, 11, 12). Thus, mycobacterial molecules could interact directly with T cells affecting functions such as adhesion and migration.

The ability to adhere to and migrate through ECM is a basic function of T lymphocytes. Recruitment and retention of T lymphocytes within inflamed tissues are dependent on adhesion to the ECM (13, 14, 15, 16, 17, 18). The ECM consists of a network of proteins, glycoproteins, and proteoglycans. Fibronectin (FN), a high molecular mass glycoprotein, is one of the major components of the ECM (19, 20, 21, 22). FN is secreted by many different cells and is present in plasma, connective tissue, and on cell surfaces (23, 24). Adhesion of T cells to FN and other ECM components is mediated by integrins. Integrins are heterodimeric transmembrane proteins with important roles in cell-cell and cell-matrix interactions (25, 26). Each integrin consists of noncovalently linked α and β subunits. At present, 24 αβ pairs have been identified in mammals, and T cells can express at least 12 of the 24 known heterodimers (16, 27, 28). Among the β2 integrins present in T cells, αLβ2 (leukocyte function-associated Ag-1) is abundant and broadly expressed. T cells can also express the β7 integrins (α4β7 and αEβ7) and, in common with other cell types, the ECM binding β1 integrins (α1-α6β1) (29, 30, 31, 32). α4β1 (VLA-4) and α5β1 (VLA-5) are the two major FN receptors expressed on T cells (33, 34). α5β1 integrin and FN form a prototypic integrin/ligand pair that mediate FN adhesion and FN matrix assembly, which are important for many in vivo cellular functions. β1 integrins are preferentially expressed by CD45RO+ memory T cells (28), and α5β1 is found on ∼50% of peripheral human T cells (35). α5β1 integrin expression and FN are found in granuloma tissues (36, 37).

Most integrins are expressed on cell surfaces in an inactive state in which they do not bind to ligands and do not signal. Integrins are tightly regulated, and both affinity modulation (conformational changes) and avidity modulation (clustering) determine their activation state (38, 39). Once activated, integrins bind their ligands and induce lymphocyte functions such as adhesion, migration, differentiation, and costimulation. Nonactivated lymphocytes are nonadherent and, in response to chemokines or Ag, become adherent to other cells and components of the ECM. Bacterial molecules, such as Yersinia pseudotuberculosis invasin and Escherichia coli intimin, can directly bind β1 integrins on T cells and trigger T cell migration to ECM proteins (40, 41, 42). We have found that the mycobacterial glycolipid phosphatidylinositol mannoside (PIM) interacts directly with the α5β1 integrin VLA-5 on CD4+ T lymphocytes, resulting in FN binding and T cell adhesion. The present work constitutes the first report of a M. tuberculosis molecule directly interacting with a receptor on T cells in the absence of APC. Implications for granuloma formation and the pathogenesis of mycobacterial infection will be discussed.

Materials and Methods

Abs, purified proteins, and glycolipids

To assess purity of CD4+ T cell populations after cell sorting, PE-conjugated Leu-4 (CD3-PE), FITC-conjugated Leu-3a (CD4-FITC), Leu-2a (CD8-FITC), FITC-labeled anti-Vδ2 TCR (clone B6.1), PE-conjugated anti-αβTCR, and FITC- or PE-conjugated isotypic control Abs were purchased from BD Pharmingen. Blocking mAbs against integrin chains (Chemicon International) were as follows: anti-CD49d (clone P1H4), anti-CD49e (clones P1D6, JBS5, and SAM-1), anti-CD29 (clones 6S6 and P4C10), anti-CD11a (clone 38). Anti-CD49d (clone 9F10) was obtained from eBioscience. PE-conjugated anti-CD49e (clone SAM-1) was used to assess α5 expression. Abs used for immunoprecipitation were as follows: mAb anti-CD49e (clone HA5, Chemicon International; clone 1, BD Biosciences), mAb anti-FN Ab (clone P1H111; Santa Cruz Biotechnology), and mouse IgG (Sigma-Aldrich). Immunoprecipitated proteins were detected with anti-CD49e polyclonal Ab (Chemicon International), anti-FN mAb (clone 3E3; Chemicon International), and HRP-conjugated mouse anti-rabbit or HRP-conjugated goat anti-mouse IgG + IgM (Jackson ImmunoResearch Laboratories), respectively. For detection of PIM by Western blotting and ELISA, we used polyclonal rabbit anti-M. bovis-bacillus Calmette Guérin (BCG) serum (DakoCytomation), followed by HRP-conjugated donkey anti-rabbit IgG (Amersham Biosciences). Preliminary experiments demonstrated that anti-M. bovis-BCG serum reacted with purified PIM1, PIM2, and PIM6 blotted onto nitrocellulose membranes or coated onto 96-well plates.

Purified α5β1 integrin and plasma FN were purchased from Chemicon International. The following mycobacterial TLC-purified glycolipids were obtained through the Tuberculosis Vaccine Testing and Research Materials Contract at Colorado State University: PIM from M. tuberculosis H37Rv, mannose-capped lipoarabinomannan (ManLAM) from M. tuberculosis H37Rv, and non-mannose-capped lipoarabinomannan (smegLAM) from M. smegmatis. Phosphatidylinositol from soybean (SPI) was purchased from Sigma-Aldrich.

Cells and medium

Unless otherwise specified, cells were cultured at 37°C and 5% CO2. T cells were maintained in X-VIVO 15 (BioWhittaker) supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin (BioWhittaker). When indicated, medium was supplemented with 10% pooled human serum. PBMC were obtained from healthy volunteers under an Institutional Review Board-approved protocol and prepared from whole blood with Ficoll-Paque Plus (Amersham Biosciences). To purify CD4+ T cells, PBMC were first depleted of monocytes by adherence. Remaining monocytes/macrophages and cytotoxic lymphocytes, including CD3−CD16+ NK cells and CD8+ T cells, were eliminated by treatment with the lysosomotropic agent leucine methyl ester (Leu-OMe, 1 mM). CD4+ T cells then were purified by negative selection with Ab-coated magnetic microbeads (Miltenyi Biotec). Purity was confirmed by FACS. One cycle of selection was sufficient to obtain 95–97% CD4+ T cells and >99% CD3+ T cells. Alternatively, T cells were purified by BD FACSAria Cell-Sorting System (BD Biosciences), yielding >99% CD4+CD3+ T cells.

Culture of mycobacteria and preparation of mycobacterial fractions

Cell wall/membrane material and cytosol from M. tuberculosis H37Ra and BCG was prepared and fractionated by biphasic extraction with Triton X-114, followed by preparative electroelution as described before (43). Thirty fractions were obtained. Fractions containing low molecular mass material, assessed by Western blotting with anti-M. bovis-BCG serum, were pooled, and further separated by electrophoresis in Novex Tricine SDS loading buffer on a Novex 16% Tricine two-dimensional gel (Invitrogen Life Technologies). Electrophoresis was performed in Novex Tricine SDS running buffer for 2 h at 125 V. Material was eluted in 60 mM Tris 40 mM CAPS (3-(cyclohexylamino)-1-propanesulfonic acid, 10% methanol v/v (pH 11.0)) using a mini whole gel eluter (Bio-Rad). Fourteen fractions were collected, analyzed by Western blotting with anti- M. bovis-BCG serum, and stored at −80°C until use.

Western blot and TLC analysis of mycobacterial products

Cell wall/membrane fractions were separated by SDS-PAGE in Novex 16% Tricine gels and transferred to nitrocellulose membranes. TLC-purified PIM and LAM were used as controls. Membranes were probed with polyclonal rabbit anti-M. bovis-BCG serum (DakoCytomation), followed by labeling with anti-rabbit IgG-HRP (Amersham Biosciences). Detection of bands was achieved with chemiluminescence using SuperSignal West Pico (Pierce). TLC analysis of fractions obtained by electroelution was performed on aluminum backed plates of silica gel 60 F254 (Merck). The solvent used was CHCl3/CH3OH/NH4OH/H2O (65:24:0.4:3.6 v/v). Visualization of resolved lipids was achieved with a charring spray (10% CuSO4 in 8% phosphoric acid) or a carbohydrate detection spray (1% α-napthol in 5% H2SO4), and heating at 110°C.

