Dust extracts were prepared as previously described (Wyatt et al. 2020). Briefly, settled dusts were collected from swine confinement facilities and stored at −20°C. To prepare 100% aqueous hog dust extract, 5-g dust was suspended in 50 mL of balanced salt solution for 1 h, then filtered and sterilized through a 0.22-mm filter (Ulu et al. 2021). The 100% aqueous dust solution was aliquoted and stored at −20°C until used. A 12.5% HDE solution was prepared for instillation by diluting 100% HDE (containing 22.1–91.1 EU/mL) in sterile phosphate buffered saline (PBS). Study animals were terminally anesthetized with isoflurane (Catalog # NDC-11695-6776-2, Henry Schein, Melville, NY, USA). All procedures were approved by the University of California, Riverside, (UCR) Institutional Animal Care and Use Committee (McCole 20190032; Nordgren; 20200014).

This study is a continuation of published work from our laboratory using 8-week-old C57BL6/J mice (n = 11) randomly divided into two treatment groups that received either a 50-μL intranasal dose of PBS or 12.5% HDE, containing 22.1–91.1 EU/mL, for daily for 3 weeks (five consecutive days per week) under light sedation (repetitive exposure) as previously described (Poole et al. 2009). Mice were acclimatized for 1 week upon arrival before experimentation and were maintained on a 12:12-h light:dark cycle. Water and standard rodent chow were ad libitum. Body weight was measured daily, and fecal samples were collected weekly.

Stool samples were collected at 3 weeks and sent to the University of California, San Diego (UCSD) Microbiome Core. The UC San Diego Microbiome Core performed nucleic acid extractions utilizing previously published protocols (https://microbiomejournal.biomedcentral.com/articles/10.1186/s40168-021-01083-0). Briefly, samples were purified using the MagMAX Microbiome Ultra Nucleic Acid Isolation Kit (Thermo Fisher Scientific, USA) and automated on KingFisher Flex robots (Thermo Fisher Scientific, USA). Blank controls and mock communities (Zymo Research Corporation, Tustin, CA, USA) were included and carried through all downstream processing steps. DNA was quantified using a PicoGreen fluorescence assay (Thermo Fisher Scientific, Waltham, MA, USA), and metagenomic libraries were prepared with the KAPA HyperPlus kit (Roche Diagnostics, Basel, Switzerland) and automated on EpMotion automated liquid handlers (Eppendorf, Hamburg, Germany). Sequencing was performed on the Illumina NovaSeq 6000 sequencing platform with paired-end 150-bp cycles at the Institute for Genomic Medicine (IGM), UC San Diego. The 16S rRNA gene amplification was performed according to the Earth Microbiome Project protocol (https://microbiomejournal.biomedcentral.com/articles/10.1186/s40168-021-01083-0). Briefly, Illumina primers with unique forward primer barcodes were used to amplify the V4 region of the 16S rRNA gene (515fB-806r, https://journals.asm.org/doi/10.1128/msystems00009-15). Amplification was performed with single reactions per sample (https://www.future-science.com/doi/10.2144/btn-2018-0192), and equal volumes of each amplicon were pooled for sequencing. The 16S libraries were sequenced at the UC San Diego Institute for Genomic Medicine on the Illumina MiSeq sequencing platform with paired-end 150 bp cycles. The UCSD Microbiome Core performed sample extractions and library preparation utilizing protocol and primers published on the Earth Microbiome Project website. This publication includes data generated at the UC San Diego IGM Genomics Center utilizing an Illumina NovaSeq 6000 that was purchased with funding from a National Institutes of Health SIG grant (#S10 OD026929). The Bioinformatics Core at the UCR implemented QIIME analyses to analyze microbial taxonomy from Week 0 and Week 3 fecal samples as previously described. Subsequently, microbial differences were examined using linear discriminant analysis of effect size (LEfSe) analyses implemented within the online Galaxy module (http://huttenhower.sph.harvard.edu/galaxy/) explained by Segata et al. (2011). Relative abundance percentages were then calculated from taxonomic data generated from QIIME.

