Eotaxin Represents the Principal Eosinophil Chemoattractant in a Novel Murine Asthma Model Induced by House Dust Containing Cockroach Allergens¹

Asthma has become a major public health issue in urban communities. In particular, low-income and minority children have the highest morbidity and mortality rates for asthma in the United States (1). A number of studies have shown that environmental allergen exposure, particularly that of the cockroach, is the major contributing factor for asthma exacerbations among inner-city children (2, 3, 4).

Asthma represents a unique form of chronic airway inflammation characterized by reversible airway obstruction and airway hyperreactivity (AHR)³ (5). Upon exposure to the allergen, inflammatory cells including lymphocytes, macrophages, neutrophils, and eosinophils infiltrate the airway. Among these, eosinophils are predominant effector cells for tissue damage and pulmonary dysfunction (5, 6). The severity of AHR strongly correlates with the intensity of pulmonary infiltration by eosinophils (7, 8, 9).

When infiltrated into the lung after allergen challenge, eosinophils induce various inflammatory changes in the airways via release of a wide variety of immunomodulator molecules (6, 10) including major basic protein (11). The localization of eosinophils to the bronchial mucosa potentially primes the subsequent immune responses and augments allergic pulmonary inflammation by the secretion of various cytokines (5, 12). The critical role of the eosinophil in the immunopathology of asthma has been proven in clinical observations and animal models (6).

Selective recruitment of eosinophils into the airways during allergic inflammation suggests that eosinophil-specific chemoattractants are produced and released throughout the course of pulmonary inflammation. Since first identified and sequenced in a guinea pig model (13), the C-C chemokine eotaxin has been considered the major eosinophil chemoattractant in various animal models of eosinophilic pulmonary inflammation (14, 15) and in human tissues after allergen sensitization (16, 17, 18).

For the last 10 years, mouse models of asthma have been developed to study the inflammatory mechanisms of asthma (19). In an effort to develop allergic asthma-like pulmonary inflammation in healthy animals, mice were sensitized and/or challenged with various kinds of allergens. Among them, OVA (20) and purified indoor allergens such as cockroach (21) and dust mite (22) are commonly used allergens for murine models. However, the allergens used for these animal models may not represent the same constituents to which asthmatics are exposed throughout their daily life in terms of quality and quantity of allergens. To date, no murine model of asthma-like pulmonary inflammation has been developed by using allergens derived directly from house dust. Although allergens represent an important component of asthma, other initiating factors are certainly present in the environment. Endotoxin is a component from the cell wall of Gram-negative bacteria, and its role in the pathogenesis of asthma has been suggested but is not well defined.

In an effort to establish a novel murine asthma model, we collected house dust from an inner-city house in Detroit, Michigan and used a simple aqueous extract for sensitization and intratracheal challenge. We investigated AHR, pulmonary cellular infiltration, and chemokine levels in response to the house dust extract (HDE). We also examined the role of the eotaxin in this model by neutralizing eotaxin with a specific anti-chemokine Ab. This model demonstrates the critical role for eotaxin in the recruitment of eosinophils during asthma-like pulmonary inflammation.

Female BALB/c mice (18–20 g) were obtained from Harlan Sprague-Dawley (Indianapolis, IN) and maintained under standard laboratory conditions. The mice were housed in a temperature-controlled room with a 12-h dark/light cycle and allowed food and water ad libitum. The experiments described below were performed in accordance with the National Institutes of Health guidelines, and approval was obtained from the University of Michigan Animal Use Committee.

Household dust was collected from 10 different houses in Detroit, Michigan. The dust samples were collected from a 1-m² area by using an electric vacuum cleaner with a dust collector (Indoor Biotechnologies, Charlottesville, VA). Sterile PBS (2 ml) was added to the dust, which was then mixed end-over-end on an orbital rotator (Roto-Shake Genie; Scientific Industries, Bohemia, NY) at 4°C for 16 h. The resuspended dust was centrifuged for 10 min at 4°C, 1000 × g, and the supernatant was collected for allergen measurement by an ELISA.

