Abstract
Dogs with immunoglobulin E (IgE) allergy for ragweed that are sensitized by intrapulmonary exposure to ragweed can be used to study the pulmonary immune response that is important in allergic asthma. Using this model, we tested the hypothesis that T lymphocytes are activated locally in the airways shortly after allergen exposure of the lungs. The airways of six allergic dogs and three non-allergic dogs were exposed to ragweed by segmental allergen challenge (SAC). T-cell subsets and T-cell activation in blood and bronchoalveolar lavage (BAL) fluid were measured by flow cytometry before SAC and at 4, 24 and 72 hr thereafter. SAC caused a statistically significant increase in the percentage of major histocompatibility complex (MHC) class II-positive CD4 and CD8 T cells in BAL fluid and a significant increase in the mean fluorescent activity of MHC class II from 4 hr after SAC onward. This activation was significantly different from that found in cells from lung lobes challenged with saline, or from lung lobes in non-allergic dogs challenged with ragweed. The percentage of CD45RAbright CD8 cells increased significantly in allergic dogs after both ragweed and saline challenges. This was significantly higher than in non-allergic dogs. We conclude that T-cell activation in the airways of dogs can be measured after in vivo activation of the cells by measuring MHC class II and CD45RA expression in BAL fluid T cells. Furthermore, in allergic dogs, T cells are activated locally in the lungs within 4 hr after exposure to ragweed allergen. These results suggest a role for T lymphocytes in the development of late-phase allergic reactions in the airways.
Introduction
Some of the offspring of allergic beagle dogs develop immunoglobulin E (IgE) antibodies to allergens and immediate-type allergic reactions after exposure to allergens.1–3 The intrapulmonary deposition of ragweed in sensitized allergic dogs results in pulmonary sensitization. The dogs develop bronchial hyper-reactivity, and upon re-exposure of the airways to ragweed, local inflammatory reactions occur that are characterized by histamine release and a rapid increase of neutrophilic granulocytes in the lungs, followed by a local increase of eosinophilic granulocytes and lymphocytes.3,4 Furthermore, the bronchoalveolar lavage (BAL) fluid of allergen-exposed dogs contains soluble factors that induce increased secretion of mucin in in vitro cultures of airway epithelial cells.5 Thus, the characteristics of these dogs make them an excellent animal model for the study of the allergic pulmonary immunity that is important in asthma.
Little is known about the precise role of lymphocytes in the sequence of events that occur after IgE-dependent mast cell degranulation in the airways and the subsequent local eosinophilic inflammation. In allergic asthma in humans, local lymphocytes are thought to be predominantly of the T helper type 2 (Th2) type.6 These cells are thought to sustain the inflammatory reactions in the chronic phase of the disease by the release of interleukin-4 (IL-4) and IL-5, cytokines which up-regulate adhesion molecules and activate eosinophils, respectively.7 In the acute phase of the allergic response, shortly after allergen exposure, the role for lymphocytes is less clear. They may be directly activated by antigen-presenting cells and by products released from mast cells.8 Under experimental conditions, lymphocytes in human airways are activated within 4–6 hr after controlled allergen exposure.9,10 However, the study of such events in humans is hampered by the burden of repeated bronchoscopy and bronchoscopy shortly after allergen exposure.
We have assumed that the dog model for allergic asthma can be used to test the hypothesis that T cells are activated locally in the airways shortly after allergen exposure. Therefore, leucocytes, T-cell sub-populations, and membrane surface markers for T-cell activation were measured in BAL fluid. Doveren et al.11 showed that major histocompatibility class II (MHC II) expression is increased in dog lymphocytes after in vitro stimulation of the cells, and others have confirmed this finding.12 Furthermore, Chang et al. showed that MHC II expression on lymphocytes and other leucocytes in BAL fluid was increased in rejection crises after lung transplantation.13 Another candidate marker is the CD45RA antibody, which is supposed to identify naive cells analogous to a similar specificity in humans.14 Therefore, we analysed whether T-cell activation could be detected with the use of anti-MHC II and anti-CD45RA antibodies in flow cytometry. A serious drawback in analysing T lymphocytes in dogs is that neutrophilic granulocytes express CD4.15 It was necessary therefore to include simultaneous staining for CD3 when analysing CD4 T cells. This is especially true when BAL fluid lymphocytes are to be analysed under conditions of local inflammation in the airways, samples in which the forward- and side-scatter properties of the cells do not allow a distinct definition of lymphocytes.
In this study sub-segments of the lungs of allergic and non-allergic dogs were challenged with ragweed or saline. BAL fluid was collected before challenge and at 4, 24 and 72 hr thereafter. We observed a local CD4 and CD8 T-cell activation in the airways of ragweed-challenged allergic dogs based on changes in MHC II and CD45RA expression from 4 to 72 hr after challenge. The activation was significantly different from any activation in non-allergic dogs, and, for MHC II, from activation after saline challenge. We concluded that the dog model for allergic asthma can be used to study T-cell activation in the airways after allergen exposure, and that T cells are activated shortly after local allergen challenge.
Materials and Methods
Animals
Six allergic dogs (beagles, 18–21 months of age) and three non-allergic dogs (24–28 months of age) were studied. The dogs were raised as previously described.3,16 Six dogs were sensitized by regular intraperitoneal or subcutaneous injections with ragweed (ragweed short, Ambrosia artemisiifolia, Bayer, Emeryville, CA) and aluminium hydroxide as adjuvant.3 Furthermore, a single dose of ragweed was instilled into six different lung lobes of each dog via a bronchoscope 4–8 months before the experiments. Three dogs not previously exposed to ragweed were used as non-allergic controls. Serum levels of immunoglobulins were measured as described previously.3 At the time of the study the serum levels were as follows: allergic dogs and non-allergic dogs: total IgE, 52 ± 10 (mean ± SEM) and 44 ± 22 au/ml (arbitrary units); IgE anti-ragweed, 22 ± 5 and 3 ± 3 au/ml; IgG anti-ragweed 133 ± 38 and 62 ± 11 au/ml, respectively.
