INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Toluene diisocyanate (TDI) is a well-known cause of occupational asthma. TDI-induced occupational asthma is characterized by specific airway hyperresponsiveness to TDI as well as an increase in nonspecific hyperresponsiveness (1, 2). A second important feature of TDI-induced occupational asthma is inflammation of the airways identified by an influx of lymphocytes, eosinophils, and neutrophils (3). However, the pathogenesis of TDI-induced asthma is diverse. In general, in only 20% of the subjects with TDI-induced occupational asthma an increased level of serum TDI-specific IgE antibodies can be detected (6). For the remaining 80% of the subjects an IgE- independent mechanism has been proposed. Most studies describing human cases of TDI-induced occupational asthma demonstrate that TDI is capable of inducing different types of immune reactions (7).

Previously, we have described an IgE-independent, murine model for TDI-induced occupational asthma (8). Short-term TDI exposure (skin sensitization twice daily on two consecutive days followed by intranasal challenge on Day 8 with 1% TDI) did not lead to the production of IgE antibodies, whereas a significant, nonspecific tracheal hyperreactivity was found 24 h after the challenge in TDI-sensitized and challenged mice when compared with nonsensitized mice. Adoptive transfer studies in which lymphoid cells from TDI-sensitized mice were transferred into naive recipients suggested an important role for T lymphocytes in the induction of this tracheal hyperreactivity. The characteristics of our model for TDI-induced occupational asthma resembled several findings found in murine models for pulmonary delayed-type hypersensitivity (DTH) reactions developed earlier in our laboratories (9, 10). We therefore hypothesized that the induction of a DTH-like reaction could be an important mechanism of action of TDI.

In literature some guinea pig models have been described to investigate the IgE-mediated mechanism of TDI-induced occupational asthma (11). The mouse, however, is also a suitable species to investigate IgE-dependent airway responses. In our laboratory Hessel and colleagues (15) developed a murine model to investigate human allergic asthma. Human allergic asthma is characterized by airway hyperresponsiveness to bronchoconstrictive mediators (16), the presence of IgE antibodies (17), and an influx of inflammatory cells, mainly eosinophils (18). In the mouse, sensitization with ovalbumin (seven intraperitoneal injections on alternate days) induced high ovalbumin-specific IgE levels in the serum. Moreover, repeated inhalation challenge with ovalbumin induced a significant in vivo airway hyperresponsiveness, which is also described by other investigators, and an influx of eosinophils in the lung (19, 20). In the present study we have extended the sensitization regimes to induce the production of TDI-specific IgE antibodies. Furthermore, we investigated whether the presence of TDI-specific IgE antibodies had an influence on the airway reactivity, using in vitro (21) and in vivo techniques (22).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Mice (male BALB/c 6 to 8 wk of age) were supplied either by the Central Animal Laboratory, Utrecht, The Netherlands or by the National Institute of Public Health and the Environment, Bilthoven, The Netherlands. They were housed in groups not exceeding 6 per cage and maintained under standard conditions. All experiments were assessed by the animal ethics committee at Utrecht University and the National Institute of Public Health and the Environment.

Sensitization Procedures

Short (2-d) exposure.  Mice were sensitized twice daily on Day 0 and Day 1 either with 1% TDI (sensitized group) dissolved in acetone:olive oil (4:1) or with vehicle control (nonsensitized group) which was applied epicutaneously to the shaved abdomen and thorax (100 µl) and four paws (100 µl in total).

Long (6-wk) exposure.  Mice were sensitized once a day on Days 0, 7, 14, 21, 28, and 35 (6 consecutive weeks) either with 1% TDI (sensitized group) dissolved in acetone:olive oil (4:1) or with vehicle control (nonsensitized group) which was applied epicutaneously to the shaved abdomen and thorax (100 µl) and four paws (100 µl in total). During all sensitization procedures the mice were anesthetized with sodium pentobarbitone (50 µl; 30 mg/kg, intraperitoneally).