The household dust sample used for all sensitizations and airway challenges was collected from a house that showed a high level of the cockroach allergens Blattella germanica 1 (Bla g1) and 2 (Bla g2). A total 4.3 g of dust was collected and resuspended with 30 ml of sterile PBS for extraction as described above. After extraction, the supernatants were aliquoted and stored at −20°C until use.

Assays for six different indoor allergens including two house cockroach allergens (Bla g1 and Bla g2), two house mite allergens (Dermatophagoides pteronyssinus, Der p1 and Dermatophagoides farinae, Der f1), one cat allergen (Felis demesticus, Fel d1), and one dog allergen (Canis familiaris, Can f1) were performed by ELISA. All Abs and standards for these ELISAs were purchased from Indoor Biotechnologies. For Bla g1, Bla g2, and Can f1, 96-well plates (Nunc Immunoplate Maxisorb; Nunc, Neptune, NJ) were coated with anti-Bla g1 mAb, anti-Bla g2 mAb, or anti-Can f1 mAb, respectively, and incubated overnight at room temperature. The plates were then washed using a wash buffer containing 0.05% Tween 20 (FisherBiotech, Fair Lawn, NJ) in PBS. Nonspecific binding sites were blocked by incubating the plates with Blocker Casein (Pierce, Rockford, IL) in PBS for 1 h. This and subsequent incubations were conducted at room temperature on a shaker. After washing, samples were added and the plates incubated for 1 h. Standard curves were prepared using the appropriate recombinant allergen. All standard and sample dilutions were made in dilution medium (DM; 10% casein in PBS supplemented with 0.01% normal rabbit plasma and 0.05% Tween 20 (FisherBiotech)) for nonspecific blocking. Plates were washed and rabbit polyclonal anti-cockroach antiserum (diluted in DM) for Bla g1 and Bla g2, and rabbit polyclonal anti-Can f1 antiserum for Can f1, were used as detection Abs. Plates were incubated for 1 h. After washing, HRP-conjugated goat anti-rabbit IgG (BioSource International, Camarillo, CA) was added and plates were incubated for 1 h. After a final wash, 3,3′,5,5′ tetramethylbenzidine (Genzyme Diagnostics, San Carlos, CA) was added, plates were incubated in the dark for 15 min, and the reaction was stopped with 1.5 N H₂SO₄. Plates were read using dual wavelengths (465 and 590 nm) on the Bio-Tek microplate reader (Bio-Tek Instruments, Winooski, VT), and allergen concentrations were estimated using the recombinant allergen standard curve.

For Der p1, Der f1, and Fel d1, 96-well plates were coated with anti-Der p1, anti-Der f1, or anti-Fel d1 mAb, respectively, and incubated overnight at room temperature. The plates were then washed and processed as above. Biotinylated anti-Der p1 Ab (diluted in DM), biotinylated anti-Der f1, or biotinylated anti-Fel d1 mAb was used as detection Ab followed by HRP-conjugated streptavidin (Jackson ImmunoResearch Laboratories, West Grove, PA) and 3,3′,5,5′ tetramethylbenzidine for color development. The lower limits of detection were 0.008 U/ml for Bla g1, 0.31 ng/ml for Bla g2, 3.13 ng/ml for Der p1, 0.39 ng/ml for Der f1, 0.03 mU/ml for Fel d1, and 3.13 IU/ml for Can f1.

