2ß1 and 4ß1 integrins to lipid rafts in Jurkat T lymphocytes exposed to collagen type IV and fibronectin
* Signal Transduction Laboratory, Graduate Program in Immunology, Clinical Research Center, and Departments of
Physiology and Biophysics and
Biochemistry, Faculty of Medicine, University of Sherbrooke, Quebec, Canada
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2ß1 and 4ß1 integrins in low-density, detergent-resistant microdomains (DRM) of Jurkat lymphocytes. CnIV-coated microspheres induced (optical trapping) the redistribution of GM1-associated fluorescence from the cell periphery to the area of contact. This was not observed in cells treated with ß-methyl cyclodextrin (MCD). Fn- or bovine serum albumin-coated microspheres did not modify the peripheral distribution of fluorescence. These observations were confirmed by confocal microscopy. Western blot analysis of cells exposed to surfaces coated with CnIV revealed that the 2-subunit was initially present at low levels in DRM, became strongly associated after 40 min, and returned to basal levels after 75 min. Fn induced a slight recruitment of the ß1-integrin 4-subunit in DRM after 5 and 10 min, followed by a return to basal levels. Neither CnIV nor Fn triggered significant changes in the distribution of the ß1-subunit in DRM. Fn- and CnIV-coated microspheres or surfaces coated with these ligands triggered a MCD-sensitive mobilization of Ca2+. MCD did not alter the state of the Ca2+ reserves. The differential distributions of the 2ß1 and 4ß1 integrins in DRM may provide one additional step in the regulation of outside-in signaling involving these integrins.
Key Words: optical trapping • Western blots • confocal microscopy • cholera toxin • GM1 ganglioside • fluorescence • calcium mobilization
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß subunits [11
], which provide a two-way relay between the extracellular environment and the cell cytoplasm [12
]. In a one-directional response, the affinity/avidity of integrins is regulated by inside-out signals [13
] and is generated by growth factors or chemokines [3
, 14
15
16
] or by occupation of the T cell receptor in T lymphocytes [17
18
19
]. In the other directional response, the ligation of activated integrins triggers outside-in signals that are essential for orchestration of actin-based cytoskeletal reorganization and cell motility, cell survival and gene expression [12
, 20
, 21
], and immune responses [9
]. Integrins regulate lymphocyte trafficking [22
], the formation of the immune synapse [23
, 24
], and the exit of lymphocytes from the blood and their migration to the injured tissues in inflammatory responses [22
].
The ligands of integrins fall into two categories: Cell-surface proteins of the immunoglobulin (Ig) supergene family, such vascular cell adhesion molecule 1 (VCAM-1) and mucosal addressin cell adhesion molecule-1 (MadCAM-1), comprise one category, whereas some protein components of the extracellular matrix (ECM) and the complement cascade belong to the other category [6
]. A restricted number of ß1-integrins can recognize ligands belonging to both categories, suggesting their involvement in cell-cell and cell-ECM interactions [25
]. This is the case of the 4ß1 integrin, which binds VCAM-1 on activated endothelial cells (EC) [26
] and fibronectin (Fn), a component of the ECM. VCAM-1 and Fn trigger outside-in signaling such as Ca2+ mobilization [27
, 28
] and up-regulation of protein tyrosine kinase activity in lymphocytes [29
30
31
], whereas ligation of VCAM-1 triggers signaling in the EC [27
, 32
]. Most members of the ß1-integrin family recognize protein ligands of only one category. This is the case of 2ß1, which binds to collagen type IV (CnIV) and laminin, and 5ß1, which binds Fn [6
]. Ligation of the 2ß1 integrin triggers Ca2+ mobilization [28
].
The mechanism involved in integrin-dependent outside-in signaling remains under investigation. Except for the ß4-subunit, the cytoplasmic tail of - and ß-subunits is relatively short (30–50 amino acid residues) [11
] and is not associated with intrinsic protein kinase activity. Integrin-dependent signal transduction must rely on efficient conformational rearrangements induced by inside-out signals to allow the binding of integrin ligands and the recruitment of cytoplasmic components to assemble the machinery of outside-in signaling [1
, 3
, 12
, 33
]. The organization of ß1-integrins in the plasma membrane may be facilitated by clustering into microdomains [34
] as in the case of lymphoid cells [35
36
37
] and the Vß3 integrin in some tumor cell lines [38
].