Challenge Procedure

Sensitized and nonsensitized groups were challenged intranasally with 1% TDI dissolved in ethyl acetate:olive oil (1:4) either on Day 8 (short-exposure protocol) or on Day 42 (long-exposure protocol). Twenty µl of the TDI solution was applied intranasally under light anesthesia (sodium pentobarbitone; 50 µl; 30 mg/kg, intraperitoneally). Furthermore, mice were also challenged on the ears on Day 8 (short-exposure protocol) or on Day 42 (long-exposure protocol); TDI (20 µl; 0.5%; dissolved in acetone) or vehicle control (20 µl) was applied topically to both sides of the ears.

Assessment of Antibody Production

Preparation of TDI-MSA conjugates.  Twenty milligrams of mouse serum albumin (MSA) was added to 4 ml 0.05 M sodium phosphate buffer, pH 7.4. To this solution, 38 µl TDI was added. The mixture was stirred at 37° C for 4 h. The reaction was stopped by addition of 5 µl monoethanolamine. After 1 h, the solution was filtered and dialyzed (23). This procedure resulted in the conjugation of 20 to 40 moles of TDI per mol of MSA (2, 11, 12, 23, 24).

Enzyme-linked immunosorbent assay (ELISA). Polystyrene microtiter plates were coated with 50 µg/ml TDI-MSA for the TDI-specific IgE/IgG measurements or with 2.5 µg/ml rat anti-mouse IgE/IgG for the measurement of total IgE/IgG in 0.1 M carbonate buffer, pH 9.6 (50 µl per well). The coated plates were incubated in a humid chamber overnight at 4° C. Plates were washed five times with phosphate-buffered saline (PBS)/Tween 20 (0.05% vol/vol) and incubated with 1% bovine serum albumin (BSA) in PBS/Tween 20 (0.05% vol/ vol; 50 µl per well) for 30 min at room temperature. After washing the plates five times, 50 µl diluted serum was added and incubated for 2.5 h at room temperature. The plates were again washed five times and 50 µl of a 1:500 dilution of peroxidase-labeled goat anti-mouse IgE/IgG was added. After a 2.5 h incubation at room temperature, the plates were washed five times and 50 µl o-phenylenediamine dihydrochloride (OPD) was added. After a further incubation period of 15 min at room temperature, the reaction was stopped by adding 25 µl per well 4 M H₂SO₄ and absorbance values were read at 492 nm using a microplate reader. Values are given as optical density by a dilution of 200, 100, 1 × 10⁷, 3,200 for the total IgE, TDI-specific IgE, total IgG, and TDI-specific IgE, respectively. Sera samples were tested after the short-exposure protocol on Day 9 (skin sensitization on 2 consecutive days and intranasal challenge on Day 8), after the long-exposure protocol on Day 42 (skin sensitization on 6 consecutive weeks and intranasal challenge on Week 7), and during the long-exposure protocol after the fourth week of sensitization.

Measurement of In Vivo Airway Reactivity

In vivo airway reactivity was measured using an air-overflow pressure method described by Hessel and coworkers (22). Mice were anesthetized with urethane (2 g/kg, intraperitoneally) and placed on a heated blanket (30° C). The trachea was cannulated and a small polyethylene catheter was placed in the jugular vein for intravenous administrations. The spontaneous breathing was suppressed by tubocurarine chloride (3.3 mg/kg, intravenously). Immediately hereafter the tracheal cannula was attached to a Fleisch flow head (Godart, Utrecht, The Netherlands), which was connected to a Gould pneumotachograph (Godart), which in turn was coupled to a respiration pump (Sanders Brinie, Enschede, The Netherlands). The inflation volume of the pump was 0.8 ml/beat with 190 beats/min. A pressure transducer (Validyne, Northridge, CA) was located between the flow head and the respiration pump in order to measure changes in the bronchial resistance to inflation. Pressure and flow signal were recorded breath-by-breath on a Gould Bruch 2400 recorder (Godart). To assess reactivity, dose-response curves to serotonin (5-hydroxytryptamine [5HT]; 20- 1,280 µg/kg, intravenously) were determined 3, 24, and 48 h after the challenge with TDI following both the long- and short-exposure protocol. Determination of dose-response curves to carbachol was not possible because of high mucus production leading to the death of the animals before the curves could be completed.