Whole blood (1 mL) was collected from the inferior vena cava from PBS controls and HDE-treated mice. Serum was collected by centrifuging the whole blood (Thermo Fisher Scientific, Sorvall Legend Micro 21R Centrifuge, Waltham, MA, USA) at 4°C for 15 min. Serum samples were analyzed by the UCR Metabolomics Core (Riverside, CA, USA).

DNA from luminal contents was extracted using the Qiagen DNeasy PowerSoil Kit (Catalog # 47014; Hilden, Germany) according to the manufacturer’s protocol. iQ SYBR Green Supermix (Bio-Rad, Hercules, CA, USA) was used for real-time quantitative PCR on a C1000 Thermal cycler equipped with a CFX96 real-time PCR system and the Bio-Rad CFX Manager 3.1 Software. The real-time PCR included an initial enzyme activation step (3 min, 95°C) followed by 45 cycles consisting of a denaturing (95°C, 10 s), an annealing (53°C–60°C, 10 s), and an extending (72°C, 10 s) step. The PCR included the following steps for each primer: Akkermansia muciniphila (A. muciniphila; forward: 5′-CAGCACGTGAAGGTGGGGAC-3′ and reverse: 3′-CCTTGCGGTTGGCTTCAGAT-5′) and Universal primer (forward: 5′-CTCCTACGGGAGGCAGCAG-3′and reverse: 5′-TTACCGCGGCTGCTGGCAC-3′). Results were analyzed by the ΔΔCT method.

Data were analyzed by unpaired t test followed by Mann–Whitney post hoc tests. Statistical significance was considered as p < 0.05 for all analyses. Data were analyzed using GraphPad Prism 9 software (San Diego, CA, USA).

The differences in alpha diversity, a measure of the richness and evenness of a bacterial sample within a habitat, between PBS controls and HDE groups were compared using Pielou’s evenness indices and Faith’s phylogenetic diversity (Faith’s PD) metric. Interestingly, evenness was higher in the repetitive HDE group than in saline controls (Figure 1A; Kruskal–Wallis pairwise, p = 0.047, n = 5/group). Moreover, a suggestive increase in biodiversity was observed in the repetitive HDE group as measured by Faith’s PD (Figure 1B; Kruskal–Wallis pairwise, p = 0.174, n = 5/group). The Unweighted Unifrac Distance metric assessed beta diversity and examined community composition, and no dissimilarities between PBS and HDE Week 3 samples fecal samples were shown (Figure 2; PERMANOVA, p = 0.565; n = 5/group).

To assess the impact of agricultural pollutants on host gut microbial abundance, we collected fecal samples at baseline and Day 15 from saline controls and HDE-exposed mice. Repetitive HDE exposure for 15 days did not significantly alter microbial abundance (Figure 3A,B). However, at taxonomic level 2 (phyla), data show a suggestive reduction in Verrucomicrobia (unpaired t test, p = 0.1111, n = 4–5/group) and an increased trend in Firmicutes (unpaired t test, p = 0.1508, n = 5) in HDE-treated mice in comparison to PBS controls (Figure 3C), without reaching significance.

Examination of microbial analysis at the species level (Taxonomic level 7) by 16S rRNA sequencing revealed a partial but statistically insignificant reduction of the beneficial species, A. muciniphila (Figure 4A; unpaired t test, p = 0.1111, n = 4–5/group). Next, we investigated the effects of HDE on the mRNA expression of key microbial species in luminal contents. A. muciniphila showed a marginal reduction in proximal colon luminal contents of HDE-exposed mice compared to saline controls (Figure 4B; unpaired t test, Mann–Whitney U, p = 0.0571, n = 4/group).