Plasma membrane microdomains have emerged as a new concept to explain the preferred localization of some of its components [39 , 40 ]. Operationally defined plasma membrane lipid rafts are composed primarily of sphingolipids and cholesterol arranged in a liquid-ordered phase [41 ]. Evidence suggests that lipid rafts provide a platform for the recruitment of proteins involved in the formation of the initial steps of receptor signaling. Lipid rafts can act as molecular filters, in which case some plasma membrane proteins having an affinity for lipid rafts are brought into molecular contacts, whereas others are excluded [42 ]. For instance, receptor ligation leads to an association of discrete lipid raft domains into one larger entity or a coalescence of lipid rafts [43 ]. Some proteins involved in signaling may have an affinity for lipid rafts following coalescence. Lipid rafts possess a reduced solubility in nonionic detergents at low temperature and have been designated as detergent-resistant microdomains (DRM), detergent-insoluble glycolipid-enriched (DIG), glycolipid-enriched membranes (GEM), or triton-insoluble floating fraction (TIFF). They can be isolated by buoyancy on sucrose gradients [43 ], allowing identification of DRM-associated proteins [40 , 44 ]. The GM1 ganglioside, a ligand of the B subunit of the cholera toxin (CTx) [45 ], can be used as a marker of DRM [46 ], and fluorescent derivatives of the CTx B subunit can be used to study the spatio-temporal dynamics of distribution of lipid rafts in live cells.
We show in this report that three different approaches (optical trapping, confocal microscopy, and Western blot analysis) lead to the conclusion that the 2ß1 and 4ß1 integrins are differentially recruited to lipid rafts in Jurkat cells exposed to ligands (CnIV, Fn) of these integrins. Evidence is also presented that these ligands trigger a mobilization of Ca2+ that is abolished by ß-methyl cyclodextrin (MCD) treatment of Jurkat cells.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2-integrin subunit mAb (clone 16B4) was obtained from Serotec (Raleigh, NC), whereas polyclonal antibodies directed against the human 4 (raised in the goat)- and ß1 (raised in the rabbit)-integrin subunits were from Santa Cruz Biotechnology (Santa Cruz, CA). HRP-conjugated secondary antibodies directed against mouse or rabbit Ig were purchased from Amersham Biosciences (Montreal, QC, Canada). Protein A/G Plus-Agarose beads (Santa Cruz Biotechnology) were used for immunoprecipitations. Other chemicals were from Sigma-Aldrich or from local suppliers. Protein quantification was done using the Bradford reagent (Bio-Rad, Richmond, CA).
Cells and cell cultures
Jurkat E6.1 T cells were a gift of Dr. Arthur Weiss (Howard Hughes Medical Institute, Department of Medicine, University of California, San Francisco). They were cultured as described [47
].
Preparation of biotinylated ECM proteins
A N,N'-dimethylformamide (DMF) solution (20 µl) of N-hydroxysuccinimidyl-biotin (6.5 µg) was added under stirring to a solution of CnIV, Fn, or bovine serum albumin (BSA; 500 µg protein in each case) in N NaHCO3 (500 µl). The mixture was stirred at room temperature for 30 min, at which point a second portion of N-hydroxysuccinimidyl-biotin (6.5 µg) in DMF (20 µl) was added. Stirring was maintained for an additional 90 min at room temperature. The solution was dialyzed overnight (4°C) against several changes of phosphate-buffered saline (PBS).
Microsphere coating
The streptavidin-coated microspheres (6x105 microspheres) were washed in Gey’s balanced salt solution (GBSS) and then incubated for 30 min at room temperature with a solution (1 ml) of biotin-labeled CnIV (10 µg), Fn (10 µg), or BSA (10 µg) under a gentle rocking motion (Nutator instrument, Clay Adams, Sparks, MD). The microspheres were washed with GBSS and used the same day.
Optical trapping
Jurkat cells were suspended in GBSS (3x105 lymphocytes/ml) and incubated (15 min at 4°C) with an Alexa 594-labeled CTx B subunit (3 µg/ml). The cells were washed (GBSS) and allowed to adhere (15 min, 37°C) to circular (22 mm) glass coverslips previously coated (15 min, 37°C) with poly-L-lysine (100 µg/ml). Each coverslip was mounted on a homemade two-piece holder, which was placed on the support of a temperature controller (Intracel, Shepreth, UK). The unit was placed on the stage of an inverted microscope (Nikon Eclipse 300), and changes in cell fluorescence were observed using a 1.2 numerical aperture, 100X objective. Pictures were taken every 20 s using a CoolSNAP fx charged-coupled device (CCD) camera (Roper Scientific, Trenton, NJ) under the control of MetaMorph Imaging software (Universal Imaging Corporation, West Chester, PA). Each reagent-coated microsphere was brought into contact with a single Jurkat cell by optical trapping using a 1.0 W continuous wave power diode laser beam (980±10 nm). The system was operated under a computer-controlled Laser Tweezer 1064/1500 CRI Microscope Work Station (Cell Robotics International, Albuquerque, NM).