Experimental design
Sub-segments of the right intermediate lobe and the left diaphragmatic lobe were lavaged 1–3 days before allergen challenge, and a blood sample was obtained. On the day of the challenge, 3 ml of 0·154 m NaCl were instilled into a sub-segment of the left cardiac and diaphragmatic lobes (a different sub-segment to the one lavaged on the control day) to serve as a control challenge. The instillation was performed through a fibreoptic bronchoscope under video inspection; a polyethylene catheter (1·57 mm, outer diameter) was passed through the bronchoscope to reach the desired segments, and the solution was administered via the catheter. The bronchoscope was retracted and carefully rinsed with saline. Next, ragweed [3 ml 0·154 m NaCl containing 0·30 mg ragweed (Ambrosia artemisiifolia, Bayer, batch J88F9911)] was instilled into a subsegment of the right cardiac and diaphragmatic lobes. The dogs recovered completely from anaesthesia within about 10 min. The lobes challenged with ragweed and saline were subsequently lavaged at 4, 24 and 72 hr after the challenge as indicated in the tables and figures. Not all conditions were sampled and analysed in each dog. Blood samples were taken from the jugular vein prior to anaesthesia and at each lavage.
Bronchoscopy, bronchoalveolar lavage and sample processing
Bronchoscopy and BAL were performed as described.17 The animals were anaesthetized with isoflurane, and an endotracheal tube was inserted. A flexible fibreoptic bronchoscope (Model BF-4B2, Olympus Corp, Lake Success, NY) was placed in wedge position in a segment of the lobe indicated. BAL was performed by instillation of five portions of 10 ml 0·154 m NaCl. Each portion was immediately aspirated with low suction and placed on ice. The bronchoscope was retracted between lavages of allergen- and saline-challenged segments, and rinsed with saline. Blood and BAL fluid were immediately processed for analysis of cells.
The portions of BAL fluid from each sub-segment were pooled, and cells were collected by centrifugation at 400 g for 10 min. The cells were washed with Dulbecco's phosphate-buffered salt solution (Sigma, St Louis, MO) containing in addition 0·5% (w/v) bovine serum albumin (Sigma), 0·5 mm ethylenediaminetetraacetic acid (EDTA; Sigma), and 0·01% NaN3 (Sigma) (pH 7·3, PBA), and resuspended in PBA. Cells were counted with a Beckman Coulter counter ZA1 (Beckman Coulter, Fullerton, CA). Some of the cells were resuspended at 0·5×106/ml for the preparation of cytospins. Cells from the control segments were combined for analysis of cell sub-populations by flow cytometry. Similarly, cells from the ragweed-challenged segments were combined except when the cell counts showed differences larger than 15%, in which case cells from the segment with the lowest cell counts were used for flow cytometric analysis. BAL-fluid cells were resuspended at 5×106/ml in PBA containing in addition 10% (v/v) normal canine serum (PBAS). Blood was collected in vacuum tubes containing heparin or sodium citrate, and blood smears were made. Leucocytes were counted with a Coulter counter. Peripheral blood mononuclear cells (PBMC) were purified by centrifugation on Histopaque [specific gravity (SG) 1·077, Sigma]. Occasionally, discontinuous gradients with Histopaque SG 1·077 and SG 1·119 (Sigma) were used as described.18 The PBMC were collected from the interphase; the cells were washed twice with PBA and resuspended at 0·5×106/ml in PBA and 1×106/ml in PBAS for cytospin preparations and flow cytometry, respectively. Where necessary, erythrocytes in BAL-fluid cell and blood cell preparations were lysed with buffered NH4Cl (Sigma), and the cells were washed twice and resuspended in PBA or PBAS. Cytospin preparations were made with a Shandon (Pittsburgh, PA) cytocentrifuge. Cells were centrifuged for 4 min at 250 g.
For flow cytometry, antibody mixtures at appropriate antibody dilutions were prepared in PBAS. Equal volumes of cell preparations and antibody solutions were mixed and incubated for 30 min. The cells were washed with cold PBA. Antibody solutions for the next step were added followed by incubation for 30 min. After a wash, the cells were incubated and washed a third time where indicated. The cells were analysed immediately after labelling. All incubations were performed on ice in the dark and the cells were kept on ice until the flow cytometric assay.
The medium for washing the cells during the purification procedure prior to cell culture consisted of Tris (Sigma) -buffered Hanks' balanced salt solution (HyQ, Logan, UT) supplemented with 2% (v/v) heat-inactivated fetal bovine serum (Sigma), 10 U/ml penicillin (Sigma), and 0·01 mg/ml streptomycin (Sigma). After the final wash, cells were suspended in cell-culture medium [Iscove's modified Dulbecco Medium (IMDM), Gibco, Rockville, MD, supplemented with 2 mm sodium pyruvate (Sigma), 0·04 mm β-mercaptoethanol (Sigma), penicillin and streptomycin, 5% (v/v) heat-inactivated normal canine serum and 5% (v/v) heat-inactivated fetal bovine serum (Sigma)].
Antibodies and other reagents
The following antibodies were used: mouse anti-canine CD3 (CA17.2A1, Instruchemie, Bilthoven, the Netherlands), mouse anti-canine CD3-fluourescein isothiocyanate (FITC; CA17.2A1, Serotec, Raleigh, NC), rat anti-canine CD4 (YKIX302.9, Serotec), rat anti-canine CD4-FITC (YKIX302.9, Instruchemie), rat anti-canine CD8R-phycoerythrin (PE; YCATE55, Serotec), mouse anti-MHC II (CVS20, Serotec), rat anti-canine MHC II (YKIX.344, Instruchemie), mouse anti-CD14R-PE-Cy5 (TUK4, Instruchemie), mouse anti-CD21-PE (B-Ly4, Pharmingen, San Diego, CA), rat anti-canine CD45 (YKIX716.1, Serotec), mouse anti-canine CD45RA (CA.1D3, P. Moore), rat anti-mouse IgG1-FITC (A85-1, Pharmingen), rat anti-mouse IgG1-biotin (LO-MG1-1, Biosource, Camarillo, CA), rat anti-mouse IgG1-PE (X56, Becton Dickinson, San Jose, CA), mouse anti-rat IgG2a-biotin (RG7/1.30, Pharmingen), mouse anti-rat IgG2b-biotin (RG7/11.1, Pharmingen), and corresponding isotype controls (Becton Dickinson and Pharmingen). SA-RPE-Cy5 was purchased from DAKO (Carpenteria, CA), and SA-FITC and SA-RPE were purchased from Becton Dickinson. Unlabelled and FITC- and PE-labelled control beads were from Becton Dickinson. Normal canine serum was obtained from Harlan (Indianapolis, IN). Propidium iodide and phytohaemagglutinin (PHA) were supplied by Sigma. Table 1 shows the combinations of antibodies applied in the incubations.
