Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Accumulation of infiltrating cells at allergic, inflammatory sites depends on the interactions between the adhesion molecules of the infiltrating cells and endothelial cells at such sites (1). One of the major populations of cells that accumulate at sites of allergic inflammation are lymphocytes. These cells express an array of adhesion molecules, among which are integrins. The very late activating (VLA) family of integrins is composed of an  chain noncovalently associated with a  chain. There are several different integrins, determined by the pair of chains that creates them. The ₄ chain can associate with either the ₁ (CD29) or the ₇ chain to create VLA-4 or ₄₇ integrin, respectively (2). Both integrins bind to fibronectin and vascular cell adhesion molecule-1 (VCAM-1). However, ₄₇ also preferentially binds to the mucosal vascular addressin called MAdCAM-1, and its major role is therefore believed to be mediating the homing of lymphocytes to mucosal tissue (3). In contrast, VLA-4, by binding to cytokine-induced VCAM-1 and/or to matrix proteins with higher preference than ₄₇, may play a main role in the entry of lymphocytes into inflamed tissue (4).

The interaction of VLA-4 with VCAM-1 has been shown to be important for site-specific lymphocyte adherence in models of tissue inflammation, including rheumatoid arthritis (5, 6), cytokine activated skin sites (7), contact hypersensitivity skin sites (8), bancroftian filariasis (9), and the brain in experimental autoimmune encephalitis (10). Unlike intercellular adhesion molecule-1 (ICAM-1) or E-selectin, the use of VLA-4 by T cells for tissue binding is specific, in that it requires prior or concurrent T-cell activation for effective adherence. Conversely, binding to VCAM-1 or to fibronectin through VLA-4 with T-cell receptor (TCR) engagement has been shown to be necessary for T-cell activation and proliferation (11).

Previous studies have suggested that the increased adhesion seen with antigen-specific (14) or nonspecific activation (15) is due to affinity modulation of the VLA-4 receptor or to the development of CD4⁺ memory T cells. Modulation of the frequency or a change in the density of expression of the VLA-4 receptor was not felt to be important in this process, since receptor density was unchanged. However, in these reports, the period of activation was relatively short, ranging from 15 to 60 min. Other studies, of stable, activated human T cells in rheumatoid arthritis, have shown both increased expression and function of VLA-4 receptors, but have not looked for changes in these receptors during stimulation (6).

Although the T cell is felt to be central to the pathogenesis of allergic disease and asthma, the mechanism responsible for selective migration of allergen-specific T cells to sites of allergic inflammation has yet to be elucidated.

In the present study we show that CD49d upregulation is an important component of T-cell activation in an allergen-driven model of inflammation. Allergen stimulation upregulates CD49d expression in a time-dependent manner, and coordinates with expression of the DR subset of human leukocyte antigens (HLA-DR), whereas the expression of other adhesion molecules is not modulated. Furthermore, the increased CD49d expression correlates with increased adhesion to the CS-1 fragment of fibronectin, providing evidence that the increased CD49d is functionally active.

This study demonstrates the modulation of CD49d on T cells in allergic disease, and suggests a mechanism for T-cell accumulation in allergen-exposed allergic lung. We hypothesize that the increase in CD49d expression on allergic T cells, and their increased adhesivity induced by allergen exposure, may be the critical first step in initiating specific lymphocyte influx into the allergic lung in asthma.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Subjects

Subjects included both atopic (n = 8) and nonatopic (n = 6) individuals. Nonatopic individuals included three male and three female subjects with an average age of 35 yr, characterized by negative skin tests and absence of symptoms to allergen. Atopic individuals included six male and two female subjects with an average age of 37 yr. Atopic individuals were characterized by positive skin tests and proliferation assays, and symptoms of allergic rhinitis and/ or asthma on exposure to allergen. Subjects for allergen challenge followed by bronchoalveolar lavage (BAL) had mild asthma controlled with theophylline or -agonist medications. All medication was withheld for 24 h prior to bronchoscopy. Normal subjects for BAL were receiving no medications and had no lung disease. Control subjects with other inflammatory lung disease underwent bronchoscopy for diagnostic purposes. All subjects signed informed consent agreements to participate in the study.