Laser scanning confocal microscopy (LSCM)
Jurkat cells were labeled with an Alexa 594-CTx B subunit and were allowed to adhere to 22 mm circular glass coverslips coated with CnIV (15 µg/ml), Fn (50 µg/ml), or BSA (10 µg/ml) as described in the case of optical trapping experiments. Changes in fluorescence were recorded on a time-scale basis at 37°C. The system consisted of a Thermo Noran (Middleton, WI) Oz Intervision confocal laser-scanning imaging system equipped with a Nikon TE300 Eclipse epifluorescence-inverted microscope, a 40X Nikon Oil Plan achromat objective, an image processor, and a Unix-based Silicon Graphics system (Mountain View, CA). The krypton/argon laser line was directed to the sample through an optical fiber optic and a primary dichroic filter. Illumination light was scanned along the xy-axes with an acousto-optic deflector, whereas a galvanometer-driven mirror scanned the point of illumination light along the z-axis. Optical sections were recorded at 0.2 µm intervals along the z-axis. The image size was 512 x 480 pixels. The CTx B subunit-labeled cells were excited at 568 nm through a 10-µm pinhole aperture, and the emitted fluorescence was measured through a long-pass filter (>590 nm). Digitized images were obtained with 256X line-averaging and enhanced using the Intervision software (Thermo Noran). The data were analyzed on a Silicon Graphics O2 Workstation. Image processing and surface quantification of pixel intensities were done using the NIH Image freeware (<http://rsb.info.nih.gov/nih-image/>) and a Mac OS 9.2 MacIntosh computer. Representations of fluorescence intensities, according to a palette of pseudocolors, were generated by means of the NIH Image freeware, which was also used to quantitate pixel intensities.
Treatment of the cells with MCD
Jurkat lymphocytes (5x106 cells/ml) in GBSS medium were treated for 30 min at 37°C with MCD (10 mM). The cells were washed and used in the experiments. Cell viability (trypan blue exclusion) was greater than 95%.
DRM isolation
Six-well culture plates (Sarstedt, Montreal, QC, Canada) were coated (2 h, 37°C) with mixtures (1 ml/well) of Fn (50 µg) and poly-L-lysine (250 µg), CnIV (15 µg) and poly-L-lysine (250 µg), or poly-L-lysine (250 µg) and were then washed three times with PBS. The wells were then treated with BSA (1% w/v in PBS) for 1 h at 37°C and washed. Jurkat cells (50x106 lymphocytes/experiment) suspended in GBSS were distributed in each coated-culture plate (500 µl/well) and left for various periods of time. The cells were lysed (20 min, 4°C) with a buffer containing Tris-HCl (50 mM, pH 7.5), NaCl (150 mM), 1% (v/v) Triton X-100, phenylmethanesulfonyl fluoride (1 mM), and a commercial (EDTA-free Complete®, Boehringer Mannheim, Montreal, Canada) cocktail of protease inhibitors. A 50% (w/v) solution of sucrose in Triton X-100-free Tris (50 mM)/NaCl (150 mM) buffer (pH 7.5) was added to a final concentration of 5% (w/v). The mixture (4.0 ml) was deposited in ultracentrifugation tubes (12 ml) and successively displaced with 30% (w/v, 4 ml) and 50% (w/v, 5.3 ml) solutions of sucrose in protease inhibitor- and detergent-free Tris/NaCl buffer (pH 7.5). The tubes were centrifuged for 16 h at 100,000 g (4°C). Fractions of 1 ml were recovered from the top of the tubes. The 5% (density 1.016 g/ml)–30% (density 1.129 g/ml) interface (fractions 3 and 4) corresponded to DRM.