Table 1.
Antibody combinations in flow cytometry
| Antibody combinations | |||||
|---|---|---|---|---|---|
| Incubation step | CD3,CD4,CD8 | CD3,CD4,MHC II | CD8,MHC II | CD3,CD4,CD45RA | CD3,CD8,CD45RA |
| 1 | anti-CD3 | anti-MHC II* | anti-MHC II | anti-CD45RA | anti-CD45RA |
| anti-CD4 | anti-CD4 | ||||
| 2 | anti-mouse IgG1biotin | anti-mouse IgG1PE | anti-mouse IgG1FITC | anti-mouse IgG1PE | anti-mouse IgG1biotin |
| anti-rat IgG2abiotin | anti-CD8PE | anti-rat IgG2abiotin | |||
| 3 | SA-RPECy5 | SA-RPECy5† | SA-RPECy5† | SA-RPECy5† | |
| anti-CD4FITC | anti-CD3FITC | anti-CD3FITC | anti-CD3FITC | ||
| anti-CD8PE | anti-CD8PE | ||||
For the analysis of blood lymphocytes, MHC II expression was occasionally analysed by incubation of anti-MHC II followed by anti-CD4FITC, anti-CD8PE, and anti-mouse IgG1biotin-SA-RPE-Cy5.
These incubations were preceded by a 5-min incubation with 10% (v/v) normal mouse serum to block any interaction of anti-mouse IgG1 with the anti-CD3FITC.
Analysis of cell populations
Blood smears (from EDTA anti-coagulated blood) and cytospin preparations were stained with May–Grünwald–Giemsa stain. Cell counts were evaluated for at least two different preparations from each sample and at least 500 cells were counted per sample.
Flow cytometry was performed with a FACStar Plus (Becton Dickinson) equipped with an Argon laser and appropriate filters for FITC, PE and RPECy-5 labels. cellquest software (Becton Dickinson) was used to acquire and analyse the data. The CD45-positive cells were identified, and lymphocytes were gated on the basis of forward and side light scatter; propidium-iodide-positive cells were excluded from the analysis. For cells found in the BAL fluid, a refinement of the lymphogate was obtained on the basis of the dotplots for the CD3 signal versus side scatter. As dog neutrophils are positive for CD4, the CD4 lymphocytes were always measured in combination with staining for CD3. No CD8-positive cells were CD3-negative; nevertheless, in several cases, CD8 T cells were defined as CD3+ CD8+ cells by performing simultaneous staining with anti-CD3. The samples were measured in duplicate; to avoid carry-over of samples, the lines of the flow cytometer were rinsed with PBS between each duplicate run.
Cell cultures
PBMC were cultured in polystyrene round-bottom plates (96 wells) or in six-well flat-bottom plates (both Corning, Pittsburgh, PA) starting at a cell density of 2×105 cells/ml. PHA was used as a stimulus. The cells were cultured at 37° in humidified air containing 5% CO2.
Statistics
Differences between groups were analysed by the Student's t-test. Paired samples were analysed by the paired t-test. The baseline values for cell counts were from the samples from the right intermediate lobe and a left diaphragmatic lobe. The mean difference in cell numbers in the lavage fluids of these two lobes was less than 10% of the cell numbers (n = 9). Therefore, the samples were pooled and used as baseline in the calculations. The low variation in cell counts in different lobes of one dog allowed us to compare subsequent lavages of challenged segments as samples paired to that baseline. Furthermore, the lavages of the saline challenged segment and the ragweed challenged segment in one dog were treated as paired samples. Data were calculated with graphpad prism 3.01.19 Two-sided P-values were calculated, and values below 0·05 were considered statistically significant.
Results
Recovery of BAL fluid
The recovery of the BAL fluid from allergic dogs was 78·3 ± 1·8% (mean ± SEM) and 81·5 ± 1·6% of the instilled fluid for the allergen- and saline-challenged segment, respectively. In the non-allergic dogs, the recoveries were 81·3 ± 2·2% and 83·7 ± 1·1%, respectively. These results did not differ significantly.
Total leucocytes and differential cell counts
Leucocyte counts in BAL fluid are shown in Table 2. At baseline, there were no significant differences between the leucocytes in allergic and non-allergic dogs. Allergen challenge in allergic dogs resulted in a statistically significant increase of total cells in BAL fluid at 24 and 72 hr after challenge. This was significantly different from the total cells in the saline-challenged segments of allergic dogs. In the non-allergic dogs, the allergen challenge did not significantly increase total cells in BAL fluid. The increase of total cells in the allergen-challenged segments of the allergic dogs was significantly higher than in the allergen-challenged segments of the non-allergic dogs.
Table 2.