To further investigate the microbiome and characterize the bacterial genome following exposure to HDE, metagenomic next-generation sequencing was conducted on fecal samples collected from saline controls and repetitive HDE study animals. Metagenomic sequencing revealed that fold change values, calculated by normalizing HDE-treated samples to PBS control (with PBS serving as the baseline reference), showed a reduction in A. muciniphila expression in fecal samples at 3 weeks post exposure (Figure 5A). Additionally, analysis of baseline (pre-HDE exposure) metagenomic data compared to HDE Week 3 data revealed an increase in A. muciniphila abundance (Figure 5B). The observed increase in A. muciniphila from baseline to Week 3 in HDE-treated mice was followed by a relative reduction compared to PBS controls. We next examined individual species that were altered at Week 3 HDE exposure. Metagenomic sequencing showed a partial reduction of Clostridium species ASF356 (Figure 6A; unpaired t test, p = 0.1349, n = 5–6/group), L. bacterium M8 (Figure 6B; unpaired t test, p = 0.0635, n = 5–6/group), while there was a significant reduction in Anaerotruncus species G3 (Figure 6C; unpaired t test, p = 0.0159, n = 5–6/group), L. bacterium A2 (Figure 6D; unpaired t test, p = 0.0159, n = 5–6/group), and L. bacterium M18 (Figure 6E; unpaired t test, p = 0.0079, n = 5–6/group). These data indicate shifts in microbial communities associated with the production of metabolites important for maintaining barrier integrity.

Having identified HDE-induced changes in bacterial species, we next characterized gut fungal communities following HDE exposure. ITS sequencing was conducted on fecal samples collected from saline and repetitive HDE study animals. HDE exposure did not significantly impact overall fungal communities (Figure 7, heatmap), and deeper analysis at the Class, Order, Family, and Genus levels did not show any alterations in fungal abundance (Figure 8A–D, p < 0.05).

Given the observed fluctuations in A. muciniphila and the alterations detected in metagenomic profiles, examining shifts in the metabolite landscape is a critical step to determine how alterations in gut microbial communities translate to functional metabolic changes. Serum metabolomic analyses revealed that HDE exposure reduced the expression of key metabolites involved in energy metabolism, leucine (Figure 9A, p = 0.018, n = 11) and nicotinic acid (Figure 9B, p = 0.049, n = 11). Additionally, HDE increased the expression of the essential amino acid, inosine (Figure 10A, p = 0.043, n = 11), and riboflavin (Vitamin B2) (Figure 10B, p = 0.0270, n = 11).

Certain metabolites serve as intermediates for SCFA production and the regulation of lung and gut homeostasis (Fusco et al. 2023; Shastak and Pelletier 2023). In the lungs, riboflavin and nicotinic acid are integral to cellular metabolism and other processes that are vital for cellular functions in the respiratory epithelium and immune cells (Li, Chen, et al. 2024; Shastak and Pelletier 2023). Altering vitamin production modifies cellular repair mechanisms and antioxidant defenses which increases susceptibility to oxidative stress and tissue damage (Li, Ren, et al. 2024; Shastak and Pelletier 2023). Inosine modulates immune responses and has been shown to exert protective effects against pulmonary inflammation (Kim et al. 2022). Its increase in HDE-exposed mice might, therefore, be an anti-inflammatory and protective response in the lungs (Liaudet et al. 2001; Mao et al. 2022).

Additionally, leucine is critical for protein synthesis and regulating immune functions via the mTOR signaling pathway (Yang et al. 2023). Its loss can disrupt the synthesis of structural proteins, weaken the pulmonary epithelium and immune responses (Das et al. 2014). Moreover, these specific changes in metabolite levels can impact lung homeostasis, making the lungs more vulnerable to inflammatory diseases and reducing their capacity to respond effectively to environmental irritants.

