INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

There is substantial evidence of T-lymphocyte activation and cytokine expression in both human subjects and animals undergoing airway challenge with a sensitizing antigen (1). Cells expressing the cytokines IL-4 and IL-5 are present within the airway wall and in bronchoalveolar lavage (BAL) fluid and have been linked to the late response, airway hyperresponsiveness, and eosinophilic infiltration of the airway wall (6). Interleukin (IL)-4 is involved in promoting IgE responses to sensitization (9) and is a major determinant of the biasing of T cells towards the production of Th2 cytokines, thereby forming part of a feed-forward mechanism accounting for allergic inflammation. IL-4 is also capable of up-regulating the VLA-4-VCAM-1 pathway, which may play a part in both eosinophil and T-cell migration (10, 11). IL-5 is essential for eosinophil activation (12), differentiation, and survival (13) and also promotes eosinophilia through effects on adhesion molecules (14). The sources of IL-4 and IL-5 include T cells (3, 15) and mast cells (16); eosinophils are also a source of IL-5 (17).

After allergen sensitization and challenge the actively sensitized Brown Norway (BN) rat develops late airway responses (18), eosinophilic airway inflammation, and airway hyperresponsiveness (19). However, it is difficult to determine the relative contributions of mast cells and T cells to the observed changes in airway structure and function because both cell types are activated after allergen exposure. In an attempt to isolate the role of T cells we and others developed a rat model of allergen-induced airway responses using the technique of adoptive transfer (20, 21). The recipients of purified CD4+ T cells from donors sensitized with ovalbumin (OA) developed late airway responses after OA challenge (22). Immunocytochemical staining for major basic protein detected eosinophilia at 8 h after challenge and was associated with the expression of IL-5 and IL-4 (23). There was no demonstrable OA-specific IgE in the recipient rats indicating that T cells could evoke late airway responses without the necessity for the triggering of mast cells through IgE crossbridging (22).

Although late allergic responses and airway hyperresponsiveness frequently occur together they have not been shown to be linked from a mechanistic point of view. For this reason we decided to examine the possibility that CD4+ T-lymphocytes may be responsible for the airway hyperresponsiveness that is present at 32 h after challenge. In previous studies of the late response (22) we did not explore the possible modulatory role of CD8+ T cells in the airway response to allergen challenge. However, it is possible that CD8+ T cells are inhibitory of CD4+ T-cell-mediated effects. We postulated that T-cell-induced hyperresponsiveness would be associated with the expression of Th2 type cytokines (IL-4 and IL-5) and eosinophilia and that inhibition of hyperresponsiveness by CD8+ T cells would be associated with the expression of the Th1 cytokine, interferon- (IFN-). Therefore, we tested the effects of T cell transfers, both purified CD4+ and CD8+ singly and in combination, from sensitized donors to naive recipients undergoing OA challenge. Responsiveness to methacholine was evaluated before and 32 h after allergen exposure. Immunocytochemistry was used to detect eosinophils and in situ hybridization was used to determine the expression of IL-2, IL-4, IL-5, and IFN- mRNA in cells retrieved by BAL.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and Sensitization

Highly inbred male Brown Norway (Rij substrain) rats 6 to 8 wk of age and weighing 180 to 230 g were purchased from Harlan Sprague- Dawley Inc. (Walkerville, MD). All procedures in this study were approved by the University Animal Care Committee.

T-cell donor rats were sensitized with a single subcutaneous injection of 1 ml of normal saline containing 1 mg OA (Grade V; Sigma Immunochemicals, St. Louis, MO) and 4.28 mg of aluminum hydroxide gel (Anachemia Chemicals, Montreal, PQ, Canada) as an adjuvant. Simultaneously, 0.5 ml of Bordetella pertussis vaccine containing 6 · 10⁹ heat-killed bacilli (IAF; Laval-Des-Rapides, Montreal, PQ) was injected intraperitoneally.