Cellular Sources

Peripheral blood T cells were obtained from atopic (n = 8) and nonatopic (n = 6) subjects. Blood was diluted 1:1 in phosphate-buffered saline (PBS) and was centrifuged over Ficoll-Paque at a concentration of 1.077 g/ml (Pharmacia, Piscataway, NJ) at 400 × g for 30 min. Mononuclear cells aspirated from the interphase were washed three times in PBS, resuspended in RPMI 1640 (Gibco/BRL, Gaithersburg, MD) and placed into cultures at 2 × 10⁶ cells/ml.

T-cell lines specific for Mus m I, Der p I, and Timothy were established from atopic subjects with positive skin tests and proliferation assays to each allergen. The cell lines were obtained by stimulation of peripheral blood T cells (2 × 10⁶/ml) every 12 to 14 d with a 1:3 to 1:5 ratio of irradiated Epstein-Barr virus (EBV)-transformed allogeneic B cells as antigen-presenting cells (APC), plus allergen at a concentration deemed optimal on the basis of the subject's previously established proliferation assay (usually from 1 to 10 µg/ml). Cell lines were examined after two to four passages with allergen. Mus m I-specific T-cell clones were derived from T-cell lines after four passages with allergen. Cells were set up at limiting dilutions of 0.3 cells/well, and were cultured with allergen along with APCs as described earlier, plus 10 U/ml of human interleukin-2 (IL-2) (Advanced Biotechnologies, Inc.) and 2 U/ml of recombinant human IL-4 (Genzyme, Cambridge, MA).

Lung T cells were obtained by BAL from mouse-allergic individuals (n = 3) with mouse-allergen-induced asthma, and from normal subjects (n = 4). Baseline cells were first lavaged from the right middle lobe. Allergic individuals were provoked by allergen instillation into the left lingula at a dose determined by the subject's skin-test reactivity. Allergen-challenged BAL cells were obtained 48 h later by lavage of the left lingula. Lavage cells were filtered through gauze and washed twice in PBS. Red cells in the pooled pellet were lysed with Gey's solution, and the remaining cells were washed twice in PBS before resuspension in RPMI 1640 (Gibco/BRL) for analysis.

Reagents

Cells were cultured in RPMI 1640 medium with 10% fetal calf serum (FCS), 200 U penicillin/ml, 200 µg streptomycin solution/ml, 4 mM L-glutamine, 50 µg/ml gentamicin, 0.1 mM nonessential amino acids, 10 mM 4-(2-hydroxyethyl)-1-piperazine-N'-2-ethanesulfonic acid (Hepes) and 1 mM sodium pyruvate (all from Gibco/BRL).

Timothy (Miles, Inc.), Der p I (Center Laboratories), or Mus m I (obtained from Dr. J. Ohman) were used to generate allergen-specific T-cell lines or clones. Human plasma fibronectin (Gibco/BRL) and the CS-1 (1- to 25-amino acid) fragment of fibronectin (Sigma) were used in adhesion tests. Monoclonal antibodies against CD3, CD49d (VLA-4), HLA-DR, lymphocyte function-associated antigen (LFA)-1b, ICAM-1 (Becton-Dickinson), and LFA-3 (the gift of Dr. D. Hamilos) were used in cell staining.

T-cell Stimulation

T-cell lines and clones were set up at 1 to 2 × 10⁶ cells/ml in fresh culture medium. Allergen was added at 1 to 10 µg/ml as described earlier. IL-2 was added during the stimulation of T-cell lines. Epstein-Barr virus-transformed allogeneic B cells, prepared as described subsequently, were added at a ratio of 1:3 to 1:5 as APCs. T cells were aspirated 24 h, 48 h, or 6 d after allergen stimulation. Control cells (0 h) consisted of cells aspirated prior to allergen stimulation.

