A Case for Antibiotic Perturbation of the Microbiota Leading to Allergy Development

Lisa A Reynolds; B Brett Finlay


Expert Rev Clin Immunol. 2013;9(11):1019-1030. 

In This Article

Systemic Effects or Local Microbiota Effects?

It has been proposed that immune responses between mucosal surfaces, such as the intestines and lung, are linked, by an as-yet unclear mechanism.[65,66] A major challenge for examining the effects of antibiotic treatment is to distinguish between the systemic effects of the intestinal microbiota, versus local impacts of the respiratory microbiota. The presence of a respiratory microbiota in healthy human lungs has been demonstrated by culture-independent techniques,[67,68] although challenges in analyzing these populations include the low density and the potential for contamination from the oral microbiota. It is unclear how stable these communities are in healthy lungs, and whether a respiratory microbiota is present in laboratory mice. Because of the difficulty of analyzing respiratory microbiota populations, few studies have assessed how antibiotics given orally affect these communities, and to what extent these communities can affect allergic airway disease. Oral antibiotics could systemically affect microbial communities if they are well absorbed by the gastrointestinal tract, or if they are administered in the drinking water of mice there is the potential they could be inhaled. Ampicillin, metronidazole, neomycin and vancomycin given in drinking water to mice alter bacterial communities found in nasal washes to different extents.[38]

Although evidence at this stage does not determine causality, the abundance and composition of microbes in the upper bronchi of humans has been shown to correlate with the occurrence of asthma.[67,69] Healthy lungs are dominated with Bacteroidetes, particularly Prevotella species, in contrast to asthmatic lungs, in which two separate studies have shown Proteobacteria species to be associated with an asthmatic phenotype.[67,69] How these communities are affected by oral antibiotic treatment, and whether their local effects are a contributing factor to asthma development and exacerbations will be an important area for future focus.

There is also ample evidence, both from allergy models and infectious studies, of the intestinal microbiota having system-wide effects on immune responses at systemic mucosal surfaces, including the lung, for which several mechanisms have been proposed (Figure 3). In mice, oral antibiotic treatment results in reduced adaptive immunity to an intranasal influenza virus infection, in terms of reduced frequencies of CD4+ and CD8+ virus-specific T cells in the lung, and decreased systemic antibody responses.[38] The authors propose that reduced TLR signaling in antibiotic-treated mice lowers the transcription and translation of pro-IL-1β and pro-IL-18.[38] Thus, following viral activation of inflammasomes, there is less pro-IL-1β and pro-IL-18 available for cleavage to their active forms.[38] The reduced proinflammatory cytokine signalling may result in less DC migration to the mediastinal lymph nodes, reducing the priming of adaptive antiviral immunity.[38] It is still unclear, however, whether it is reduced TLR signalling specifically in the intestines, or in the respiratory tract that impacts lung antiviral immunity.

Figure 3.

Systemic effects of intestinal microbiota disruption. Several mechanisms have been described by which the intestinal microbiota can regulate immune responses at systemic sites. Signals from the microbiota appear to regulate bone marrow hematopoiesis,31 immunoglobulin secretion,27,31,57,58 susceptibility to infection,38,39,71 epigenetic modifications28 and metabolite production.62 Adding to the complexity, each of the listed factors will impact on each other, and on the composition of the microbiota. A multifaceted approach is required to generate a more complete understanding of the immune processes regulated by the microbiota.

In many of the models whereby the absence of microbiota signaling results in exaggerated allergic symptoms (early-life antibiotic treatment in mice,[57,58] adult antibiotic treatment in mice,[31] MyD88-deficient mice,[31] GF mice,[27,31] heightened serum IgE levels are seen. These heightened IgE levels appear to contribute to the development of basophil precursors in the bone marrow, promoting their surface expression of the IL-3R subunit CD123, thereby enhancing their proliferative capacity, resulting in elevated circulating basophil numbers.[31] Additionally, IgE can cause aggregation of the FceR1 on mast cells independently of the presence of antigen, which can promote mast cell survival, histamine content and cytokine production.[70]

The function of neutrophils extracted from the bone marrow of broad-spectrum antibiotic-treated mice is also altered compared with those from untreated mice; neutrophils from antibiotic-treated mice are impaired in their ability to kill the bacterial pathogens Streptococcus pneumoniae and Staphylococcus aureus in in vitro assays.[71] Peptidoglycan, derived from the microbiota is present systemically in the sera of untreated animals, and following antibiotic treatment, is reduced.[71] Recognition of peptidoglycan by Nod1 is required for normal bacteria killing of neutrophils, and in the absence of the microbiota, administration of Nod1 ligands is sufficient to restore neutrophil function.[71] It is as yet unknown whether it is the translocation of peptidoglycan, or other microbial products or metabolites, across the epithelial mucosa into systemic circulation that are able to inhibit IgE production in conventional mice.

The microbiota ferment dietary components to produce a variety of epigenetically active metabolites such as folate and short-chain fatty acids (SCFAs).[62] It has therefore been proposed that early-life modulation of the microbiota alters the metabolomic profile of the host, causing epigenetic modifications that can alter disease phenotype later in life.[72] Vancomycin-treated mice have a much altered metabolic profile,[73] which may be in part responsible for the increased susceptibility to allergic airway inflammation seen following its administration[57,58] through as yet uncharacterized pathways. SCFAs derived from components of the microbiota can stimulate TGF-β production by epithelial cells,[24] which may contribute to Treg differentiation at mucosal sites, and reduce inflammatory responses. SCFAs can bind to the receptor Gpr43 (widely expressed on immune cells, epithelial cells and in adipose tissue), and mice deficient in Gpr43 develop exacerbated airway inflammation following ovalbumin-driven experimental asthma induction,[74] supporting a role for SCFAs in the prevention of allergic inflammatory responses.

A comparison of histone deacetylase and histone acetyltransferase activity in the peripheral blood mononuclear cells of asthmatic and non-asthmatic children, revealed higher net histone acetylation in asthmatic subjects, as well as a correlation between cellular acetylation and AHR severity within asthmatic subjects.[75] Although the nature of this study cannot determine causality, the ability of the microbiota to mediate epigenetic modifications early in life will be an important area for future investigations. Taking these studies together, it is clear that the effects of antibiotic treatment on the intestinal microbiota are not limited to the intestine, but have system-wide effects, the mechanisms of which are beginning to be described.