The effect of exposure to household dust on the development of atopy and asthma has recently become a focus of investigation; however this avenue of study has resulted in more controversy than consensus. Epidemiological evidence shows that the presence of cockroach allergens in inner city homes is a strong risk factor for the development of atopic disease [29, 30]. Cockroach allergen sensitization was found to be responsible for exacerbations in a large proportion of asthmatic children in inner city Washington, D.C . Further, asthma exacerbations were not the result of sensitization to other allergens such as cat dander and dust mite allergens, which were also present in these homes [1, 21]. Exposure to innate immune activators such as TLR ligands, as well as environmental pollutants such as diesel particulates can also drive the adaptive immune response toward tolerance or sensitization [10, 31]. This area of study is gaining significant interest in light of the increasing incidence of atopy and allergy in the developed world.
This study sought to examine whether the presence of LPS in the household environment modifies the allergic response after sensitization to cockroach allergens has been established. It is possible that our method of LPS reduction also removed other allergens from the HDE mixture which may contribute to the allergic response. While this is a real concern, it is a shortcoming of all column purification methods. We are confident that our methods resulted in physiologically relevant data given that the LPS-reduced HDE was still able to induce a robust adaptive immune response.
Recent studies have employed house dust extracts to determine their roles in asthma severity, and have suggested that these extracts can induce allergic phenotypes or can skew towards tolerance, depending upon the dose and timing of the exposure [31, 32]. However, these studies focused on the immunomodulatory effects of the house dust extracts on mice presensitized to ovalbumin, and thus study house dust as an exacerbating factor, rather than a causative agent. In contrast, the current study directly examines the synergistic effects of CRA and LPS contained in the HDE in mice presensitized to CRA, which is the major allergen in the HDE.
LPS contamination in allergen preparations has been raised as an issue in many models of allergy, however we believe that approaching these studies using household components with naturally occurring LPS more closely and accurately reflects the environment in which one develops asthma. It is not appropriate to refer to the HDE as "contaminated" with LPS, similar to how one would not say it is contaminated with cockroach allergens. In an OVA study, Watanabe et.al. demonstrated that LPS contamination of commercially available OVA preparations significantly attenuated cellular influx, AHR and IgE production . The differences observed between these models might be explained by the intrinsic protease activity of the cockroach allergen Bla g2, which may enhance allergen sensitization by increasing epithelial permeability, and allergen exposure to dendritic cells [34, 35]. This highlights the need to study the effect of LPS in the development of asthma-like inflammation using a relevant allergen mixture.
In the lung, TNFα is primarily produced by alveolar macrophages and secreted within 1 hour of LPS challenge [36, 37]. Our results show that mice challenged with the LPS-reduced HDE produce significantly less TNFα compared to those challenged with the HDE, verifying that removal of LPS from the HDE has the expected immunological effect. Several studies have implicated TNFα and TNFα signaling in the development of adaptive immunity, particularly in the context of class switch to IgE production [8, 9]. Therefore, we would expect that we did not see differences in CRA-specific IgE, given that our study focused on the effect of LPS at challenge only. In contrast, the studies by Eisenbarth et. al. and Watanabe et.al. focused on LPS at the time of sensitization, which likely has more effect on the production of antigen-specific antibodies [8, 33]. This was also demonstrated in a rat model where aerosol exposure to LPS before or shortly after sensitization reduced the development of OVA-specific IgE . Levels of CRA-specific IgG1 and IgG2a are of significant interest, however reliable assays for these antibodies are not currently available.
The increase in total IgE and IgG1 in the HDE-challenged mice represents an increase in polyclonal, antigen non-specific IgE. LPS alone has been shown to induce expression of activation-induced cytidine deaminase, which is required for activation of downstream signals responsible for class switch recombination to both IgG1 and IgE in B cells . Other models have shown polyclonal IgE to have a role in mast cell survival and cytokine release .
Our data demonstrate that airways hyperresponsiveness was not diminished in the LPS-reduced HDE challenged groups despite significantly attenuated TNFα. We postulate that AHR is governed by other mediators and is not directly induced by TNFα. The role of IgE was explored in other studies that demonstrated that AHR may be modified independent of IgE alterations. Hamelmann et. al. showed robust AHR induction in both mast cell and B cell deficient mice and it has been shown that mast cell activation is not required for the induction of AHR[40, 41]. Further, methacholine acts directly upon muscarinic receptors in order to induce smooth muscle contraction. We have recently shown that pulmonary exposure to high concentrations of LPS can induce modest differences in muscarinic receptor expression. Therefore, it is possible that the concentrations of LPS used in this study were not sufficient to alter muscarinic receptor expression, and therefore did not alter AHR.
The role of LPS in driving allergic responses toward a Th2 phenotype has been demonstrated in several models and our data show this to be true in the context of total antibody production. However, our data show increases in both Th1 and Th2 cytokines in the LPS-reduced HDE group. One possible explanation for this is that the source of these cytokines is not only the T cells present in the lung. Other cell types such as eosinophils, epithelial cells and mast cells are contributing to the pool of cytokines present in the BAL fluid . Therefore, without assessing the direct cellular source of these cytokines, it is not possible to conclude whether LPS-induced Th2 cell differentiation is impaired in this model.
Interestingly, challenge with intact HDE results in decreased BAL cytokine production. We postulate that this is modulation of the innate immune response through induction of microbial cross-tolerance. We and others have shown that cross-tolerance to TLR ligands can be induced in the lung and in pulmonary macrophages [44, 45]. It is likely that simultaneous exposure to various TLR ligands present in the HDE can induce cross-tolerance and therefore attenuate the cytokine response. Removal of the potent TLR4 agonist LPS, may reduce the tolerogenic response, driving increased cytokine production.
Limited epidemiological data exist investigating whether removal of house dust can limit future asthma exacerbations. One such study carried out in the UK found no correlation between removal of house dust and amelioration of asthma symptoms in presensitized individuals. However, interventions to reduce dust in homes of children who were highly susceptible to developing atopy was effective in reducing asthma diagnosis [46, 47]. Taken together our data show that in a clinically relevant model of house dust induced asthma-like inflammation, the removal of only LPS at challenge does not protect against the development of the inflammatory response. Further studies will be needed to clarify whether LPS in the home environment exacerbates asthma in presensitized individuals.