Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

T lymphocytes are key regulators of the inflammatory immune response underlying the pathogenesis of asthma. T cells are increased in the bronchial mucosa and bronchoalveolar lavage (BAL) fluid (BALF) of asthmatics (1, 2), and their numbers correlate with severity of disease (3, 4). These T cells are phenotypically activated (5) and are characterized as T-helper (Th)2 CD4⁺ cells because of their production of the cytokines interleukin (IL)-4, -5, and -13 (6). L-4 and -13 induce the production of immunoglobulin (Ig)E by B cells (7), and IL-5 regulates the growth, differentiation, and activation of eosinophils (10). In allergen-sensitized mice, eosinophilic airway inflammation is dependent upon the presence of CD4+ T lymphocytes (11, 12), and depletion of CD4⁺ T lymphocytes prevents the development of airway hyperreactivity and pulmonary eosinophilia in response to an inhaled allergen (13).

Although most T lymphocytes express T-cell receptors (TCRs) composed of  and  chains (TCR cells), another class of T lymphocytes is characterized by the expression of TCRs containing  and  chains. These TCR cells comprise less than 10% of T lymphocytes in the peripheral blood and secondary lymphoid tissue of rodents and humans; however, increased percentages of TCR cells are localized in the skin, intestine, and lung (14, 15). This epitheliotropism, coupled with differences in antigen specificity and responsiveness of TCR cells compared with TCR cells, suggests that TCR lymphocytes may function as a "first line of defense" of epithelial surfaces against invading pathogens (16). Intraepithelial TCR lymphocytes may be involved in airway inflammation, in that their numbers are increased in the nasal mucosa of humans with allergic rhinitis (17) and in BALF of patients with severe asthma (18). Nevertheless, the function of TCR cells in normal immune responses to protein antigens remains to be fully established. In murine models of asthma, TCR lymphocytes have been ascribed both anti-inflammatory (19, 20) and proinflammatory roles (21). Accordingly, the present study was designed to further assess the roles of TCR and TCR lymphocytes during the acute allergic inflammatory stage of an ovalbumin (OVA)-induced murine model of asthma. We found diminished injury in animals lacking TCR cells and the complete absence of injury in animals lacking TCR cells. These results indicate that both TCR and TCR cells have proinflammatory roles in the pathogenesis of acute allergic airway inflammation in immunocompetent hosts.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mice and Sensitization Protocol

These studies employed adult male and female C57BL/6J- wild-type, C57BL/6J-Tcrb^tm1Mom (TCR^/), and C57BL/ 6J-Tcrd^tm1Mom (TCR^/) mice. Wild-type animals were obtained from Jackson Laboratories (Bar Harbor, ME); both knockout mouse lines were originally obtained from Drs. P. Mombaerts and S. Tonegawa at the Massachusetts Institute of Technology (Cambridge, MA) (22, 23) and backcrossed to C57BL/6J mice for at least 10 generations. A colony is now maintained at the University of Connecticut Health Center, housed in microisolators under specific pathogen-free conditions. All animal manipulations were approved by the Animal Care Committee at the University of Connecticut Health Center or Boehringer-Ingelheim Pharmaceuticals, Inc. (Ridgefield, CT).

Our sensitization and challenge protocol was a modification of that used by other investigators (24, 25). Briefly, challenged mice received three weekly intraperitoneal injections of 8-25 µg OVA (grade V; Sigma Chemical Co., St. Louis, MO; adsorbed to 2 mg aluminum hydroxide) followed 1 wk later by 10 consecutive d of aerosol challenge with 1% OVA in normal saline. Aerosols were generated by a Lovelace nebulizer into a 7.6-liter inhalation exposure chamber, with a chamber airflow of 10 liters/min and daily exposure times of 45 min. Mice were confined in close-fitting tubes, and only their noses were exposed to the aerosol stream. The estimated daily inhaled OVA dose was 80 µg/mouse. Two series of paired experiments were performed, one in wild-type and knockout animals, and another in wild-type animals that received intraperitoneal injections of hamster IgG (control; Jackson ImmunoResearch, Inc., West Grove, PA), hamster anti-TCR (H57.597) (26), or hamster anti-TCR (GL3) (27) (each 0.5 mg/200 µl PBS) at Days 3 and 3 of the 10-d aerosol exposures. Additional control groups included: (1) naive mice, not exposed to OVA; (2) immunized mice, which received the three intraperitoneal OVA injections but no aerosols; and (3) aerosolized mice, which received only the OVA aerosols without prior immunization. Our preliminary characterization of this model demonstrated that serum IgE, airway eosinophil levels, and methacholine hyperreactivity peak between 7 and 10 d of aerosol exposure.

