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

Early-life Antibiotic Use: Microbiota Disruption During a Critical Developmental Period

Many human studies have suggested that antibiotic use is associated with the development of allergic asthma and food allergies. Accumulating evidence suggests it is exposure to antibiotics while in utero or during the first year of life that is associated with the development of allergy and asthmatic symptoms later in life.[46–54] Moreover, the length and frequency of antibiotic exposure during the first year of life positively correlate with a higher risk of allergy development.[49,52,53] By contrast, some studies have found no association between antibiotic use and allergic asthma development in children where at least one parent is atopic.[52,55] These reports support the hypothesis that only the disruption of a healthy microbiota predisposes children toward allergy development.

Asthma diagnosis in infants is challenging, and in some cases, early-life wheezing can be falsely diagnosed as a respiratory tract infection. The misdiagnosed respiratory tract infection may be treated with antibiotics, thus leading to a potential for protopathic bias in correlative studies.[56] Human epidemiological studies have been revealing, yet due to this potential for protopathic bias, and the many confounding variables, animal studies have been extremely valuable in determining the impact of early-life antibiotics on the development of allergic diseases.

Mice that are exposed to clinically relevant doses of vancomycin in utero via their mother's drinking water, and throughout life, have profound shifts in their fecal microbiota communities, which become dominated by Lactobacillaceae family members.[57] These mice exhibit high serum IgE levels, and enhanced disease severity in an ovalbumin-driven model of experimental allergic asthma, evidenced by increased total BALF cell counts, increased eosinophil numbers in BALF, worse lung pathological scores and increased airway hyperresponsiveness (AHR).[57] The timing of vancomycin exposure is critical for the development of exacerbated airway disease, as mice given vancomycin as adults do not display any differences in pathology to untreated mice, whereas mice exposed to vancomycin only during their first 3 weeks of life do develop increased airway inflammation.[58]

A lack of microbial exposure during early life can have persistent effects on immune cell frequencies later in life (Figure 2). Early-life vancomycin treatment also results in reduced proportions of Tregs in the colonic LP, a phenotype which is persistent even following vancomycin withdrawal, which may be a contributing factor to the elevated IgE levels in these mice,[58] although the mechanism behind this relationship remains to be defined.

Figure 2.

Persistent immune disruptions following absent or reduced microbial signaling during early life.

Long-term immune changes are also observed following a lack of microbial exposure during early life. The elevated iNKT cell frequencies seen in adult GF mice persist even when the mice are exposed to conventional conditions at 5 weeks of age.[28] Regions of the Cxcl16 gene are hypermethylated under GF conditions, and exposing 1-day-old mice, but not 5-week-old mice, born to GF mothers to conventional conditions is sufficient to reverse this hypermethylation, and restore iNKT cell levels to that of conventionally housed mice.[28] The mechanisms by which the absence of a microbiota results in persistent epigenetic modifications are to date unknown, but at least in this example, are MyD88-independent.[28] Enhanced Cxcl16 mRNA expression in the lung epithelium does not appear to mediate the enhanced airway inflammation seen in early-life vancomycin-treated mice,[58] suggesting perhaps that epigenetic modifications of this locus do not occur when the microbiota populations are altered, rather than absent.

It is not clear how persistent the effects of an absence of early-life bacteria are on antibody production later in life. One report suggested that conventionalizing 5- to 6-week-old GF mice by co-housing with conventionally housed mice for 3 weeks returned serum IgE levels to normal levels.[27] By contrast, in a separate report where GF mice were conventionalized by similar methods, elevated serum IgE levels were still seen compared with mice raised in conventional conditions.[31] Additionally, mice treated with vancomycin for the first 3 weeks of life still exhibited elevated IgE levels, and susceptibility to airway inflammation even after vancomycin treatment had ceased.[58]