| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
1 (VLA-1) and 21 (VLA-2) Integrins on CD4 and CD8 T Cells during Influenza Infection1
* Department of Microbiology and Immunology, David H. Smith Center for Vaccine Biology and Immunology, Aab Institute of Biomedical Sciences, University of Rochester, Rochester, NY 14642;
Novartis Institutes for Biomedical Research, Cambridge, MA 02139; and
Alnylam Pharmaceuticals, Cambridge, MA 02139
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
1 integrin family of collagen-binding receptors, 
11 and 21 (CD49a, VLA-1 and CD49b, VLA-2, respectively), on CD4 and CD8 T cells during the response to influenza infection in the lung. Flow cytometry showed that whereas T cells infiltrating the lung and airways can express both CD49a and CD49b, CD49a expression was most strongly associated with the CD8+ subset. Conversely, though fewer CD4+ T cells expressed CD49a, most CD4+ cells in the lung tissue or airways expressed CD49b. This reciprocal pattern suggested that CD4 and CD8 T cells might localize differently within the lung tissue and this was supported by immunofluorescent analysis. CD8+ cells tended to localize in close proximity to the collagen IV-rich basement membranes of either the airways or blood vessels, whereas CD4+ cells tended to localize in the collagen I-rich interstitial spaces, with few in the airways. These observations suggest that CD4 T cell interaction with the tissue microenvironment is distinct from CD8 T cells and support the concept that CD4+ T cells in peripheral tissues are regulated differently than the CD8 subset. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
11 (VLA-1) and 21 (VLA-2). VLA-1 preferentially binds to types I and IV collagen, whereas VLA-2 preferentially binds to type I collagen (3, 4, 5). Type I collagen is the most abundant collagen type, comprising 90% of the total body collagen, and is found in bone, skin, connective tissues, and internal organs (6). Type IV collagen is the major component of basement membranes and its distribution is restricted to this location (6). VLA-1 and VLA-2 have also been shown to be receptors for a variety of other less abundant ECM components including laminin (7). Respiratory infection with influenza virus is a major cause of illness and death in humans (8, 9, 10). Understanding the essential elements of immune protection from this virus is of obvious importance. Influenza virus is also unique among viruses in its ability to experience antigenic shift as well as antigenic drift (11, 12, 13). These changes produce new flu strains that avoid existing neutralizing Abs, thereby easily establishing infection. Fortunately, many individuals have heterosubtype-specific CD4+ and CD8+ T cells that recognize conserved internal epitopes common to different serotypes (14, 15). Thus, whereas these individuals may become reinfected with flu, the severity of disease and duration of infection is reduced when heterosubtypic immunity is present (14, 15).
Localization of memory CD4+ and CD8+ T cells to extralymphoid tissues is now understood to be an important attribute of immune protection (16, 17, 18). Heterosubtypic immune protection mediated by CD8+ T cells against respiratory infections has been shown to depend on the number of virus-specific cells in the airways (14, 17, 18). We have shown that VLA-1 plays an important role in regulating the ability to retain influenza-specific CD8 T cells to the lung tissue and airways (18). In fact, inhibition or deletion of CD49a, the 1 chain of VLA-1, reduced the number of flu-specific memory CD8 T cells in the lung and abrogated secondary immune protection (18). Whether VLA-1 plays a similar role on CD4 T cells is not known.
VLA-2, in contrast, has been shown to participate in the regulation of cell migration within collagen matrices (19, 20, 21). However, its role on T cells is less well-described. Understanding the patterns of VLA-1 and VLA-2 expression by extralymphoid T cells during the course of an immune response would enhance our understanding of how these integrins might function to regulate the antiviral immune response and the formation of tissue memory. Therefore, we have examined the overall patterns of VLA-1 and VLA-2 expression on Ag-specific and total CD4 and CD8 T cells responding to influenza infection in the lung. In addition, immunofluorescence was used to determine the distribution of collagen in the lung and the localization of CD4 and CD8 T cells relative to the matrix environment.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Female C57BL/6 (B6) mice were purchased from Taconic Farms at 6 wk of age. Congenic Thy1.1+B6.PL and Thy1.2+OT-II CD4+ TCR-transgenic mice (22) were obtained from colonies maintained at the University of Rochester. All animals were housed in the University of Rochester vivarium facilities under specific pathogen-free conditions using microisolator technology. Primary inoculation with influenza virus was performed in animals 8–12 wk of age. All experimental protocols have been reviewed and approved by the Institutional Animal Care and Use Committee and the animal facility is Association for Assessment and Accreditation of Laboratory Animal Care accredited.