Western blotting
Antibodies (2 µg) directed against the various ß1-integrin proteins were added to the cell lysate (200 µg protein), and the mixture was incubated under rotary agitation for 5 h at 4°C. A suspension (20 µl) of Protein A/G Plus-Agarose beads (Santa Cruz Biotechnology) was then added, and mixing was continued overnight at 4°C. The beads were washed three times with a Tris-HCl (50 mM, pH 7.5)/NaCl (150 mM) buffer containing 0.5% (v/v) Nonidet P-40 (NP-40) buffer and once with NP-40-free buffer. The beads were treated with Laemmli’s loading buffer and heated in boiling water for 5 min in the presence of reducing agent (DL-dithiothreitol). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed using 7.5% acrylamide gels. Protein transfer and Western blot analyses were done as described [47
] using an enhanced chemiluminescence (ECL) detection kit (Amersham Biosciences).
Mobilization of Ca2+ in Jurkat cells exposed to CnIV or Fn
Jurkat cells (5x106 lymphocytes/ml) were loaded with Fura2/AM as described previously [47
]. In the case of optical trapping experiments, a sample of the Fura2-loaded cell population was added to 22 mm coverslips coated with poly-L-lysine mounted in a holder, as described in the case of the studies of the spatial distribution of GM1-associated fluorescence. After establishing a contact between the CnIV- or the Fn-coated microspheres, the optical trap was turned off, and the cells were exposed at time intervals of 10 s to a luminous source (100 W mercury lamp). Alternative cell excitation at 340 and 380 nm was achieved using appropriate filters mounted on a wheel controlled by the MetaFluor Imaging System (Universal Imaging Corporation) software. Fluorescence emission at 510 nm was automatically collected every 10 s through a long-pass filter. A CoolSNAP fx CCD camera (Roper Scientific) under the control of the MetaFluor Imaging System was used to collect the images. Ca2+ mobilization was also studied in Fura2-loaded Jurkat cells exposed to 22 mm coverslips coated with CnIV (15 µg/ml) or Fn (50 µg/ml). The cells were allowed to adhere (2 min) to the coverslips mounted in the holder described for optical trapping experiments, and changes in fluorescence were recorded at 37°C and analyzed as above.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2ß1 integrin-dependent process. This interpretation was tested by repeating the experiments using Jurkat cells treated with MCD, which depletes cellular cholesterol and destabilizes DRM [49
]. Results showed that the GM1-associated fluorescence initially distributed at the cell periphery (Fig. 1E) did not coalesce to the region of contact with the CnIV-coated microsphere after a period of 20 (Fig. 1F)
or 40 min (not shown).
 View larger version (81K): [in a new window] |
Figure 1. Distribution of GM1-associated fluorescence in one Jurkat cell exposed to a microsphere coated with CnIV. An optical trapping device was used to induce contact between one CnIV-coated microsphere and one GM1-labeled (Alexa 594-CTx B subunit conjugate) Jurkat cell. Fluorescence was recorded automatically as a function of time. (A) Position (arrow) of the microsphere in contact with a Jurkat cell in a micrograph recorded under visible light. Distribution of the fluorescence in the same cell at the onset (B) of monitoring and 20 min (C) later. (D) Position (arrow) of the microsphere in contact with a MCD-treated Jurkat cell in a micrograph recorded under visible light. Distribution of the fluorescence in the same cell at the onset (E) of monitoring and 20 min (F) later. The cells were maintained at 37°C in GBSS. Data are representative of 10 independent experiments. The original scale bars under each micrograph correspond to 5 µm.
|
4ß1 and 5ß1 integrins did not appear to induce detectable coalescence of lipid rafts in Jurkat cells treated with MCD (Fig. 2E
and 2F)
.
 View larger version (72K): [in a new window] |
Figure 2. Distribution of GM1-associated fluorescence in one Jurkat cell exposed to a microsphere coated with Fn. An optical trapping device was used to induce contact between one Fn-coated microsphere and one GM1-labeled (Alexa 594-CTx B subunit conjugate) Jurkat cell. Fluorescence was recorded automatically as a function of time. (A) Position (arrow) of the microsphere in contact with the cell in a micrograph recorded under visible light. Distribution of the fluorescence in the same cell at the onset (B) of monitoring and 20 min (C) later. (D) Position (arrow) of the microsphere in contact with a MCD-treated Jurkat cell in a micrograph recorded under visible light. Distribution of the fluorescence in the same cell at the onset (E) of monitoring and 20 min (F) later. The cells were maintained at 37°C in GBSS. Data are representative of 10 independent experiments. The original scale bars under each micrograph correspond to 5 µm.
