Butyrate regulates neutrophil homeostasis and impairs early antimicrobial activity in the lung

Abstract

Short-chain fatty acids (SCFAs) are metabolites that are produced after microbial fermentation of dietary fiber and impact cell metabolism and anti-inflammatory pathways both locally in the gut and systemically. In preclinical models, administration of SCFAs, such as butyrate, ameliorates a range of inflammatory disease models including allergic airway inflammation, atopic dermatitis, and influenza infection. Here we report the effect of butyrate on a bacteria-induced acute neutrophil-driven immune response in the airways. Butyrate impacted discrete aspects of hematopoiesis in the bone marrow resulting in the accumulation of immature neutrophils. During Pseudomonas aeruginosa infection, butyrate treatment led to the enhanced mobilization of neutrophils to the lungs as a result of increased CXCL2 expression by lung macrophages. Despite this increase in granulocyte numbers and their enhanced phagocytic capacity, neutrophils failed to control early bacterial growth. Butyrate reduced the expression of nicotinamide adenine dinucleotide phosphate, oxidase complex components required for reactive oxygen species production, and reduced secondary granule enzymes, culminating in impaired bactericidal activity. These data reveal that SCFAs tune neutrophil maturation and effector function in the bone marrow under homeostatic conditions, potentially to mitigate against excessive granulocyte-driven immunopathology, but their consequently restricted bactericidal capacity impairs early control of Pseudomonas infection.

Fig. 1

Effect of butyrate on acute Pseudomonas aeruginosa immunity. (A) Quantification of the total number of cells in BALF 18 hours post-inoculation with 1 × 10⁶ CFU of P. aeruginosa strain PAO1 in Ctrl and But mice. (B) Frequency and quantification of cell types in cytospins of BALF from control and butyrate-treated mice 18 hours after PAO1 infection; (C) Quantification of P. aeruginosa CFU in BALF 18 hours post-challenge. (D, E) Cytokine and chemokine production in BALF 18 hours following infection. (F) MFI of CXCR2 on lung AM and IM from Ctrl and But mice. (G) MFI of CXCR2 on Ctrl and But lung macrophages US or stimulated for 2 hours with LPS in vitro. (H) transmigration of Ctrl and But neutrophils toward a CXCL2 or CCL2 chemokine gradient after 3 hours. Results are a mean of two independent experiments. Values are expressed as mean ± standard error of mean; n = 6–14. Statistical significance was determined with one-way analysis of variance in (A–E, G, H) and Student’s t test (unpaired, two-tailed) in (F). * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001. AM = alveolar macrophages; BALF = bronchoalveolar lavage fluid; But = butyrate-treated; CFU = colony-forming units; Ctrl = control; IFN = interferon; IL = interleukin; IM = interstitial macrophages; LPS = lipopolysaccharide; MFI = mean fluorescence intensity; PAO1 = P. aeruginosa strain 1; TNF = tumor necrosis factor; US = unstimulated.

Fig. 2

Butyrate affects neutrophil maturation in the BM during homeostasis. (A) Relative representation and quantification of CMP and GMPs in BM of Ctrl and But mice. (B) Frequency, and quantification of proNeu, preNeu, immNeu, and matNeu in BM of Ctrl and But mice. Relative representation and quantification of immNeu and matNeu (C) in BM 18 hours pi and (D) in BALF 18 hours pi. Results are representative of a mean of two independent experiments; E, maturation of FACS-sorted control preNeus cultured for 18 hours with 300 uM butyrate and G-CSF into immNeu and matNeu. Results are expressed as mean ± standard error of mean; n = 9–12 per group in (A–D) and n = 5 per group in (E). Statistical significance was determined with Student’s t test (unpaired, two-tailed) in (A–D) and One-way analysis of variance in (E). * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001. BALF = bronchoalveolar lavage fluid; BM = bone marrow; But = butyrate-treated; CMP = common myeloid progenitor; Ctrl = control; G-CSF = granulocyte colony-stimulating factor; GMPs = granulocyte monocyte progenitors; immNeu = immature neutrophils; matNeu = mature neutrophils; PAO1 = P. aeruginosa strain 1; PBS = phosphate buffered saline; pi = post-infection; preNeu = preNeutrophils; proNeu = proNeutrophils.