BAL Analysis

At 24 h after the last inhalation, BAL was performed under terminal ketamine/xylazine anesthesia. Lungs from each animal were lavaged in situ with five 1-ml aliquots of sterile saline. Total leukocyte counts were performed with a hemacytometer using trypan blue dye exclusion as a measure of viability. The BALF cell differential was determined by analysis of cytocentrifuged slide preparations stained with Wright-Giemsa. Lymphocytes were further characterized by fluorescence flow cytometry using the following monoclonal antibodies conjugated with biotin, phycoerythrin (PE), fluorescein isothiocyanate (FITC), allophycocyanine (APC), or Cychrome: anti-CD45-FITC (clone 30-F11), anti-TCR-APC or -biotin (H57.597), anti- TCR-PE (GL3), anti-CD3-biotin (500A2), anti-CD8- Cychrome (53-6.7), or anti-B220-APC (RA3-6B2) (all from PharMingen, San Diego, CA); and anti-CD4-PE (clone GK1.5; Becton-Dickinson Collaborative Technologies, Bedford, MA). Biotin-conjugated antibodies were detected with streptavidin-PE or -Cy5 (Jackson ImmunoResearch, Inc.) or -Cychrome (PharMingen). For fluorescence flow cytometry, BALF was washed in phosphate-buffered saline (PBS) containing 0.2% bovine serum albumin (BSA) and 0.1% NaN₃. Aliquots containing 10⁴ to 10⁵ cells were incubated with 100 µl of appropriately diluted antibodies for 30 min at 4°C. After staining, the cells were washed twice with the PBS solution, and relative fluorescence intensities were determined on a four-decade log scale by flow cytometric analysis using a FACScan or FACScalibur (Becton Dickinson, San Jose, CA). BALF protein content was determined by the method of Lowry and colleagues (28), using BSA as a standard.

Lung Histology

After the mice were killed, unmanipulated lungs were removed from animals not subjected to methacholine inhalation or BAL, fixed with 10% buffered formalin, and processed in a standard manner. Tissue sections were stained with hematoxylin and eosin (29).

Serum IgE Levels

Total IgE levels in venous blood samples were measured by enzyme-linked immunosorbent assay (ELISA) (30). IgE was captured from serum (diluted 2- to 40-fold) using Immulon 2 microtiter plates (Dynatech Laboratories, Chantilly, VA) coated with antimouse IgE (Clone R35-72 at 2 µg/ ml in PBS). Detection was performed with biotinylated antimouse IgE (Clone R35-92 at 2 µg/ml; antibodies from PharMingen) and avidin-conjugated horseradish peroxidase (1:2,000 dilution; Zymed Laboratories, San Francisco, CA).

Airway Hyperreactivity

Airway responses to methacholine were assessed by two methods. In one series of experiments, measurements of pulmonary resistance (RL) were determined via standard protocol (31). Naive and OVA-challenged mice (24 h after last inhalation) were anesthetized with pentobarbital (75-mg/kg, intraperitoneal injection). The abdominal inferior vena cava was cannulated, and a tracheostomy catheter was placed. The chest was opened by a small anterior incision, and the animal was placed in a whole-body plethysmograph. Mechanical ventilation was established with a small rodent respirator (Model 683; Harvard Apparatus, Natick, MA) delivering 10 ml/kg tidal volume and 140 breaths/min; a positive end-expiratory pressure of 3 cm H₂O was provided. Values for RL were calculated by analysis of electrical signals proportional to lung volume, airflow, and transpulmonary pressure. Changes in lung volume were determined from the measured changes in plethysmographic pressure and were differentiated over time to obtain flow measurements. Transpulmonary pressure was obtained from the difference between measured pressures at the airway opening and within the plethysmograph. After the establishment of baseline lung function, the animal received sequentially increasing intravenous doses of methacholine (Sigma; 3 to 3,000 µg/ml in 1 ml/kg body-weight increments). Maximal RL responses were determined from measurements averaged over 6-s intervals. Pulmonary function was allowed to return to baseline before each subsequent dose.