|
 View larger version (57K): [in a new window] |
Figure 3. Distribution of GM1-associated fluorescence in one Jurkat cell exposed to a microsphere coated with BSA. An optical trapping device was used to induce contact between one BSA-coated microsphere and one GM1-labeled (Alexa 594-CTx B subunit conjugate) Jurkat cell. Fluorescence was recorded automatically as a function of time. (A) Position (arrow) of the microsphere in contact with the cell in a micrograph recorded under visible light. Distribution of the fluorescence in the same cell at the onset (B) of monitoring and 20 min (C) later. (D) Position (arrow) of the microsphere in contact with a MCD-treated cell in a micrograph recorded under visible light. Distribution of the fluorescence in the same cell at the onset (E) of monitoring and 20 min (F) later. The cells were maintained at 37°C in GBSS. Data are representative of three independent experiments. The original scale bars under each micrograph correspond to 5 µm.
|
 View larger version (60K): [in a new window] |
Figure 4. LSCM analysis of the time-dependent, three-dimensional distribution of GM1-associated fluorescence. Jurkat lymphocytes labeled with an Alexa 594-CTx B subunit conjugate were exposed to coverslips coated with ß1-integrin protein ligands. (A) Representation in pseudocolors of pixel intensities recorded at the onset of monitoring of one cell exposed to CnIV. (B) Similar representation recorded 20 min after the onset of the experiment. (C). Representation in pseudocolors of pixel intensities recorded at the onset of monitoring of one cell exposed to Fn. (D) Similar representation recorded 20 min later. (E) Representation in pseudocolors of pixel intensities recorded at the onset of monitoring of one cell exposed to BSA (control). (F) Similar representation recorded after 20 min. (B). The cells were maintained at 37°C in GBSS medium. The black rectangles shown under each figure correspond to an original scale of 5 µm. The palette of pseudocolor intensities is indicated at the bottom of the figure on a scale of 0–255 arbitrary units. Pixel intensities correspond to the sum of five sections (0.2 µm each) recorded along the z-axis. Data are representative of two independent experiments.
|
 View larger version (11K): [in a new window] |
Figure 5. Quantification of relative pixel intensities of fluorescence recorded by LSCM at the region of contact of Jurkat cells with protein-coated coverslips. Results recorded after 20 min of observations are shown relative to the value (arbitrarily set to one) measured at the onset of each set of experiments illustrated in Figure 4
. Data are shown in the case of Jurkat cells exposed to CnIV, Fn, or BSA. The data show the mean and SEM indicated by the vertical bars. Data are representative of two separate experiments and analysis of three cells in each experiment. Statistical analysis was done using Student’s t-test for paired data. ns, Nonsignificant.
|
2-, 4-, and ß1-integrin subunits in DRM in Jurkat cells exposed to CnIV or Fn2-, 4-, and ß1-integrin subunits in DRM was analyzed in Jurkat cells exposed to surfaces coated with CnIV or Fn. Western blot analysis of GM1-positive (revealed by dot blotting of aliquots using an Alexa 594-labeled CTx B subunit) DRM isolated at the interface of the 5%–30% sucrose interface was performed. In the case of cells exposed to CnIV, results showed that the 2-subunit was initially present at low levels in DRM (Fig. 6A
). There was an increase in the recruitment of the 2-subunit in DRM after 10 and 20 min, and a maximum was reached after 40 min, followed by a return to basal levels (Fig. 6A)
. Semiquantification of densitometric measurements relative to densitometric measurements at the beginning (t=0 min) of the experiments showed ratios of 2.7 (10 min), 2.9 (20 min), and 12 (40 min; Fig. 7A
). In contrast to the 2-subunit, the 4-subunit was found in DRM at the start of the experiments of Jurkat cells exposed to Fn (Fig. 6B)
. There were slight increases in recruitment in DRM after 5 and 10 min of exposure of the cells to Fn. The relative ratios were 1.8 and 1.7, respectively (Fig. 7B)
. Further exposure of the cells to Fn resulted in the return of the DRM-associated 4-subunit to initial levels (Fig. 6B)
. In the case of the ß1-integrin, results showed that it was initially present in DRM (Fig. 6C
and 6D) . Exposure of Jurkat cells to CnIV did not induce a recruitment of the subunit to DRM as a function of time (Fig. 6C
and 6D)
as shown by densitometric analysis (Fig. 7C)
. Exposure of Jurkat cells to Fn did not modifiy the distribution of the ß1 subunit in DRM as a function of time (Figs. 6D
and 7D)
.