Viruses and infection
The H3N2 A/Hong Kong/X31 (X31) influenza virus was grown and titered in embryonated chicken eggs and harvested as allantoic fluid preparations (23). The WSN-OVAII influenza virus was used as described (24). For primary infections, mice were sedated with avertin (2,2,2-tribromoethanol) before intranasal challenge with 105 EID50 of X31 or 500 PFU WSN-OVAII in 30 µl of PBS.
Induction of experimental allergic airway response to OVA
Thy1.1+B6.PL mice received an adoptive transfer of 1 x 106 CD4+Thy1.2+ congenic OVA323–339-specific TCR-transgenic spleen cells. One and 8 days later, the mice were given OVA (4 mg) in alum (10 mg) in 0.5 ml of PBS i.p. At 15 days after transfer, the animals were exposed two times for 1 h each with 4 h between exposures to OVA aerosol (5 mg/ml) in an inhalation chamber using a small volume nebulizer. Animals were sampled 5 days later to obtain lymphocytes from airway, lung tissue, and spleen.
Integrin-specific Abs
The function blocking hamster mAb Ha31/8 against 1 (25), and the anti-keyhole limpet hemocyanin Ha 4/8 control mAb were all affinity purified in an azide-free, low endotoxin. For flow cytometry, the Abs were conjugated to Alexa 488 or biotin according to the directions of the manufacturer (Molecular Probes).
Cell isolations
Following cardiac perfusion of the mice with PBS, the trachea was cannulated and bronchoalveolar lavage cells were collected by lavage with 1 ml of HBSS three times. These were then resuspended in complete MEM and plated in a cell culture-treated petri dish for 45 min at 37°C to remove adherent cells. Animals were then perfused with cold PBS through the left ventricle of the heart, and the lavaged and perfused lung tissue was minced and disrupted using a dounce homogenizer. After passing through a 90-µm nylon mesh, lung tissue lymphocytes were isolated from the homogenate using histopaque-1083 (1083-1; Sigma Diagnostics). Single-cell suspensions were prepared from spleen and lymph nodes by disruption in a dounce homogenizer followed by passage through 90-µm nylon mesh. Splenocytes and PBL were depleted of erythrocytes using a buffered ammonium chloride solution (Gey’s solution). All cell counts were obtained by trypan blue exclusion.
Flow cytometric analysis
Lymphocyte populations were stained as aliquots of 2 x 105 cells with various combinations of mAbs to CD8 (53-6.72 or CT8a), Thy1.1 (OX-7), Thy1.2 (30-H12), CD44 (IM7), and anti-CD62L (MEL-14) conjugated to FITC, PE, biotin, allophycocyanin, or PE-Cy5. The conjugated mAbs were purchased from BD Pharmingen or Caltag Laboratories and are referenced in their current catalogs. Tetrameric complexes of H-2Db/influenza polymerase (PA)224–233 (DbPA), H-2Db/influenza nucleoprotein (NP)366–374 (DbNP), and were prepared by the Trudeau Institute Molecular Biology Core Facility and used as described previously (26, 27). Data were collected using either a BD Biosciences FACSCalibur in four-color mode, or an LSR-II in 11-color mode, and analyzed using FlowJo software (Tree Star).