The effect of ragweed and saline challenge on leucocytes in BAL fluid
| Ragweed challenge | Saline challenge | ||||||
|---|---|---|---|---|---|---|---|
| Time | Dog | Total leucocytes | Neutrophils | Eosinophils | Total leucocytes | Neutrophils | Eosinophils |
| Before | allergic | 253 ± 52 | 7·85 ± 1·9 | 10·1 ± 4·8 | same values as shown at the left | ||
| non-allergic | 183 ± 35 | 4·66 ± 1·0 | 20·7 ± 5·9 | ||||
| 4 hr | allergic (n = 3) | 414 ± 101 | 166 ± 79¶ | 51 ± 22 | nd | nd | nd |
| non-allergic | 301 ± 85 | 40·9 ± 23 | 53 ± 28 | nd | nd | nd | |
| 24 hr | allergic | 977 ± 296*†‡ | 473 ± 204*†‡ | 232 ± 74*†‡ | 220 ± 56 (n = 3) | 17·9 ± 2·9 (n = 3) | 24·9 ± 15·2 (n = 3) |
| non-allergic | 177 ± 30 | 33·5 ± 11·1 | 60 ± 28 | nd | nd | nd | |
| 72 hr | allergic | 1090 ± 101*†§ | 140 ± 43*†§ | 560 ± 43*†§ | 193 ± 23 | 8·2 ± 1·6 | 13·3 ± 5·7 |
| non-allergic | 271 ± 68 | 22·5 ± 7·8 | 34 ± 11 | 237 ± 45 | 15·9 ± 4·7 | 32·1 ± 2·2 | |
Results are from six allergic and three non-allergic dogs except where indicated. All results are expressed as mean ± SEM in 106 cells/ml BAL fluid; nd, not done.
Higher than in non-allergic dogs, P < 0·03.
Higher than before allergen challenge in allergic dogs, P < 0·01.
Higher than saline challenge in allergic dogs, P < 0·05 and < 0·005, respectively.
Higher than saline challenge in allergic dogs, P < 0·05 and < 0·005, respectively.
Higher than before challenge, P < 0·05.
The number of neutrophils in the allergen-challenged segments of the allergic dogs increased significantly at 4, 24 and 72 hr after challenge, with the peak value at 24 hr (Table 2). The neutrophil levels in the allergen-challenged segments were higher than in the saline segment. Furthermore, the neutrophil levels in the allergen-challenged segment of the allergic dogs were higher than in the corresponding segments of the non-allergic dogs. The non-allergic dogs showed only a small increase of the number of neutrophils both in the allergen- and saline-challenged segments.
The number of eosinophils increased significantly in the allergen-challenged segment at 24 and 72 hr after allergen challenge in the allergic dogs (Table 2). This increase was significantly higher than the changes in the non-allergic dogs and in the saline-challenged segments of the allergic dogs. The non-allergic dogs showed small changes in the number of eosinophils in the allergen-challenged and saline-challenged segments.
The number of lymphocytes increased slightly in the allergen-challenged segment of five allergic dogs at 24 hr and all dogs at 72 hr after the challenge (Fig. 1). The increase at 72 hr differed significantly from the saline-challenged segments at 72 hr (P < 0·02). The increase also differed significantly from any change in the allergen-challenged segment in the non-allergic dogs (P < 0·05). There was no significant difference in lymphocyte counts between the allergen- and saline-challenged segments of non-allergic dogs (not shown in Fig. 1).
Figure 1.
The effect of allergen challenge on the number of lymphocytes in BAL fluid. The results are expressed as mean ± SEM. ▪, ragweed-challenged allergic dogs; □, saline-challenged allergic dogs; ▴, ragweed-challenged non-allergic dogs. Lymphocytes were significantly increased in the allergen-challenged segment of allergic dogs at 72 hr: P < 0·02 (paired t-test) compared to before challenge; P < 0·02 (paired t-test) compared to the saline-challenged segment; P < 0·05 (t-test) compared to ragweed-challenged in non-allergic dogs. The number of lymphocytes after saline challenge in non-allergic dogs at 72 hr was 33·7 × 106 cells/l.
The number of alveolar macrophages in the allergic dogs was 200 ± 45×106/ml BAL fluid (mean ± SEM) before allergen challenge, and these numbers did not significantly change upon ragweed or saline challenge. Thus, the increase of total leucocytes was predominantly caused by an increase of neutrophilic granulocytes at the early time-points and of eosinophilic granulocytes at the later time-points.
Lymphocyte sub-populations before allergen challenge
Representative dot-plots for blood and BAL fluid cells are shown in Fig. 2. To substantiate the validity of the analysis, the following checks were calculated.20 In the peripheral blood mononuclear cell population, the CD3+ plus CD21+ cells accounted for 86·0 ± 1·6% (mean ± SEM, n = 9) of the cells in the lymphogate. Other cells in the lymphogate may have been natural killer cells, but were not analysed further. The CD3+ CD4+ plus CD3+ CD8+ cells amounted to 87·4 ± 1·1% (n = 11) of the total CD3+ cells. In the BAL fluid, the CD3+ cells accounted for 46·0 ± 1·9% (n = 21) of the cells in the lymphogate, and the CD3+ CD4+ plus CD3+ CD8+ cells amounted to 71·0 ± 1·4% (n = 21) of the CD3+ cells. CD21+ cells in BAL fluid were 2·2 ± 0·4% (n = 13) of the lymphogate. The results point to a substantial number of CD3 cells in BAL fluid that are negative for CD4 and CD8.
Figure 2.
Dotplots for peripheral blood mononuclear cells (a–c) and BAL fluid cells (d–f) with identical FACS instrument settings. (a,d) Scatterplots with lymphogates; (b,e) CD3 versus CD4 fluorescent signal; and (c,f) CD3 versus CD8 fluorescent signal. The MFI of CD3 was higher in blood than in BAL fluid. BAL fluid cells within the lymphogate contain a CD4+ population that is negative for CD3.
In the non-allergic dogs, the ratio of CD4 : CD8 (always measured as CD3+ CD4+/CD3+ CD8+) before allergen challenge was lower in the BAL fluid than in blood (Fig. 3a). In the allergic dogs, this ratio was higher in BAL fluid than in the blood in four out of six animals, with a trend to a significant difference between allergic and non-allergic dogs (P = 0·059).
Figure 3.
(a) The CD4 : CD8 ratio was measured as CD3 CD4 : CD3 CD8 ratio. The ratio was lower in BAL fluid than in blood in the non-allergic dogs. Four of six allergic dogs had a higher ratio in BAL fluid than in blood. (b) The mean fluorescence intensity of MHC II was lower on BAL fluid cells than on blood lymphocytes, both for CD4 and CD8 cells, P < 0·01 (paired t-test). The results are shown for antibody CSV20.