Fig. 3

Butyrate alters neutrophil transcriptional landscape under homeostatic conditions. Differential gene expression heat-map of FACS-sorted (A) immature and (B) mature neutrophils from Ctrl and But mice under homeostasis; C, gene expression (fold change) of Camp, Ngp, and Ltf from Ctrl and But total neutrophils; D, immunoblot of uninfected ex vivo total neutrophils and quantification of LTF band normalized to β-actin loading Ctrl; E, cytokine expression of LTF in bronchoalveolar lavage fluid of Ctrl and But mice 18 hours post-infection. Results are a mean of two independent experiments for (C, E) or are representative of data from two independent experiments (D). Values are expressed as mean ± standard error of mean; n = 5–7 per group in (C, E). For immunoblotting n = 3–4 mice per group were pooled. Statistical significance was determined with Student’s t test (unpaired, two-tailed) in (C, D) and with one-way analysis of variance in (E). ** p ≤ 0.01, *** p ≤ 0.001. But = butyrate-treated; Camp = cathelicidin; Ctrl = control; Ltf = lactoferrin; mRNA = messenger RNA; Ngp = neutrophil granule protein; PAO1 = P. aeruginosa strain 1; PBS = phosphate buffered saline.

Fig. 4

Butyrate reduces bacterial clearance capacity but enhances phagocytosis. (A) Quantification of Pseudomonas aeruginosa after co-culture with Ctrl or butyrate neutrophils for 2 hours to assess for in vitro killing capacity. (B) Myeloperoxidase and elastase expression in BALF of uninfected or P. aeruginosa- challenged Ctrl and But mice. (C) Phagocytosis of PE-labeled immunoglobulin G beads by total, immature and mature bone marrow neutrophils after 2 hours incubation expressed as fold change over control neutrophils. Results are a mean of two independent experiments. Values are expressed as mean ± standard error of mean, n = 6–12 per group. Statistical analysis was determined with Student’s t test (unpaired, two-tailed) in (A) and One-way analysis of variance in (B, C). * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001. BALF = bronchoalveolar lavage fluid; But = butyrate-treated; Ctrl = control; immNeu = immature neutrophils; matNeu = mature neutrophils; mRNA = messenger RNA; PAO1 = P. aeruginosa strain 1; PE = phycoerythrin; PBS = phosphate buffered saline.

Fig. 5

Butyrate impairs ROS-mediated bacterial clearance in neutrophils. A, fold increase of MFI expression of DHR 123 in total, immature and mature BM neutrophils of Ctrl or But mice as a ROS measurement following treatment with PBS or 20 nM PMA for 45 minutes; B, MFI of DHR123 in Ctrl and butyrate BM neutrophils following co-incubation with Pseudomonas aeruginosa for 2 hours; C, gene expression of Nox2, Cyba, Ncf1, and Ncf4 of BM neutrophils; D, western blot of uninfected ex vivo neutrophils and quantification of NOX2 band normalized to loading control. Results are a mean of two independent experiments (A–C) or are representative of data from two independent experiments (D). Values are expressed as mean ± standard error of mean, n = 7–10 per group in (A, B), n = 6 per group in (C). For immunoblotting n = 3–4 mice per group were pooled. Statistical analysis was determined with one-way analysis of variance in (A) and Student’s t test (unpaired, two-tailed) in (B–D). * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001. BM = bone marrow; But = butyrate-treated; Ctrl = control; DHR = dihydrorhodamine; immNeu = immature neutrophils; matNeu = mature neutrophils; MFI = mean fluorescence intensity; mRNA = messenger RNA; ROS = reactive oxygen species; PAO1 = P. aeruginosa strain 1; PE = phycoerythrin; PMA = Phorbol 12-Myristate 13-Actetate; PBS = phosphate buffered saline.