In a second series of experiments, gas trapping was assessed in naive and OVA-challenged mice in response to aerosolized methacholine (0 to 300 mg/ml), administered with inhalation chamber and nebulizer as described earlier. Each mouse was killed after an 8-min exposure to a single dose of methacholine. The lungs and trachea were removed, trimmed of nonpulmonary tissue, and attached to a brass anchor. The anchored lungs were then immersed in saline and suspended from a hook at the top of a Mettler balance. Because lung-tissue weight approximates that of saline, attachment of the excised lungs yielded a negative weight display that represented the amount of gas trapped in the lungs (32). This value has been shown to correlate well with changes in dynamic compliance and total RL in small rodents (33). The data were expressed as excised lung gas volume (ELGV) in terms of ml air/kg body weight.

Statistical Analysis

Statistical comparisons between groups were made with analysis of variance (ANOVA) using StatView 4.5 (Abacus Concepts, Inc., Berkeley, CA). Dose-response data and serum IgE levels were compared by repeated-measures ANOVA. Bronchial responsiveness to methacholine was also determined from the interpolated dose associated with a 270% increase in RL (i.e., the mean effective dose that produces 270% increase in RL [ED₂₇₀] [31, 34]) and was compared by ANOVA. Paired t tests were used to assess changes in IgE concentrations before and after OVA aerosol exposure.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

No airway inflammation was detectable in BALF from any of the naive, aerosol exposed-only, or immunized-only mice. In these animals, BAL routinely revealed 4 to 10 × 10⁴ total leukocytes per mouse, of which over 98% were alveolar macrophages and < 2% were neutrophils and eosinophils. Also, OVA immunization followed by aerosol challenge with heterologous antigen (BSA, Fraction V; Sigma) did not elicit airway inflammation or eosinophilia (data not shown). Our preliminary studies demonstrated that peak airway eosinophilia required 7 to 10 days of aerosol exposure. As shown in Table 1, BAL analysis of challenged wild-type animals (immunized and aerosolized for 10 d) showed a marked increase in total leukocytes, eosinophils, lymphocytes, macrophages, and protein content, as compared with any of the nonchallenged control animals (P < 0.0001). It was notable that, in contrast to naive mice, TCR cells were recovered in BALF from every challenged wild-type mouse, where they represented 6 to 10% of airway T lymphocytes.

Airway inflammation was also seen in challenged TCR^/ mice; however, the responses were markedly reduced compared with challenged wild-type animals (Table 1). BALF total leukocytes, eosinophils, and lymphocytes were increased from levels in nonchallenged TCR^/ animals but were significantly less in challenged TCR^/ mice than in challenged wild-type animals (P < 0.0001 for each). Along with decreases in absolute cell numbers, the relative distributions of major populations of leukocytes (eosinophils, lymphocytes, and macrophages) differed in challenged TCR^/ versus challenged wild-type animals. Moreover, the presence of TCR cells affected the fraction of B cells among BAL lymphocytes. B cells accounted for 38.9 ± 1.4% of lymphocytes in challenged wild-type mice compared with 18.2 ± 2.5% in challenged TCR^/ mice (P < 0.0001). Unlike the TCR^/ mice, challenged TCR^/ mice had no BAL or histopathologic evidence of pulmonary inflammation.

