Background
Asthma is a common chronic respiratory disease worldwide, typically starting in childhood and characterized by symptoms such as shortness of breath, chest tightness, wheezing, and coughing, which may vary in frequency and severity over time [
1]. It is estimated that asthma-related symptoms impact a substantial global population, with the most recent Global Burden of Disease Study (2019) reporting an asthma prevalence of nearly 262 million individuals worldwide [
2]. The condition may be particularly severe in some children with asthma, especially in low- and middle-income countries [
3]. Cluster analysis has identified distinct asthma phenotypes among patients, which are influenced by various factors such as age at asthma onset, sex, body mass index (BMI), and inflammatory profiles [
4]. The etiology of asthma is complex and likely involves a variety of genetic, environmental, infectious, and nutritional factors [
5]. Some environmental factors that may contribute to the onset and exacerbation of asthma include allergens, viral infections, tobacco smoke exposure, and air pollution [
6]. The development of asthma may also be related to individual susceptibility. Despite significant progress in understanding and managing asthma, it remains a major public health problem with substantial morbidity, mortality, and economic burden [
7].
The gut microbiota plays a critical role in regulating human health through various mechanisms such as metabolic and immune regulation [
8]. Environmental factors, including antibiotic use and birth mode, have been shown to impact the gut microbiota composition and increase the vulnerability to immune-related disease [
9,
10]. Dysbiosis of the gut microbiota, characterized by an altered microbial community composition and imbalance, has been correlated to various diseases, such as obesity, hypertension, diabetes, and cancer [
11‐
13]. Recently, several cross-sectional studies have shown the association between gut microbiota and asthma. For example, Demirci et al. found that
Akkermansia muciniphila and
Faecalibacterium prausnitzii were decreased in the asthma children group compared to the healthy children group [
14]. A Danish prospective cohort study on asthma indicates that a higher abundance of
Veillonella and lower abundance of
Roseburia,
Alistipes, and
Flavonifractor at age 1 year were associated with an increased risk of developing asthma by age 5 years [
15]. While these traditional observational studies have suggested associations between gut microbiota and asthma, these studies are limited by factors such as participant selection bias, confounders, and reverse causality. Therefore, the causal relationship between gut microbiota and asthma remains uncertain. It is imperative to clarify a causal relationship to better understand the pathogenesis of asthma and guide microbiota-oriented interventions in clinical practice.
Mendelian randomization (MR) is a statistical method that infers the causal relationship between exposures and outcomes by using genetic variations as instrumental variables (IVs) [
16]. MR integrates summary data from genome-wide association studies (GWAS), similar to natural randomized controlled trials. As the assignment of genotypes from parents to offspring is random, MR studies are less prone to confounders and reverse causality than traditional observational studies [
17]. MR has emerged as a powerful tool for identifying causal relationships between risk factors and diseases and is widely used in epidemiological research to explore the potential causal associations between two traits [
18].
Recently, MR analysis has been widely used to identify the causal associations between gut microbiota and the risk of many diseases, such as cardiovascular diseases, autoimmune diseases, and psychiatric disorders [
19‐
21]. To our knowledge, no MR study has extensively examined the causal association between gut microbiota and asthma. Therefore, in this study, we conducted the two-sample bi-directional MR analysis to examine the potential causal relationships between gut microbiota and asthma as well as its phenotypes (i.e., obesity related asthma, non-allergic asthma, allergic asthma, and eosinophilic asthma). We also used the MR method to evaluate the causal effect of gut metabolites on asthma and its phenotypes.
Discussion
In the present study, we conducted MR analyses to examine the causal associations between gut microbiota, metabolites and asthma. Our analysis utilized summary data from the MiBioGen consortium’s largest GWAS meta-analysis of gut microbiota and the FinnGen consortium’s asthma summary data. Our results provide evidence for the causal effects of specific microbiota on asthma and its phenotypes, as well as for reverse causality. Additionally, the risk of eosinophilic asthma was also potentially associated with the lower indolepropionate. To our knowledge, this study represents the first comprehensive MR analysis to explore the potential role of gut microbiota and metabolites in the development of asthma at the gene prediction level, which may contribute to strengthening the theoretical basis for the “gut-lung” axis.