Isometric Measurement of In Vitro Tracheal Reactivity

Tracheal reactivity was measured using the method of Garssen and coworkers (21). Mice were killed by an overdose of sodium pentobarbitone (0.3 ml; 60 mg/kg, intraperitoneally). The trachea, which was resected in toto, was carefully cleaned of connective tissue using a binocular microscope. A 9-ring piece of the trachea (taken from just below the larynx) was then transferred to a 10-ml organ bath containing a modified oxygenated Krebs solution (118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl₂ · 6 H₂O, 0.5 mM MgCl₂ · 6 H₂O, 25.0 mM NaHCO₃, 1.0 mM NaHPO₄ · H₂O, 11.1 mM glucose). The trachea was directly slipped onto two supports of an organ bath, one of which was coupled to the organ bath and the other to an isometric transducer. The solution was aerated (95%:5%; O₂:CO₂) at a constant temperature (37° C). Isometric measurements were made using a force displacement transducer (Harvard Bioscience, Boston, MA) and a 2-channel recorder (Servogar type SE-120; Plato BV, Diemen, The Netherlands) and were expressed as changes in milligram force. Optimal preload for the mouse trachea was determined to be 1,000 mg. The trachea was allowed to equilibrate for at least 1 h before drug effects were elicited. During the equilibrium phase the fluid in the bath was changed every 15 min. To assess reactivity, concentration-response curves to carbachol (10⁸ to 10⁴ M) and 5HT (10⁸ to 10⁴ M) were determined 2 to 3, 24, and 48 h after challenge in both the long- and short-exposure protocol.

Histopathology

At 0.5, 2, and 24 h after and before the intranasal challenge on Day 42 lungs and trachea were removed from TDI-sensitized and nonsensitized mice after lethal anesthesia with 50 µl of a cocktail consisting of 7 ml of 50 mg/ml ketamine hydrochloride (Ketalar), 3 ml 2% xylazine hydrochloride (Rompun), and 1 ml of 1 mg/ml atropine, injected intramuscularly (pentobarbital sodium [Nembutal] causes vasodilatation which negatively influences the histology preparations). Before removing the lungs, the mice were perfused with 5 ml PBS (37° C) via the right heart ventricle. After intratracheal fixation of the lungs with a mixture of 0.8% formalin and 4% acetic acid, the trachea was ligated. The inflated lungs and trachea were fixed for at least 24 h in the fixative, dehydrated, and embedded in paraplast (Monoject, Kildare, Ireland). Four-µm-thick sections were stained with hematoxylin and eosin.

Measurement of Cutaneous Reactions

An increase in ear thickness was measured 2, 24, and 48 h after topical challenge with 0.5% TDI in acetone after the short- and long-exposure protocol. Immediately after an intraperitoneal overdose of sodium pentobarbitone, the thickness of the TDI-treated ear and the vehicle-treated ear were measured using an engineer's micrometer (No. 293-561, Mitutoyo, Japan) (25). Results are expressed as the difference in ear thickness ( ear thickness, mm) between the two ears.

Chemicals

The following chemicals were purchased: TDI, olive oil, MSA, carbachol, 5HT, and OPD (Sigma Chemical Co., St. Louis, MO); rat anti-mouse IgE, rat anti-mouse IgG, rat anti-mouse IgE-peroxidase-labeled, and rat anti-mouse IgG-peroxidase-labeled (Monosan, Uden, The Netherlands); Tween 20 (Janssen Pharmaceutical, Beerse, Belgium); monoethanolamine (Merck, Amsterdam, The Netherlands); and sodium pentobarbitone (Sanofi, Maassluis, The Netherlands).