Almost 100% of the blood CD4 and CD8 T cells were MHC II positive (Table 3). In BAL fluid from the allergic dogs, the percentages of MHC II-positive cells were 79·9 ± 3·3% (mean ± SEM) and 65·2 ± 6·2 for CD4 and CD8 T cells, respectively, with a trend for a significant difference between CD4 and CD8 cells (P = 0·062). The MHC II expression on T cells from the non-allergic dogs was similar to that of the allergic dogs, but the low number of non-allergic dogs in the study may have obscured any differences. The percentages of MHC II-positive cells were always lower in BAL fluid than in the blood.
Table 3.
MHC II and CD45RA expression on blood and BAL fluid lymphocytes
| Allergic dogs | Non-allergic dogs | |||
|---|---|---|---|---|
| Blood | BAL fluid | Blood | BAL fluid | |
| MHC II CD4 | 99·2 ± 0·4* | 79·9 ± 3·3†‡ | 96·2 ± 1·5 | 87·2 ± 4·1 |
| (% of CD4) | ||||
| MHC II CD8 | 98·0 ± 1·0 | 65·2 ± 6·2 | 96·2 ± 1·6 | 70·5 ± 5·9 |
| (% of CD8) | ||||
| CD45RAbright CD4 | 49·7 ± 7·3 | 3·2 ± 0·5 | 47·1 ± 9·7 | 3·9 ± 0·1 |
| (% of CD4) | ||||
| CD45RAbright CD8 | 80·7 ± 2·9 | 37·3 ± 4·8§ | 82·1 ± 4·7 | 68·7 ± 6·3 |
| (% of CD8) | ||||
All results are expressed as mean ± SEM; for MHC II antibody CVS20 was used throughout.
Values in BAL fluid were lower than those in blood in all cases both for MHC II and CD45RA bright, except for one value for CD45RAbright CD8 in the BAL fluid from one non-allergic dog that fell within the range of values of blood lymphocytes.
Higher than MHC II CD8 in allergic dogs, P = 0·062.
Lower than percentage CD45RA bright CD8 in non-allergic dogs, P < 0·01.
The CD45RAbright-positive cells are supposed to represent naive cells. In the blood, 49·7 ± 7·3 and 80·7 ± 2·9% of the CD4 and CD8 T cells, respectively, were CD45RAbright, without a difference between allergic and non-allergic dogs (Table 3). Both CD4 and CD8 cells had lower percentages of naive cells in the BAL fluid than in the blood (Table 3) with CD45RAbright CD4 cells below 5% of the total CD4 cells. The percentage of CD45RAbright CD8 T cells in BAL fluid from the allergic dogs was 37·3 ± 4·8%, which was significantly lower than that in the non-allergic dogs (P < 0·01). This points to more experienced CD8 T cells in the airway compartment of the allergic dogs.
The mean fluorescent intensity (MFI) of the CD3 signal was lower on BAL fluid lymphocytes than on blood lymphocytes both in allergic and non-allergic dogs, indicating that BAL fluid lymphocytes were more activated (Fig. 2). Furthermore, the MFI of the MHC II signal was significantly lower in BAL fluid than in the blood (Fig. 3b).
The effect of allergen challenge on lymphocyte sub-populations
Allergen challenge resulted in increased percentages of MHC II-positive CD4 and CD8 T cells in the BAL fluid of the allergic dogs at all time-points (Fig. 4a,b). At 4 hr, the increase for MHC II+ CD4+ cells was significantly higher (P < 0·01) than the small changes in the non-allergic dogs, and there was a trend for a significant difference for the MHC II+ CD8+ cells (P = 0·065). At 72 hr, the increases were significantly higher than the small changes in MHC II CD4 and MHC II CD8, respectively, in the non-allergic dogs. There were small increases in the percentage of MHC II CD4 and MHC II CD8 in the saline-challenged segments of the allergic dogs as well, but these increases were smaller than those in the ragweed-challenged segments. Furthermore, allergen challenge increased the MFI of the MHC II signal both on the CD4 (P < 0·05) and the CD8 (P < 0·02) T cells (shown for t =72 hr in Fig. 5). The increase on CD4 T cells in allergic dogs was significantly greater than any increase in non-allergic dogs (Fig. 5a, P < 0·05). For CD8 T cells there was only a trend for such a difference. Similar effects of the allergen challenge on the MFI of MHC II were observed at 4 and 24 hr after allergen challenge. Thus, activated T cells were found in BAL fluid from allergic dogs after ragweed challenge from 4 hr after challenge onward. The results for MHC II reported in the tables and figures were obtained with the antibody CVS20. In some experiments, the antibody YKIX.344 was applied in parallel and yielded essentially the same results.
Figure 4.
(a) Changes of the percentage CD3 CD4 MHC II after allergen and saline challenge. The results did not deviate from a normal distribution and are expressed as mean ± SEM. ▪, ragweed-challenged allergic dogs; □, saline-challenged allergic dogs; ▴, ragweed-challenged non-allergic dogs. At all time-points the percentage MHC II-positive CD3 CD4 cells after ragweed challenge in allergic dogs was higher than before the challenge: *, **, ***, P < 0·02 (t-test). The increases after ragweed challenge in allergic dogs were significantly larger than those after ragweed challenge in non-allergic dogs: *, ***, P < 0·01 (t-test), and those after saline challenge in allergic dogs: **, P < 0·03 (paired t-test); whereas there was a trend for a significant difference with saline challenge in allergic dogs at 72 hr: ***, P = 0·065 (paired t-test). (b) Similar results were obtained for the CD8 MHC II. At all time-points the percentage MHC II-positive CD8 cells after ragweed challenge in allergic dogs was higher than before challenge: *, **, ***, P < 0·01 (t-test). At 4 hr the increases after ragweed challenge in allergic dogs tended to be larger than those after ragweed challenge in non-allergic dogs: *, P = 0·06. At 72 hr the increases after ragweed challenge in allergic dogs were significantly larger than those after ragweed challenge in non-allergic dogs: ***, P < 0·05 (t-test). The increases after ragweed challenge in allergic dogs were significantly larger than those after saline challenge in allergic dogs: **, P < 0·02 (paired t-test) and ***, P < 0·01 (paired t-test). Antibody CSV20 was used throughout.
Figure 5.