Similar findings were observed in mice depleted of TCR or TCR cells by treatment with monoclonal antibodies (Figure 1). Immunized C57BL/6J mice treated with control hamster IgG before and during the OVA aerosol challenges developed marked airway eosinophilia to the same degree as did the challenged wild-type mice described earlier. These control antibody-treated mice also had increased lymphocytes recovered on BAL, although to a lesser degree than the challenged wild-type animals (12.5 ± 4.7 versus 33.7 ± 1.7 × 10⁴ cells/animal; P = 0.0044). The alveolar macrophage population was also not expanded in the control antibody-treated mice. These differences may reflect nonspecific anti-inflammatory Fc receptor-mediated events. Like the TCR^/ mice, mice treated with anti-TCR (GL3) had no detectable TCR cells in the airway or spleen, as determined by fluorescence flow cytometric analysis (data not shown). Anti-TCR-treated mice had significantly reduced airway eosinophilia compared with the control antibody-treated animals (P = 0.005), although the reduction in BALF eosinophils relative to control levels was not as great in the anti-TCR-treated mice as it was in the TCR^/ mice (22.6 ± 3.6 versus 4.8 ± 1.2 × 10⁴ cells/animal; P = 0.0016). Mice treated with anti-TCR (H57) had negligible BALF eosinophils or lymphocytes after OVA challenges, as did the TCR^/ animals.

View larger version (17K): [in this window] [in a new window]   Figure 1.   Comparison of BALF leukocytes in challenged TCR knockout mice and mice treated with antibodies directed against TCR. Solid bars represent challenged wild-type animals (CRL), TCR/ animals (TCR), or TCR/animals (TCR). Striped bars represent challenged animals receiving control hamster IgG (CRL), anti-TCR (TCR), or anti-TCR (TCR). Control antibody-treated animals demonstrated similar numbers of BALF eosinophils as did wild-type mice but fewer BALF lymphocytes (P = 0.004) and macrophages (P = 0.017). Relative to TCR/ mice, anti-TCR antibody-treated mice had increased BALF eosinophils (P = 0.002) but similar numbers of lymphocytes and macrophages. No differences in any cell types were observed between the TCR/ and the anti-TCR antibody- treated mice. Data represent means ± standard error of the mean (SEM) for eosinophils (top panel), lymphocytes (middle panel), and macrophages (bottom panel). Significant difference as noted earlier: *P < 0.05; **P < 0.01.

Qualitative histologic evaluations made on uninflated, formalin-fixed lungs (not shown) agreed with the above BAL findings. As previously reported (24), challenged wild-type mice developed typical "asthmatic" airway pathologic changes, with marked airway inflammation characterized by dense peribronchial infiltrates of lymphocytes, plasma cells, and eosinophils; moderate bronchial smooth-muscle hypertrophy; and prominent inflammation of vessels surrounding large airways. In comparison, challenged TCR^/ mice had substantially diminished peribronchial and perivascular inflammatory responses and less airway smooth-muscle hypertrophy than did the challenged wild-type animals. Challenged TCR^/ mice had no histopathologic evidence of pulmonary inflammation. Thus, histologic examinations supported the notion that airway inflammatory responses were attenuated in challenged TCR^/ and abrogated in challenged TCR^/ mice.

Changes in serum IgE levels paralleled the airway inflammation data (Figure 2). In wild-type mice, the intraperitoneal immunizations elevated serum IgE levels from 0.16 ± 0.06 to 3.38 ± 0.50 µg/ml at 2 d after the last intraperitoneal injection. A further significant increase was observed after aerosol challenge, to 4.88 ± 0.56 µg/ml (P = 0.024). These responses were profoundly reduced in TCR^/ mice (0.08 ± 0.02 µg/ml in naive, 1.53 ± 0.40 µg/ml in immunized, and 1.59 ± 0.29 µg/ml in challenged animals; P = 0.003 versus wild-type) and were minimal in TCR^/ mice ( 0.06 µg/ml in naive, 0.36 ± 0.15 µg/ml in immunized, and 0.38 ± 0.23 µg/ml in challenged animals; P = 0.0002 versus wild-type). Moreover, in contrast to the elevation of IgE following OVA aerosol exposure in challenged wild-type animals, neither the challenged TCR^/ nor TCR^/ mice demonstrated increased serum IgE when compared with levels detected before aerosol exposure.