Statistics

All experiments were designed as completely randomized multifactorials with 4 to 14 mice per group. Effective concentration that produced 50% of the response (EC₅₀) and maximal contractile response (Emax) values for the carbachol-induced tracheal contractions of each experimental animal were calculated separately by nonlinear least-squares regression analysis (simplex minimalization) of the measured contractions versus carbachol concentration using the sigmoid concentration-response relationship and including a threshold value. The data were analyzed by two-way analysis of variance (ANOVA) followed by a post hoc comparison between groups. In the figures and tables group means ± SEM are given and a difference was considered significant when p < 0.05. All data manipulation, nonlinear fittings, ANOVA, and post hoc comparisons were carried out with a commercially available statistical package (SYSTAT, version 5.03; SYSTAT, Inc., Evanston, IL).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Antibody Responses

Antibody concentrations were measured in TDI-sensitized mice and nonsensitized mice following the short-exposure protocol (sensitization twice daily on Day 0 and Day 1), the long-exposure protocol (sensitization on 6 consecutive weeks followed by intranasal challenge on Week 7), and after the fourth week of sensitization. In the sera of TDI-sensitized and challenged mice after the short-exposure protocol no increases in total IgE, TDI-specific IgE, and total IgG were found, whereas an increase in TDI-specific IgG was detected when compared with the nonsensitized and challenged mice (Figures 1A-1D). However, sensitization on 6 consecutive weeks and challenge on Week 7 caused an increase in all antibodies measured. Figures 1A and 1C show that both the total IgE and the total IgG levels were significantly increased in the sera of TDI-sensitized mice when compared with nonsensitized mice following the long-exposure protocol (p < 0.01 and p < 0.05, respectively). More importantly, TDI-specific IgE and TDI-specific IgG were also significantly elevated in sera of TDI-sensitized mice when compared with the control mice (p < 0.01, Figures 1B and 1D). The formation of antibodies with time was followed by measuring the levels after sensitization for 4 consecutive weeks. After the fourth week of TDI sensitization the total IgE level had already increased to the same level as after 6 wk of sensitization and challenge on Week 7 (Figure 1A). In contrast, the TDI-specific IgE levels had not reached the maximal level after the fourth week of sensitization when compared with the increase found after the long-exposure protocol (Figure 1B).

View larger version (34K): [in this window] [in a new window]   Figure 1.   Antibody levels in sera of TDI-sensitized and nonsensitized mice. (A) Total IgE, (B) TDI-specific IgE, (C ) total IgG, and (D) TDI-specific IgG were determined by spectrophotometry in sera of mice after the short-exposure protocol (sensitization on Day 0 and Day 1 followed by challenge on Day 8 [1 wk; nonsensitized: hatched bars; TDI-sensitized: crossed bars]), and in TDI-sensitized (closed bars) and nonsensitized (open bars) mice after the long- exposure protocol (sensitization on 6 consecutive weeks followed by challenge on Week 7 [7 wk]) and after fourth week of sensitization (4 wk). The dotted lines represent the respective antibody levels in the sera of nontreated mice. Results are expressed as mean ± SEM for n = 4 mice/group. Significant differences between TDI-sensitized and nonsensitized mice (p < 0.05 and p < 0.01) are denoted by asterisk and double asterisk, respectively.