Preparation of APCs

Peripheral blood mononuclear cells (PBMC) were isolated on a gradient of Ficoll-Paque 1.077 g/ml (Pharmacia). B cells were isolated by E-rosetting and transformed with supernatant containing Epstein-Barr virus-released myeloma B958. Immortalized B cells were cultured in complete RPMI 1640 with 10% nu serum. Epstein-Barr virus-transformed B cells were gamma-irradiated at 12,000 rads, and constituted the source of APCs that were added together with allergen to T-cell cultures to establish T-cell lines or clones.

Cell Staining and Flow-Cytometric Analysis

Cells were stained according to standard procedure with primary antibodies against CD3, CD49d, HLA-DR, CD11b, CD58, or CD54, in PBS containing 2% FCS and 0.02% sodium azide. Control cells were incubated with irrelevant antibodies conjugated with phycoerythrin or fluorescein to account for nonspecific binding. After 30 min of incubation at 4°C, cells were washed three times. Subsequently, the secondary antibody against mouse IgG, conjugated with phycoerythrin or fluorescein, was added, and cells were incubated for another 30 min at 4°C. Cells were washed three times and finally resuspended in 4% paraformaldehyde. Unstimulated control peripheral blood lymphocytes were directly stained in blood.

Flow-cytometric analysis was performed on an EPICS Profile I Analyzer (Coulter, Inc., Hialeah, FL). Cells were gated on typical lymphocyte scatter. Mean channel fluorescence (MCF) was analyzed on CD3⁺ T cells for the molecule of interest.

T-cell Adhesion to CS-1 and Fibronectin

Flat-bottom wells of 96-well culture plates (Costar) were coated overnight at 4°C with 50 µl of fibronectin or CS-1 at 10 µg/ml. Unbound ligand was washed away with PBS containing 0.2% human serum albumin (PBS/HSA). Remaining binding sites were blocked with 1% bovine serum albumin (BSA) at 37°C for 2 h. After the last wash, 50-µl aliquots of PBS/HSA were dispensed and the plate was incubated for 30 min at 4°C. T cells were labeled with ⁵¹Cr (Amersham) in complete RPMI 1640/10% FCS for 1.5 h at 37°C, and were washed afterwards with cold PBS/HSA. As a last step, cells were resuspended in cold PBS/HSA at a concentration of 1 × 10⁶/ml, from which 50-µl aliquots were dispensed into the wells of the plate kept at 4°C. Samples were generated in triplicate. Adhesion was allowed to occur for 1 h at 4°C and 10 min at 37°C. Unbound cells were washed away with PBS/HSA at room temperature. Adherent cells were lysed with 1% Nonidet P-40 (Sigma) and ⁵¹Cr release was measured in a gamma counter (1470 Wizard; Wallac). Percent of adherent cells was calculated according to the following equation:

(1)

where: background = cells incubated in wells coated with 1% BSA.

Statistical Analysis

Baseline MCF (0 h) for T lymphocytes was compared with that of unstimulated peripheral blood lymphocytes (unstimPBL) through one-way analysis of variance (ANOVA) with Dunnett's multiple comparison procedure, using JMP statistical computer software. The MCF of cells at 24 h, 48 h, or 6 d after stimulation was compared with the baseline MCF, using an individual linear contrast, if the main effect for day was significant at the P = 0.05 level. The statistical analysis also included two-way ANOVA of MCF of cells at Day 0, 24 h, 48 h, or 6 d, in which observations are blocked on experiment. The analysis of the studies of MCF are based on the log₁₀ of the raw data. One-way ANOVA was performed in comparing the results of adhesion of stimulated T cells with adhesion of the cells before restimulation. Statistical significance is included with each figure or table.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

₄ (CD49d) Receptor Density on T Cells following Allergen Challenge

We examined the effect and time course of in vitro stimulation of allergen-specific T cells with a dose of allergen that had previously been shown to elicit maximal response in a T-cell proliferation assay (data not shown). Since most T cells from blood and BAL were positive for CD49d expression, we found that the percent of positive cells measured after allergen stimulation was an insensitive measure of change (data not shown).

