Dysbiosis of the Female Murine Gut Microbiome Exacerbates Neutrophil-mediated Vascular Allograft Injury by Affecting Immunoregulation by Acetate : Transplantation

Journal Logo

Original Basic Science

Dysbiosis of the Female Murine Gut Microbiome Exacerbates Neutrophil-mediated Vascular Allograft Injury by Affecting Immunoregulation by Acetate

Rey, Kevin M. PhD1; Tam, Franklin F. BSc1; Enns, Winnie BSc1; Rahim, Javaria F. BSc1; Safari, Kwestan BSc1; Guinto, Elizabeth BSc1; Van Rossum, Thea PhD2; Brinkman, Fiona S.L. PhD1; Choy, Jonathan C. PhD1

Author Information
doi: 10.1097/TP.0000000000004161

Abstract

INTRODUCTION

The gut microbiota is a population of bacteria, viruses, and fungi that inhabits the host intestinal tract and produces physiologically important metabolites.1 Dysbiosis of the microbiota affects many immune-related processes and can be caused by environmental influences, such as antibiotic use, especially early in life.2,3 In order to influence immune responses distal to the intestinal tract, metabolites produced by the gut microbiota are absorbed through the intestinal epithelium and circulate systemically. One class of metabolites, short chain fatty acids (SCFAs), is well known to impact the immune system.4 The most abundant of these SCFAs are acetate, butyrate, and propionate.4 Acetate inhibits the pathologic activation of neutrophils, and butyrate increases Treg.5-8 SCFAs are produced through the fermentation of glycans in mucin and dietary fiber.9 Microbes, such Akkermansia muciniphila, metabolize mucin, and Lactobacillus bacteria are main contributors to the production of SCFAs by the metabolism of dietary fiber.10-12

The gut microbiota and its composition influences the outcome of organ transplantation. This microbial community can be perturbed by transplant-related factors such as antibiotics, immunosuppressive drugs, and surgical stress.13 Clinical studies suggest that the gut microbiota may effect poor outcomes in liver transplantation.14,15 Also, the composition of this microbial community is persistently altered after kidney transplantation and is associated with diarrhea, acute rejection, and diabetes.16-19 In experimental animal studies, the presence of donor and recipient microbiota increases rejection of cardiac and skin allografts through activation of a type I interferon response.20 Also, antibiotics given peritransplant improve cardiac graft survival, possibly by eliminating bacteria that are cross-reactive to allospecific T cells.21 Certain components of the microbiota are also immunoregulatory by inducing development of B cells that produce interleukin-10, affecting Treg and improving the immunosuppressive potency of tacrolimus.22-24

Allograft vascular injury is a main component of acute heart transplant rejection that can be independent of parenchymal immune injury.25 Immunological drivers of allograft vascular injury include neutrophils, donor-reactive antibodies, and T cells.26 Notably, neutrophils directly injure vascular smooth muscle cells in the media of allograft arteries in a NADPH oxidase 2-dependent manner that is independent of adaptive immune responses.27 We previously found that altering the gut microbiota of graft recipients by treatment with broad-spectrum antibiotics early in life (for the first 3 wk only) leads to increased neutrophil-mediated injury of vascular allografts performed later in life at a time point that is comparable to early adulthood.28 Here, we characterize metagenomic changes in the gut microbiota that are caused by treatment with broad-spectrum antibiotics early in life and how they relate to immunological injury of vascular allografts. Dysbiosis of the gut microbiota caused by early life disruption reduces or eliminates a handful of bacterial species and alters its metabolic gene content related to acetate production. Normalizing the gut microbiota of antibiotic-treated mice or increasing their dietary intake of acetate prevents neutrophil-mediated allograft vascular injury. Our findings are one of the first to describe how the metagenome of the gut microbiota is related to immune drivers of transplant rejection and identify specific bacterial species and acetate as influencers of transplant rejection.