The allergen challenge resulted in significant increases of the MFI of MHC II both on (a) CD3 CD4 and (b) CD8 cells in BAL fluid. Results are shown for antibody CSV20 obtained at 72 hr after challenge.
The ragweed challenge resulted in an increase of the CD45RAbright CD8+ cells in the BAL fluid of the allergic dogs (Fig. 6). At 24 and 72 hr this was significantly different from the slight decrease of CD45RAbright CD8 in the non-allergic dogs. An increase of CD45RAbright CD8 was also observed in the saline-challenged segment of the allergic dogs, without a statistically significant difference between ragweed- and saline-challenged segments. The changes in the CD45RAbright CD4 cells were only 2–4%, which prohibited a reliable analysis.
Figure 6.
Changes of the percentage of CD45RAbright CD8 after allergen and saline challenge. The results are expressed as mean ± SEM. ▪, ragweed-challenged allergic dogs; □, saline-challenged allergic dogs; ▴, ragweed-challenged non-allergic dogs. The increases after ragweed challenge in allergic dogs were significantly different from those in non-allergic dogs; *,**, P < 0·05. Saline challenge in allergic dogs resulted in changes similar to those after ragweed challenge.
The allergen challenge did not result in statistically significant differences between allergic and non-allergic dogs with respect to the changes in the CD4 : CD8 ratio.
MHC II expression after in vitro stimulation of T lymphocytes
To investigate whether an increase in the MFI of MHC II represents activation of the lymphocytes, T cells were stimulated in vitro with PHA and the MHC II expression was measured. Figure 7 shows that at 24 hr after PHA stimulation, the MFI of MHC II-positive cells had increased from 158 to 220 au.
Figure 7.
Peripheral blood mononuclear cells were cultured with (solid line) and without (dotted line) PHA for 24 hr. MHC II on CD3-positive cells was measured with antibody CSV20. PHA stimulation resulted in an increase of the MFI of MHC II from 158 to 220 au.
Discussion
The beagle model for allergic asthma provides the opportunity to study pulmonary allergic responses under biological and physiological conditions that are closer to the conditions in humans than several models in rodents.1–4 Furthermore, the model allows longitudinal studies in individual animals. Our data show activation of local T lymphocytes in the lungs after ragweed deposition in a sub-segment of the lungs of allergic dogs sensitized to ragweed. The T-cell activation was significantly different from any activation in the saline-challenged segment of the same allergic dogs. It was also significantly different from any activation after ragweed challenge in non-allergic dogs. T cells were activated as early as 4 hr after ragweed challenge until at least 72 hr after challenge.
The T-cell activation was evaluated by measuring the changes in the percentages of cells positive for MHC II and by measuring the MFI of MHC II expression. In addition, significant changes in the CD45RA-positive populations were shown. Markers for T-cell activation in dogs are limited. Furthermore, neutrophils from dogs express CD4,15 which makes it necessary to perform simultaneous staining for CD3 and CD4 when mixed cell samples such as BAL-fluid cells are to be analysed under conditions of an inflammatory reaction. A triple staining for CD3 and CD4 and activation markers has not been reported earlier for dog lymphocytes; this was done by the application of mouse anti-canine CD3FITC in combination with rat anti-canine CD4 and monoclonal anti-rat antibody, and monoclonal rat anti-mouse subclass-specific antibody with an anti-activation-marker antibody (Table 1). The MFI of MHC II on dog T cells was interpreted as a marker for T-cell activation as demonstrated by in vitro cell culture experiments in which MFI increased after stimulation of the cells with PHA (refs. 11, 12 and the present study).
Our results showed that T cells are activated in the airways of IgE-allergic animals within 4 hr after allergen exposure. It is generally accepted that mast-cell-dependent reactions cause the immediate pathology of the IgE-dependent hypersensitivity reactions. Such reactions are generally followed by late-phase reactions, which start to develop at 4–6 hr after allergen exposure. It is not known which factors cause the late-phase allergic reactions in the airways.21 Eosinophilic and neutrophilic granulocytes probably play an important role. Lymphocytes specifically react with allergens and thereby may impose the antigenic specificity in the events leading to the development of the late-phase reactions. They may be activated by allergen-derived peptides presented by antigen-presenting cells, or by products released from activated mast cells. Activated lymphocytes may secrete cytokines, such as interferon-γ, tumour necrosis factor-α, IL-4, IL-5, IL-13 and IL-16, which may activate airway epithelial cells and contribute to the chemotaxis and activation of inflammatory cells.22,23 In humans with allergic asthma, T-cell activation in the airways occurred within 4–6 hr after controlled allergen exposure.9,10 In this study, allergic, ragweed-sensitized dogs showed a similar activation of T lymphocytes within 4 hr after ragweed exposure, providing further evidence for a possible role of T cells in the late-phase allergic reactions in the airways.
It cannot be determined whether the changes in the MHC II and CD45RAbright expression were due to the migration of cells from the blood into the airway compartment or to the activation of resident cells. At 4 hr, there was no net increase of lymphocyte numbers in BAL fluid. This does not exclude, however, that cells from the blood may have replaced resident cells. At 24 hr, an increase, though not statistically significant, of lymphocytes had occurred. Therefore, both activation of resident cells and migration of cells from the blood may have contributed to the changes in local T-cell sub-populations.
The increase of MHC II expression after ragweed challenge was higher in allergic dogs than in the non-allergic dogs as well as in the saline-challenged segments of allergic dogs. In contrast, the number of the CD45RAbright cells increased both in the saline- and ragweed-challenged segments of allergic dogs. Apparently, the local pulmonary sensitization that had been induced by the earlier ragweed instillation in the allergic dogs had conditioned the airways to react with significant changes in memory and naive phenotypes of the cells upon the non-specific triggering signal of the saline instillation. Alternatively, local reactions induced by ragweed in one segment result in distant effects with respect to cell migration and/or activation in other segments. The cells in the saline segments did not show, however, activation as determined by the MHC II expression. The combined results of the MHC II and CD45RA expression support the assumption that T cells migrate into the airway compartment of allergic dogs. These cells become activated in the allergen-challenged segments, and MHC II expression is down-regulated on these cells in the saline segments.