In contrast, we found that receptor density, as measured by MCF, in flow-cytometric analysis, did rise significantly with allergen challenge in a consistent, time- dependent manner. Allergen-specific T-cell lines exhibited a baseline receptor density (0 h) that was significantly higher, by 2-fold, than that seen on unstimulated circulating peripheral blood T cells (unstimPBL) obtained from healthy subjects (Figure 1A). After 24 h of allergen stimulation, receptor levels rose by 150% and remained significantly elevated after 48 h. The MCF of CD49d cells before allergen stimulation was also significantly greater for T-cell clones than for unstimPBL T cells. Receptor density per cell rose by 130% after 24 h of allergen stimulation, and returned to baseline after 48 h (Figure 1B).

View larger version (47K): [in this window] [in a new window]   View larger version (48K): [in this window] [in a new window]   View larger version (41K): [in this window] [in a new window]   Figure 1.   CD49d receptor density on allergen-challenged T cells. Changes in receptor density were measured by MCF, and are presented as FACS analysis graphs (upper graphs) and bar graphs, which represent corresponding quantitative changes (lower graphs). (A) T-cell lines derived from allergic individuals were stimulated with allergen and analyzed with flow cytometry. Results shown represent seven individual experiments, and are compared with the MCF of unstimulated peripheral blood lymphocytes (unstimPBL) obtained from healthy donors. Unstimulated PBL values represent 16 individual experiments. Bars on the ordinate show values for time 0, 24 h, and 48 h after allergen stimulation. (B) T-cell clone cells derived from three separate rodent-allergic individuals were stimulated with allergen in a similar manner, and were analyzed for changes in receptor density over 48 h. (C) Peripheral-blood T cells obtained from five allergic individuals were assessed for CD49d receptor density at baseline. No difference was seen in comparison with cells from 16 healthy donors. Cells were then stimulated with allergen and analyzed for changes in receptor density at 24 h, 48 h and 6 d after allergen stimulation in vitro. Results represent geometric mean ± SEM; N.S. = P value not significant.

The upregulation of CD49d MCF was also seen when peripheral blood T cells from allergic individuals were stimulated with allergens (Figure 1C). After 24 h of allergen stimulation, CD49d density rose significantly, by 180%. By 48 h, MCF was 240% of that seen at baseline, and began to fall slightly by Day 6, but was still significantly elevated. Peripheral blood T cells cultured in 10% FCS or 10% autologous serum without added allergen did not upregulate in response to culture conditions alone (data not shown).

CD49d Expression on T Cells Isolated from BAL before and after In Vivo Allergen Provocation

Lung T cells obtained from BAL fluid (BALF) of normal subjects and from endobronchially challenged mouse-allergic individuals were examined with flow cytometry (Figure 2). Baseline expression of CD49d on lung T cells from allergic individuals was nearly 6-fold greater than on peripheral blood T cells, and 3-fold greater than on lung T cells from BAL of normal subjects. After allergen challenge, CD49d receptor density was upregulated by values ranging from 135% to 158%. The difference in receptor levels did not reach statistical significance, in part due to the sample size and in part because of the presence of a population of allergen-nonspecific T cells. Lung cells were also obtained from individuals with other inflammatory lung disease (n = 4) and examined for CD49d expression. The two subjects with idiopathic pulmonary fibrosis (IPF) had too few T lymphocytes (< 4% CD3⁺ cells) to evaluate. The other two subjects, with granulomatous lung disease (hypersensitivity pneumonitis and beryllium sensitization), were evaluable. CD49d expression on these subjects' CD3⁺ T cells was 21.5 ± 9.2, which is not different from that seen for normal subjects.