Homotypic T cell adhesion (HTCA) assay

CD4+ T cells were resuspended in X-VIVO-15 supplemented with 10% pooled human serum at 5 × 105 cells/ml. For experiments comparing different medium, cells were washed with buffer containing EDTA and resuspended in RPMI 1640 medium supplemented either with 50 μg/ml BSA, 50 μg/ml FN, or 10% pooled human serum. Cells were incubated in 48-well tissue culture plates at 150,000 cells/well. Mycobacterial cell wall/membrane fractions were obtained as mentioned before and homogenized (6 mg dried weight) by addition of 20% DMSO in RPMI 1640 medium, vortexing, and sonication for 30 min. The suspension was centrifuged and supernatant added to T cells in final dilutions of 1/30 or 1/300. Cytosol (6 mg protein/ml) was added in final concentrations of 200 and 20 μg/ml. RPMI 1640 medium served as control. Electroelution fractions were used in final dilutions of 1/10, 1/100, and 1/1000. Purified PIM, ManLAM, smegLAM, and SPI were added in concentrations between 0.01 and 10 μg/ml. T cells were observed under microscope every 30 min in a Nikon Diaphot 200 inverted microscope (×100, ×200) with contrast phase and photographed (2 and 24 h) with a RT color Spot camera (Diagnostics Instruments) and the SPOT Advanced software. Relative levels of aggregation were determined visually. Alternatively, cells were transferred to polystyrene tubes with minimal pipetting, centrifuged, fixed with 1% paraformaldehyde, and acquired with a FACSCalibur instrument (BD Biosciences). Samples were analyzed with FlowJo software (Tree Star). Presence of aggregates by microscopy correlated with detection of events with increased forward (FSC) and side scatter (SSC) by FACS. Two populations of events were detected in cultures exhibiting HTCA, and the corresponding gates were drawn: low FSC-SSC events (gate 1) corresponding to single cells and high FSC-SSC events (gate 2) corresponding to cell aggregates. Percentage of events in gate 2 was a measure of intensity of HTCA (percent HTCA).

T cell adhesion to FN-coated plates

CD4+ T cells in X-VIVO were labeled with 10 μM calcein-AM (Molecular Probes) for 1 h and 10 min. Cells were washed and resuspended in Dulbecco PBS (DPBS) supplemented with Ca2+ and Mg2+ (BioWhittaker). For blocking experiments, labeled cells were pretreated with anti-integrin Abs (20 μg/ml), control IgG (20 μg/ml), RGD motif-containing peptide (1 mM GRGDSPK; Sigma-Aldrich), or control peptide (1 mM GRADSPK; Sigma-Aldrich) for 30 min. Cells then were incubated in FN (1 μg/well)-coated 96-well flat-bottom microtiter plates (Immulon 4 HBX). SmegLAM, ManLAM, PIM, or SPI were added to cells at final concentrations of 10 or 50 μg/ml. Cells treated with buffer only served as control for spontaneous adhesion. Plates were spun for 2 min, 1200 rpm at 4°C. After 2 or 6 h of incubation at 37°C in the dark, nonadherent cells were removed by carefully washing with room temperature DPBS-Ca2+Mg2+. A standard curve was generated with known numbers of labeled cells. Cells used for the standard curve and cells used in the adhesion assay were labeled with calcein-AM at the same time and washed extensively. Fluorescence was detected with 485-nm excitation and 535-nm emission filters in a 96-well plate fluorescence reader (Spectrafluor Plus; Tecan). Adherent cell numbers were calculated by extrapolating OD values to the standard curve.

Coprecipitation of α5β1 integrin and mycobacterial PIM

Purified human α5β1 integrin (2.5 μg; Chemicon International) and purified plasma FN (2.5 μg; Chemicon International) were incubated separately with mycobacterial PIM (5 μg) in serum-free RPMI 1640 medium supplemented with 10 mM HEPES (pH 7.4) and 0.02% BSA. Monoclonal anti-α5 integrin chain (2.5 μg of clone HA5 plus 2.5 μg of clone 1) or anti-FN (5 μg of clone P1H111) Abs were added, respectively, and incubated with mixing at 4°C for 4 h. As control, PIM was incubated with monoclonal anti-α5 and anti-FN Abs or an irrelevant IgG Ab in absence of purified α5β1 integrin or FN. Complexes were immunoprecipitated with protein G-Sepharose beads (Amersham Biosciences). Beads were washed three times with RPMI 1640 supplemented with 10 mM HEPES and 0.02% BSA, twice with PBS, and once with 50 mM Tris (pH 8). Bound materials were extracted in reducing SDS Tris-glycine sample buffer and analyzed by SDS-PAGE on Novex 4–20% Tris-glycine gels (Invitrogen Life Technologies), followed by transfer to nitrocellulose membranes. Purified proteins (α5β1 integrin and FN) and PIM were included as controls. Membranes then were probed with anti-M. bovis-BCG polyclonal Ab, followed by anti-rabbit-HRP, and bands were detected by chemiluminescence using SuperSignal West Pico (Pierce). After stripping, membranes were reprobed with either anti-α5β1 polyclonal Ab or anti-FN mAb, followed by anti-rabbit-HRP or anti-mouse-HRP, and bands were detected as described above.

Solid-phase binding of PIM to purified α5β1 integrin

A modified solid-phase ligand receptor binding was developed according to the methods of Frankel et al. (41). Purified integrins were diluted with PBS containing divalent cations (0.2–5 μg/ml), and 50-μl aliquots were added to the wells of Immulon HBX 96-well plates (Dynatech Laboratories). Plates were incubated 1 h at 37°C, followed by overnight incubation at 4°C, and then blocked with 200 μl of 5% BSA, 150 mM NaCl, and 10 mM Tris-HCl (pH 7.4) for 1 h at room temperature. Wells were washed twice with 200 μl of 150 mM NaCl, 5 mM MnCl2, 25 mM Tris-HCl (pH 7.4), and 1 mg/ml BSA (buffer A). Fifty-microliter aliquots of 1–5 μg/ml PIM in buffer A were added. The plates then were incubated at 30°C for 3 h. Unbound PIM was aspirated, and the wells were washed six times with buffer A. Anti-M. bovis-BCG then was incubated for 1 h at room temperature, followed by anti-rabbit-HRP Ab. Wells were extensively washed, and bound PIM was detected by addition of 50 μl of 3,3′,5,5′-tetramethylbenzidine (Sigma-Aldrich). The level of nonspecific binding was measured in every experiment by determining the level of binding to wells coated with BSA alone. These values are shown Results. As control, α5β1 bound to plastic was detected with anti-CD49e mAb (clone 1; BD Biosciences), followed by anti-mouse-HRP Ab. In blocking experiments, 10 μg/ml mAb JBS5 or control Ab was incubated for 1 h, and unbound Ab was removed by washing before addition of PIM. For each condition, binding was calculated as a percentage over the maximum binding obtained without Ab addition and at the highest α5β1 integrin concentration. Absorbance was measured at 450 nm (A450) in a microplate reader (VersaMax; Molecular Devices).

Results

M. tuberculosis cell wall and membrane fractions induce HTCA

To determine whether components of M. tuberculosis bacilli could directly affect human T cell adhesion, we used highly purified peripheral blood CD4+ T cells in a HTCA assay. We focused our study on CD4+ T cells because they are central to the immune response to M. tuberculosis and are readily purified from whole blood. Depletion of macrophages from PBMC was performed by adhesion to plastic followed by lysis in 1 mM l-leucine methyl ester. CD4+ T cells then were isolated by negative selection with mAb-coated magnetic beads. This yielded CD4+ T cells of ≥95% purity with 99% of cells expressing CD3.