The ragweed instillation in non-allergic dogs caused some increase of neutrophilic and eosinophilic granulocytes in subsequent lavage fluids, the significance of which may have been obscured by the low number of dogs tested. An inflammatory reaction in the lungs of these dogs may have been the result of the reaction of IgG antibodies with ragweed, and/or of a possible endotoxin contamination of the ragweed preparation. Furthermore, the saline instillation also resulted in slight increases in leucocyte counts. This may have been the result of introducing endotoxin during the bronchoscopy procedure or a reaction to the bronchoscopy procedure itself. The significance of these small changes may be evaluated in studies with larger numbers of animals.
A limitation in the design of the study was that we did not analyse saline challenges at 4 hr after challenge, and in non-allergic dogs at 24 hr after saline exposure. Although this limited the insight in the events we have assumed paucity of reactions under those conditions on the basis of our experience with this dog model. Indeed, at 24 hr, the saline challenge resulted only in small increases in neutrophils and eosinophils even in allergic dogs. Furthermore, at 4 hr we analysed the reactions after allergen challenge in only three allergic dogs; and similarly at 24 hr, after saline challenge in allergic dogs only three animals were tested. Those limitations were simply related to restrictions in available reagents and time. On the basis of the present results it will be possible to develop better focused study designs.
In addition to the T-cell activation after ragweed challenge, several interesting observations were made with respect to the steady state before allergen challenge. The CD4: CD8 ratio was measured as CD3 CD4:CD3 CD8 ratio. The triple staining for CD3, CD4 and CD8 had not been reported earlier for dog T cells. It was achieved by the application of anti-canine CD3 in combination with monoclonal rat anti-mouse antibody and directly labelled anti-canine CD4 and anti-canine CD8 antibodies. The CD4: CD8 ratio in BAL fluid from non-allergic dogs was lower than in the blood, which is in line with the observation by Dirscherl et al.24 In the allergic dogs, however, a very high CD4:CD8 ratio was found in four out of six dogs. This points to active involvement of CD4 T cells in sustaining the local pulmonary immunity and or the local allergic condition resulting from earlier local ragweed deposition. It would be interesting to study the cytokine profile of these CD4 T cells with respect to Th1 and Th2 cytokine production.
The percentage of MHC II-positive T cells was lower in BAL fluid than in blood. This confirms earlier reports12,24 and points to a down-regulation of the activation state of BAL fluid T cells in this respect, in comparison with blood T cells. Furthermore, the MFI of MHC II of BAL fluid T cells was much lower than that of blood T cells. The lower activation state of BAL fluid T cells, as deduced from MHC II expression, is in contrast to the activated state of T cells as deduced from the lower MFI of CD3 on BAL fluid T cells than on blood T cells. Thus, different regulatory pathways cause the actual degree of activation of BAL fluid T cells.
The CD45RA expression on BAL fluid T cells is lower than in blood.24 We have shown here that in particular the CD4 T cells have a low CD45RA expression; the CD45RAbright CD4 T cells are less than 5% of the total CD4 T cells. This indicates that the CD4 T cells in the airway compartment have differentiated into the memory phenotype both in normal and allergic dogs. It is not known whether the CD45RAbright sub-population of CD8 T cells in dogs may contain a memory-effector subset, as has been described for humans and for other animals.25 It is probable that the CD8 T cells in the airway compartment of dogs are mainly of the naive phenotype.
We conclude that T cells are activated in the airways of dogs after in vivo challenge. Furthermore, the dog model for allergic asthma showed that T cells are activated within 4 hr after local allergen challenge, which makes the model suited for further analysis of the possible involvement of T lymphocytes in the allergic broncho-obstructive reactions.
Acknowledgments
The study was supported by grants from the Netherlands Asthma Foundation (grant no. 99.92) and ALK ABELLO BV, the Netherlands. We thank Dr P. Moore for the gift of anti-CD45RA, Dr B. Muggenburg for his assistance in some of the lavage procedures, Ms M. Nysus for her help with animal care, and Mr R. Jaramillo for his help with the flowcytometer. Dr V. Rutten (Utrecht, the Netherlands) and Mr R. M. R. Reijneke co-operated in preliminary experiments on analysing the dog BAL lymphocyte subsets. The Lovelace Respiratory Research Institute is fully accredited by the Association of the Assessment and Accreditation of Laboratory Animal Care International. The animals were cared for in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The study was approved by the Animal Care and Use Committee of the Lovelace Respiratory Research Institute.