For the development of the model, mice were sensitized by an i.p. injection of the HDE mixed with an adjuvant and then given two separate pulmonary challenges. Specifically, on day 0, mice were sensitized by i.p. injection of 50 μl of HDE emulsified in 50 μl TiterMax Gold (CytRx, Norcross, GA) for a total volume of 100 μl. TiterMax has been shown to induce a strong immune response with minimal tissue swelling (23). On days 14 and 21, mice were given an airway challenge of 50 μl of HDE while under anesthesia with methoxyflurane (Metofane; Shering-Plough, Union, NJ) (24). Briefly, an anesthetized mouse was suspended on its back on an inclined board by its teeth. The body weight was supported by taping the base of the tail to the board. While under anesthesia the jaw was opened and the tongue gently extended with forceps. This positions the epiglottis with the trachea open. An aliquot of the HDE (50 μl) was placed at the base of the oropharynx, which temporarily occludes air flow and is subsequently inhaled. The technique produces a reliable delivery of fluid droplets to the lung. The unimmunized group did not receive the i.p. injection on day 0 but received two airway challenges on days 14 and 21. For the specificity of the model, mice were also challenged with the house dust and received one airway challenge on day 7. For the second airway challenge the mice received either purified cockroach allergen, Bla g2 or purified dust mite allergen, Der p1. Forty-eight hours from last airway challenge, AHR to acetyl β-methacholine (Mch) was measured, and the mice were sacrificed for histologic examination and collection of bronchoalveolar lavage (BAL).

AHR of mice to increasing doses of aerosolized Mch (Sigma, St. Louis, MO) was measured in unrestrained and conscious animals by a whole-body plethysmography (WBP) system (Buxco, Troy, NY). After the box was calibrated, the pressure difference between the main chamber of the WBP containing the mouse and a reference chamber was measured. The difference of the box signal results from changes in volume and resultant pressure in the main chamber during each respiratory cycle of the animal. Inspiration and expiration were processed as a waveform of the box pressure-time signal and recorded by a computer data-acquisition system. Changes in early expiration during bronchoconstriction will alter the waveform of the box pressure-time and can be quantified. These quantified alterations are reflected by the main indicator of airway obstruction, enhanced pause (Penh). Penh is strongly correlated with airway resistance of the animal and is widely used in murine asthma models (25).

After mice were acclimated to the main chamber, baseline Penh was measured for 5 min. Either aerosolized PBS or Mch (Sigma) in increasing concentrations (6, 12, 25, and 50 mg/ml) was nebulized through the inlet of the main chamber for 2 min, and response to each dose was measured for 5 min. The average Penh for 5 min was used to compare the results among experimental groups. Airway responsiveness was expressed as a percent increase of Penh for each dose of Mch compared with Penh for PBS challenge.

For blood counting, 20 μl of EDTA (Sigma) anti-coagulated blood was collected from the tail as previously described (26) at 48 h after the second airway challenge. A Hemavet Mascot Multispecies Hematology System Counter 1500R (CDC Technologies, Oxford, CT) was then used to perform a complete blood count. At the time of sacrifice, blood was also collected from the retro-orbital venous plexus into tubes containing 50 U of porcine derived heparin (Elkins-Sinn, Cherry Hill, NJ), and plasma was stored at −20°C.

After blood collection, mice were euthanized by cervical dislocation. For the BAL, the trachea was exposed and then intubated with a polyethylene catheter. BAL fluid was harvested by lavaging with two separate aliquots of 1 ml of HBSS (Life Technologies, Grand Island, NY) through the trachea. The first wash was centrifuged, and the BAL supernatant was stored as above. The second wash was centrifuged, the cell pellet from the first wash was pooled with the cell pellet from the second, and a total cell count was obtained using a Coulter counter model ZF (Coulter Electronics, Hialeah, FL). Cytospin slides were prepared and stained with Diff-Quick (Baxter, Detroit, MI), and differentials were obtained after counting 300 cells.

The left lung lobe and trachea were removed, fixed in formalin, and processed for routine histology in paraffin.

Eotaxin was measured using matched Ab pairs via ELISA (R&D Systems, Minneapolis, MN) using our previously described methods (26). Endotoxin was measured with a chromogenic Limulus assay following the manufacturer’s instructions (BioWhittaker, Walkersville, MD).