View larger version (18K): [in this window] [in a new window]   Figure 2.   Serum IgE levels in immunized and challenged mice. Relative to levels in nonimmunized naive controls (see RESULTS), serum IgE levels were significantly increased 2 d after the third intraperitoneal OVA immunization (striped bars) in wild-type (P = 0.0014) and TCR/ (P = 0.010) mice, but not in TCR/ animals. The IgE produced in response to immunization was less in TCR/ mice than in wild-type mice (P = 0.002). Subsequent 10-d aerosol OVA challenge (solid bars) elicited a further increase in serum IgE in wild-type mice (*P = 0.024) but not in TCR/ or TCR/ mice. Data represent means ± SEM for total serum IgE measured by capture ELISA technique; n = 6-8 animals in each group.

These immune responses were accompanied by changes in lung function. Lung resistance increased in a dose- response relationship to increasing concentrations of intravenous methacholine, and the response was greater in challenged than naive animals (P = 0.045 by repeated-measures ANOVA). The maximum RL level was 42% greater in the challenged animals (8.42 ± 0.46 versus 5.94 ± 0.83 cmH₂O/ ml/s; P = 0.037; Figure 3); however, sensitivity to methacholine, as defined by the ED₂₇₀ methacholine dose, was unchanged (96.7 ± 20.9 mg/ml in challenged versus 140.2 ± 9.0 mg/ml in naive mice; P = 0.15). Gas trapping, as determined by ELGV measurements, proved to be a more sensitive assay for demonstrating changes in lung function associated with airway inflammation. Aerosolized methacholine elicited dose-dependent increases in air trapping in lungs of both naive and challenged mice, reaching significance at 30 and 100 mg/ml (Figure 4A). No difference in responsiveness was observed between naive wild-type and TCR^/ mice. The maximum degree of air trapping was not changed in either group of challenged mice; however, challenged animals demonstrated heightened sensitivity to methacholine. Relative to congenic naive controls, the challenged wild-type mice had significantly increased ELGV measurements to 3 and 10 mg/ml methacholine (P < 0.05 for each; Figure 4B). The increase in TCR^/ mice was less than that in wild-type animals, such that challenged TCR^/ animals had no increased sensitivity to 3 mg/ml methacholine and a reduced response to 10 mg/ml methacholine (P < 0.05). Thus, the decreased immune responses observed in challenged TCR^/ mice were associated with diminished physiologic perturbations, as indicated by the degree of nonspecific airway hyperreactivity to methacholine. Methacholine responses were not studied in TCR^/ mice because every animal studied failed to mount an inflammatory immune response to OVA.

View larger version (18K): [in this window] [in a new window]   Figure 3.   RL dose-response relationships to intravenous methacholine in naive and challenged wild-type mice. Methacholine elicited significant increases in RL in both naive (open circles) and challenged (filled circles) wild-type mice. The maximal responses were greater in challenged than in naive mice. Data represent means ± SEM. *Difference at P < 0.05.

View larger version (12K): [in this window] [in a new window]   Figure 4.   Air trapping in response to aerosolized methacholine in naive and aerosol OVA-challenged wild-type and TCR/ mice. (A) Methacholine elicited significant increases in ELGV in naive mice at doses of 30 and 300 mg/ml. No differences were observed between naive wild-type (circles) and naive TCR/ animals (squares). Data represent means ± SEM for weight-corrected ELGV. (B) Aerosol OVA-challenged mice demonstrated increased responsiveness to methacholine as compared with naive congenic controls. The response was greater in challenged wild-type than in TCR/ mice. Challenged wild-type animals (filled bars) had significantly increased responses to 3 and 10 mg/ml methacholine, whereas challenged TCR/ mice (open bars) had an increased response only to 10 mg/ml methacholine (P < 0.05). Data represent means ± SEM of three to five animals, expressed as the percent of the ELGV response in paired naive control animals to each methacholine concentration.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