References
- 1.Theodorou A, Weger N, Kunke K, Rhee K, Bice D, Muggenburg B, Lemen R. Ragweed sensitization alters pulmonary vascular responses to bronchoprovocation in beagle dogs. J Appl Physiol. 1997;83:912–17. doi: 10.1152/jappl.1997.83.3.912. [DOI] [PubMed] [Google Scholar]
- 2.Bice DE, Seagrave J, Green FHY. Animal models of asthma: potential usefulness for studying health effects of inhaled particles. Inhal Toxicol. 2000;12:829–62. doi: 10.1080/08958370050123207. [DOI] [PubMed] [Google Scholar]
- 3.Redman TK, Rudolph K, Barr EB, Bowen LE, Muggenburg BA, Bice DE. Pulmonary immunity to ragweed in a beagle dog model of allergic asthma. Exp Lung Res. 2001;27:433–51. doi: 10.1080/019021401300317143. [DOI] [PubMed] [Google Scholar]
- 4.Bice DE, Collie DDS, Muggenburg BA, Hahn FF. Effects of anti-ragweed immunity in the lung on inflammation and airway reactivity. Am J Respir Crit Care Med. 1997;155:A157. [Google Scholar]
- 5.Longphre M, Li D, Gallup M, et al. Allergen induced IL-9 directly stimulates mucin transcription in respiratory epithelial cells. J Clin Invest. 1999;104:1375–82. doi: 10.1172/JCI6097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Humbert M, Menz G, Ying S, Corrigan CJ, Robinson DS, Durham SR, Kay AB. The immunopathology of extrinsic (atopic) and intrinsic (non-atopic) asthma. Immunol Today. 1999;20:528–33. doi: 10.1016/s0167-5699(99)01535-2. [DOI] [PubMed] [Google Scholar]
- 7.Bochner BS, Undem BJ, Lichtenstein LM. Immunological aspects of allergic asthma. Annu Rev Immunol. 1994;12:295–335. doi: 10.1146/annurev.iy.12.040194.001455. [DOI] [PubMed] [Google Scholar]
- 8.Mekori YA, Metcalfe DD. Mast cell–T cell interactions. J Allergy Clin Immunol. 1999;104:517–23. doi: 10.1016/s0091-6749(99)70316-7. [DOI] [PubMed] [Google Scholar]
- 9.Gratziou C, Carroll M, Montefort S, Teran L, Howarth PH, Holgate ST. Inflammatory and T-cell profile of asthmatic airways 6 hours after local allergen provocation. Am J Respir Crit Care Med. 1996;153:515–20. doi: 10.1164/ajrccm.153.2.8564090. [DOI] [PubMed] [Google Scholar]
- 10.Out TA, De Pater-Huijsen FL, Nocker RET, De Riemer MR, Van der Zee JS, Jansen HM. Segmental allergen challenge in asthma affects CD27+CD103+ T cells in BAL fluid and blood within four hours after challenge. Am J Respir Crit Care Med. 1999;159:A20. [Google Scholar]
- 11.Doveren RF, Buurman WA, Schutte B, Groenewegen G, Van der Linden CJ. Class II antigens on canine T lymphocytes. Tissue Antigens. 1985;25:255–65. doi: 10.1111/j.1399-0039.1985.tb00448.x. [DOI] [PubMed] [Google Scholar]
- 12.Chang SC, Hu CP, Perng RP, Shiao GM, Lin CY. MHC class II antigens on canine bronchoalveolar lavage cells. Immunol Lett. 1990;26:11–15. doi: 10.1016/0165-2478(90)90169-q. [DOI] [PubMed] [Google Scholar]
- 13.Chang SC, Hsu HK, Lin CY. Usefulness of bronchoalveolar lavage cell profile in early detection of canine lung allograft rejection. Immunol Lett. 1991;29:265–70. doi: 10.1016/0165-2478(91)90181-9. [DOI] [PubMed] [Google Scholar]
- 14.Tipold A, Moore PF, Jungi TW, Sager H, Vandervelde M. Lymphocyte subsets and CD45RA positive T cells in normal canine cerebrospinal fluid. J Neuroimmunol. 1998;82:90–5. doi: 10.1016/S0165-5728(97)00192-6. [DOI] [PubMed] [Google Scholar]
- 15.Moore PF, Rossitto PV, Danilenko DM, Wielenga JJ, Raff RF, Severns E. Monoclonal antibodies specific for canine CD4 and CD8 define functional T-lymphocyte subsets and high-density expression of CD4 by canine neutrophils. Tissue Antigens. 1992;40:75–85. doi: 10.1111/j.1399-0039.1992.tb01963.x. [DOI] [PubMed] [Google Scholar]
- 16.Collie DDS, Deboer DJ, Muggenburg BA, Bice DE. Evaluation of association of blood and bronchoalveolar eosinophil numbers and serum total immunoglobulin E concentration with the expression of nonspecific airway reactivity in dogs. Am J Vet Res 1997. 1997;58:34–9. [PubMed] [Google Scholar]
- 17.Bice DE, Muggenburg BA. Pulmonary immune memory: localized production of antibody in the lung after antigen challenge. Immunology. 1996;88:191–7. doi: 10.1111/j.1365-2567.1996.tb00004.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Strasser A, Kalmar E, Niedermüller H. A simple method for the simultaneous separation of peripheral blood mononuclear and polymorphonuclear cells in the dog. Vet Immunol Immunopathol. 1998;62:29–35. doi: 10.1016/s0165-2427(97)00150-5. [DOI] [PubMed] [Google Scholar]
- 19.Motulsky H. Analyzing Data with Graphpad Prism. San Diego, CA: Graphpad software Inc.; 1999. [Google Scholar]
- 20.Byrne KM, Kim HW, Chew BP, Reinhart GA, Hayek MG. A standardized gating technique for the generation of flow cytometry data for normal canine and normal feline blood lymphocytes. Vet Immunol Immunopathol. 2000;73:167–82. doi: 10.1016/s0165-2427(99)00163-4. [DOI] [PubMed] [Google Scholar]
- 21.Bently AM, Kay AB, Durham SR. Human late asthmatic reactions. Clin Exp Allergy. 1997;27:S71–S86. doi: 10.1111/j.1365-2222.1997.tb01830.x. [DOI] [PubMed] [Google Scholar]
- 22.Robinson D, Hamid Q, Bently A, Ying S, Kay AB, Durham SR. Activation of CD4+ T cells, increased Th2-type cytokine mRNA expression, and eosinophil recruitment in bronchoalveolar lavage after allergen inhalation challenge in patients with atopic asthma. J Allergy Clin Immunol. 1993;92:313–24. doi: 10.1016/0091-6749(93)90175-f. [DOI] [PubMed] [Google Scholar]
- 23.Virchov JC, Jr, Walker C, Hafner D, Kortsik C, Werner P, Matthys H, Kroegel C. T cells and cytokines in bronchoalveolar lavage fluid after segmental allergen challenge in asthma. Am J Respir Crit Care Med. 1995;151:960–8. doi: 10.1164/ajrccm/151.4.960. [DOI] [PubMed] [Google Scholar]
- 24.Dirscherl P, Beisker W, Kremmer E, Mihalkov A, Voss C, Ziesenis A. Immunophenotyping of canine bronchoalveolar lavage and peripheral blood lymphocytes. Vet Immunol Immunopathol. 1995;48:1–10. doi: 10.1016/0165-2427(94)05414-n. [DOI] [PubMed] [Google Scholar]
- 25.Hamann D, Roos MT, Van Lier RA. Faces and phases of human CD8 T-cell development. Immunol Today. 1999;20:177–80. doi: 10.1016/s0167-5699(99)01444-9. [DOI] [PubMed] [Google Scholar]