 View larger version (25K): [in a new window] |
Figure 6. Time-course analysis of the distribution of the ß1-integrin subunits in DRM of Jurkat cells exposed to CnIV or Fn. The cells were exposed to surfaces coated (Coat) with CnIV or Fn for various periods of time at 37°C in GBSS and then lysed. DRM were isolated on sucrose-density gradients. The 2-, 4-, and ß1-integrin subunits were immunoprecipated, and the complexes were separated by SDS-PAGE. Western blottings (WB) were performed, and detection was done by ECL. The figure displays the results obtained (A and C) in the case of cells exposed to surfaces coated with CnIV and (B and D) in the case of cells exposed to Fn. The sizes of the two protein bands in the case of the 2-subunit were 158 and 126 kDa and 140 kDa in the case of the 4- and ß1-subunits, as determined from reference-colored protein standards (Sigma-Aldrich).
|
 View larger version (17K): [in a new window] |
Figure 7. Semiquantitative analysis of the time-dependent distribution of - and ß1-integrin subunits in DRM of Jurkat cells exposed to surfaces coated with CnIV or Fn. ß1-Integrin subunits revealed by Western blotting were scanned, and densitometric measurement was performed using the NIH Image software. Data are shown relative to band intensities observed at the start (t=0 min) of each experiment. Values are shown for (A) the 2-subunit (CnIV-coated), (B) the 4-subunit (Fn-coated), (C) the ß1-subunit (CnIV-coated), and (D) the ß1-subunit (Fn-coated).
|
 View larger version (27K): [in a new window] |
Figure 8. Patterns of Ca2+ mobilization in Fura2-loaded Jurkat cells using ß1-integrin ligand-coated microspheres (optical trapping) or sufaces. (A and B) Time-dependent Ca2+ responses triggered by CnIV-coated microspheres. (C and D) Time-dependent Ca2+ responses triggered by Fn-coated microspheres. In these experiments, the cells bathing in a 1 mM Ca2+-containing medium (GBSS) were maintained at 37°C. Individual cells were exposed to one ß1-integrin ligand-coated microsphere, and changes in the ratio of fluorescence measured at 340 and 380 nm [wavelength of emission (Em), 510 nm] were automatically recorded at 10-s intervals. (E) Profile of the Ca2+ response of untreated cells exposed to a glass coverslip coated with CnIV. (F) Results of a similar experiment using Jurkat cells previously treated with MCD (10 mM). (G) Profile of the Ca2+ response of untreated cells exposed to a glass coverslip coated with Fn. (H) Results of a similar experiment using Jurkat cells previously treated with MCD (10 mM). (I) Effects of thapsigargin (Tg, 2 µM) on Jurkat cells treated with MCD (10 mM). In these experiments, the cells bathed in a (1 mM) Ca2+-containing medium were maintained at 37°C. Changes in the ratio of fluorescence measured at 340 and 380 nm (Em, 510 nm) were automatically recorded at 10-s intervals. Data correspond to the Ca2+ response of 20–30 cells per field of acquisition.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3ß1 [50
], 6ß1 [51
], vß3 [38
], Lß2 [lymphocyte function-associated antigen-1 (LFA-1)] [52
], and Mß2 (Mac-1) [53
] integrins have been reported to be enriched in DRM. However, it is not known whether each member of the various integrin superfamilies can be recruited in DRM. We have addressed this question by exposing Jurkat T lymphocytes to ligands of ß1-integrins. The 2ß1 and 4ß1 integrins were selected because of their involvement in key events of lymphocyte responses. These integrins share the properties to bind to protein components of the ECM and are important participants in the migration of lymphocytes to injured peripheral tissues. In addition, the 4ß1 integrin is involved in the binding to activated endothelium, an early step leading to lymphocyte extravasation through the vasculature lining [54
]. Single-cell assays were performed using an optical tweezer to simulate real-time contact between a delimited region of individual Jurkat cells and ligands (CnIV and Fn) of ß1-integrins bound to microspheres. Results clearly showed that CnIV-coated microspheres induced a time-dependent coalescence of GM1-associated fluorescence at the region of contact with the microsphere (Fig. 1C) . Microspheres coated with Fn did not trigger apparent changes in the peripheral distribution of fluorescence (Fig. 2) , suggesting that CnIV and Fn induced a differential reorganization of lipid rafts. Cell treatment with MCD [49 ] prevented the dynamic redistribution of CnIV-dependent fluorescence, adding further evidence that lipid rafts were involved in the coalescence of individual CTx subunit B-labeled clusters (Fig. 1C) . Single-cell assays were also performed using LSCM to analyze the spatio-temporal distribution of GM1-associated fluorescence in Jurkat cells in response to CnIV and Fn bound to coverslips. The LSCM experiments revealed that CnIV-coated surfaces induced a dynamic recruitment of GM1-associated fluorescence (Fig. 4B) , which was not seen when the coverslips had been coated with Fn (Fig. 4D) or BSA (Fig. 4F) . Data obtained by LSCM were consistent with those obtained by optical trapping, further arguing that CnIV and Fn induced differential reorganization of lipid rafts in Jurkat T cells.