Measurement of Pulmonary Mechanics

Measurement of airway responses was performed as previously described (20) in animals that were anesthetized intraperitoneally with urethane (1 g/kg). The end of the endotracheal tube (6-cm length of PE240) of intubated animals was placed inside a small Plexiglas box and a Fleisch no. 0 pneumotachograph coupled to a differential pressure transducer (Micro-Switch 163PC01D36; Honeywell, Scarborough, ON) was attached to the other end of the box to measure airflow. Volume was obtained by integration of the flow signal. Changes in esophageal pressure were measured using a saline-filled catheter connected to a differential pressure transducer (Transpac II disposable transducer; Sorenson, Salt Lake City, UT) referenced to the Plexiglas box. Transpulmonary pressure was obtained by subtraction of esophageal pressure from the pressure in the Plexiglas box. Pulmonary resistance (RL) was determined by fitting the equation of motion of the lung to the data by multiple linear regression analysis using a commercial software package (RHT Infodat Inc., Montreal, PQ).

Immunomagnetic Separation of CD4+ and CD8+ T-Cell Subsets and Adoptive Transfer

Fourteen days after sensitization, mononuclear cells were obtained from four or five cervical lymph nodes of donor rats by mincing of tissue and subsequent passage through a stainless steel sieve. Cells were passed through a nylon mesh to remove debris. Negative selection using a magnetic cell sorter (MACS; Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) was performed. Briefly, undesired cells in the cell suspension were labeled with a mixture of primary-specific mAbs and indirectly labeled with MACS rat antimouse IgG₁ microbeads. Next, cells were treated with either W3/25 (MCA P55, mouse antirat CD4 mAb, IgG₁) or OX-8 (MCA P48, mouse antirat CD8 mAb, IgG₁) in combination with OX-33 (MCA P49, mouse antirat k chain mAb, IgG₁) and ED9 (MCA P340, mouse antirat myeloid differentiation antigen, IgG₁) in a dilution of 1:100. The labeled cells were then passed through the MACS column and the effluent containing the desired cells was collected. ED9, which recognizes a membrane antigen on rat macrophages, monocytes, dendritic cells, and granulocytes, was used to remove nonlymphocytic cells in the cell suspension, whereas OX-33 was used to remove B cells. Flow cytometry was performed on a sample of cell isolates to establish the fractions of the cells that were CD4+ or CD8+. We used W3/13HLK to establish the fraction of cells that were T cells after purification by MACS. W3/25, MRC OX8, and W3/13HLK were purchased from Cedarlane Laboratories (Hornby, ON); MRC OX33 and ED9 were purchased from Prince Laboratories Inc. (Toronto, ON).

Adoptive transfer of the T-cell subsets was performed on Day 14, immediately after the immunomagnetic T-cell separation. Either CD4+ or CD8+ enriched populations of T cells or whole T cells were resuspended in sterile PBS. Two million CD4+, 2 million CD8+, or 5 million T cells that contained both CD4+ and CD8+ cells were transferred by intraperitoneal injection to naive recipient BN rats. The cell numbers refer to the total cells transferred and are not corrected for the purity of the preparations.

Ovalbumin Challenge and Measurement of Airway Responsiveness to Methacholine

After adoptive transfer, 2 d were allowed to elapse prior to antigen challenge to permit the transferred T cells to stabilize their distribution. Then recipient rats were challenged with either aerosolized OA or bovine serum albumin (BSA) (5% wt/vol), using a Hudson nebulizer (Model 1400; Hudson, Temecula, CA) with an airflow of 10 L/min and an output of 0.15 ml/min for 5 min.

Thirty-two hours after the challenge, airway responsiveness to methacholine (MCh) was measured for comparison with baseline responsiveness, which was determined 2 d before the transfer of T cells. Responsiveness was quantitated using RL. Animals were exposed to progressively doubling concentrations of aerosolized MCh until RL underwent a doubling from the baseline value. The concentration of MCh required to effect this change was called the EC₂₀₀RL and was calculated by interpolation. A Hudson nebulizer was also used but the duration of nebulization for each concentration of MCh was 30 s.