Purified T cells then were incubated with either the cell wall/membrane or cytosolic fractions of French-pressed M. tuberculosis-H37Ra and M. bovis BCG bacilli and assessed for T cell-to-T cell adhesion (i.e., HTCA). These experiments demonstrated that H37Ra and BCG cell wall/membrane fraction induced HTCA as detected by microscopy (Fig. 1⇓C). The cytosolic fraction did not induce homotypic adhesion when added up to 200 μg protein/ml (Fig. 1⇓B). For more objective quantification of HTCA, flow cytometry at low flow rates was used to detect T cell aggregates as increased forward and side scatter (FSC vs SSC) (Fig. 1⇓, D–F). T cell aggregates formed within 1 h of addition of cell wall/membrane fractions and remained stable 24 h (Fig. 1⇓F). To exclude the possibility that small numbers of non-T cells were responsible for the observed aggregation, two approaches were used. First, flow-sorted CD4+ T cells (>99% purity) also aggregated with cell wall/membrane fraction. Second, when autologous monocytes were added to purified T cells, in fact, decreased aggregation was measured (data not shown). Therefore, mycobacterial cell wall/membrane associated component(s) induced T cell-to-T cell adhesion in the absence of APC, indicating a direct effect on CD4+ T cells.

FIGURE 1.
FIGURE 1.

Cell wall/membrane fraction of M. tuberculosis induces HTCA. Pelleted M. tuberculosis bacilli were passed through a French press and separated into cell wall/membranes and cytosol. Cell wall/membranes were homogenized in DMSO-PBS and insoluble material removed by centrifugation. Purified CD4+ T lymphocytes (1.5 × 105 cells/well) were cultured in 48-well plates with medium alone (A and D), medium with cell wall/membrane homogenates (C and E), or medium with cytosol (B). HTCA was detected by microscopy (A–C) or flow cytometry (D–F) after 2 h of incubation with cell wall/membranes or cytosol (200 μg protein/ml). Flow cytometry revealed two cell populations based on FSC and SSC: single cells (G1) and aggregated cells (G2) (D and E). HTCA by cell wall/membrane fractions, as measured by flow cytometry, persisted for 24 h (F). One of three representative experiments is shown. Photo magnification is ×200.

M. tuberculosis fractions that cause HTCA contain PIM

To identify the molecule(s) responsible for HTCA, fractions from M. tuberculosis cell walls were prepared by biphasic extraction with TX-114. Detergent extracted material was resolved by electrophoresis in a 12.5% SDS polyacrylamide slab gel, and size separated components eluted with a whole gel eluter. Fractions containing low molecular mass components (<15 kDa) with low affinity for gel silver stain (fractions 22–29) induced HTCA by microscopy and flow cytometry (Fig. 2⇓, A and B). Improved resolution of low molecular mass bands was obtained with 16% Tricine gels. Fractions 22–29 from the first elution were pooled and re-electroeluted from a preparative 16% Tricine mini gel into 14 new fractions. As shown in Fig. 2⇓C, Western blotting with anti-BCG serum identified two bands < 6 kDa in fractions with HTCA activity (fractions 11–14). Fractions 1–10 contained higher molecular mass material and did not have HTCA activity.

FIGURE 2.
FIGURE 2.

HTCA by mycobacterial cell wall/membranes is associated with the presence of PIM. Cell walls were fractionated by biphasic extraction with Triton X-114, followed by preparative electroelution. A, Twenty-nine 1-ml fractions were collected, resolved in a 12.5% analytic SDS-PAGE gel, and silver stained (fractions 17–29 shown). Arrow indicates low molecular mass bands. B, CD4+ T cells (1.5 × 105/well) were incubated with electroeluted fractions at 1/100 dilution overnight, and HTCA evaluated by flow cytometry. Percentage of HTCA (events in aggregate gate G2) is shown in one representative experiment for fractions 17 and 21–29. C, Fractions inducing HTCA were pooled and electroeluted from a 16% Tricine preparative min-gel. Fourteen fractions were collected and analyzed by Western blotting with anti-BCG serum. Low molecular mass bands (4–6 kDa, arrows) were detected in fractions with HTCA activity (11–14). Inactive fractions contained higher molecular mass bands (1–10). D, TLC of active fractions from two different electroelutions (EE1 and EE2) compared with PIM standards. Lane 1, PIM1 std; lane 2, EE1; lane 3, PIM2 std; lane 4, EE2; and lane 5, PIM6 std.

Localization to cell wall and membranes, partition to detergent phase, very low molecular mass, and poor labeling by Silver stain (unless oxidizing step performed) strongly suggested that HTCA was due to a glycolipid, with PIM as the major candidate. Fractions with HTCA activity were analyzed by TLC. As shown in Fig. 2⇑D, TLC detected PIM1, PIM2, and PIM6 in pools of active fractions isolated by electroelution from M. tuberculosis H37Ra lysate. To prove that PIM could induce HTCA, PIM1 and PIM2 from M. tuberculosis H37Rv were tested in our HTCA assay. TLC-purified PIM induced HTCA in concentrations ≥ 1 μg/ml within 1 h of addition (Fig. 3⇓A).

FIGURE 3.
FIGURE 3.

PIM but not LAM induces HTCA. A, CD4+ T cells (1.5 × 105/well) were treated for 2 h with different concentrations of TLC-purified PIM from M. tuberculosis (0.01–10 μg/ml). HTCA (events in gate 2) was measured by flow cytometry. B, CD4+ T cells were left untreated or treated with 1 or 10 μg/ml of either PIM or ManLAM from M. tuberculosis (H37Rv), smegLAM from M. smegmatis, or SPI. T cell aggregation was detected by flow cytometry and expressed as percent HTCA (events in gate 2). Data from duplicate samples from a representative experiment of three are shown as mean ± SD, and significantly different percentages of events in gate 2 compared with control are indicated (∗, p < 0.05; ∗*, p < 0.005).

To determine the role of mannosylation of phosphatidylinositol in HTCA, SPI was tested and found to be inactive (Fig. 3⇑B). Next, the role of oligo- vs multiglycosylation of phosphatidylinositol in HTCA was assessed. LAM is the multiglycosylated form of PIM and was tested as ManLAM from M. tuberculosis H37Rv and smegLAM. Neither ManLAM nor smegLAM induced HTCA in concentrations from 100 ng/ml to 40 μg/ml (Fig. 3⇑B). In conclusion, bioactivity of low molecular mass fractions obtained from TX-114 extraction, followed by electroelution from M. tuberculosis or M. bovis lysates, was attributable to PIM. Neither the nonmannosylated phosphatidylinositol nor the multiglycosylated forms of PIM induced HTCA, indicating that both the presence of mannose residues and the size of the glycosidic chain were important for T cell adhesion.

PIM-induced HTCA depends on FN, divalent cations and an intact cytoskeleton

To characterize PIM-induced HTCA, CD4+ T cells were treated with a combination of PIM1, PIM2, and PIM6 in the presence or absence of serum or FN. HTCA was dependent on the presence of serum or plasma FN (Fig. 4⇓). Both human and bovine serum supported HTCA by PIM. Medium supplemented with BSA did not support the formation of aggregates by PIM (Fig. 4⇓). Medium supplemented with serum or FN in the absence of PIM did not induce HTCA. Cells that were stripped of surface FN with EDTA-containing buffer, treated with PIM in the absence of serum and washed, aggregated upon resuspension in serum-containing medium (data not shown). This suggested that PIM interacted directly with the T cells and not with FN.

FIGURE 4.
FIGURE 4.

PIM-induced HTCA depends on the presence of serum or FN. CD4+ T cells were washed in EDTA-containing buffer and resuspended in RPMI 1640 medium with 50 μg/ml BSA (A and D), RPMI 1640 medium with 50 μg/ml FN (B and E), or RPMI 1640 medium with 10% pooled human serum (C and F). Cells were incubated (150,000 cells/well) in 48-well plates and treated with (D–F) or without (A–C) a mixture of PIM1, PIM2, and PIM6 (10 μg/ml). After overnight incubation, cells were observed by contrast phase microscopy and pictures taken (×200). One of three representative experiments is shown.