MATERIALS AND METHODS

Disruption of the Gut Microbiota

C57Bl/6 (H-2b) and Balb/c (H-2d) mice were purchased from Jackson Laboratories and bred in-house. All study protocols were reviewed and approved by the Simon Fraser University Animal Care Committee based on guidelines by the Canadian Council on Animal Care. The gut microbiota was disrupted in mice using an antibiotic cocktail in the drinking water that contained ampicillin (0.5 g/L), vancomycin (0.25 g/L), neomycin sulfate (0.5 g/L), and metronidazole (0.5 g/L) that was given until 3 wk of age. This treatment leads to the near complete elimination of gut bacteria for the duration of antibiotic exposure in young mice and then to colonization of the intestinal tract by bacterial populations after cessation of antibiotics.28 In some experiments, untreated and antibiotic-treated mice were co-housed starting at 3 wk of age. Acetate was administered by adding 100 mmol/L magnesium acetate in drinking water starting 14 d before transplant and continuing until the end point. The gut microbiota was sampled by collecting fecal pellets at 8 to 12 wk of age.

Aortic Interposition Grafting

Aortic transplantation was performed as described previously in mice that were 8 to 12 wk of age.28,29 Briefly, a section of infrarenal aorta from Balb/c donors was interposed into the infrarenal aorta of sex-matched C57Bl/6 recipient mice. C57Bl/6 donors were used as syngraft controls.

Histology and Immunohistochemistry

Artery grafts were perfusion fixed with 4% paraformaldehyde. For histological examination of medial injury, 2 cross-sections per artery that were at least 100 μm apart were H&E stained and the average thickness of the media measured in Image J (National Institutes of Health). The medial thickness was normalized to arterial radius.28

Immunohistochemistry was performed using antibodies toward myeloperoxidase (Abcam, Cambridge, United Kingdom) to visualize neutrophils, CD4 (4SM95; eBioscience, Waltham, Massachusetts), CD8 (4SM15, eBioscience), Foxp3 (FJK-16S, eBioscience), and Mac-3 (BD Biosciences, Franklin Lakes, New Jersey) to visualize macrophages as described previously.29 Staining was visualized by aminoethyl carbazole chromagen (red; Vector Laboratories, Burlingame, CA) and counterstained with hematoxylin (blue). For quantification, the total number of cells in each cross-section was manually counted in a blinded manner and reported per mm2 of media or adventitia. Two cross-sections that were at least 100 μm apart were evaluated for each artery and an average value from the cross-sections calculated and used as a single data point.

Microbiome 16S rRNA Sequencing and Analysis

DNA from fecal samples was extracted, amplified, and 16S rRNA gene sequenced as described previously.28 Analysis of sequences was performed using QIIME2, version 2019.2.30 Reads were filtered for quality and denoised using divisive amplicon denoising algorithm.31 Phylogenetic trees for rarefaction were constructed using multiple alignment using fast Fourier transform and randomized axelerated maximum likelihood.32,33 Samples were each rarefied to 1.2 × 104 reads, and the resulting amplicon sequence variants were classified using a Bayesian classifier trained on reads extracted from the SILVA 132 release.34 Genetic content was inferred using PICRUSt2 and grouped into MetaCyc pathways and enzyme commission (EC) reactions.35-37

Metagenomic Shotgun Sequencing and Analysis

Whole genome shotgun sequencing was performed at the Michael Smith Genome Sciences Center (Vancouver, British Columbia, Canada) on the Illumina HiSeq platform using V2.5 chemistry. Before analysis, raw reads were filtered for quality and presence of mouse sequences using KneadData.38 Resulting data were analyzed using MetaPhlAn3 using the included ChocoPhlAn3 pipeline based on UniProt and NCBI repositories.38,39 HUMAnN3 was used to determine genetic content; genes were detected using the UniRef90 database and regrouped under MetaCyc pathways and EC reactions.38,40 Gene abundance data were normalized to reads per kilobase. Pathways and EC reactions were filtered for those that appeared in at least 3 mice across all treatment groups.