Bronchoalveolar Lavage Cell Preparation for Immunocytochemistry and In Situ Hybridization

BAL was performed after the completion of physiologic measurements. The lungs were lavaged through the endotracheal tube with 25 ml of chilled normal saline, and cell number was counted on a fresh specimen using a hemacytometer. Cytospin slides were prepared using Cytospin model II (Shandon, Pittsburgh, PA) onto poly-L-lysine-coated glass slides and were fixed with 4% paraformaldehyde. BAL cells were immunostained with BMK13 mAb, which recognizes major basic protein (MBP) (BMK 13 was kindly provided by Dr. Redwan Moqbel, University of Alberta) using the alkaline phosphatase anti-alkaline phosphatase (APAAP) method. MBP-positive cells were quantified by an investigator blinded to group status. A minimum of 1,000 BAL cells was counted, and the proportion of cells expressing MBP immunoreactivity was determined.

In situ hybridization was performed as previously described (24). Antisense and sense riboprobes were prepared from cDNAs coding for rat IL-2, IL-4, IL-5, and IFN- mRNA (3). For detection of cytokine mRNAs, cytospin preparations from BAL were permeabilized with Triton X-100 and proteinase K (1 µg/ml) in 0.1 M TRIS containing 50 mM EDTA for 20 min at 37° C. To prevent nonspecific binding of ³⁵S-labeled RNA probes, the preparations were incubated with 10 mM N-ethyl maleimide and 10 mM iodoacetamide for 30 min at 37° C, followed by incubation in 0.5% acetic anhydride and 0.1 M triethanolamine for 10 min at 37° C. Prehybridization was performed with 50% formamide and 2× standard saline citrate for 15 min at 40° C. For hybridization, antisense or sense probes (10⁶ cpm/section) were diluted in hybridization buffer. Dithiothreitol (100 mM) was present in the hybridization mixture to ensure blocking of any nonspecific binding of the ³⁵S-labeled probes. Posthybridization washing was performed in decreasing concentrations of standard saline citrate at 45° C. Unhybridized single-strand RNA was removed by RNase A (20 µg/ml). After dehydration, the slides were immersed in NBT2 emulsion and exposed for 10 d. In addition to using sense probes as a negative control we also treated cytospins with RNAse prior to hybridization with antisense probes. This additional negative control served to confirm the specificity of our probes and signals. The autoradiographs were developed in Kodak D-19, fixed, and counterstained with hematoxylin.

Slides were coded, and positive cells were counted blindly using a magnification ×100 with an eyepiece graticule. The results were expressed as the mean number of positive cells per 1,000.

Statistical Analysis

Measurements of airway responsiveness before and after OA challenge were compared using Wilcoxon's signed rank test. Comparison of data for expression of various cytokines was done using ANOVA followed by Fisher's least significant difference test. Differences were considered to be statistically significant when p values were less than 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Purity of T-Cell Preparations after Immunomagnetic Cell Sorting

The purity of the T-cell preparations obtained by MACS was determined by flow cytometry. Of the cells retrieved after removal of B cells and myeloid cells, 93 and 96% were T cells in two separate experiments. With removal of CD8+ cells the purity achieved for the CD4+ T-cell preparations was greater than 98% (n = 3). The CD8+ enriched cells were less pure and were 58 and 64% of the total cells in two separate experiments. However, this cell fraction contained less than a few percent of CD4+ cells, B cells, and myeloid cells.

Changes in Airway Responsiveness after OA Challenge

The baseline values of EC₂₀₀RL were comparable among the various groups, with the exception of animals that received total T cells (Figure 1). These animals had greater airway responsiveness (lower values of EC₂₀₀RL) relative to other groups prior to T-cell transfers or OA exposure.

View larger version (17K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   Figure 1.   Physiologic changes of airway hyperresponsiveness observed in (A) recipients of antigen-primed CD4+ T cells, (B) CD8+ T cells, and (C ) total T cells before and after inhalation challenge with ovalbumin (OA) or bovine serum albumin (BSA). The values shown are mean ± SE. A significant change was demonstrated by Wilcoxon's signed rank test (p < 0.05), illustrating that antigen-induced airway hyperresponsiveness can be transferred by antigen specific CD4+ T cells in BN rats.