Several studies have reported a potential link between asthma and dysbiosis or altered microbiota in the gut [
43‐
46]. The differences in microbial diversity and composition between healthy individuals and asthma patients indicate a potential involvement of gut microbiota in the development of asthma [
47,
48]. However, there is no clear causal relationship between gut microbiota dysbiosis and asthma risk. The use of glucocorticoids in asthma patients may cause alterations in the gut microbiome, and differences in gender ratios and ethnicities between studies may affect the composition of the gut microbiome [
49‐
51]. Furthermore, while studies have found that asthma patients tend to have a phenotype of gut microbiome dysbiosis [
52], the results regarding changes in specific strains have been inconsistent, making it difficult to infer a causal link between gut microbiota and asthma risk.
In our study, we aimed to identify specific gut microbiota that are causally associated with asthma and its phenotypes. We identified 47 potential candidates, of which three showed a significant causal relationship with asthma. Our study found that the class
Bacilli and order
Lactobacillales were associated with a lower risk of asthma. This finding aligns with existing research, such as the study by Spacova et al., which demonstrated the beneficial effects of
Lactobacillus rhamnosus, a member of
Lactobacillales, in preventing airway function deterioration in a murine asthma model [
53]. Our results contribute to the growing body of evidence on the role of specific microbiota, including
Lactobacillales, in asthma pathogenesis.
Our study revealed that the relative abundance of genus
Lachnospiraceae_UCG001 was suggestive causally associated with a higher risk of asthma, while the genetically predicted risk of asthma was positively correlated with the relative abundance of genus
Lachnospiraceae_UCG004. These findings are consistent with a previous study that has shown increased levels of
Lachnospiraceae in allergic subjects [
54]. However, the direction of associations between
Lachnospiraceae and asthma has not been consistent. We also observed a negative correlation between the genetically predicted risk of asthma and the relative abundance of the genus
Lachnospiraceae_NK4A136_group. Similarly,
Lachnospiraceae has been found to be associated with a decreased risk of eczema and inhalant allergic sensitization [
55]. A recent study has also manifested that decreased levels of
Lachnospiraceae in infancy are associated with allergic disease [
56]. Our study suggests that inconsistencies in previous clinical studies may be due to insufficient classification of the genera level of gut microbiota. Notably, members of the
Lachnospiraceae family have been found to encode B cell “superantigens” that stimulate potent IgA responses resulting in bacterial IgA coating [
57]. As major producers of short-chain fatty acids,
Lachnospiraceae are involved in regulatory T cell development in the gut, and gut regulatory T cells [
58], perhaps through IL-10 expression, may be protective against the development of asthma. The
Lachnospiraceae family includes three main genera:
Ruminococcus, Lachnospira, and
Anaerofilum. Arrieta et al. conducted a study on the gut microbiome of infants at risk for asthma in the Canadian Healthy Infant Longitudinal Development (CHILD) Study [
59]. They reported a significant decrease in the relative abundance of the genus
Lachnospira in children at risk of asthma, which was also confirmed in a mouse model of experimental asthma [
60]. The authors suggested that inoculation of germ-free mice with these bacterial taxa ameliorated lung inflammation in their adult progeny. Subsequently, another study extended their previous work and found a reduction in the abundance of
Lachnospira in the 3-month fecal microbiota of asthmatic children, which was considered a potential indicator of asthma diagnosed in preschool-age children [
61]. This reduction was accompanied by reduced levels of fecal acetate and dysregulation of enterohepatic metabolites [
62]. However, our findings were inconsistent with those studies as we found that asthma was positively correlated with the relative abundance of genus
Lachnospira. We hypothesize that the positive and negative effects of
Lachnospira on asthma may be species- and strain-specific, and our study only analyzed data from adults. In addition to
Lachnospiraceae, we also found that genus
Ruminococcus2 was significantly causally associated with a lower risk of asthma, while the genetically predicted risk of asthma was negatively correlated with the relative abundance of genus
Ruminococcus_torques_group, suggesting that genus
Ruminococcus may have a protective effect against asthma. These findings are in line with previous research that has shown a low relative abundance of the genus
Ruminococcus in stools collected during early childhood is linked to an increased risk of asthma [
63]. In addition, a reduction of
Ruminococcus was also negatively correlated with the total fecal IgE levels and strongly associated with children who have mite-sensitized asthma [
64]. Furthermore, our study revealed suggestive causal effects of genus
Oxalobacter on a higher risk of asthma, and class
Actinobacteria, family
Pasteurellaceae, order
NB1n, and order
Pasteurellales on a lower risk of asthma. Prior studies, including Chung KF [
65] and Perez-Garcia et al. [
66], have highlighted the involvement of
Actinobacteria and
Pasteurellaceae in asthma risk, indicating these taxa’s potential as biomarkers and therapeutic targets. Our study contributes to the understanding of these relationships by quantifying their effects on asthma risk and underscores the complexity of the microbiome’s role in respiratory health.