In Vivo Airway Reactivity

In vivo airway reactivity to 5HT was measured in TDI-sensitized and nonsensitized mice both after the short- and long- exposure protocol. In Figure 2 the in vivo airway reactivity to 5HT is depicted for TDI-sensitized mice and nonsensitized mice 3 and 24 h after the challenge following the long-exposure protocol. Three hours after the challenge on Day 42 the TDI-sensitized group was clearly more sensitive to 5HT when compared with the nonsensitized group (Figure 2A). The increase in airway hyperresponsiveness was significant at the doses of 20 to 480 µg/kg 5HT (p < 0.05). This hyperresponsiveness could not be found when mice were skin-sensitized on 6 consecutive weeks with TDI but not challenged on Day 42 (data not shown). Figure 2B indicates that at 24 h after the challenge this in vivo airway hyperresponsiveness to 5HT of TDI-sensitized mice following the long-exposure protocol had disappeared. In vivo airway reactivity was also measured after TDI sensitization following the short-exposure protocol. In contrast to the long-exposure protocol, no difference in responsiveness to 5HT between TDI-sensitized and the nonsensitized mice could be found 3 and 24 h after the challenge on Day 8 (Figure 3). When the airway reactivity was tested 48 h after the challenge there was also no difference between TDI and nonsensitized mice both after the long- and the short- exposure protocol (data not shown). In Table 1 the respective ED₅₀ values of TDI-sensitized and nonsensitized mice are listed. A significant leftward shift in ED₅₀ values could only be found 3 h after the challenge in TDI-sensitized mice when compared with the nonsensitized mice following the long- exposure protocol.

View larger version (28K): [in this window] [in a new window]   Figure 2.   In vivo airway reactivity (A) 3 h and (B) 24 h after the challenge. Dose-response curves to 5HT were measured in TDI-sensitized (closed bars) and nonsensitized (open bars) mice 3 and 24 h after the challenge on Day 42 (according to the long-exposure protocol). Results are expressed as mean ± SEM for n = 6-9 mice/ group. Significant differences between TDI-sensitized and nonsensitized mice (p < 0.05 or p < 0.01) are denoted by asterisk or double asterisk, respectively.

View larger version (30K): [in this window] [in a new window]   Figure 3.   In vivo airway reactivity (A) 3 h and (B) 24 h after the challenge. Dose-response curves to 5HT were measured in TDI-sensitized (closed bars) and nonsensitized (open bars) mice 3 and 24 h after the challenge on Day 8 (according to the short-exposure protocol). Results are expressed as mean ± SEM for n = 6 mice/group.

In Vitro Tracheal Reactivity

To investigate the correlation between in vitro and in vivo airway reactivity, tracheal reactivity in vitro was also measured. Figure 4A shows the tracheal reactivity to carbachol 3 h after the challenge on Day 42 (long-exposure protocol). The TDI-sensitized mice were hyperreactive to carbachol when compared with the nonsensitized mice (Emax: nonsensitized 2,247 ± 146 mg; TDI-sensitized 3,019 ± 154 mg, mean ± SEM for n = 6 mice/group, p < 0.05). Twenty-four and 48 h after the challenge no difference in tracheal reactivity to carbachol between the TDI-sensitized and nonsensitized mice was found (Emax, 24 h after the challenge: nonsensitized 2,100 ± 301 mg; TDI-sensitized 2,347 ± 151 mg; Emax, 48 h after the challenge: nonsensitized 2,650 ± 124; TDI-sensitized 2,765 ± 159 mg, mean ± SEM for n = 5-14 mice/group). In contrast to the concentration-response curves to carbachol, reproducible concentration-response curves to 5HT in vitro were difficult to obtain in nontreated and treated BALB/c mice. However, in each experiment after long-term exposure to TDI the TDI-sensitized mice were more reactive to 5HT 3 h after the challenge when compared with the nonsensitized mice (Emax experiment 1: TDI-sensitized 2,103 ± 426 mg; nonsensitized 660 ± 230 mg; experiment 2: TDI-sensitized 1,486 ± 375 mg; nonsensitized 934 ± 193 mg; experiment 3: TDI-sensitized 2,389 ± 188 mg; nonsensitized 1,921 ± 235 mg; experiment 4: TDI-sensitized 2,599 ± 224 mg; nonsensitized 2,152 ± 299 mg, mean ± SEM for n = 4-6 mice/group in each experiment). Due to the degree of variation following this procedure, the mean values of EC₅₀ or Emax of the TDI-sensitized mice were not significantly different from the EC₅₀ or Emax values of the nonsensitized mice. In comparison, a significant tracheal hyperreactivity to carbachol (Figure 4B) and 5HT (Emax: nonsensitized 1,837 ± 118 mg; TDI-sensitized 2,430 ± 197 mg, mean ± SEM for n = 12 mice/group, p < 0.05) was found in TDI-sensitized mice 24 h after the challenge after sensitization according to the short-exposure protocol. At 2 and 48 h there was no significant difference in tracheal reactivity between the TDI-sensitized and nonsensitized mice as previously described (8).