Groups of normal BALB/c mice were sensitized and challenged with the HDE as described above. Twenty-four hours after the second challenge (day 22), mice were given 10 μg of rat anti-murine eotaxin mAb (R&D Systems) or 10 μg of control rat IgG (Jackson ImmunoResearch Laboratories) by intratracheal challenge. Forty-eight hours after the last airway challenge, AHR was measured and the cellular constituents in the BAL fluid were determined.

Summary statistics were expressed as mean ± SEM in all figures. Differences between all treatment groups were compared by ANOVA. A Tukey test for pair-wise comparisons was performed when the overall F value was statistically significant (p 0.05). Eotaxin levels that were not detectable were assigned a value equal to half the lower limit of detection for that assay.

Cockroach infestation and sensitization to these allergens are strongly correlated with the development of allergic respiratory disease, especially asthma (3). A major objective of the present study was to establish a novel murine model of asthma using house dust containing high levels of endogenous cockroach allergens instead of immunizing the mice with purified or recombinant allergens. In an effort to locate a house containing dust with a high level of cockroach allergen, house dust was collected from 10 houses in Detroit, Michigan. The dust samples were collected from a 1-m² area in the kitchen by using a dust collector attached to an electric vacuum cleaner. After being transported to our laboratory, each house dust sample was extracted with sterile PBS and screened for six different indoor allergens; German cockroach (B. germanica, Bla g1 and Bla g2), house dust mite (D. pteronyssinus, Der p1 and D. farinae, Der f1), cat (F. domesticus, Fel d1), and dog (C. familiarizs, Can f1). Based on this screening, we selected the house that contained the highest concentration of cockroach allergen, and a large quantity of house dust was collected extensively from the house. A total of 4.3 g of house dust was collected and extracted with 30 ml sterile PBS.

We assayed for six different indoor allergens in this extract by using an ELISA (Table I). Our HDE contained high concentrations of cockroach allergen (378 U/ml Bla g1 and 6249 ng/ml Bla g2), whereas Der p1, Der f1, Fel d1, and Can f1 allergen levels were very low. In addition to quantification of various allergens, the endotoxin level of the HDE was measured by a Limulus assay, which detected 270 pg/ml endotoxin in this extract. This aqueous HDE was used for immunization and intratracheal instillation for all experiments throughout this study. Initial dose-response studies showed significant increases in pulmonary inflammation from the mice immunized and challenged with 1/10 diluted extract as well as with undiluted extract (data not shown). This dilution (1/10) was used for all subsequent studies.

An increase in the number of eosinophils in the airways has been considered a hallmark sign of allergic asthma (5, 27). To characterize pulmonary inflammation in this mouse model, BALB/c mice were sensitized and intratracheally challenged with the HDE containing high concentrations of cockroach allergens. The number of eosinophils and neutrophils in the BAL was quantitated at 12-h intervals after the last challenge. The eosinophil counts progressively increased in the lung lavage from this model (Fig. 1). Eosinophil infiltration started within 12 h after the last intratracheal challenge and reached maximum levels at 48 h (Fig. 1). After 48 h, the number of eosinophils dropped but remained elevated compared with the 12-h time point. Based on this observation, animals were sacrificed 48 h after the last intratracheal administration of the allergen in subsequent studies. Interestingly, a significant number of polymorphonuclear cells (PMNs) in the lung lavage were observed at the early time point (12 h). Early recruitment of PMNs is probably due to the endotoxin present in the HDE. To further characterize the model, immunized and unimmunized mice were compared. Eosinophils and PMNs accumulated in the BAL were measured 48 h after the last pulmonary challenge (Fig. 2). The number of eosinophils in immunized mice was significantly higher than in the unimmunized group, indicating that prior sensitization was required for the development of the asthma-like inflammatory response. Additionally, there were significantly more lymphocytes recovered from the BAL fluid of the immunized mice (Fig. 2).