OVA immunization plus aerosol challenge resulted in marked airway eosinophilia, increased bronchial reactivity to methacholine, and increased serum IgE levels in wild-type C57BL/6J mice, similar to observations by others (24, 25). In these challenged wild-type animals, air trapping proved to be a more sensitive method for determining bronchial hyperreactivity than did lung resistance measurements, possibly because of nonairway tissue factors affected by the intravenous administration of methacholine in the latter method. There was a 10-fold increase in methacholine sensitivity in the challenged wild-type animals, as identified by the threshold dose associated with a significant increase in air trapping from baseline. Challenged TCR^/ mice showed no evidence of airway inflammation or augmentation of serum IgE levels in response to aerosolized antigen. This finding is consistent with the fact that mice lacking TCR cells are deficient in B-cell responses to T-cell-dependent antigens such as OVA (35). Challenged TCR^/ mice had airway hyperreactivity and inflammation, but at levels at least 3- to 10-fold less than challenged wild-type animals. Qualitatively similar findings were observed in BALF differential leukocyte counts from challenged anti-TCR-treated and from anti-TCR- treated mice.

In our model of murine asthma, it appears that the presence of TCR lymphocytes augments IgE responses to both systemic immunization and aerosol challenge. Three-fold higher levels of serum IgE were present in immunized wild-type than in immunized TCR^/ mice. This observation implies that TCR cells influence either primary activation of TCR cells or Ig class switching by B lymphocytes, both of which can influence IgE production. Although repeated parasitic infection has been shown to induce class-switched Ig production in the absence of TCR cells (36), this function of TCR cells has been difficult to observe during immune responses in normal mice. The presence of TCR cells in challenged TCR^/ mice was not sufficient to compensate for the absence of TCR cells in response to intraperitoneal immunization in TCR^/ mice. Thus, TCR lymphocytes appear to have an integral role in systemic IgE production. In addition, our data demonstrate that the absence of TCR cells prevented the further increase in IgE after aerosol exposure to OVA. The 44% increase in serum IgE following pulmonary administration of OVA in wild-type mice strongly implicates TCR cells as potent mediators of the inflammatory response within the lung.

Our observation that TCR cells may have a proinflammatory role in acute allergic airway inflammation confirms and extends the recent work of Zuany-Amorim and colleagues (21). In their model, intraperitoneal OVA- immunized and intranasal OVA-challenged BALB/c mice developed lung inflammation, high levels of IL-4 and -5 in BALF, elevated serum IgE levels, and increased respiratory reactivity to methacholine. In comparison with the challenged wild-type mice, TCR-deficient BALB/c mice showed fewer pulmonary eosinophils and T lymphocytes, minimal BALF IL-5 levels, and decreased postimmunization/prechallenge serum IgE levels. The functional response did not correlate with changes in airway inflammation, inasmuch as it was similar in the challenged wild-type and TCR-deficient mice. In contrast, in our study the ELGV measurement was found to accurately reflect differences in airway inflammatory responses in challenged wild-type and TCR^/ animals. Zuany-Amorim and associates (21) attributed the reduced pulmonary inflammation to a consequence of the blunted peripheral response to OVA in their BALB/c-TCR^/ animals. Our IgE measurements extend the previous work by suggesting pulmonary as well as systemic sites of action of TCR cells.