Isolation of DRM from lysates of Jurkat cells exposed to CnIV or Fn and Western blotting analysis was used to obtain semiquantification of the differential recruitment of ß1-integrins to lipid rafts. Relative densitometric quantification (Fig. 7)
clearly indicated differential dynamic recruitment of the 2- and 4-subunits (Fig. 6)
. Differences were noted in terms of kinetic, maximal levels, and amounts of DRM-associated -subunits relative to the onset of cell simulation. The distribution of the 2-subunit in DRM was transient, reaching maximal levels after 40 min and returning to initial levels after 75 min. These results are in apparent discrepancy with those obtained by visualization of the redistribution of the GM1 ganglioside (Fig. 1)
with respect to the time required to reach maximal effect. Whereas lipid raft coalescence appeared to reach a maximum after 20 min in the case of optical trapping experiments, DRM isolation studies showed a maximal recruitment of the 2-subunit to lipid rafts 40 min poststimulation. This difference may be explained by suggesting that the 2ß1 integrin may continue to be recruited to lipid rafts for an extended period of time or that the kinetics of 2ß1 integrin recruitment in lipid rafts depends on the surface-bound ligand available for integrin binding. Two protein bands of relative mass, 158 and 126 kDa, were revealed by the anti-2-subunit antibody used in Western blotting experiments. The presence of two protein bands in 2-subunit immmunoprecipitates of adult fibroblasts grown in collagen has been previously observed [55
].
The distribution of the 4-subunit with DRM was also transient. In this case, small increases in the levels of association with DRM (Fig. 7)
were observed after 5 and 10 min of cell stimulation, followed by a return to the initial levels. These observations and those of optical trapping experiments (Fig. 2)
suggested that exposure of Jurkat cells to Fn induced a weak translocation of the 4ß1 integrin to lipid rafts. A number of possibilities can be considered to explain these results. A first possibility is that Fn stimulation alone may not be sufficient to induce the translocation of the 4ß1 integrin to DRM. In this connection, Leitinger and Hogg [35
] have reported that activation of the Lß2 (LFA-1) integrin can induce the constitutive activation of the 4ß1 intcgrin by a cross-talk mechanism. A second possibility is that the 4ß1 integrin may lack a required shell of sphingolipids and cholesterol to efficiently be recruited to lipid rafts [56
]. In this connection, Leitinger and Hogg [35
] and Shamri et al. [57
] have observed that the 4ß1 intcgrin is not predominantly found in lipid rafts in unstimulatd Jurkat and peripheral blood lymphocytes. A third possibility is related to VCAM-1 as a physiological ligand [6
] of the 4ß1 integrin. It is not known at present whether exposing Jurkat cells to VCAM-1 would induce lipid raft coalescence and the recruitement of the 4ß1 integrin in DRM. A fourth possibility is that scaffold proteins could preferentially associate with ß1-integrins to favor the recruitment of the 4ß1 intcgrin to DRM. In this respect, it has been recently reported that the heat-shock protein 90 interacts preferentially with acylated G12 and targets it to lipid rafts as opposed to G13 [58
].
The ß1-integrin subunit was already present in high levels in DRM at the beginning of the experiments (Fig. 6)
. Exposing Jurkat cells to CnIV did not significantly increase its redistibution to DRM (Figs. 6
and 7)
, although Fn caused a slight increase in the level of distribution in DRM after 5 min of stimulation. The role of the ß1-subunit within the lipid raft microdomains is difficult to interpret, owing to the fact that this subunit is shared by other -subunits. Jurkat cells express six members (1, 2, 4, 5, 6, and 7) of the ß1-integrin superfamily [59
], and the behavior of these integrins with respect to their recruitment to lipid microdomains is unknown.