Flow Cytometry

Splenocytes and blood cells were stained with antibodies recognizing CD4 (RM4.5, BDBiosciences), CD8 (53–6.7, BD Biosciences), Foxp3 (MF23, BD Biosciences), Ly6G (1A8, BD Biosciences), and CD11b (M1/70, eBioscience). Data were acquired on a BD LSRFortessa X-20 (BD Biosciences) and analyzed using FlowJo.

Statistical Analyses

Differences between groups were determined using a Mann-Whitney U-test. Volcano plots display results from Kruskal-Wallis rank-sum tests with Benjamin Hochberg correction. Ordination plots were generated using the Vegan package in R. Bray-Curtis or weighted-Unifrac distances were calculated, and these matrices were used to produce a 2-dimensional nonmetric multidimensional scaling (NMDS) plot. Covariance ellipses were calculated using Vegan. Significant differences were defined as having a P < 0.05.

RESULTS

Disruption of the Gut Microbiota With Antibiotics Early in Life Exacerbates Neutrophil Accumulation in Female Recipients of Vascular Transplants

The effects of early life disruption of the gut microbiota were examined by administering a cocktail of antibiotics in the drinking water of recipient mice for the first 3 wk of life and then maintaining them on normal water thereafter. Aortic transplants were then performed into these mice at 8 to 12 wk of age and collected at day 7 posttransplantation.28 When stratified by sex, antibiotic treatment early in life markedly and significantly exacerbated neutrophil accumulation in both the media and adventitia of allograft artery transplants in female recipients, whereas there was minimal effect in males (Figure 1A). There was no effect on accumulation of CD4 T cells, CD8 T cells, Treg, or macrophages in either females or males (Figure 1B–E). Also, immune cell accumulation was minimal in syngrafts (Figure 1 and Figure S1A, SDC, https://links.lww.com/TP/C423). When the systemic levels of leukocytes were examined in female mice, there was no effect of antibiotic treatment on CD4 T cells, CD8 T cells, Treg, monocytes, or neutrophils (Figure S1B–D, SDC, https://links.lww.com/TP/C423). All subsequent studies were performed in female mice.

F1
FIGURE 1.:
Antibiotic treatment early in life exacerbates neutrophil accumulation in female graft recipients. Recipient mice were untreated or treated with antibiotics for the first 3 wk of life (ABT). Sex-matched aortic interposition allografts were performed between 8 and 12 wk in male and female recipients. Arteries were recovered at 7 d posttransplantation and stained (red) for (A) neutrophils, (B) CD4 T cells, (C) CD8 T cells, (D) Treg, and (E) macrophages. Cell counts are normalized to surface area of media and adventitia. Insets are isotype control staining. Magnification = × 400, scale bar = 50 μm. * P < 0.05. ABT, antibiotic treated; UT, untreated.

Antibiotic Treatment Early in Life Changes the Inferred Metagenome of the Gut Microbiota

We examined how antibiotic treatment early in life alters the taxonomic and metagenomic landscape of the gut microbiota of adult mice. Fecal samples were collected from mice at 8 to 12 wk of age, DNA was isolated, and bacterial composition was determined by 16S rRNA sequencing. Antibiotic treatment early in life altered the gut microbiota of adult mice. This was characterized by an increase of the bacterial classes Clostridia and Erysipelotrichia, a decrease of Bacteroidia, and an absence of Verrucomicrobiae (Figure 2A). NMDS analysis of weighted-UniFrac distances showed a clear separation between untreated and antibiotic-treated mice (Figure 2B). There was also a slight reduction in microbial diversity caused by antibiotic treatment early in life (Figure 2C).

F2
FIGURE 2.:
Antibiotic treatment early in life changes the gut microbiota in adult female mice. 16S rRNA sequencing was performed on fecal samples from 8- to 12-wk-old female mice. A, Bacterial classes in the fecal samples of UT and mice treated with antibiotics for the first 3 wk of life (ABT) were identified using QIIME2. Individual mice are represented as separate columns. B, The difference in bacterial composition of the gut microbiota was determined by NMDS plot of weighted-UniFrac distances calculated from bacterial taxa identified by QIIME2 and plotted with NMDS. Covariance ellipses show 95% CI and stress <0.001 (C). The diversity of the gut microbiota was assessed by Shannon diversity of untreated and ABT mice. ABT, antibiotic treated; NMDS, nonmetric multidimensional scaling; UT, untreated.