Additional support for an intrapulmonary proinflammatory function for TCR lymphocytes is our observation that their removal with monoclonal antibodies after systemic immunization but before OVA aerosolization attenuated BAL airway eosinophilia. As noted in Table 1, significantly fewer TCR cells were recovered in BALF from challenged TCR^/ mice than from challenged wild-type animals. This observation is consistent with the decreased allergic IgE responses and airway inflammation observed in the challenged TCR^/ animals. It should be noted that, as a percentage of total lymphocytes, the relative proportion of TCR cells was greater in the challenged TCR^/ than the challenged wild-type animals (57.0 ± 3.2% versus 43.7 ± 1.5%; P = 0.0002). Nevertheless, the reduced number of BALF TCR cells in challenged TCR^/ mice raises the consideration that the reduced airway eosinophilia and other inflammatory responses demonstrated by challenged TCR^/ mice resulted from the reduction in TCR cells rather than the absence of TCR cells in these animals. This potential explanation is not supported by our studies using antibody-treated animals. BALF TCR cell counts were similar in challenged control antibody mice (3.2 ± 0.5 × 10⁴) and in challenged anti-TCR mice (2.2 ± 0.3 × 10⁴), but BALF eosinophilia was still significantly reduced in the anti-TCR animals. Moreover, BALF eosinophilia was similar in challenged wild-type animals and challenged control antibody-treated animals despite the 4.6-fold reduction in BALF TCR cells in the antibody-treated mice. Thus, it appears that it is the absence of BALF TCR cells, rather than the reduction in TCR cells, that is responsible for the diminished airway inflammatory responses in challenged TCR^/ mice. In addition, if we add together the BALF eosinophil or serum IgE levels from challenged TCR^/ and TCR^/ mice, these combined responses are lower than those actually observed in challenged wild-type mice. The more-than-additive responses in wild-type animals identifies a synergism in the proinflammatory functions of TCR and TCR lymphocytes in acute allergic pulmonary inflammation.

Our findings, coupled to those of Zuany-Amorim and coworkers (21), are opposite to the anti-inflammatory functions ascribed to TCR cells by McMenamin and coworkers (19, 20) in another inhaled OVA model. These investigators reported that repeated OVA inhalations by C57BL/ 6J mice (19) or Brown Norway rats (20) induce a state of antigen-specific, IgE isotype-specific tolerance in the animals. The adoptive transfer of interferon (IFN)--producing splenic TCR lymphocytes from OVA-tolerant animals selectively suppresses Th2-dependent IgE production to intraperitoneal OVA/alum challenge in naive recipients (19, 20). However, Seymour and colleagues have shown that the development of the chronic tolerant state can occur independent of TCR cell influence (37). The apparently conflicting findings of anti-inflammatory and proinflammatory roles for TCR cells are not truly at odds with each other but, rather, are a reflection of the kinetic development of acute and chronic airway responses to antigen. It is known that TCR lymphocytes can differentiate into either Th1 or Th2 phenotypes (38, 39). In the chronic tolerant state (Th1 conditions), the production of IFN- by anti-inflammatory Th1-like TCR cells suppresses Th2-dependent anti-OVA IgE production (19, 20). In contrast, in the acute allergic state (Th2 conditions), proinflammatory TCR augment the systemic (21) and pulmonary Th2-inflammatory responses of TCR and/or TCR cells, possibly through the production of IL-4. Collectively, these observations demonstrate the vast repertoire of TCR lymphocyte responses and strongly indicate that TCR cells can be potent regulators of the inflammatory response. It is possible that the decreased injury seen in TCR knockout and antibody-depleted animals may represent an accelerated development of inhalational tolerance in the absence of TCR cells. Although this potential mechanism cannot be excluded by the present study, its implication that TCR cells delayed or inhibited the development of inhalational tolerance would contradict the conclusions of McMenamin and associates (19, 20).

In summary, allergic airway inflammation and hyperresponsiveness are dependent upon T-lymphocyte function. TCR cells are required if allergic airway sensitization is to occur, but TCR cells are potent contributors to the pathogenesis of the acute allergic response. There are at least two possible mechanisms by which TCR cells could exacerbate airway inflammation. First, TCR cells may function during the innate immune response to OVA as promoters of Th2-type TCR cell differentiation and subsequent IgE production, possibly by presenting antigen to CD4⁺ TCR cells (40). Second, TCR cells may themselves function as Th2 cells and participate directly in allergic airway disease in the immune response to foreign protein antigens. Although we favor the first possibility to explain the role of TCR cells during the systemic immune response, we have also shown that TCR cells contribute to disease pathogenesis within the lung. TCR cells infiltrate the airways upon exposure to aerosolized antigen. Their presence profoundly influences the progression of the pulmonary eosinophilic inflammation and bronchial hyperreactivity characteristic of asthma.