Concerning the gut, the altered levels of A. muciniphila in mice exposed to HDE may coincide with observed alterations in gut-derived metabolites: riboflavin, nicotinic acid, inosine, and leucine. A. muciniphila is primarily known for its role in degrading mucin to maintain intestinal barrier integrity (Lukovac et al. 2014). Its reduced levels in HDE-exposed mice may impact the production of SCFAs, which other microbes utilize for metabolism and immune function (Rodrigues et al. 2022). Additionally, reduced A. muciniphila may have contributed to the observed increase in gut permeability and the subsequent translocation of LPS into systemic circulation of HDE-exposed mice (Crawford et al. 2024). This can potentially exacerbate inflammatory responses within the gut and other metabolically active organs. Such dysregulation could contribute to lung inflammation, highlighting a critical interaction between gut microbial dynamics and lung health (Enaud et al. 2020). In our study, the associated decrease in metabolites like leucine and nicotinic acid, both essential for cellular energy metabolism, may further impair tight junction regulation and barrier integrity via reactive oxygen species pathways (Song et al. 2020; Lin et al. 2022). In addition to the decrease in A. muciniphila, other beneficial microbes essential to metabolite production, such as Clostridium sp. ASF356, Lachnospiraceae bacterium, and Anaerotruncus species were reduced in HDE fecal samples. Although the role(s) of Anaerotruncus species in the gut has not been thoroughly investigated, studies have found that these species are associated with nucleoside metabolism and amino acid fermentation (Raimondi et al. 2021). Therefore, reduced Anaerotruncus abundance may impact the fermentation of dietary fibers and proteins important for epithelial barrier function (Fu et al. 2023). Clostridium species are important for branched-chain amino acid (BCAA) metabolism (Ren et al. 2024). These species can convert leucine into SCFAs like isobutyrate and isovalerate, which are important for water and electrolyte balance in the colon (Mager et al. 2020). Moreover, Lachnospiraceae members have been shown to participate in BCAA metabolism, thereby influencing energy homeostasis (Li, Chen, et al. 2024). Both Clostridium and Lachnospiraceae are known to participate in nicotinic acid (niacin) metabolism, and a reduction in these bacteria could decrease niacin availability and promote intestinal inflammation and barrier dysfunction (Karunaratne et al. 2020), which aligns with the inflammatory effects of HDE exposure.

In our study, inosine and riboflavin increased following HDE exposure. Inosine regulates intestinal barrier function by reducing inflammation and acting on the peroxisome proliferator-activated receptor (PPAR)-y signaling pathway, prompting goblet cell and mucus production (Tang et al. 2024). Studies have also indicated riboflavin as an important regulator of tight junction expression in the intestine and a mediator of intestinal inflammation, possibly due to its antioxidant functions (Zhang et al. 2024; Xu et al. 2022). Lachnospiraceae bacteria and Clostridium species are involved in the breakdown and/or production of inosine via bacterial nucleosidases (purine metabolism) and riboflavin biosynthesis and uptake (LeBlanc et al. 2013; Magnúsdóttir et al. 2015). Although these bacteria decreased, an increase in metabolites may demonstrate a protective response against HDE exposure. Additionally, the increased riboflavin found in HDE-treated mice may mitigate intestinal epithelial cell permeability by influencing the transcription of tight junction genes (Chen et al. 2015). Moreover, riboflavin deficiency can impact the production of other metabolites by interrupting riboflavin kinase activity, which phosphorylates riboflavin, a precursor to flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD) production (Chengalroyen et al. 2024; Moreno et al. 2022). Therefore, increased riboflavin in HDE mice may also represent a compensatory effect to prevent the loss of other metabolites. This possible interplay between microbial alterations, metabolite dysregulation, and immune response underscores the complex mechanisms through which environmental factors like HDE may influence both gut and lung health, potentially driving inflammatory processes.

Significant reductions in Lachnospiraceae bacterium A2 and M18 were observed in the gut microbiota of mice exposed to HDE. Members of the Lachnospiraceae family are known to play pivotal roles in the fermentation of dietary fibers, producing SCFA and other metabolites critical for intestinal homeostasis and metabolic balance. More specifically, Lachnospiraceae are involved in the metabolic processes that contribute to the biosynthesis of B vitamins and amino acids, essential components for cellular energy metabolism and protein synthesis (Vacca et al. 2020).

The reduction in these specific Lachnospiraceae strains may disrupt metabolic pathways, decreasing the production of these critical metabolites and exacerbating the physiological responses observed in HDE-exposed mice. This association highlights the link between dietary fiber fermentation by the gut microbiota and metabolite production, suggesting that HDE can alter microbial communities and their metabolic functions. The gut microbiomes’ ability to modulate systemic immunity and inflammation through metabolic intermediates highlights the potential of microbial and metabolic alterations to impact gut and lung function under inflammatory conditions like those induced by HDE exposure.

The study focused on a relatively short exposure period (3 weeks). However, a longer duration may reveal different effects on microbial communities and systemic health. Additionally, a single time point analysis of the microbiome (at Day 15) may have excluded dynamic changes in the microbiota that occur over time. Moreover, the serum metabolomics findings (systemic) do not represent the potential localized gut changes (fecal metabolomics). Finally, while microbial composition was analyzed, functional implications (e.g., metabolic activity and immune response) of the observed changes were not directly assessed.