Hyperresponsiveness developed after OA challenge of the recipients of CD4+ cells from OA-sensitized donors; the EC₂₀₀RL fell from 5.5 mg/ml (geometric mea) to 3.8 mg/ml after OA challenge (p < 0.05). The BSA-challenged animals showed no change (baseline EC₂₀₀RL, 5.3 mg/ml and postchallenge 4.7 mg/ml; p = NS). The recipients of CD8+ cells showed no significant changes in responsiveness to MCh, irrespective of whether they were challenged with OA or BSA. The recipients of total T-cell preparations had no significant changes in responsiveness after OA challenge. However, recipients of total T cells that were challenged with BSA had a significant reduction in responsiveness, but these changes were associated with relatively low values of baseline EC₂₀₀RL to begin with.

Bronchoalveolar Lavage Leukocytes after OA Challenge

The volume of BAL fluid recovered ranged from 20 to 23 ml and did not differ significantly between groups. The total cell counts performed on BAL fluid did not show significant differences between the treatment groups by ANOVA (Figure 2, panel A). The macrophages and lymphocytes were not significantly different between treatment groups (Panels B and D). Neutrophils were significantly lower in CD8+ T-cell recipients that underwent OA challenge compared with other treatment groups ( Panel C).

View larger version (45K): [in this window] [in a new window]   Figure 2.   Cellular profile in BAL fluids performed at 32 h after challenge. Total cell number (A) was counted on a fresh specimen using a hemacytometer. The numbers of (B) macrophages, (C ) neutrophils, and (D) lymphocytes were assessed by May-Grünwald Giemsa staining on 200 cells. Neutrophils were lower in CD8+ transferred and OA-challenged rats than in their BSA-challenged controls.

Eosinophil numbers were determined by immunostaining for MBP. There was an increased proportion of eosinophils in the BAL fluid of CD4+ recipients compared with all other groups (Figure 3). The recipients of total T-cell transfers that were OA-challenged also showed an increase in eosinophils compared with their BSA controls, although significantly less than the OA-challenged CD4+ recipients.

View larger version (34K): [in this window] [in a new window]   Figure 3.   Eosinophil count in BAL fluids by immunocytochemistry. Eosinophil counts were assessed by APAAP using BMK 13, an anti-human MBP mAb cross-reacting with rat MBP and expressed as the number of positive-stained cells/1,000 cells.

Cytokine Expression in Bronchoalveolar Lavage Fluid Leukocytes

Cytokine expression was determined by in situ hybridization. The most striking findings related to IL-5 (Figure 4). Significantly more IL-5-positive cells were present among the BAL cells from CD4+ recipients that were OA challenged compared with other groups. Expression of IL-5 was low in BSA-challenged rats. CD8+ recipients and total T-cell recipients were comparable to each other, but they had fewer cells that were positive for IL-5 than did CD4+ recipients that were OA-challenged. IL-4 mRNA positive cells were significantly higher in CD4+ recipients that underwent OA challenge than in other groups, but there was no significant difference between IL-4 positive cells in the OA- and BSA-challenged CD4+ recipients. Interferon- was elevated in CD8+ recipients compared with CD4+ and total T-cell recipients, but again it was not specific for OA challenge. IL-2 was not different between groups.

View larger version (38K): [in this window] [in a new window]   Figure 4.   Cytokine mRNA profile in BAL fluids performed at 32 h after challenge. The expression of (A) IL-2, (B) IL-4, (C ) IL-5, and (D) IFN- was determined by in situ hybridization and quantitated as the number of positive cells/1,000 cells. The significant differences after ANOVA and multiple comparisons by Tukey testing are illustrated.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The principal results of the current experiments are that CD4+ T cells harvested from sensitized rats can transfer airway sensitivity to OA to naive recipients. Thirty two hours after inhalational challenge with OA CD4+ T cell recipients showed eosinophilic airway inflammation and airway hyperresponsiveness to inhaled MCh. Enhanced responses to MCh were not seen in either recipients of CD8+ cells or in those animals receiving a mixture of CD4+ and CD8+ cells. There was an increased expression of IL-5 in cells in the bronchoalveolar lavage fluid of CD4+ recipients after OA challenge compared with the other treatment groups. IL-4-expressing cells were also increased in CD4+ recipients that underwent OA challenge compared with all other treatment groups, with the exception of the BSA-challenged control rats. CD8+ recipients had low eosinophil numbers and relatively few cells expressing IL-5 in the BAL fluid. The total T-cell recipients resembled CD8+ recipients in the pattern of IL-5 expression, but they had more eosinophils than the CD8+ group. Interferon--expressing cells were higher in CD8+ recipients than total T-cell recipients and the BSA-challenged CD4+ recipients.