View larger version (16K): [in this window] [in a new window]   Figure 4.   Tracheal reactivity to carbachol (A) 2 to 3 h after the challenge following the long-exposure protocol and (B) 24 h after the challenge following the short-exposure protocol. Concentration-response curves to carbachol were measured in the tracheae of TDI-sensitized (closed circles) and nonsensitized (open circles) mice. Results are expressed as mean ± SEM for n = 6-12 mice/ group. Significant differences between TDI-sensitized and nonsensitized mice (p < 0.05) are denoted by asterisk.

Histopathology

Histopathology of the lungs and trachea was performed to investigate the effect of TDI sensitization and challenge according to the long-exposure protocol (Table 2). Skin sensitization on 6 consecutive weeks with 1% TDI did not result in any histological changes. Only in 2 of 6 mice bronchial mucus was present in the TDI-sensitized and nonsensitized group 0.5 h after the challenge. However, the intranasal challenge with 1% TDI caused slight to moderate hypertrophy of bronchial and bronchiolar epithelium in the large airways in half of the number of nonsensitized mice and slight to marked hypertrophy in all TDI-sensitized mice within 2 h after the challenge. Twenty-four hours after challenge a moderate to strong tracheitis/bronchitis was accompanied by necrosis of the epithelium resulting finally in an extensive desquamation of the epithelium. However, the observed histological changes were present in both TDI-sensitized and nonsensitized mice, which indicates that they are caused by the irritating effect of the intranasal challenge rather than by TDI sensitization.

Cutaneous Responses

In addition to measuring pulmonary responses induced by long exposure to TDI, the development of cutaneous immune responsiveness was followed. Figure 5A shows the increase in ear thickness 2 to 3, 24, and 48 h after topical challenge with 0.5% TDI. At all time points measured, the increase in ear thickness in the TDI-sensitized group after the long-exposure protocol was significantly higher when compared with the nonsensitized group (p < 0.05). In addition, the cutaneous responses of TDI-sensitized and nonsensitized mice following the short-exposure protocol were examined. In comparison to the long-exposure protocol it is striking that 2 to 3 h after the challenge no increase in ear swelling could be measured in TDI-sensitized mice after the short-exposure protocol when compared with the nonsensitized mice (Figure 5B). However, 24 h after the challenge a significant increase in ear swelling in TDI-sensitized mice when compared with the nonsensitized mice was found. This increase was largely resolved 48 h after the challenge (Figure 5B).