We conducted an analysis to examine the possible associations between gut metabolites and asthma, as they play a crucial role in the interplay between gut microbiota and asthma. While previous studies have suggested potential roles for gut metabolites in asthma, our MR study failed to demonstrate the causality of genetically predicted gut microbiota with asthma. However, our study did reveal that eosinophilic asthma was associated with lower levels of indolepropionate (
p = 3.81 × 10^-2), albeit this
p-value is nominal and has not been adjusted for multiple comparisons. This finding, therefore, should be viewed as exploratory, prompting further investigation into the role of indolepropionate as a potential biomarker for eosinophilic asthma. Indolepropionate has been shown to activate mouse pregnane X receptor (PXR) and induces anti-inflammatory effects [
67]. Previous studies have revealed that a higher level of indolepropionate was associated with a lower risk of type 2 diabetes and increased insulin secretion [
68,
69]. Another large population-based study showed that increased physical activity was significantly associated with high levels of indolepropionate [
70]. Considering these multifaceted implications, the association of indolepropionate with asthma, particularly eosinophilic asthma, warrants a deeper investigation.
Our MR study presents several noteworthy advantages. Firstly, we employed a distinctive two-sample bi-directional MR design to investigate the potential causal association between gut microbiota and metabolites with asthma, thereby providing a robust theoretical foundation for the “gut-lung axis” mechanisms. Secondly, we utilized one of the largest available GWAS summary datasets, ensuring sufficient statistical power to detect causal effects accurately. Lastly, we comprehensively analyzed four distinct asthma phenotypes, enabling us to evaluate the common gut microbiota causally related to different asthma phenotypes and identify novel insights into the gut microbiota-mediating pathogenesis of asthma.
However, there are also several limitations in this study. Firstly, the study included a relaxed cutoff for instrumental variables selection (p < 1 × 10−5) for gut microbiota and metabolites due to limited SNPs meeting the genome-wide significance threshold (p < 5 × 10−8), potentially leading to weak instrument bias. Nonetheless, we addressed these limitations using F-statistics and sensitivity analyses to ensure the validity of the results. Secondly, we only described gut microbiota at the genus level or above due to the lack of data at the species level, highlighting the need for metagenomic sequencing techniques to obtain more specific and accurate results. Thirdly, our study was constrained by the availability of demographic information in the underlying data sources, which precluded us from conducting further subgroup analyses to explore age-specific or gender-specific causal relationships between gut microbiota and asthma. Fourthly, The Finnish population has unique genetic characteristics due to historical events, which may limit the generalizability of our findings to other populations. While this specificity offers valuable insights into asthma genetics, it also means that our results may not fully represent the genetic associations found in more genetically diverse populations. Fifthly, MiBioGen, predominantly of European ancestry, includes approximately 28% with other/multiple ancestries, while all FinnGen individuals are of European descent. We acknowledge that varying genetic ancestries can lead to differences in LD patterns, potentially influencing the robustness of our MR results. Furthermore, the gut metabolites GWAS in our study had a relatively small sample size and limited loci studied. Therefore, further research with larger GWAS statistics is necessary to provide a more precise evaluation of the association between gut metabolites and asthma. Additionally, we recognize a limitation regarding the MR assumption of genotype independence from microbe-asthma confounders. Anthropomorphic traits like BMI, which influence both microbial abundance and asthma risk and have genetic components, might introduce unmeasured confounding in our analysis.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.