Histology and immunofluorescence
Mice were infected with HK/X31 (30 µl) intranasally. On sequential days following infection, mice were placed under deep anesthesia using avertin. The peritoneal cavity was opened and spleens were removed, washed with PBS, placed in OCT and frozen using a mixture of dry ice and 2-methyl butane (–45°C). The rib cage was opened and a blunted 23-gauge needle was placed in an incision made in the trachea. The canula was tightly fixed in place using suture. Warmed OCT (0.8 ml) was slowly injected using a 1-ml syringe to inflate the lungs. The OCT was held in the inflated lungs by tying the trachea with suture. The lungs were then carefully excised and placed in OCT as described for the spleen. Tissues were stored in a –80°C freezer. Sections (5–10 µm) were cut using a cryostat. For staining, sections were thawed and residual OCT was removed by incubating for 5 min with 
1 ml of PBS. The sections were fixed in a mixture of methanol-acetone (1:1) for 5 min. Following fixation, sections were left to dry for 15 min. The sections were then rehydrated in a 5-min incubation step with PBS-Tween 20 (0.05%). After this point, all incubation and washing steps were made using PBS-Tween 20. An FcR blocking step was performed using unlabeled anti-CD16/32 (BD Pharmingen; 1:200) for 20 min. The sections were washed twice for 5 min and stained with PE-labeled rat anti-mouse CD8a (Caltag Laboratories; 1:100), allophycocyanin-labeled rat-anti-mouse CD4 (BD Pharmingen; 1:100), and unlabeled goat anti-type IV collagen (Southern Biotechnology Associates; 1:200) for 60 min. Sections were then washed and stained with a secondary donkey anti-goat FITC for 45 min. The sections were then washed and mounted using a cover slip. Fluorescence microscopy was performed using a Nikon Eclipse E600 fluorescence microscope equipped with a 100 W mercury lamp (Chiu Technical) and a SPOT RT Color digital camera (Diagnostic Instruments).
Computerized image analysis of histological sections
Images for CD4 T cells, CD8 T cells, and type IV collagen were obtained as grayscale images, then assigned to blue, red, and green channels, respectively. The intensity histograms for red and blue channels were stretched to cover the entire eight-bit range. Segmentation processing was used to distinguish the individual cells and collagen structures. All algorithms as well as the graphical user interface were developed in MATLAB and MATLAB is required for their use. Image contrast was enhanced using a negative Laplacian filter and noise was reduced using a 7 x 7 median filter applied consecutively five times. Intensity holes in the resulting image were filled. The image was then inverted and a watershed transform applied. The watershed image was converted to binary. Nonuniform background was approximated by multiplying the filtered image by the inverse watershed image, which created a map of all low intensity regions between signal peaks (all other regions became 0). The resulting image was then binned to 1/1600 of its original size with the 10th percentile value of all nonzero values assigned to each bin. If no nonzero values existed for a bin, the bin value became zero and was considered an empty bin. Values for empty bins were determined by interpolation using a soap film partial differential equation. The image was then rescaled back to the original size using a two-dimensional interpolation by cubic spline. This resulted in a strong approximation of any nonuniform background fluorescence. The background was then subtracted from the filtered image and segmented by a user-defined threshold. Final segmentation was achieved by multiplying the segmented image by the binary watershed image. Segments were then uniquely identified with different numbers assigned to each. Segments represented the locations of cells and were used for subsequent measurements. After segmentation, the type IV collagen tissue channel (green) was converted to binary. The tissue segments were fused together to create a binary collagen mask by dilating the segments with two structuring elements (s.e.) first using 5 x 3 s.e., then a 3 x 5 s.e. Locations of infiltrating CD4 (blue) and CD8 (red) T cells relative to type IV collagen tissue (green) were determined using the binary collagen mask. The localization of individual cells was accomplished by determining the area and center of mass for each cellular event. Area equals the number of pixels covered by a cell. Center of mass was determined by measuring the average position of pixels (weighted by each pixel’s respective intensity) within a cell. Net intensity is the summation of all intensity values within a cell minus the local background. Local background equaled the mean intensity of the perimeter pixels of a cell multiplied by the area (in pixels) of the cell. Circularity was determined by evaluating the radial variability of a cell. This was accomplished by measuring the distance from the center of mass of a cell to each of its perimeter pixels. The mean (µ) and SD (
) of the distance set were found. Circularity equals the mean (µ) minus two SDs (2), then that quantity was divided by the mean (µ). Negative values were reassigned to zero and the result multiplied by 100 to represent circularity as a pseudopercentage. Two-color events within two pixels (0.7353 µm) of one another (measuring from the center of mass of a cell to the nearest edge of the collagen mask) were considered positive for both colors. The resulting data set representing the location and area of each cell was stored as list mode in flow cytometry standard format data and analyzed using FlowJo cytometry software (Tree Star). The percentages of blue (CD4) or red (CD8) cells within a 0.7353-µm radius of the green (collagen IV) mask were calculated after gating on the red plus green or blue plus green positive events.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Before infection, examination of CD4 and CD8 T cells in the lymphoid compartment and lung tissue, not including the airways, revealed few cells expressing either VLA-1 or VLA-2 (Fig. 1). In general, few T cells can be isolated from the airways of mice maintained in clean animal facilities and without prior deliberate infection. However, even among the few T cells that could be obtained from pooling the bronchoalveolar lavage of several mice, there was a proportion (27%) of both CD4 and CD8 T cells that expressed CD49a, the -chain of the VLA-1 receptor (Fig. 1A). Fewer T cells expressed VLA-2, with a small number (4–7%) of CD49a–CD49b+ single-positive CD4+ T cells, or CD49a+CD49b+ double-positive CD8+ T cells in the airways (Fig. 1A). Very few CD4 and CD8 T cells in the mediastinal lymph node (MLN) and spleen expressed either integrin, with only 2% CD49a+ CD49b– T cells in the MLN, and slightly higher (7%) in the spleen.