Eosinophil recruitment in the lung of house dust-challenged mice was further confirmed by histological studies (Fig. 3). Mice were either unimmunized or immunized according to the protocol. Lungs were processed for histology 48 h after the last challenge, and after BAL was performed. In the unimmunized mice, the pulmonary histology was essentially normal (Fig. 3,A). In contrast, the immunized mice had a significant influx of inflammatory cells including lymphocytes and eosinophils. These cells were located in the peribronchial space as well as in the perivascular space (Fig. 3,B). Higher power magnification demonstrated that the cells were both eosinophils and lymphocytes (Fig. 3 C). It is important to note that the histology was prepared after the lungs were lavaged, which accounts for the lack of eosinophils within the airways.

To confirm whether pulmonary inflammation was a localized reaction rather than a part of systemic inflammation, peripheral blood was collected and a complete blood count performed 48 h after the second airway challenge. There were no significant differences in circulating eosinophil numbers between normal, immunized, and unimmunized mice (Table II).

The number of eosinophils infiltrated into the lung is closely correlated with AHR in many mouse models of asthma (5, 8). Penh from WBP represents an accepted measure of AHR in unrestrained and conscious animals, and Penh changes are closely related to the pulmonary recruitment of eosinophils in asthmatic animals (25). We measured the responses to inhaled Mch in immunized and unimmunized animals (48 h after last intratracheal challenge) to investigate how pulmonary function is affected by house dust containing high concentrations of cockroach allergens. The control group (unimmunized but intratracheally challenged) showed no significant changes in Penh in response to increasing doses of aerosolized Mch (Fig. 4). In contrast, the Penh of the immunized group was substantially increased with escalating doses (25 and 50 mg/ml) of aerosolized Mch compared with control mice.

We next investigated the specificity of pulmonary inflammatory reaction in this mouse model by using purified or recombinant allergens. Mice were immunized and challenged once with the HDE. For the second challenge, the same quantity of purified cockroach allergen, Bla g2 (31 ng) found in the HDE was used. For a control, recombinant house dust mite allergen, Der p1, (31 ng) was used for the second challenge. We used an irrelevant Ag for the control rather than only using normal saline to further define that the response was specific to the cockroach allergen. The Bla g2-challenged mice showed significantly more eosinophils in lung lavage than Der p1-challenged mice (Fig. 5,A). These mice also had enhanced AHR with substantially higher Penh values (Fig. 5 B).

Selective recruitment of eosinophils into the lungs of immunized mice suggests the presence of eosinophil-specific chemoattractants in the lung during pulmonary inflammation. Eotaxin, a CC chemokine, is the hallmark eosinophil chemoattractant released in the lung in many animal models of eosinophilic airway inflammation (13, 28). We measured eotaxin levels in the lung lavage to investigate the involvement of this chemokine in eosinophil accumulation in the lung using the HDE-immunized mouse model (Fig. 6). BAL eotaxin levels in the immunized mice were significantly higher than in the unimmunized mice.

The role of eotaxin in this asthmatic pulmonary inflammation was evaluated by administration of neutralizing Ab to the immunized mice. The immunized mice were treated with either 10 μg of rat anti-mouse-eotaxin mAb or 10 μg of rat IgG. The role of eotaxin in the pulmonary inflammation was evaluated by quantitating the number of eosinophils in the lung lavage (Fig. 7). The number of eosinophils in the BAL of immunized and anti-eotaxin Ab-treated animals was substantially lower than immunized and control Ab-treated mice. This provides strong evidence that eotaxin represents a critical mediator of asthmatic pulmonary inflammation induced by HDE containing a high concentration of house cockroach allergens.