FN is a ligand for β1 integrins and their interactions depend on the presence of divalent cations (review in Plow, 2000) (44). PIM-mediated HTCA was inhibited in the presence of the divalent cation chelator EDTA at 5 mM (Fig. 5⇓). PIM-induced HTCA also was abolished by pretreatment of T cells with cytochalasin D (0.8 μM), an antagonist of actin polymerization, indicating that an intact cytoskeleton was required for HTCA induced by PIM (Fig. 5⇓). Thus, HTCA required divalent cations and an intact cytoskeleton. PIM-induced HTCA depended on FN, suggesting the involvement of β1 integrins.

FIGURE 5.
FIGURE 5.

PIM induced HTCA is blocked by cytochalasin D and EDTA. CD4+ T cells (1.5 × 105/well) were preincubated for 30 min in medium with or without 0.8 μM cytochalasin D or 5 mM EDTA and then treated with or without a mixture of PIM1, PIM2, and PIM6 (10 μg/ml) for 2 h. HTCA was detected by flow cytometry. Percentage of HTCA (events in aggregate gate 2) was calculated. Data from duplicate samples from a representative experiment of two are shown as mean ± SD, and significantly different percentages of events in aggregate gate are indicated (∗, p < 0.05).

PIM induces VLA-5-dependent/RGD-dependent T cell adhesion to immobilized FN

FN is a major ECM protein, also found in granulomas, that interacts with β1 integrins on the surface of different cells, resulting in their adherence to ECM. To determine the effect of PIM on T cell adhesion to FN-coated plates and involvement of β1 integrins, purified CD4+ T cells were labeled with calcein-AM (10 μM) and incubated on FN-coated plates (1 μg/well) in the presence or absence of a combination of PIM1, PIM2, and PIM6 (50 μg/ml). PMA induces T cell adhesion to FN and served as positive control. CD4+ T cells treated for 2 h with PIM adhered to plate-bound FN (Fig. 6⇓C). PMA-induced adhesion peaked at 1–2 h after addition of T cells to immobilized FN and was down-regulated by 6 h. PIM-induced T cell adhesion to FN was maintained for 6 h (Fig. 6⇓D). Neither PIM nor PMA induced T cell adhesion to BSA-coated plates (Fig. 6⇓E). In addition, neither SPI nor smegLAM or ManLAM induced T cell adhesion to immobilized FN, indicating that the presence of mannose residues and the size of the glycosidic chain determined this biological activity of PIM (Fig. 6⇓E).

FIGURE 6.
FIGURE 6.

PIM induces T cell adhesion to immobilized FN. A–C, CD4+ T cells resuspended in PBS with Ca2+Mg2+ and incubated in FN (1 μg/well)-coated 96-well plates in medium alone (A), PMA (10 ng/ml) (B), or PIM (50 μg/ml) (C). Plates were washed after 2 h to remove unbound cells and examined by phase-contrast microscopy (photos, ×100). D, CD4+ T cells were labeled with calcein-AM (10 μM), resuspended in DPBS with Ca2+Mg2+, and incubated for 2–6 h in FN (1 μg/well)-coated 96-well plates with medium alone, PMA (10 ng/ml), or PIM (50 μg/ml). Unbound cells were removed by washing, and fluorescence of adherent cells were detected with a microplate reader (485-nm excitation, 535-nm emission filters). Number of adherent cells was calculated by interpolation of OD values in the standard curve. Data from triplicate samples from a representative experiment of five are shown as mean ± SD; significantly different numbers of adherent cells compared with control are indicated (∗, p < 0.005; ∗*, p < 0.0005). E, Fluorescently labeled cells were incubated in plate-bound FN (pb FN, 1 μg/well)-coated wells or wells blocked with BSA (pb BSA, 5 mg/ml) in the presence or absence of 10 ng/ml PMA or glycolipids (50 μg/ml), SPI, ManLAM, smegLAM, or PIM. Unbound cells were removed after 2 h, and fluorescence was measured as described before. Data from triplicate samples from a representative experiment of four are shown as mean ± SD; significantly different numbers of adherent cells compared with control are indicated (∗*, p < 0.0005).

VLA-4 (α4β1) and VLA-5 (α5β1) are the major FN receptors on T cells. To further define the involvement of β1 integrins in PIM-induced T cell adhesion to immobilized FN, blocking mAbs were used to determine the role of the α4, α5, and β1 integrin chains. First, we determined the ability of the different mAbs to block T cell adhesion to FN induced by PMA. As shown in Fig. 7⇓A, all tested mAbs partially blocked T cell adhesion to FN induced by PMA. In contrast, only anti-α5 mAb JBS5 blocked adhesion of T cells to immobilized FN induced by PIM and did so completely (Fig. 7⇓B). Cells pretreated with one of two anti-β1 mAbs (clone 6S6) tended toward lower PIM-induced adhesion compared with control Ab-treated cells but did not reach statistical significance. mAbs to the α4 chain of VLA-4, the other major receptor for FN, did not block PIM-induced T cell adhesion to immobilized FN (Fig. 7⇓A) or HTCA (data not shown).

FIGURE 7.
FIGURE 7.

Abs to α5 integrin chain and RGD motifs block PIM- and PMA-induced T cell adherence to immobilized FN. Fluorescently labeled cells pretreated (30 min) with mAb against different integrin chains (20 μg/ml) or with control Ab (20 μg/ml) were incubated in FN (1 μg/well)-coated wells in the presence or absence of PMA (10 ng/ml) (A) or a mixture of PIM1, PIM2, and PIM6 (50 μg/ml) (B). Alternatively, cells were pretreated with or without GRGDSPK or GRADSPK peptides (1 mM) (C). After 2 h of incubation, unbound cells were removed by washing, and adherent cells were quantified by assessing fluorescence in a 96-well plate reader (485-nm excitation, 535-nm emission filters). Number of adherent cells was calculated by interpolation of OD values in a standard curve. Data from triplicate samples from a representative experiment (A and B, n = 4; C, n = 2) are shown as mean ± SD; significantly different numbers of adherent cells compared with control Ab-treated (A and B) or untreated (C) cells are indicated (∗, p < 0.05; ∗*, p < 0.001; ∗*∗, p < 0.0001; ∗*∗*, p < 0.00001).

VLA-5, but not VLA-4, bind FN by interacting with the short amino acid sequence RGD. Soluble, small RGD-containing peptides can block integrin binding to FN and interfere with cell adhesion. As shown in Fig. 7⇑C, preincubation of T cells with the RGD-containing peptide GRGDSPK, but not with the analog peptide GRADSPK, inhibited PIM-induced T cell adhesion to FN. In addition, PIM-induced T cell adhesion to immobilized FN correlated with α5β1 expression on T cells in different donors (data not shown), suggesting that PIM effects depend on the expression of VLA-5 on T cells. Thus, PIM induced T cells to adhere to immobilized FN through its α5β1 integrin VLA-5, and binding was dependent on RGD motifs.

PIM binds directly to purified α5β1 integrin

T cells pretreated with PIM and then washed extensively to eliminate unbound PIM were still able to aggregate when serum was added to the washed cells, suggesting direct binding of PIM to the T cells. In addition, results described above indicate a role for α5β1 integrin (VLA-5) in T cell binding to FN. To determine whether PIM was interacting directly with α5β1 integrin, immunoprecipitations were performed with purified soluble α5β1 integrin. A mixture of PIM1, PIM2, and PIM6 was allowed to interact with either medium alone, purified FN, or α5β1 integrin, followed by immunoprecipitation with bead-bound mouse IgG, anti-FN, or anti-α5 integrin chain mAbs. Immunoprecipitated material was analyzed by Western blot with anti-BCG serum. Membranes were stripped and reprobed with either anti-FN or anti-α5 Abs. As shown in Fig. 8⇓A, PIM coprecipitated with α5β1 and not with FN. Control experiments demonstrated that PIM did not precipitate above background levels with anti-α5 Abs in the absence of α5β1 integrin. These results demonstrate that PIM can bind directly to VLA-5 and does not bind to FN. Further confirmation of the results from the immunoprecipitation experiments was obtained by developing a solid-phase binding assay. In this assay, plates were coated with purified α5β1 integrin, incubated with the mixture of PIM, washed extensively, and then probed for the presence of bound PIM with the anti-BCG serum. As shown in Fig. 8⇓B, PIM were able to bind in a dose-responsive manner to increasing concentrations of plate bound α5β1 integrin. Furthermore, the anti-α5 Ab JBS5 blocked the interaction between α5β1 integrin and PIM (Fig. 8⇓C). Overall, our results suggest that binding of PIM to VLA-5 may promote FN binding and thus CD4+ T cell adhesion to ECM.