|
70% CD49a+. A much smaller proportion (28%) of the CD4+ T cells in the airways also expressed CD49a. Instead, a high proportion (65%) of the CD4 T cells expressed CD49b, with fewer (48%) of the CD8 T cells expressing this molecule. Interestingly, there was a converse relationship between CD49a and CD49b expression between the CD4 and CD8 T cells such that CD4 T cells favored the expression of CD49b only (45%) or CD49b and CD49a (20%), whereas the CD8 T cells in the airways expressed CD49a (33%) alone or in conjunction with CD49b (37%). Few CD4 T cells expressed CD49a alone (8.4%) and few CD8 T cells expressed CD49b alone (11%). These observations suggest that the expression of these integrins could direct the differential localization of CD4 and CD8 T cells within the matrix of the lung environment, dependent on the relative distributions of collagen subtypes.
|
Expression of VLA-1 and VLA-2 among the Ag-specific CD4 and CD8 T cells elicited by influenza infection
In light of the low proportions of T cells expressing either CD49a or CD49b in the lymphoid organs, experiments were performed to identify the T cells responding to the infection in an Ag-specific manner. The question being addressed was whether the few collagen-binding integrin+ T cells in the lymphoid sites essentially comprise the subsets of T cells specific for viral Ags. The primary CD8 T cell response to infection with influenza A/HK/X31 is dominated by two viral epitopes within the influenza NP (NP366–374) and the acid PA (PA224–233) (27). In our hands, on day 7 of the infection, CD8 T cells responding to these epitopes represent 3.9% and 6.9% of the CD8+ T cells in the airways (Fig. 2, A and B). The patterns of CD49a and CD49b expression on these Ag-specific CD8 subsets in the airways and spleen were compared after gating on the CD8+ NP or PA tetramer+ cells. Among the tetramer+ CD8 T cells, the patterns of CD49a and CD49b expression resembled that of the total CD8 population in the organ from which the cells were obtained. (Figs. 1B and 2, C–F). For example, whereas a large proportion of the CD8 population expressed either CD49a (69%) or CD49b (48%) in the airways on day 8, 46 and 33% of the DbNP+ and 51 and 43% of the DbPA+ CD8+ cells expressed CD49a or CD49b, respectively, on day 7. The increase in values from days 7 to 8 may be attributed to selection within the tissue. In the spleen (Fig. 2, E and F), lower proportions of both the DbPA+ and, particularly, DbNP+ cells expressed CD49a or CD49b. From these observations, several conclusions can be drawn. One is that the few integrin-expressing CD8+ T cells in the lymphoid organs do not necessarily represent Ag-specific T cells responding to the infection. Instead, the data are consistent with integrin expression being a reflection of the environment, regardless of whether they have responded to the infection.
|
The peak of the cellular response to WSN-OVAII infection occurs at day 9 (24). This infection efficiently recruits donor CD4+Thy1.2+OT-II cells into the airways and lung parenchyma (Fig. 3, A and B), and promotes expansion of the cells in the lymph node and spleen (Fig. 3, C and D). Of these OVA-specific virus-primed cells, substantial proportions expressed CD49b alone, with fewer CD49a+CD49b+ double-positive cells, and very few CD49a single positives (Fig. 3, E–H). CD49a was virtually absent on the donor cells in the MLN and spleen, with small proportions of the cells expressing CD49b alone (Fig. 3, G and H). These patterns were similar to the total CD4 infiltrates observed in X31 infection (Fig. 1). Comparable results were obtained using DO11.10 TCR-transgenic CD4+ T cells transferred into BALB/c recipients infected with WSN-OVAII (data not shown), suggesting the collagen integrin expressing populations do not simply reflect the activation status of the cells, and may be independent of genetic background.