View larger version (34K): [in this window] [in a new window]   Figure 2.   VLA-4 receptor density on allergen-challenged BAL T cells. Cells were obtained by endobronchial lavage of the right middle lobe from three rodent-allergic individuals before stimulation (0 h). Concurrently, allergen was instilled into the left lingula. Forty-eight hours later, the left lingula was lavaged and CD3+ cells were analyzed for changes in receptor density as assessed by MCF. Cells were also obtained from four normal subjects (normal) in the same way as from allergic subjects, but with no allergen challenge. Changes are presented as FACS analysis graphs (upper graphs) and bar graphs, which represent corresponding quantitative changes (lower graphs). Results are compared with VLA-4 receptor density on unstimPBL T cells from 16 healthy donors, or with the MCF of CD3+/VLA-4+ BAL T cells before provocation. Results represent geometric mean ± SEM; N.S. = P value not significant.

In some cases, CD49d was examined on T cells in paired blood and BAL samples, before and after challenge. CD49d MCF rose by 200%, from 15 at baseline to 33.5 2 d after allergen challenge. This pattern was similar to that seen with PBMCs exposed to allergen in vitro.

Adhesion of T Cells to Fibronectin and CS-1 Fragment

Only 4.72% of allergen-specific T cells adhered to the CS-1 fragment before their stimulation. Twenty-four hours after stimulation with allergen, the percentage of T cells that adhered to CS-1 increased significantly, to 21.21%. Forty-eight hours later, this increase was still significant, with 13.95% of T cells being adherent (Figure 3). Binding to fibronectin was greater at baseline, with 31.18% of T cells being adherent. However, binding was not increased during the course of allergen stimulation. Twenty-four hours after allergen stimulation, 29.98% of T cells were bound, and after 48 h of allergen stimulation 34.7% of T cells were fibronectin adherent (Figure 4).

View larger version (27K): [in this window] [in a new window]   Figure 3.   Adhesion of Timothy-allergen-specific T-cell lines to the CS-1 fragment. Adhesion to the CS-1 fragment of fibronectin was determined in Timothy-pollen-allergen-specific T-cell lines derived from six atopic individuals at 0 h, 24 h, and 48 h after allergen stimulation. Results are shown as percent of cells that adhered to CS-1. Results are presented as geometric means ± SEM.

View larger version (45K): [in this window] [in a new window]   Figure 4.   Adhesion of Timothy-allergen-specific T-cell lines to fibronectin. Adhesion to fibronectin was determined in Timothy-pollen-allergen-specific T-cell lines derived from six atopic individuals at 0 h, 24 h, and 48 h after allergen stimulation. Results are shown as percent of cells that adhered to fibronectin. Results are presented as geometric means ± SEM; * changes not significant at any time point.

HLA-DR Expression on In Vitro or In Vivo Allergen-stimulated T Cells

We chose HLA-DR, which is well described as an activation marker, to monitor the effect of allergen stimulation. As expected, both the number of blood T cells expressing HLA-DR and the HLA-DR MCF increased following allergen stimulation in T-cell lines (from 48% to 93%) and peripheral-blood leukocytes (from 5% to 16%). T-cell clones already expressed high levels of HLA-DR (89%), and showed only a modest increase in HLA-DR expression (to 93%) after allergen challenge (data not shown). Expression of HLA-DR on BAL-derived T cells was increased at baseline (98%), with no further increase seen 48 h after allergen stimulation (data not shown).

HLA-DR receptor density also rose significantly with repeated allergen stimulation, with effects limited to the most allergen-selected T cells (e.g., lines and clones). No change in HLA-DR receptor MCF after allergen challenge was seen with either BAL-derived T cells or stimulated peripheral blood T cells (data not shown).