FIGURE 8.
FIGURE 8.

Mycobacterial PIM binds directly to human α5β1 integrin. A, A mixture of PIM1, PIM2, and PIM6 (5 μg) was incubated with buffer alone (lanes 1 and 3), purified FN (FN, 2.5 μg; lane 2), or purified α5β1 integrin (2.5 μg; lane 4), followed by mAb against FN (5 μg; lanes 1 and 2) or mAbs against α5 chain (5 μg; lanes 3 and 4). Complexes were immunoprecipitated (IP) with protein G-Sepharose beads. Immunoprecipitated complexes along with control purified PIM (PIM1, PIM2, and PIM6; lane 5) were resolved in 4–20% gradient Tris-glycine gels, transferred to nitrocellulose membranes, and probed with anti-BCG polyclonal Ab. After stripping, membranes were reprobed with mAb anti-FN or polyclonal anti-α5 integrin chain, followed by anti-mouse-HRP or anti-rabbit-HRP (IB). One of five representative experiments is shown. B, Solid phase binding of PIM to α5β1 integrin. Purified α5β1 integrin was immobilized in 96-well plates (0.2–5 μg/ml). Wells were blocked for 1 h, and purified PIM were incubated (1 or 5 μg/ml) for 3 h at 30°C. After washing, bound PIM were detected with rabbit anti-BCG polyclonal Ab, followed by anti-rabbit-HRP Ab. OD was determined at 450 nm. Shown are average OD450values ± SD of one representative experiment of five. C, Blocking of PIM binding to α5β1 integrin. Solid phase binding of PIM to α5β1 integrin was performed as described above, except for that 10 μg/ml mAb JBS5 or control Ab were incubated with plate bound integrin for 1 h before addition of PIM. For each condition, binding was calculated as a percentage over the maximum binding obtained without Ab addition and at the highest α5β1 integrin concentration. One of two representative experiments is shown.

Discussion

CD4+ T cells are essential for immune responses to M. tuberculosis. T cells recognize MHC-peptide complexes on APC and upon activation secrete cytokines or express CTL activity. During chronic, persistent M. tuberculosis infection, these effector functions largely take place in granulomas, highly organized structures with a central aggregate of activated macrophages surrounded by lymphocytes and fibrosis. The mechanisms of migration, distribution, and retention of T cells in mycobacterial granulomas are poorly understood. In addition to the host, the mycobacteria themselves likely contribute to the regulation of T cell traffic through granulomas.

Interactions between M. tuberculosis and CD4+ T cells are mediated primarily through macrophages and other APC. Our results demonstrate that mycobacterial glycolipids such as PIM can also directly interact with and affect CD4+ T cell function. The intracellular localization of M. tuberculosis during primary and persistent infection does not preclude direct interactions between mycobacterial constituents and T cells. Mycobacterial lipid-containing vesicles can traffic outside phagosomes into the extracellular environment and be phagocytosed by bystander cells (6, 7, 12). In addition, the lipid content of these vesicles may allow fusion with plasma membranes and thus affect nonphagocytic cells. Ilangumaran et al. (45) demonstrated that PIM can integrate into lymphomonocytic plasma membranes, and Shabaana et al. (46) showed that LAM inserts into Th cell rafts and modify cytokine production.

Our experiments determined that the cell wall/membrane fraction of M. tuberculosis induced HTCA and biochemical fractionation established that PIM was responsible for this effect. PIM, described by Ballou et al. (47) in the 1960s, consist of phosphatidylinositol with one to six mannosyl residues attached to the inositol moiety of phosphatidylinositol. The predominant PIM species in mycobacteria are PIM2 and PIM6, each with variable states of acylation. TLC analysis of our cell wall/membrane fractions with HTCA activity found three dominant PIM species: PIM1, PIM2, and PIM6. It is not clear whether these PIM species differ in their effects on host cells. All mycobacteria, including pathogenic M. tuberculosis, synthesize PIM. PIM are essential for mycobacterial growth and viability (48). Although there are no variations in PIM structure, the amounts of PIM exposure on cell surfaces differ among mycobacterial strains and species and appears inversely correlated with the abundance of capsular polysaccharides (49). Our results indicate that PIM isolated from two M. tuberculosis strains, H37Ra or H37Rv, or M. bovis-BCG induce HTCA and CD4+ T cell adhesion to FN. PIM are abundant and exist in the plasma membrane and cell wall, and glycosylation of this product is extended to form LAM (50). Neither smegLAM nor ManLAM induced HTCA, suggesting that long carbohydrate chains interfered with HTCA. Furthermore, SPI did not induce HTCA, confirming the importance of the oligomannosyl core of PIM. Integration of PIM into plasma membrane’s lipid rafts has been shown to depend on both the acyl chains and glycan core (45). In summary, PIM-mediated induction of T cell adhesion depended not only on the presence but also the length of the core glycan with multiglycosylation of the phospholipid hindering HTCA activity.

Adhesion of lymphocytes results from a cascade of intracellular signals delivered upon engagement of cell surface molecules with their natural ligands. Homotypic adhesion assays have been used extensively as a model to study integrin activation and function (51, 52). T cell-to-T cell adhesion requires active metabolism, reorganization of cytoskeleton, and relocalization of cell surface molecules and can be triggered by agonist Abs against α4 (CD49d), α5 (CD49e), β1 (CD29), αL (CD11a), and β2 (CD18) integrin chains (51, 52, 53, 54, 55, 56). HTCA requires interactions between specific integrins on opposing cells. PIM-induced HTCA depended on FN or serum and was blocked with mAbs to α5. PIM-induced HTCA required divalent cations, suggesting an integrin-ligand interaction (44). T cell aggregation induced by PIM depended on an intact cytoskeleton as described previously in systems of integrin-induced cell-to-cell adhesion (57). HTCA occurred within 30 min after PIM addition. This rapid response is consistent with PIM insertion into plasma membranes, which is maximal after 5 min (45). The prompt formation of T cell aggregates suggests that PIM induced rapid changes in VLA-5, resulting in integrin activation.

FN, a major ECM component, is a ligand for the α4β1 and α5β1 integrins expressed on T cells. PIM increased T cell adhesion to immobilized FN, which was blocked by Abs against the α5 chain but not against the α4 or β1 chains. PMA-induced adherence to FN is mediated by multiple integrins and therefore was partially blocked by anti-α5, anti-α4, and anti-β1 Abs. The lack of blocking of PIM-induced adhesion by anti-β1 blocking Abs may be related to how α5β1 is activated by PIM. PIM-induced changes in α5β1 may bypass the activation requirements of PMA-stimulated adhesion. Only one of two blocking anti-α5 mAbs blocked PIM-induced T cell adhesion. Most inhibitory anti-α5 mAbs do not compete directly with FN for binding but instead recognize sites that are linked allosterically to the ligand binding site. Abs P1D6 and JBS5 bind two different sites of the beta propeller of the α5 chain, and this may explain the differential blocking activity (58, 59). PIM-induced T cell adhesion was completely blocked by RGD containing peptides and correlated with α5β1 expression on T cells, supporting the central role of the interaction between FN and α5β1 in PIM-induced CD4+ T cell adhesion.

Integrin function is controlled by changes in affinity and avidity. Increased integrin affinity for its ligand is induced by conformational changes while avidity regulation is due to integrin clustering (25, 27, 28, 60). Our immunoprecipitation and solid phase receptor ligand-binding studies demonstrate direct binding of VLA-5 by PIM. We hypothesize that, given PIM size and ability to insert into lipid rafts, binding of VLA-5 may occur in the transmembrane region or very close to the membrane in the extracellular portion of the molecule. PIM could cause either clustering or conformational changes, resulting in activation of the integrin head for ligand binding. Other bacteria, Y. pseudotuberculosis through invasin and E. coli through intimin, have been shown to bind the extracellular domain of β1 integrins on T cells (40). On the other hand, mycobacterial glycopeptidolipids, ManLAM and PIM, have been shown to interact with the β2 integrins αMβ2 (CR3, MAC-1), and αXβ2 (CR4) on macrophages (49, 61).