Increase in responsiveness to inhaled agonists such as MCh is one of the characteristic features observed after allergen challenge of actively sensitized animals. The timing and duration of the effect seems to vary from animal to animal, but in the rat it is present 24 to 32 h after challenge (25). Allergen-induced hyperresponsiveness to MCh was also demonstrable in T-cell recipients at a time point similar to that of the actively sensitized animals, confirming the results obtained by Haczku and colleagues (21) in which CD4+ T cells isolated from intrathoracic lymph nodes of sensitized rats permitted the induction of hyperresponsiveness to inhaled acetylcholine by OA challenge in naive rats. A role for CD4+ T cells has also been demonstrated in allergen-induced hyperresponsiveness in actively sensitized mice (29). The depletion of CD4+ T-lymphocytes by administration of an anti-CD4 monoclonal antibody 3 d prior to challenge with sheep red blood cells prevented the development of hyperresponsiveness to acetylcholine as well as BAL and pulmonary tissue eosinophilia (29).

In contrast to earlier results (20) a mixture of CD4+ and CD8+ cells failed to permit the induction of hyperresponsiveness to MCh by OA challenge, suggesting that the presence of CD8+ cells neutralized the influence of those CD4+ cells responsible for the induction of hyperresponsiveness after allergen challenge. The reason for the discrepancy in the results between the two studies is not clear, but it may relate to the site of harvesting of the T cells, which in the former study was the intrathoracic lymph nodes. The method of purification of the cells also differed between the two studies; crude purification by nylon wool columns was previously employed (20). Indeed it appears as if the total T cell transfers may actually reduce responsiveness to MCh. Such an effect was unexpected and was only statistically significant in magnitude in the recipients challenged with bovine serum albumin. Animals included in both of the total T-cell groups (OA- and BSA-challenged animals) had a baseline airway responsiveness to MCh that was higher than in all other batches. This finding was not related to the T-cell adminstrations because responsiveness was evaluated prior to the administration of the cells, but whatever its mechanism the finding increases the likelihood of observing a reduction in responsiveness by chance.

The mechanism of allergen-induced increases in airway responsiveness has not been elucidated. However, inhibition of hyperresponsiveness by corticosteroids (30, 31) and by specific antibodies against various adhesion molecules such as ICAM-1, LFA-1, and VLA-4 (32) lends support to the idea that the induction of inflammation by allergen is responsible for the change in responsiveness. Allergen challenge typically triggers inflammation involving activated lymphocytes, macrophages, and eosinophils, which are all present in increased numbers at 32 h after allergen exposure (19, 32). The CD4+ T-cell recipients that underwent OA challenge differed from their BSA-challenged controls in the BAL fluid cell findings only with respect to eosinophil numbers that were elevated. These data suggest that the eosinophil is potentially responsible for hyperresponsiveness in these CD4+ T-cell-mediated responses. The hyperresponsiveness is unlikely to involve IgE because in previous experiments no evidence of IgE could be found in recipient animals, and few if any B cells were present among the transferred cells (22). A recent study using IgE-deficient mice has convincingly demonstrated the propensity of sensitization and challenge with allergen to induce hyperresponsiveness to challenge with MCh without any participation of IgE (36).