Asthma is one of the most common illnesses in industrialized countries and affects 8–10% of children and 3–5% of the adult population. In the United States, ∼15 million Americans suffer from asthma (19, 29). The morbidity and mortality due to asthma have increased for the last 30 years in the Western world; the reasons for this recent increase have not been fully defined (29). Several explanations for the increase in asthma have been proposed such as elevated levels of potential sensitizing agents, i.e., tobacco smoke and other organic chemicals, and diminishing bacterial infections that generally promote differentiation of a Th1 immune response (29). Increased levels of indoor allergens including dust mite, cockroach, and pet dander represent other possible reasons. Recent published studies have demonstrated that cockroach allergen sensitization correlates with the morbidity due to asthma because in some inner-city areas nearly 37% of asthmatic children (3) and 48% of asthmatic adults (30) are allergic to cockroach allergens. Thus, it is of great interest to develop experimental murine models that reproducibly demonstrate the major characteristics of asthma by using house dust.

Reversible airway obstruction in response to allergens, chronic airway eosinophilia, and AHR are characteristics of asthma (31). Among these, chronic eosinophilic pulmonary inflammation represents an obvious hallmark of allergic asthma (32). We have examined the asthma-like inflammatory changes in mice immunized and intratracheally challenged with house dust containing high concentrations of cockroach allergens and have investigated the role of eotaxin in the pulmonary recruitment of eosinophils in this model.

In this study, eosinophil infiltration was initiated within 12 h after the second intratracheal challenge of house dust and reached a peak at 48 h. The number of eosinophils recruited in the BAL of the immunized mice was significantly higher than that of unimmunized mice. Pulmonary eosinophilia was further confirmed by lung histology. Compared with unimmunized mice, substantial numbers of leukocytes are recruited in immunized mice and most of the infiltrated cells are either eosinophils or lymphocytes. The early rise in the number of neutrophils in BAL fluid is probably due to the endotoxin contamination in our HDE. These results were not surprising because the HDE contained high concentrations of endotoxin (270 pg/ml). We collected the house dust for these studies from the kitchen, where the cockroach allergen level is higher than other areas (33) and presumably there is a greater possibility of bacterial growth in the dust collected from that area. Previous studies have also shown that neutrophils are the first cell type recruited into the lung in response to allergen challenge (27, 34). Our results are similar to other studies regarding the kinetics of eosinophil recruitment where pulmonary eosinophilia peaked at 48 h after the last allergen challenge in cockroach allergen-sensitized mouse model (35). Pulmonary eosinophilia peaked 3 h after the last challenge in an OVA-sensitized and -challenged mouse model (15) and peaked in 24 h after the last challenge in a cockroach whole body extract-sensitized and -challenged guinea pig model (21).

Extensive studies have demonstrated that BAL fluid of patients with bronchial asthma showed increased number of eosinophils (36, 37), which correlates with AHR (5) and disease severity (38). In the present study, AHR to aerosolized Mch was induced in house dust-sensitized and intratracheally challenged mice as measured through WBP, whereas no enhanced AHR was observed in unimmunized mice. Therefore, the pulmonary recruitment of eosinophils in this mouse model paralleled the development of airflow obstruction. Thus, it is concluded that the characteristics of bronchial asthma, including eosinophilic pulmonary inflammation and AHR, were induced in mice sensitized and challenged with house dust containing a high concentration of cockroach allergens.

The role of the cockroach allergens in this model was further confirmed by Ag-specific induction of pulmonary eosinophilia and airway obstruction. In these experiments, mice from each group were treated identically except for the second intratracheal challenge, in which mice were given the same amount of either Bla g2 or Der p1 allergen. The number of eosinophils recruited in the lung and the Penh of Bla g2-treated mice were significantly higher than those of Der p1-treated mice and substantially similar to those of house dust-treated mice. Thus, it is suggested that cockroach allergens in the HDE play the major role in pulmonary eosinophilia and AHR observed in this mouse model of asthma. The allergen-specific pulmonary infiltration of eosinophils in this model is very similar to the asthmatic responses demonstrated in the aerosolized whole-body extract of cockroach-sensitized guinea pig models (21) and in a mouse model using purified cockroach allergen (35). The cockroach-sensitized guinea pigs showed a dose-response relationship between the number of eosinophils recruited in the BAL and the amount of cockroach allergen. In the mouse model, very similar kinetics of cockroach allergen-specific induction of pulmonary eosinophilia and AHR were demonstrated as shown in our model.