PIM can have diverse effects on host responses. PIM are recognized by T cells in the context of CD1d, are thought to be a TLR2 ligand, mediate attachment of M. tuberculosis to nonphagocytic cells, recruit NK T cells to granulomatous foci, stimulate early endosomal fusion, activate dendritic cells, and participate in receptor-dependent internalization of mycobacteria in human macrophages (62, 63, 64, 65, 66, 67, 68). None of these effects have been ascribed to binding to VLA-5. Chen et al. (69) demonstrated that BCG triggers intracellular signaling in transitional carcinoma cell lines by cross-linking α5β1 integrin with FN, functioning as a protein bridge linking BCG to the tumor cell surface. FN-mediated cross-linking of cell surface integrins activated intracellular signaling pathways and reduced tumor cell growth. These effects on bladder carcinoma cells may be in part mediated by direct binding of mycobacterial PIM to α5β1 integrin.

VLA-5 is important for thymocyte adhesion and migration on FN, but its role for adhesion and migration of T cells in nonlymphoid tissues is not defined (34, 70). α5β1 expression on memory CD4+ T cells suggests that it plays a role in migration and retention of memory T cells in inflammatory sites, such as mycobacterial granulomas. Increased deposition of FN and staining for α5β1 integrin has been demonstrated in lung granulomas and lymph nodes in sarcoid patients (36, 37). PIM could contribute to retention of CD4+ memory T cell in sites of mycobacterial infection and thus promote granuloma formation and enhance host responses. Alternatively, PIM could be exocytosed and deposited away from infected cells, inducing retention of CD4+ T cells in sites distant from Ag-bearing macrophages. This would constitute an advantage for M. tuberculosis by preventing the T cell-APC interaction and thus ensuring bacterial persistence. In conclusion, we have demonstrated that the mycobacterial glycolipid PIM can bind the α5β1 integrin VLA-5 on CD4+ T cells, resulting in T cell adhesion to FN. This represents the first demonstration of direct binding of a mycobacterial molecule to a receptor on T cells.

Acknowledgments

We thank the Orthopedic Department at Case Western Reserve University technical assistance in the use of microscopy facilities.

Disclosures

The authors have no financial conflict of interest.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • ↵1 This work was supported by the National Institutes of Health Grants AI-27243 and HL-55967 (to W.H.B.) and AI34343 and AI35726 (to C.V.H.), Tuberculosis Prevention and Control Research Unit Contract AI-45244/95383, Grant HHSN26620040091C, and National Institutes of Health, National Institute of Allergy and Infectious Diseases Grant N01-AI-40091.

  • ↵2 Address correspondence and reprint requests to Dr. Roxana E. Rojas, Department of Medicine and Tuberculosis, Research Unit, Case Western Reserve University School of Medicine, 10900 Euclid Avenue, Cleveland, OH 44106-4984. E-mail address: rxr38{at}cwru.edu

  • ↵3 C.V.H. and W.H.B. share senior authorship.

  • 4 Abbreviations used in this paper: ECM, extracellular matrix; BCG, bacillus Calmette Guérin; DPBS, Dulbecco PBS; FN, fibronectin; FSC, forward scatter; HTCA, homotypic T cell adhesion; LAM, lipoarabinomannan; smegLAM, LAM from M. smegmatis; ManLAM, mannose-capped LAM from M. tuberculosis; PIM, phosphatidylinositol mannoside; SPI, soybean PI; SSC, side scatter.

  • Received February 6, 2006.
  • Accepted June 21, 2006.