ICAM-1 (CD54) Expression on T Cells after Allergen Stimulation

ICAM-1 expression was not upregulated by allergen stimulation in vitro or in vivo (data not shown). Less than 1% of circulating blood T cells expressed ICAM-1, consistent with an inactivated state and the ability to circulate rather than to adhere. Of T cells in culture, an average of 30% expressed ICAM-1, but no significant upregulation of the numbers of cells expressing ICAM-1 was seen after stimulation with allergen. The higher baseline levels seen of ICAM-1 expression is consistent with the increased expression reported under activated culture conditions (16). Only 9% to 11% of BAL-derived T cells expressed ICAM-1, with no further change following allergen challenge. Incubation of peripheral blood T cells with allergen produced only a modest increase in cells expressing ICAM-1, from 0.4% to 1.3% after 48 h, and to 3.2% at 6 d. As a control, addition of phytohemagglutinin (PHA) to the same peripheral blood T cells resulted in a 30% expression of ICAM-1 by Day 1 and 45% expression by Days 2 to 6 (data not shown).

Allergen stimulation also did not upregulate the MCF for ICAM-1 on any of the T-cell lines or clones (data not shown).

LFA-1b (CD11b) and LFA-3 (CD58) Expression on T Cells before and after Allergen Challenge

Virtually all cultured T cells are positive for LFA-1b, with a characteristic bimodal distribution. There was no change in expression of this integrin or its MCF with the addition of allergen to the culture (data not shown).

The expression of LFA-3 on T cells is similar to that of LFA-1. Approximately 90% to 100% of cultured T cells express this adhesion molecule (data not shown). The number of LFA-3-positive T cells and the MCF for LFA-3 also did not change with repeated stimulation.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Allergic inflammation is determined by the accumulation of a characteristic array of cells at specific tissue sites. The accumulation of these cells, among which are lymphocytes, is regulated in part by the expression of adhesion molecules on the migrating cells and in part by the adhesion molecules on endothelial cells. Changes in the expression of adhesion molecules may affect the migratory abilities of T lymphocytes. Integrins are one of the main groups of adhesion molecules expressed by T lymphocytes. Integrins are composed of an  chain that is noncovalently associated with a  chain. There are several different integrins determined, by the pair of chains that create them (2). The ₄ chain (CD49d) can associate with either the ₁ (CD29) or the ₇ chain to create VLA-4 or ₄₇ integrin, respectively. Both integrins bind to fibronectin and VCAM-1. However, ₄₇ preferentially binds also to the MAdCAM-1, and its major role is therefore believed to be mediating the homing of lymphocytes to mucosal tissue (3). On the other hand, VLA-4, by binding with a higher preference than ₄₇ to IL-4 (17), IL-13 (18), IL-1 (19), or tumor necrosis factor- (TNF-) (20) induced VCAM-1 (21), and/or to matrix proteins, may play a main role in the entry of lymphocytes into inflamed tissue (4). Changes in the expression of VLA-4 or of one of its chains on T cells may determine the relative ability of these cells to migrate to sites of inflammation.

Our data demonstrate that allergen stimulation upregulates the expression of CD49d in a time-dependent manner on an array of allergen-specific T cells. Upregulation occurs on T-cell lines and clones derived from peripheral blood of atopic asthmatic individuals, and less significantly on T cells derived from allergic airways. Since most cells, resting as well as activated, express CD49d on their surface, the actual nature of CD49d upregulation in response to allergen challenge is not the percentage of cells expressing this receptor, but rather receptor density as assessed by MCF.

The baseline levels of CD49d receptor density on T cells also rise over time with repeated allergen stimulation. This may reflect the selection of memory T lymphocytes, which express high levels of surface VLA-4, as found by the use of an antibody that recognizes CD49d associated with ₁ chain only. These cells also show increased binding to fibronectin as compared with naive T cells (15). In our experiments, we sought to determine whether the enhanced CD49d density seen 24 h to 48 h after allergen stimulation also correlated with enhanced function. We analyzed binding of T cells to fibronectin and to the CS-1 fragment of fibronectin, which specifically binds CD49d (22). Our results showed that binding to CS-1 was significantly increased after 24 h and 48 h of stimulation with allergen and APCs, although binding to whole human plasma fibronectin was unchanged. These findings confirm that the increased density of CD49d is functional and responsible for increased binding to its specific ligand. The unchanged adhesion of allergen-stimulated T cells to human-plasma-derived fibronectin may reflect the lower content of CS-1 in plasma-derived fibronectin. Tissue fibronectin has been shown to contain 50% more CS-1 than plasma fibronectin (23), suggesting that adhesion to CS-1 is relatively more important in tissue than in the blood. It is also possible that the content of CS-1, the alternatively spliced fragment of fibronectin, may be altered in inflamed tissue by local cytokines (24), leading to increased or decreased adhesion and migration of inflammatory cells. The unchanged adhesion of allergen-stimulated T cells to whole fibronectin may also reflect the involvement of VLA-5, which also binds to fibronectin and which may dilute the effect of enhanced reaction between CD49d and the CS-1 fragment.