Although the principal difference between CD4+ and CD8+ recipients in BAL fluid cells after OA challenge was in eosinophil numbers, and this difference is consistent with the changes in responsiveness observed, discrepancies between eosinophil numbers and responsiveness have been previously observed. For example, dexamethasone in doses sufficient to markedly reduce eosinophil numbers in the lungs of allergen-exposed BN rats did not attenuate allergen-induced hyperresponsiveness (19). Similar findings were obtained with cyclosporin (30). Neutrophil numbers were suppressed in rats that received CD8+ cells and underwent OA challenge compared with BSA-challenged control rats, which suggests T cell regulation of these cells. However, these findings do not appear to have any relationship to airway responsiveness to MCh.

Cytokine expression as determined by in situ hybridization revealed a Th2 pattern similar to that which had been previously noted at 8 hours after challenge of the BN rat with OA (7, 23). The number of IL-5 mRNA-positive cells was highest in the BAL fluid of CD4+ T-cell recipients after OA challenge. The number of cells expressing IL-4 was also significantly different compared with other treatment groups, with the exception of the CD4+ recipients that were BSA-challenged. The lack of difference between OA and BSA-challenged CD4+ recipients in IL-4 positivity indicates that IL-4 alone cannot account for the striking differences in IL-5 observed between the same treatment groups. IL-2 did not show any differences between the various groups. IFN- was elevated in the CD8+ recipients compared with several of the other treatment groups. Although we anticipated a reciprocal relationship between IL-5 and IFN- through an interaction of IFN- with Th2 cytokine producing cells (37), there was no clear association seen. When one considers only the CD4+ and CD8+ groups there was a concordance between IL-5 expression, eosinophilia, and the development of airway hyperresponsiveness to inhalational challenge with MCh. The group showing high IL-5 expression had eosinophilia and hyperresponsiveness, whereas no group with low levels of IL-5 expression and eosinophilia exhibited no increase in responsiveness after OA challenge. This finding is consistent with previous studies in guinea pig and monkey in which neutralization of IL-5 using specific antibodies prevented allergen-induced airway eosinophilia and hyperresponsiveness (40).

The results of allergen challenge of the total T-cell recipients raise several interesting questions. The OA-challenged total T-cell recipients did not develop hyperresponsiveness despite intermediate levels of BAL eosinophilia. Furthermore, differences in BAL eosinophilia between OA- and BSA-challenged animals were not associated with any significant differences in IL-5 expression. This finding suggests that other factors released by OA challenge contribute to the development of eosinophilia. Perhaps differences in eotaxin or other chemotactic factor may be present.

Findings from the current experiments suggest that the inhibitory effects of CD8+ T-cell transfers on both eosinophilia and hyperresponsiveness after OA challenge may be mediated by IFN-. This cytokine has the potential to induce a down-regulation of Th2 cytokines and in so doing could reduce eosinophilia (37). Recent evidence obtained on a mouse deficient in the IFN- receptor demonstrated that the magnitude of the acute eosinophilia after OA challenge was not different from wild-type controls, although the duration of the eosinophilia was markedly prolonged (43). Transfer of CD8+ T cells from the spleen of OA-sensitized mice can inhibit IgE production by cultured spleen mononuclear cells as well as reducing serum IgE levels in sensitized intact animals (44). The CD8+ cells also abolish the enhancement of sensitivity of isolated tracheal preparations to electrical field stimulation induced by allergen exposure (44).

An additional and somewhat unexpected finding in the present study was the observation of similarity of the distributions of IL-4 and IFN- mRNA-positive cells in the different treatment groups without a strong effect of the allergen challenge. This suggests that the transferred cells may have affected cytokine expression in the airway cells prior to allergen challenge. This raises the question of the potential for allergen-primed T cells to modulate airway responses to inhaled allergens in a more nonspecific fashion than previously suspected.

In conclusion, the administration of CD4+ T cells harvested from OA-sensitized BN rats confers sensitivity to OA challenge on the naive recipients that develop eosinophilia and airway hyperresponsiveness to inhaled MCh. There is a Th2 pattern of cytokine expression by in situ hybridization in BAL cells of the OA-challenged animals that may be responsible for inducing the observed changes. CD8+ T cells appear to have the capacity to neutralize the effects of the CD4+ cells, presumably through secretion of IFN-. The exact mechanism of action of the T-cell subsets and their cytokines in modulating allergic airway responses will require further experimentation.