References

  1. ↵
    Saunders, B. M., A. M. Cooper. 2000. Restraining mycobacteria: role of granulomas in mycobacterial infections. Immunol. Cell Biol 78: 334-341.
    OpenUrlCrossRefPubMed
  2. ↵
    Flynn, J. L., J. Chan. 2005. What’s good for the host is good for the bug. Trends Microbiol. 13: 98-102.
    OpenUrlCrossRefPubMed
  3. ↵
    Cosma, C. L., D. R. Sherman, L. Ramakrishnan. 2003. The secret lives of the pathogenic mycobacteria. Annu. Rev. Microbiol. 57: 641-676.
    OpenUrlCrossRefPubMed
  4. ↵
    Rhoades, E. R., A. M. Cooper, I. M. Orme. 1995. Chemokine response in mice infected with Mycobacterium tuberculosis. Infect. Immun. 63: 3871-3877.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Algood, H. M., J. L. Flynn. 2004. CCR5-deficient mice control Mycobacterium tuberculosis infection despite increased pulmonary lymphocytic infiltration. J. Immunol. 173: 3287-3296.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Beatty, W. L., E. R. Rhoades, H. J. Ullrich, D. Chatterjee, J. E. Heuser, D. G. Russell. 2000. Trafficking and release of mycobacterial lipids from infected macrophages. Traffic 1: 235-247.
    OpenUrlCrossRefPubMed
  7. ↵
    Beatty, W. L., H. J. Ullrich, D. G. Russell. 2001. Mycobacterial surface moieties are released from infected macrophages by a constitutive exocytic event. Eur. J. Cell Biol. 80: 31-40.
    OpenUrlCrossRefPubMed
  8. ↵
    Banerjee, S., S. Kumar, B. C. Harinath. 2003. Isolation and characterisation of in vivo released 41 kDa mycobacterial antigen in pulmonary and bone and joint tuberculosis and its identification with H37Ra in vitro released antigen. Int. J. Tuberc. Lung Dis. 7: 278-283.
    OpenUrlPubMed
  9. ↵
    Harth, G., D. L. Clemens, M. A. Horwitz. 1994. Glutamine synthetase of Mycobacterium tuberculosis: extracellular release and characterization of its enzymatic activity. Proc. Natl. Acad. Sci. USA 91: 9342-9346.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Kashyap, R. S., K. M. Dobos, J. T. Belisle, H. J. Purohit, N. H. Chandak, G. M. Taori, H. F. Daginawala. 2005. Demonstration of components of antigen 85 complex in cerebrospinal fluid of tuberculous meningitis patients. Clin. Diagn. Lab. Immunol. 12: 752-758.
    OpenUrlPubMed
  11. ↵
    Nair, R. E., S. Banerjee, S. Kumar, B. C. Harinath. 2000. Isolation and characterization of a 31 kDa mycobacterial antigen from tuberculous sera and its identification with in vitro released culture filtrate antigen of mtb H37Ra bacilli. Scand. J. Infect. Dis. 32: 551-556.
    OpenUrlCrossRefPubMed
  12. ↵
    Russell, D. G., H. C. Mwandumba, E. E. Rhoades. 2002. Mycobacterium and the coat of many lipids. J. Cell Biol. 158: 421-426.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Hogg, N., A. Smith, A. McDowall, K. Giles, P. Stanley, M. Laschinger, R. Henderson. 2004. How T cells use LFA-1 to attach and migrate. Immunol. Lett. 92: 51-54.
    OpenUrlCrossRefPubMed
  14. ↵
    Kuznetsova, S. A., D. D. Roberts. 2004. Functional regulation of T lymphocytes by modulatory extracellular matrix proteins. Int J. Biochem. Cell Biol. 36: 1126-1134.
    OpenUrlCrossRefPubMed
  15. ↵
    Epler, J. A., R. Liu, Y. Shimizu. 2000. From the ECM to the cytoskeleton and back: how integrins orchestrate T cell action. Dev. Immunol. 7: 155-170.
    OpenUrlCrossRefPubMed
  16. ↵
    Pribila, J. T., A. C. Quale, K. L. Mueller, Y. Shimizu. 2004. Integrins and T cell-mediated immunity. Annu. Rev. Immunol. 22: 157-180.
    OpenUrlCrossRefPubMed
  17. ↵
    Savino, W., D. A. Mendes-Da-Cruz, S. Smaniotto, E. Silva-Monteiro, D. M. Villa-Verde. 2004. Molecular mechanisms governing thymocyte migration: combined role of chemokines and extracellular matrix. J. Leukocyte Biol. 75: 951-961.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Woods, M. L., Y. Shimizu. 2001. Signaling networks regulating β1 integrin-mediated adhesion of T lymphocytes to extracellular matrix. J. Leukocyte Biol. 69: 874-880.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Hynes, R. O., K. M. Yamada. 1982. Fibronectins: multifunctional modular glycoproteins. J. Cell Biol. 95: 369-377.
    OpenUrlFREE Full Text
  20. ↵
    Mao, Y., J. E. Schwarzbauer. 2005. Fibronectin fibrillogenesis, a cell-mediated matrix assembly process. Matrix Biol. 24: 389-399.
    OpenUrlCrossRefPubMed
  21. ↵
    Pankov, R., K. M. Yamada. 2002. Fibronectin at a glance. J. Cell Sci. 115: 3861-3863.
    OpenUrlFREE Full Text
  22. ↵
    Yamada, K. M., S. Aota, S. K. Akiyama, S. E. LaFlamme. 1992. Mechanisms of fibronectin and integrin function during cell adhesion and migration. Cold Spring Harb. Symp. Quant. Biol. 57: 203-212.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Kosmehl, H., A. Berndt, D. Katenkamp, U. Mandel, R. Bohle, U. Gabler, D. Celeda. 1995. Differential expression of fibronectin splice variants, oncofetal glycosylated fibronectin and laminin isoforms in nodular palmar fibromatosis. Pathol. Res. Pract. 191: 1105-1113.
    OpenUrlCrossRefPubMed
  24. ↵
    Ruoslahti, E., E. Engvall, E. G. Hayman. 1981. Fibronectin: current concepts of its structure and functions. Coll. Relat. Res. 1: 95-128.
    OpenUrlPubMed
  25. ↵
    Hynes, R. O.. 2002. Integrins: bidirectional, allosteric signaling machines. Cell 110: 673-687.
    OpenUrlCrossRefPubMed
  26. ↵
    Miyamoto, S., B. Z. Katz, R. M. Lafrenie, K. M. Yamada. 1998. Fibronectin and integrins in cell adhesion, signaling, and morphogenesis. Ann. NY Acad. Sci. 857: 119-129.
    OpenUrlCrossRefPubMed
  27. ↵
    Hogg, N., M. Laschinger, K. Giles, A. McDowall. 2003. T cell integrins: more than just sticking points. J. Cell Sci. 116: 4695-4705.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Miyamoto, Y. J., B. F. Andruss, J. S. Mitchell, M. J. Billard, B. W. McIntyre. 2003. Diverse roles of integrins in human T lymphocyte biology. Immunol. Res. 27: 71-84.
    OpenUrlCrossRefPubMed
  29. ↵
    Hauzenberger, D., J. Klominek, S. E. Bergstrom, K. G. Sundqvist. 1995. T lymphocyte migration: the influence of interactions via adhesion molecules, the T cell receptor, and cytokines. Crit. Rev. Immunol. 15: 285-316.
    OpenUrlCrossRefPubMed
  30. ↵
    Shimizu, Y., G. A. Van Seventer, K. J. Horgan, S. Shaw. 1990. Regulated expression and binding of three VLA (β1) integrin receptors on T cells. Nature 345: 250-253.
    OpenUrlCrossRefPubMed
  31. ↵
    Sims, T. N., M. L. Dustin. 2002. The immunological synapse: integrins take the stage. Immunol. Rev. 186: 100-117.
    OpenUrlCrossRefPubMed
  32. ↵
    von Andrian, U. H., C. R. Mackay. 2000. T cell function and migration: two sides of the same coin. N. Engl. J. Med. 343: 1020-1034.
    OpenUrlCrossRefPubMed
  33. ↵
    Hemler, M. E.. 1990. VLA proteins in the integrin family: structures, functions, and their role on leukocytes. Annu. Rev. Immunol. 8: 365-400.
    OpenUrlCrossRefPubMed
  34. ↵
    Krivacic, K. A., A. D. Levine. 2003. Extracellular matrix conditions T cells for adhesion to tissue interstitium. J. Immunol. 170: 5034-5044.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Pilling, J. E., A. Galvin, A. M. Robins, H. F. Sewell, Y. R. Mahida. 1998. Expression of α5 (CD49e) and α6 (CD49f) integrin subunits on T cells in the circulation and the lamina propria of normal and inflammatory bowel disease colonic mucosa. Scand. J. Immunol. 48: 425-428.
    OpenUrlCrossRefPubMed
  36. ↵
    Roman, J., Y. J. Jeon, A. Gal, R. L. Perez. 1995. Distribution of extracellular matrices, matrix receptors, and transforming growth factor β1 in human and experimental lung granulomatous inflammation. Am. J. Med. Sci. 309: 124-133.
    OpenUrlCrossRefPubMed
  37. ↵
    Shigehara, K., N. Shijubo, M. Hirasawa, S. Abe, T. Uede. 1998. Immunolocalization of extracellular matrix proteins and integrins in sarcoid lymph nodes. Virchows Arch. 433: 55-61.
    OpenUrlCrossRefPubMed
  38. ↵
    Sanchez-Mateos, P., C. Cabanas, F. Sanchez-Madrid. 1996. Regulation of integrin function. Semin. Cancer Biol. 7: 99-109.
    OpenUrlCrossRefPubMed
  39. ↵
    Stewart, M., N. Hogg. 1996. Regulation of leukocyte integrin function: affinity vs. avidity. J. Cell. Biochem. 61: 554-561.
    OpenUrlCrossRefPubMed
  40. ↵
    Arencibia, I., N. C. Suarez, H. Wolf-Watz, K. G. Sundqvist. 1997. Yersinia invasin, a bacterial β1-integrin ligand, is a potent inducer of lymphocyte motility and migration to collagen type IV and fibronectin. J. Immunol. 159: 1853-1859.