|
One question that arose was whether the pattern of collagen-specific integrins we observed was something attributable to the virus priming of the cells. To address this question, we primed OVA-specific OT-II CD4 T cells using a strategy typically used as a model of allergic airway disease. Congenic Thy1.1+B6.PL mice received donor Thy1.2+OT-II CD4 T cells 1 day before i.p. injection of OVA in alum. A second dose of OVA in alum was administered 1 wk later. Fourteen days after the first injection, mice were exposed to OVA aerosol for 1 h, then sampled 5 days later for the presence of donor CD4 T cells in the airways, lung, and lymphoid tissues (Fig. 4, A–C). As expected, and contrary to the virus-primed T cells, OT-II CD4 T cells primed in this manner produced IL5 and little IFN-
(data not shown), consistent with a program of Th2 differentiation (29, 30, 31). Regardless of the organ that was sampled, the pattern of CD49a and CD49b expression was very similar to that of OT-II cells primed by WSN-OVAII influenza virus infection in that many of the lung tissue and airway cells were CD49b+CD49a–, with fewer cells expressing both integrins or CD49a alone compared with CD8+ T cells (Fig. 4, D–F, Table I). In the allergic vs infection models, respectively, there were generally fewer CD4 T cells that were double positive (5 ± 3 vs 17 ± 8% in the airways; 7 ± 3 vs 21 ± 6% in the lung tissue). The low proportion of CD49a+ staining was observed in repeat experiments, and was not due to problems with the Abs because CD8+ cells in the airway and lung tissue of these mice stained very well for CD49a (see insets in Fig. 4, D and E). Though virus infection seemed to favor a higher proportion of CD49a/CD49b double-positive CD4+ T cells, it was still lower than for the CD8+ subset suggesting that the expression of these integrins is intrinsic to the CD4 or CD8 subsets and the organ from which they are obtained, and is only partially affected by the mode of priming.
|
|
The substantial changes in the expression of CD49a and CD49b between the start and the peak of the cellular response led to the question of when these integrins increased their expression. Animals were infected with influenza A/HK/X31 and sampled at several time points for analysis of the integrin expression. What is most noticeable is that, at the start of the infection, very few T cells in the lung and airways are collagen integrin positive (Fig. 5, A–D). The exception is for T cells in the airways, a modest proportion of which express CD49a (Fig. 5A). CD49a and CD49b begin to increase on both CD4 and CD8 T cells in the lung and airway between days 4 and 7. CD49a predominates on the CD8+ subset, peaking briefly at day 8, dropping at day 9, and then slowly increasing again. This trend to peak at 8 days, drop briefly, and recover is a highly reproducible phenomenon, and was noted previously for the NP-specific CD8 population (18). The significance is not clear, though it suggests there may be continual infiltration of CD8 T cells into the lung tissue and airways after day 9, the point at which the virus has been cleared from most or all of the animals (32, 33). A similar trend was not apparent for the CD4+ subset, suggesting independent regulation of these cells in the lung.
As was noted for the peak response, CD49b was a prominent feature of the CD4+ subset, increasing in proportion as early as day 2 in the airways and lung tissue, likely reflecting early non-Ag-specific T cell recruitment. CD49b then peaked between days 7 and 8 and remained at a stable level that was elevated compared with mice that were not influenza infected. Together, the observations suggest that the inflammatory process in the airways and lung may serve to increase the expression of CD49b, or alternatively, to specifically recruit or retain these cells while the infection is active. In addition, as has been noted in the past (34), influenza infection seems to permanently alter the cellularity of the airways, with the present data suggesting that the infection also phenotypically changes the cells that can be isolated after recovery from the infection.
Relationship of collagen distribution in the lung to the localization of CD4 and CD8 T cells in the lung parenchyma and proximity to airways
The CD49a/CD29 heterodimer, VLA-1, is the only known adhesion receptor for type IV collagen (35, 36), whereas CD49b/CD29 (VLA-2) is considered the major receptor for the collagen I subtype (3, 4, 5, 37). Within the lung and many other tissues, the collagen subtypes are differentially distributed, but relatively little is understood regarding the relationship between these collagens and the localization of the T cells within the lung.