The increased expression of CD49d after 24 h and 48 h of allergen exposure correlates with the increased expression of HLA-DR as a marker of activation. By Day 6, CD49d expression returns to baseline. The time course is reiterated in the rise and fall of HLA-DR expression. Coordinate upregulation of CD49d and HLA-DR expression by allergen suggests that CD49d expression is enhanced in the context of cell activation. Our observations on BAL-derived T cells provide evidence of the recruitment of these cells into lung tissue during the allergic response. We showed that T cells derived from BAL of asthmatic patients express a density of CD49d three times higher than those of T cells obtained from BAL of normal subjects. It is possible that the increased CD49d expression on these stimulated T cells permits more effective attachment to the vascular endothelium, and modulates extravascular migration. Migration of T cells through the extravascular spaces to the site of inflammation is mediated by binding between integrins and matrix proteins. The importance of the VLA-4/ CS-1 interactions in T-cell migration has been shown in studies by Ferguson and colleagues (25). These authors found that CS-1-pretreated T cells lost their ability to mediate in vivo reactions of contact hypersensitivity, providing evidence that CD49d plays an important role in the localization of T cells to sites of antigen challenge. The increased binding of T cells to the CS-1 fragment of fibronectin after antigen stimulation, as found in our experiments, suggests that CD49d may be important for efficient T-cell migration through extravascular spaces to the site of inflammation. The importance of VLA-4 in T-cell recruitment has been confirmed in an animal model in which antibody against VLA-4 prevented the antigen-induced infiltration of T cells into the mouse trachea (26). Inhibition of VLA-4 was more potent than blocking of ICAM-1 and LFA-1, which may reflect our findings of unchanged LFA-1b, LFA-3, and ICAM-1 expression after allergen challenge. In contrast to other adhesion molecules, changes in the expression of VLA-4 or of its CD49d chain may be a primary regulatory mechanism for T-cell adherence and migration. Other inflammatory diseases characterized by T-cell infiltration may also use CD49d modulation as the means of site-specific localization, as has been shown in human filarial inflammation (9). This mechanism is not unique to T cells, and may also be utilized by monocytes in their in vivo migration to inflammatory sites (27).

What is the functional importance to T cells of the increased expression of CD49d? Increased expression of CD49d on allergen-specific T cells, and their binding to the CS-1 fragment of fibronectin, may facilitate trafficking of these cells to sites of allergic inflammation. This process may be mediated by VLA-4 or by the ₄₇ integrin, of which the CD49d chain is a part. The modulation of CD49d expression may also provide a mechanism for selecting allergen-specific T cells that will enter the inflammatory site. Increased expression of CD49d may enhance the ability of allergen-specific T cells to proliferate in response to antigen-major histocompatibility complex stimulation by APCs, as has been shown with stimulation of resting CD4⁺ T cells (28).

In summary, these studies demonstrate coordinate upregulation of the integrin CD49d and HLA-DR on the surface of CD3⁺ cells that have been triggered by allergen exposure, and concomitantly increased adhesion of these cells to the CS-1 fragment of fibronectin. The significance of these findings is that they provide a potential mechanism for the selective accumulation of T cells at sites of allergic inflammation, with important implications for the pathogenesis of asthma and the allergic response.