    OpenUrlAbstract
  41. ↵
    Frankel, G., O. Lider, R. Hershkoviz, A. P. Mould, S. G. Kachalsky, D. C. Candy, L. Cahalon, M. J. Humphries, G. Dougan. 1996. The cell-binding domain of intimin from enteropathogenic Escherichia coli binds to β1 integrins. J. Biol. Chem. 271: 20359-20364.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    Van Nhieu, G. T., R. R. Isberg. 1991. The Yersinia pseudotuberculosis invasin protein and human fibronectin bind to mutually exclusive sites on the α5β1 integrin receptor. J. Biol. Chem. 266: 24367-24375.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    Gehring, A. J., K. M. Dobos, J. T. Belisle, C. V. Harding, W. H. Boom. 2004. Mycobacterium tuberculosis LprG (Rv1411c): a novel TLR-2 ligand that inhibits human macrophage class II MHC antigen processing. J. Immunol. 173: 2660-2668.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Plow, E. F., T. A. Haas, L. Zhang, J. Loftus, J. W. Smith. 2000. Ligand binding to integrins. J. Biol. Chem. 275: 21785-21788.
    OpenUrlFREE Full Text
  45. ↵
    Ilangumaran, S., S. Arni, M. Poincelet, J. M. Theler, P. J. Brennan, P. J. Nasir-ud-Din, D. C. Hoessli. 1995. Integration of mycobacterial lipoarabinomannans into glycosylphosphatidylinositol-rich domains of lymphomonocytic cell plasma membranes. J. Immunol. 155: 1334-1342.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Shabaana, A. K., K. Kulangara, I. Semac, Y. Parel, S. Ilangumaran, K. Dharmalingam, C. Chizzolini, D. C. Hoessli. 2005. Mycobacterial lipoarabinomannans modulate cytokine production in human T helper cells by interfering with raft/microdomain signalling. Cell. Mol. Life Sci. 62: 179-187.
    OpenUrlCrossRefPubMed
  47. ↵
    Ballou, C. E., E. Vilkas, E. Lederer. 1963. Structural studies on the myo-inositol phospholipids of Mycobacterium tuberculosis (var. bovis, strain BCG). J. Biol. Chem. 238: 69-76.
    OpenUrlFREE Full Text
  48. ↵
    Kordulakova, J., M. Gilleron, K. Mikusova, G. Puzo, P. J. Brennan, B. Gicquel, M. Jackson. 2002. Definition of the first mannosylation step in phosphatidylinositol mannoside synthesis: PimA is essential for growth of mycobacteria. J. Biol. Chem. 277: 31335-31344.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    Villeneuve, C., M. Gilleron, I. Maridonneau-Parini, M. Daffe, C. Astarie-Dequeker, G. Etienne. 2005. Mycobacteria use their surface-exposed glycolipids to infect human macrophages through a receptor-dependent process. J. Lipid Res. 46: 475-483.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    Morita, Y. S., J. H. Patterson, H. Billman-Jacobe, M. J. McConville. 2004. Biosynthesis of mycobacterial phosphatidylinositol mannosides. Biochem. J. 378: 589-597.
    OpenUrlCrossRefPubMed
  51. ↵
    Bednarczyk, J. L., J. N. Wygant, M. C. Szabo, L. Molinari-Storey, M. Renz, S. Fong, B. W. McIntyre. 1993. Homotypic leukocyte aggregation triggered by a monoclonal antibody specific for a novel epitope expressed by the integrin β1 subunit: conversion of nonresponsive cells by transfecting human integrin α4 subunit cDNA. J. Cell. Biochem. 51: 465-478.
    OpenUrlPubMed
  52. ↵
    Rothlein, R., T. A. Springer. 1986. The requirement for lymphocyte function-associated antigen 1 in homotypic leukocyte adhesion stimulated by phorbol ester. J. Exp. Med. 163: 1132-1149.
    OpenUrlAbstract/FREE Full Text
  53. ↵
    Bednarczyk, J. L., B. W. McIntyre. 1990. A monoclonal antibody to VLA-4 α-chain (CDw49d) induces homotypic lymphocyte aggregation. J. Immunol. 144: 777-784.
    OpenUrlAbstract
  54. ↵
    Keizer, G. D., W. Visser, M. Vliem, C. G. Figdor. 1988. A monoclonal antibody (NKI-L16) directed against a unique epitope on the α-chain of human leukocyte function-associated antigen 1 induces homotypic cell-cell interactions. J. Immunol. 140: 1393-1400.
    OpenUrlAbstract
  55. ↵
    Neelamegham, S., S. I. Simon, B. W. McIntyre, K. Zygourakis. 1996. Induction of homotypic lymphocyte aggregation: evidence for a novel activation state of the β1 integrin. J. Leukocyte Biol. 59: 872-882.
    OpenUrlAbstract
  56. ↵
    Shen, C. X., S. Stewart, E. Wayner, W. Carter, J. Wilkins. 1991. Antibodies to different members of the β1 (CD29) integrins induce homotypic and heterotypic cellular aggregation. Cell. Immunol. 138: 216-228.
    OpenUrlCrossRefPubMed
  57. ↵
    Whittard, J. D., S. K. Akiyama. 2001. Activation of β1 integrins induces cell-cell adhesion. Exp. Cell Res. 263: 65-76.
    OpenUrlCrossRefPubMed
  58. ↵
    Burrows, L., K. Clark, A. P. Mould, M. J. Humphries. 1999. Fine mapping of inhibitory anti-α5 monoclonal antibody epitopes that differentially affect integrin-ligand binding. Biochem. J. 344: (Pt. 2):527-533.
    OpenUrlAbstract/FREE Full Text
  59. ↵
    Mould, A. P., J. A. Askari, S. Aota, K. M. Yamada, A. Irie, Y. Takada, H. J. Mardon, M. J. Humphries. 1997. Defining the topology of integrin α5β1-fibronectin interactions using inhibitory anti-α5 and anti-β1 monoclonal antibodies: evidence that the synergy sequence of fibronectin is recognized by the amino-terminal repeats of the α5 subunit. J. Biol. Chem. 272: 17283-17292.
    OpenUrlAbstract/FREE Full Text
  60. ↵
    Hogg, N., J. Harvey, C. Cabanas, R. C. Landis. 1993. Control of leukocyte integrin activation. Am. Rev. Respir. Dis. 148: S55-S59.
    OpenUrlCrossRefPubMed
  61. ↵
    Schlesinger, L. S., S. R. Hull, T. M. Kaufman. 1994. Binding of the terminal mannosyl units of lipoarabinomannan from a virulent strain of Mycobacterium tuberculosis to human macrophages. J. Immunol. 152: 4070-4079.
    OpenUrlAbstract
  62. ↵
    Gilleron, M., C. Ronet, M. Mempel, B. Monsarrat, G. Gachelin, G. Puzo. 2001. Acylation state of the phosphatidylinositol mannosides from Mycobacterium bovis bacillus Calmette Guerin and ability to induce granuloma and recruit natural killer T cells. J. Biol. Chem. 276: 34896-34904.
    OpenUrlAbstract/FREE Full Text
  63. ↵
    Hoppe, H. C., B. J. de Wet, C. Cywes, M. Daffe, M. R. Ehlers. 1997. Identification of phosphatidylinositol mannoside as a mycobacterial adhesin mediating both direct and opsonic binding to nonphagocytic mammalian cells. Infect. Immun. 65: 3896-3905.
    OpenUrlAbstract/FREE Full Text
  64. ↵
    Jones, B. W., T. K. Means, K. A. Heldwein, M. A. Keen, P. J. Hill, J. T. Belisle, M. J. Fenton. 2001. Different Toll-like receptor agonists induce distinct macrophage responses. J. Leukocyte Biol. 69: 1036-1044.
    OpenUrlAbstract/FREE Full Text
  65. ↵
    Mempel, M., C. Ronet, F. Suarez, M. Gilleron, G. Puzo, L. Van Kaer, A. Lehuen, P. Kourilsky, G. Gachelin. 2002. Natural killer T cells restricted by the monomorphic MHC class 1b CD1d1 molecules behave like inflammatory cells. J. Immunol. 168: 365-371.
    OpenUrlAbstract/FREE Full Text
  66. ↵
    Sieling, P. A., D. Chatterjee, S. A. Porcelli, T. I. Prigozy, R. J. Mazzaccaro, T. Soriano, B. R. Bloom, M. B. Brenner, M. Kronenberg, P. J. Brennan, et al 1995. CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269: 227-230.
    OpenUrlAbstract/FREE Full Text
  67. ↵
    Sprott, G. D., C. J. Dicaire, K. Gurnani, S. Sad, L. Krishnan. 2004. Activation of dendritic cells by liposomes prepared from phosphatidylinositol mannosides from Mycobacterium bovis bacillus Calmette-Guerin and adjuvant activity in vivo. Infect. Immun. 72: 5235-5246.
    OpenUrlAbstract/FREE Full Text
  68. ↵
    Vergne, I., R. A. Fratti, P. J. Hill, J. Chua, J. Belisle, V. Deretic. 2004. Mycobacterium tuberculosis phagosome maturation arrest: mycobacterial phosphatidylinositol analog phosphatidylinositol mannoside stimulates early endosomal fusion. Mol. Biol. Cell 15: 751-760.
    OpenUrlAbstract/FREE Full Text
  69. ↵
    Chen, F., G. Zhang, Y. Iwamoto, W. A. See. 2005. BCG directly induces cell cycle arrest in human transitional carcinoma cell lines as a consequence of integrin cross-linking. BMC Urol. 5: 8-17.
    OpenUrlCrossRefPubMed
  70. ↵
    Cotta-de-Almeida, V., D. M. Villa-Verde, F. Lepault, J. M. Pleau, M. Dardenne, W. Savino. 2004. Impaired migration of NOD mouse thymocytes: a fibronectin receptor-related defect. Eur. J. Immunol. 34: 1578-1587.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section