Ligation of ß1-integrins using surfaces coated with appropriate ligands has been shown to trigger Ca2+ mobilization in Jurkat lymphocytes [27
, 28
]. Furthermore, Wei et al. [60
] reported that microspheres coated with an anti-CD3 antibody triggered Ca2+ mobilization in murine lymphocytes after cell contact had been established by optical trapping. We used a similar approach to determine whether ligation of the 2ß1 integrin would trigger Ca2+ mobilization in Jurkat cells. Results clearly indicated that CnIV-coated microspheres were effective in inducing Ca2+ responses in individual cells. We [61
] and others [62
] have previously reported that individual Jurkat T cells display characteristic patterns of Ca2+ responses. Results shown here (Fig. 8)
clearly indicated that this was also the case when CnIV was used as a stimulus. For instance, a biphasic profile (Fig. 8A)
or complex patterns of Ca2+ spiking were observed (Fig. 8B)
. Similar remarks could be made in the case of Jurkat cell response to Fn-coated microspheres. In this case, monophasic (Fig. 8D)
or biphasic (Fig. 8C)
Ca2+ profiles were observed, although complex patterns of Ca2+ spiking were not observed.
Data showed that the kinetics of ß1-integrin-triggered Ca2+ responses (Fig. 8)
was faster than the recruitment of the 2ß1 and 4ß1 integrins in DRM (Fig. 6)
. These observations would suggest that perturbation of the lipid raft microdomains should not affect Jurkat cell response to ligation of the ß1-integrins. This hypothesis was tested by exposing Jurkat cells to glass surfaces coated with CnIV or Fn. Results showed (Fig. 8E
and 8G)
biphasic Ca2+ responses characterized by a rapid rise in [Ca2+]i followed by a plateau. Unexpectedly, treating the cells with MCD abolished the Ca2+ responses in both cases (Fig. 8F
and 8H)
. The lack of response was not a result of a perturbation of the state of the Ca2+ reserves, as thapsigargin triggered a robust Ca2+ response in these cells (Fig. 8I)
. The sensitivity of the 2ß1 and 4ß1 integrin dependency on intact lipid rafts suggests that the basal levels of these integrins present in DRM may be sufficient to trigger Ca2+ mobilization in response to stimulation with CnIV or Fn. In addition, our data show that DRM are required for efficient Ca2+ mobilization in Jurkat cells, a process that depends on cross-linking 2ß1 and 4ß1 integrins [27
].
Results reported here showed that the 2ß1 and 4ß1 integrins that share a common ß1-subunit behaved differently in response to their ligands in a number of assays. Of significance were the findings of their differential distribution in DRM in response to ligation, suggesting that their respective -subunits play a critical role in this process. The differential recruitment of the 2ß1 and the 4ß1 integrin in DRM may have implication on their signal-transducing properties given the fact that lipid rafts can act as molecular filters [42
] by assembling or excluding some integrins and effectors of downstream signaling, such as key protein kinases and phosphatases. The differential localization of the 2ß1 and 4ß1 integrins in DRM may be an additional mechanism in the timing of lymphocyte rolling and attachment to the endothelium that leads to lymphocyte extravasation and tissue infiltration.
Received September 26, 2002; revised November 7, 2002; accepted November 9, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Vß3, integrin-associated protein (CD47), and heterotrimetric G proteins J. Cell Biol. 146,673-6823ß1 and 6ß1 physically and functionally associate with CD36 in human melanoma cells. Requirement for the extracellular domain of CD36 J. Biol. Chem. 275,35264-352754ß1 governs lymphocyte migration J. Immunol. 167,2824-28302ß1 is upregulated in fibroblasts and highly aggressive melanoma cells in three-dimensional collagen lattices and mediates the reorganization of collagen I fibrils J. Cell Biol. 115,1427-14364 integrin avidity but not of LFA-1 avidity to endothelial ligands under shear flow requires cholesterol membrane rafts J. Biol. Chem. 277,40027-4003512 but not G13 to lipid rafts J. Biol. Chem. 277,32409-32412This article has been cited by other articles:
 |
|
 |
|
  S. Krishnan, M. P. Nambiar, V. G. Warke, C. U. Fisher, J. Mitchell, N. Delaney, and G. C. Tsokos Alterations in Lipid Raft Composition and Dynamics Contribute to Abnormal T Cell Responses in Systemic Lupus Erythematosus J. Immunol., June 15, 2004; 172(12): 7821 - 7831. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Larbi, N. Douziech, G. Dupuis, A. Khalil, H. Pelletier, K.-P. Guerard, and T. Fulop Jr Age-associated alterations in the recruitment of signal-transduction proteins to lipid rafts in human T lymphocytes J. Leukoc. Biol., February 1, 2004; 75(2): 373 - 381. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