Eotaxin represents one of the most efficient eosinophil chemoattractants and plays a key role in allergic airway inflammation. In previous studies, the levels of eotaxin expressed correlated with pulmonary eosinophilia (5, 15, 39). In murine models, neutralization of eotaxin significantly inhibited eosinophilic inflammation and AHR following allergen challenge (40). In our study, significant amounts of eotaxin were expressed in immunized mice, and the level of eotaxin in BAL showed a dose-response relationship with the amount of allergen used for sensitization (data not shown). Furthermore, the number of eosinophils accumulated in BAL fluid was substantially reduced by neutralization of eotaxin with anti-eotaxin Ab. These results indicate that eotaxin is responsible for the pulmonary infiltration of eosinophils in response to cockroach allergen challenge.

The role of LPS (or endotoxin) in the pathogenesis of asthma has become a recent subject of investigation. It has been postulated that LPS exposure during early life will induce Th1 cytokines such as IFN-γ and IL-12. The cytokines will diminish the synthesis of Th2 cytokines including IL-4, -5, and -13 (41). Based on this hypothesis, LPS exposure during the priming phase of an allergic response would result in decreased asthma. Recent clinical results have shown that children with high levels of endotoxin in the house dust have less allergic sensitization (41). However, conflicting results are found when evaluating adults. In these studies the severity of asthma was positively correlated with the levels of endotoxin within the house dust, with worse asthma associated with higher levels of endotoxin (42). Additionally, increased numbers of eosinophils were found after low-level endotoxin challenge in humans (43). In a murine model of asthma induced by OVA, LPS exposure results in increased numbers of inflammatory cells within the lung (44). Therefore, the relationship between endotoxin exposure and the subsequent development of asthma is not clear. Another consideration in this model is the use of the aqueous extract of the house dust rather than the dust itself. It is possible that allergens collected by using a vacuum cleaner may not precisely represent the aeroallergens in the air, which are actually inhaled. The size and weight of some aeroallergens may prevent them from settling to the floor and/or from being trapped in the dust collection tube of the vacuum cleaner.

There are some important similarities and differences between this novel model of asthma and other models based on OVA or purified cockroach allergens. The similarities include the development of asthma-like pulmonary inflammation with typical histopathology, eosinophils in the airways, and AHR. Some important differences include the early recruitment of neutrophils and the fact that very low levels of allergen (31 ng) are required to trigger the pulmonary inflammation.

The data we presented here have demonstrated the characteristics of a mouse model of asthma in response to house dust that contains the environmental allergens found in the German cockroach as well as endotoxin. Cockroach allergen sensitization of mice elicits asthma-like pulmonary inflammation characterized by eosinophilia and AHR. The pulmonary eosinophil recruitment is dependent upon pulmonary expression of eotaxin. This novel murine model of asthma induced by house dust represents a valuable tool for further study and may permit the dissection of the allergen-endotoxin interactions important in the pathogenesis of asthma.

The Michigan Center for the Environment and Children’s Health (MCECH) is a community-based participatory research initiative investigating the influence of environmental factors on childhood asthma. MCECH involves collaboration among the University of Michigan Schools of Public Health and Medicine, the Detroit Health Department, the Michigan Department of Agriculture, Plant and Pest Management Division, and nine community-based organizations in Detroit (Butzel Family Center, Community Health and Social Services Center, Detroiters Working for Environmental Justice, Detroit Hispanic Development Corporation, Friends of Parkside, Kettering/Butzel Health Initiative, Latino Family Services, United Community Housing Coalition and Warren/Conner Development Coalition), and Henry Ford Health System.