To investigate these relationships, we performed an immunofluorescence analysis of collagen subtypes and T cell subsets during the course of an acute primary influenza infection. To demonstrate specificity, immunofluorescence staining was performed, separately, on frozen sections of healthy lung tissue for collagens types I, III, and IV. Single-color fluorescence images for each of the collagen subtypes is shown in Fig. 6. It is easy to appreciate distinctions in each of the three collagens. The type I collagen has a fibrous appearance and is widely distributed in the parenchymal spaces within the lung tissue (Fig. 6A). Type IV collagen is more compact in its distribution and is useful for "outlining" the boundaries of the blood vessels, airways, and alveoli (Fig. 6B). Type III collagen is more sparsely distributed with a punctate distribution (Fig. 6C).
|
|
|
|
Phenotypes of Ag-specific and total CD4 and CD8 T cells in the lung tissue and airways in the memory phase
Given the observed differences in both the localization of CD4 and CD8 T cells in the lung and the differential expression of CD49a and CD49b collagen-binding integrins during the acute phase of the response, we sought to determine whether these patterns persisted into the memory phase.
In the memory phase, CD49a was expressed on subpopulations of both CD4+ and CD8+ T cells in the airways (Fig. 10, A and C) and lung tissue (Fig. 10, B and D). In fact the majority (80%) of CD8+ cells in the airways, and half of the CD8 population in the lung expressed CD49a. An even higher proportion of the influenza specific subset was CD49a+ at 92 and 74% of the CD8+DbNP+ cells in the airways and lung tissue (Fig. 10, G and H). However, unlike the NP-specific CD8 T cells, CD49a was not enriched among the Ag-specific CD4+ T cells primed by influenza infection (Fig. 10, E and F). These observations help explain the profound effects that Ab blockade or gene deletion of CD49a has on the tissue memory CD8 T cells (18). The data would suggest similar blocking or gene deletion studies focused on the CD49a+CD4+ subset would yield more modest changes in the tissue memory CD4 T cells because both the proportion and number of these cells in the lung is lower than for CD8+ cells (24, 28).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The data show that CD49a and CD49b (VLA-1 and VLA-2, respectively) are primarily expressed on both CD4+ and CD8+ T cells in the nonlymphoid tissues of the lung. CD49a (VLA-1) is expressed by a larger proportion of the CD8+ cells, particularly in the airways. This proportion increases toward the memory phase. In contrast, whereas CD49b (VLA-2) was expressed by a significant proportion of both CD4 and CD8 T cells in the lung parenchyma and airways during the acute phase, the proportions were lower among memory cells, particularly for CD8+ T cells in the airways. The general expression pattern of these integrins appears to be intrinsic to the cell type and the tissue localization, and is not affected by the mode of priming. For example, the observation that more CD4+ T cells in the lung expressed CD49b, and fewer expressed CD49a prevailed whether the cells were primed by viral infection or Th2-inducing conditions. This is at odds with the simple interpretation that CD49b expression correlates with Th1 and Th2 phenotypes among CD4 T cells (38, 39). Similar expression of CD49a and CD49b was observed on CD8+ T cells responding to lymphocytic choriomeningitis virus infection in the spleen (40).
Immunofluorescence analysis of the distribution of collagen subtypes within the lung environment revealed important differences that help explain the observed localization of T cell subsets. The CD49a/CD29 integrin VLA-1 preferentially binds to type IV collagen and can also bind type I collagen and laminin. As such, it is the only known adhesion receptor for type IV collagen. Given that the lung is rich in basement membrane along the airways, it was perhaps not surprising that VLA-1+ T cells were the major population in the airways, and that immunofluorescent staining showed mostly CD8+ cells closely associated with the airway basement membranes and on the luminal side of the airway walls. This distribution also explains the effects of CD49a inhibition on the ability to retain CD8 T cells in or close to the airways in the memory phase (18). The observations suggest that one of the functions of CD49a expression may be to help localize the T cells close to the outermost layers of the lung epithelium where they could be of the most use in a re-encounter with a pathogen such as influenza.