View larger version (19K): [in this window] [in a new window]   Figure 5.   Ear swelling response 2 to 3, 24, and 48 h after the challenge in TDI-sensitized and nonsensitized mice following (A) the long-exposure protocol and (B) the short-exposure protocol. TDI-sensitized (closed bars) and nonsensitized (open bars) mice were topically challenged on Day 8 (short-exposure protocol) or on Day 42 (long-exposure protocol) on both ears; one ear with 0.5% TDI and the other with vehicle. The ear swelling was measured 2 to 3, 24, and 48 h after the challenge using a micrometer and the difference in ear thickness (× 102 mm) between the two ears is expressed as mean ± SEM for n = 6 mice/group. Significant differences between bars (p < 0.01) are denoted by double asterisk.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have previously developed a murine model to investigate TDI-induced occupational asthma (8). Skin sensitization on Day 0 and Day 1 with 1% TDI and intranasal challenge on Day 8 (1% TDI) led to in vitro tracheal hyperreactivity to carbachol and 5HT 24 h after the challenge. This model was not associated with an increase in TDI-specific IgE antibodies and can therefore be used to investigate IgE-independent TDI- induced occupational asthma which accounts for as much as 80% of the reported patients. We hypothesized that skin sensitization with TDI according to the short-exposure model resulted in a DTH-like reaction (8). DTH reactions are characterized by T helper cell type 1 (Th1) responses which are not accompanied by the production of IgE antibodies. Accordingly, in the present study we have indeed found no increase in IgE antibodies but an increase in TDI-specific IgG was found instead after TDI application following the short-exposure protocol. In the remaining 20% of the subjects with TDI- induced occupational asthma IgE and IgG antibody production does seem to play an important role in the development of occupational asthma. In the present study we developed a new sensitization protocol for TDI which resulted in TDI-specific IgE and TDI-specific IgG antibody production. Recently, Kitagaki and coworkers (26) demonstrated that repeated application of contact sensitizing agents to the same skin site resulted in a shift in the time course of antigen-specific hypersensitivity responses from a typical DTH to an immediate type reaction followed by a late reaction 24 h after the challenge. Interestingly, in our study repeated application of TDI for 6 consecutive weeks did lead to an increase in total and TDI-specific IgE antibodies which, in general, are associated with an immediate type response. In addition, both total and TDI-specific IgG were increased.

Varying the exposure regime to TDI not only resulted in different antibody responses but the development of airway responses was also differentially affected. TDI sensitization according to the short-exposure protocol resulted in in vitro tracheal hyperreactivity 24 h after the challenge, whereas no in vivo airway changes were observed. The reason for the discrepancy between in vitro and in vivo airway reactivity with regard to the short-exposure protocol could be that the in vitro tracheal reactivity method might be more sensitive when compared with the in vivo airway reactivity technique. Additionally, using the in vitro technique a specific isolated piece of trachea is taken and consequently reactivity of the bronchi and any other influences from the intact mouse, which are still present when in vivo airway changes are measured, could be negated. In contrast, the long-exposure protocol caused more immediate changes in airway parameters. Indeed, 3 h after the challenge on Day 42 (i.e., after the long-exposure protocol) in vitro tracheal hyperractivity to carbachol and 5HT were found in TDI-sensitized mice when compared with nonsensitized mice. Moreover, TDI sensitization following the long-exposure protocol resulted in in vivo airway hyperresponsiveness 3 h after the challenge.

The dissimilarity between the IgE-mediated (long-exposure) and IgE-independent (short-exposure) mechanism was also obvious when cutaneous responses were measured. TDI sensitization according to the long-exposure protocol resulted in a significant ear swelling response 2 to 3 h after the challenge which was not found after TDI sensitization according to the short-exposure protocol. Taken together, TDI sensitization and challenge according to the long-exposure protocol resulted in the production of TDI-specific IgE antibodies accompanied by more immediate reactions in the airways and skin when compared with sensitization according to the short-exposure protocol. Although so far no relationship between IgE or IgG antibodies to TDI and the respiratory response in asthmatics sensitized to isocyanates has been reported, it was suggested that the early-onset responses reflected an IgE- mediated response (27). The study of Karol and coworkers, in addition to the data presented here, demonstrated that TDI is capable of inducing different immunological reactions. We hypothesize that the short-exposure protocol resembles the IgE-independent mechanism of action of TDI and that the long-exposure protocol is a viable model to investigate IgE-mediated, TDI-induced occupational asthma. In future studies, the strict IgE dependency of the airway hyperresponsiveness and the cutaneous responses observed in this model will have to be established.

In conclusion, this study demonstrated that by altering the exposure time and/or cumulative dosage TDI is capable of inducing different immunological reactions. This important finding might be an explanation for the diversity of symptoms found in patients suffering from TDI-induced asthma. Both the short-exposure and the long-exposure model will be useful to further investigate the mechanisms of action of TDI.