In addition to physical associations with the ECM, the localization of T cells within tissues is also regulated by chemokines (41, 42, 43). It is likely that there are important differences in the distribution of chemokines in the lung that can account for the observed predominance of CD8 T cells in the lumen of the infected airways, though the identity of these chemokines has yet to be determined. Accessibility to these chemokines could also be affected by the binding to specific collagen subtypes because it is well-established that chemotactic factors are often bound to matrix components (44, 45). Thus, collagen binding could bring the T cells into closer proximity to the chemotactic ligands.
The lower proportion of CD49a+CD4 T cells in models of both virus infection and experimental allergy supports the notion that localization of CD4+ T cells in the airways may not be as beneficial to lung immunity. CD49b was a more consistent marker of CD4+ T cells in the lung. When CD49b pairs with the CD29 1 integrin chain, it forms a heterodimeric adhesion molecule specific for type I collagen (3, 4, 5), which is rich in the lung parenchyma. Consistent with this, higher proportions of the CD4+ and CD8+ T cells isolated from the lung parenchyma expressed CD49b. Together, the observations on collagen distribution and integrin expression imply that these integrins could be directing the localization of CD4 and CD8 T cell subsets within the lung. Experiments to directly test this hypothesis remain to be performed.
There are several possible explanations for the differential expression of CD49a+ and CD49b+ on T cells within the lung tissue and airways compared with the lymphoid organs. One possibility is that there are signals within the tissue and airways that increase or stabilize the expression of CD49b or CD49a, respectively. The collagen subtypes themselves could do this, and in fact although we stained for types I, III, and IV collagen there are many more collagen subtypes. These include the fibril-associated collagens with interrupted triple helices that associate as single molecules with large collagen fibrils and could modify receptor interactions (46). The role of these collagens and their distribution, for instance between lymph nodes and airways, is not well understood. An alternative possibility is that these observations reflect a selection process for T cells dependent on integrin expression. Thus, VLA-1+ cells are retained in or near the collagen IV-rich airways, whereas the abundance of type I collagen in the parenchyma may selectively retain VLA-2+ cells. It would be interesting to induce constitutive expression of VLA-1 on CD4+ T cells to see whether this affects their relative distribution in the lung tissue and airways.
Previous studies showed that Ab-mediated inhibition of CD49a reduced the number of virus-specific memory CD8 T cells in the lung and other extralymphoid tissues (18), but did not affect the acute CD4 or CD8 T cell response. Similar experiments did not dramatically affect the number of virus-primed CD4+ memory T cells in the lung (data not shown), though the numbers of CD4+ memory cells in the lung are low to begin with (24, 28). The limited impact of blocking CD49a on the CD4+ cells could be explained if the substantial CD49a-negative subset was obscuring effects on the smaller CD49a+CD4 population. Interestingly, though a large proportion of T cells in the lung express CD49b, administration of blocking Abs to CD49b did not appear to affect either the CD4 or the CD8 T cell numbers in the lung or lymphoid tissues (data not shown). The latter finding is supported by the recent observation that adoptively transferred CD49a+ and CD49b–CD4 T cells homed equally well to an influenza infected lung (38). This suggests that CD49b (VLA-2) may not function in a manner analogous to CD49a (VLA-1) in regulating the retention of memory T cells in the extralymphoid space.
The lack of an effect of CD49b inhibition on T cell retention is not surprising given the observed role of VLA-2 on other cell subsets. VLA-2-mediated ligation of type I collagen on neutrophils and some tumor cells is associated with increased motility (20, 47, 48). Most NK cells express CD49b/VLA-2 (49), and NK cells could be considered the prototypical extralymphoid lymphocyte. Their role in immune surveillance within tissues could be promoted by VLA-2-mediated stimulation of cell motility within the type I collagen-rich extralymphoid tissue space. VLA-2 could play a similar role on T cells, promoting their migration within extralymphoid spaces, particularly during the acute phase of the infection.
An intriguing aspect of the kinetics was what appeared to be a biphasic pattern of infiltration of the CD8+ T cells. Various studies support the concept that there are multiple phases of T cell infiltration into the lung during respiratory virus infections (50). Thus, in addition to some constitutive memory T cells present at the start of an infection (though low or nonexistent in "naive" mice), there follows an early non-Ag-specific infiltrate 3–5 days after infection (51, 52) and then the entry of recently divided Ag-specific cells from the lymphoid organs around day 5 onwards (50, 51, 52, 53, 54). Our earlier observations of influenza VLA-1+ DbNP-specific CD8 T cells during primary influenza infection showed that there was a steady increase in VLA-1+ cells until day 8, the peak of the cellular response, then a slight drop at day 9, the point at which virus is cleared, followed by a steady increase again through 14 days after infection. The current phenotypic and histological data reiterate this observation for the total CD8+ T cell subset (Fig. 5, A and B). After this point, the number of T cells, and of CD49a+ T cells remains increased indefinitely compared with baseline, showing that the infection can induce permanent alterations to the cellular immune components of the lung. The observations suggest that T cell infiltration is a continuous process, though at a lower rate, even after the virus is cleared, with significant implications for the formation and maintenance of the tissue memory population. It seems that the tissue memory may not be derived from T cells that were in the lung at the peak of the response, but instead comes from a separate set of cells that continue to infiltrate the tissue long after the virus is cleared. This is corroborated by a recent study using Sendai virus in which the authors showed that effector memory T cell populations in the lung airways were maintained by continual recruitment of new cells from the circulation (55).
It is clear that collagen adhesion and interactions with the ECM are important for regulation of T cell populations in the lung environment, and probably within the environments of other extralymphoid tissues. The lung as a whole is quite exceptional in this regard as it is the location where the highest proportion of integrin expressing T cells exists, and these populations have significant contributions to secondary immunity to respiratory virus infections. The observations also have implications for the regulation of allergic responses because the trafficking and retention of allergen-specific T cells in the lung could be viewed to have detrimental effects upon Ag re-encounter. Recent work suggests that CD49b is stably expressed on a subset of CD4+ memory T cells, suggesting that these are a population of memory T cells completely distinct from the CD49b negative subset (38). It is not known to what extent collagen integrin-negative memory T cells can convert to collagen integrin-positive cells during the resting memory phase. Determining the signals that regulate the expression of these adhesion molecules on T cells remains as an important area for discovery. It also remains to be determined whether these integrins function simply by affecting adhesion and retention, stimulating or retarding motility, by promoting the survival of T cells in the extralymphoid spaces, or some combination of these functions.
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported in part by grants from the American Lung Association (RG-075N) and the National Institutes of Health (AG021970 and ESO1247). This work was also supported by National Institutes of Health Training Grants HL66988 (to S.J.R.), AI07169 (to T.J.C.), and HL6688 (to S.J.A.). 
2 M.R., S.J.R., and T.J.C. contributed equally to this study.
3 Address correspondence and reprint requests to Dr. David J. Topham, David H. Smith Center for Vaccine Biology and Immunology, Aab Institute of Biomedical Sciences, University of Rochester Medical Center, 601 Elmwood Avenue, Box 609, Rochester, NY 14642. E-mail address: david_topham{at}urmc.rochester.edu
4 Abbreviations used in this paper: ECM, extracellular matrix; s.e., structuring element; MLN, mediastinal lymph node; PA, polymerase; NP, nucleoprotein.
Received for publication August 10, 2006. Accepted for publication January 8, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
11 and the 21 integrins. J. Cell Biol. 113: 1475-1483. 11, 21 and 31 integrins in cell–collagen interactions: identification of conformation dependent 11 binding sites in collagen type I. EMBO J. 11: 3865-3873. [Medline]11 and 21. Eur. J. Biochem. 215: 151-159. [Medline]11. D. Gullberg, ed. I Domains in Integrins 14 Landes Bioscience, Georgetown. 11 integrin VLA-1 regulates CD8 T cell-mediated immune protection against heterologous influenza infection. Immunity 20: 167-179. [Medline]21 (VLA-2) is a principal receptor used by neutrophils for locomotion in extravascular tissue. Blood 95: 1804-1809. 1 Integrins are critically involved in neutrophil locomotion in extravascular tissue In vivo. J. Exp. Med. 187: 2091-2096. 1 integrin-independent strategies for polarization, interaction with collagen fibers and locomotion. Eur. J. Immunol. 28: 2331-2343. [Medline]- and -chain genes under the control of heterologous regulatory elements. Immunol. Cell Biol. 76: 34-40. [Medline]