Next, genomic composition of potential metabolic pathways was inferred from 16S data based on a reference set of genomes using PICRUSt2.35 After grouping metabolic genes under MetaCyc pathways, NMDS ordination of Bray-Curtis distances showed that the total inferred metabolome of antibiotic-treated mice was significantly different than untreated counterparts (Figure 3A). Genes encoding enzymes in metabolically relevant reactions were then examined by regrouping the PICRUSt2 data under EC reactions. Inferred enzymatic reactions clustered separately between untreated and antibiotic-treated mice (Figure 3B). This analysis suggested that antibiotic treatment early in life changes the resultant gut microbiota in adults and alters its metabolic capacity. Direct comparison of pathways and EC reactions between groups showed that there may be alterations in those related to nucleotide and amino acid biosynthesis, but these did not reach statistical significance.

F3
FIGURE 3.:
Antibiotic treatment early in life alters the abundance of inferred genes related to metabolic pathways. Genes involved in metabolic pathways were inferred based on 16S sequencing results using PICRUSt2. A, The difference in the composition of inferred genes related to metabolic pathways was determined by NMDS plot of Bray-Curtis distances. Covariance ellipses show 95% CI, stress < 0.001. B, The difference in abundance of inferred genes related to EC reactions was determined by NMDS plot of Bray-Curtis distances. Covariance ellipses show 95% CI and stress <0.001. C, Schematic of mucin degradation by A. muciniphila. Top panel: Oligosaccharides are covalently bound to serine and threonine residues in mucin glycoproteins. Bacterial enzymes remove sugars from mucin proteins for use in bacterial metabolic pathways. Bottom panel: The resultant liberation of GlcNAc contributes to acetate production as a by-product of cellular metabolism pathways. D, Abundance of inferred genes that encode enzymes that metabolize mucin in UT and ABT mice. *P < 0.05. ABT, antibiotic treated; EC, enzyme commission; GlcNAc, N-acetylglucosamine; NMDS, nonmetric multidimensional scaling; UT, untreated.

Because of the role of acetate in regulating neutrophils and the production of this metabolite as a by-product of mucin metabolism, we also examined the inferred metagenomic landscape for genes involved in reactions that metabolize mucin and could potentially affect the production of SCFAs. Mucin degradation pathways have yet to be curated in the databases we queried, but studies have established key enzymatic reactions in this process (Figure 3C).41 Six such reactions were inferred from the 16S rRNA, data and of these, 4 were significantly reduced in antibiotic-treated mice, suggesting a potentially reduced capacity to metabolize mucin and produce acetate (Figure 3D). In addition to mucin, we also queried our data for inferred genes involved in the production of SCFAs from fermentation of fiber but were unable to assign genes specifically to this process.

Antibiotic Treatment Early in Life Reduces Bacterial Taxa Involved in Production of SCFAs

The bacterial composition of the gut microbiota was then examined in greater detail by whole genome shotgun sequencing, and the effect of co-housing was also determined. Fecal samples were collected at 8 to 12 wk of age. The data showed similar trends to 16S analysis. NMDS analysis of Bray-Curtis distances showed a clear separation between the composition of the gut microbiota in untreated and antibiotic-treated mice (Figure 4A). There was an increase in the bacterial class Clostridia and decrease in Bacteroidia in antibiotic-treated mice compared with untreated counterparts (Figure 4B). There was also a reduction of Verrucomicrobiae, Bacilli, and Actinomycetia in antibiotic-treated mice. Notably, co-housing mice normalized the gut microbiota, resulting in a bacterial population that was comparable to untreated mice (Figure 4A and B). Of the bacterial species detected in our samples, 8 were significantly different between the untreated and antibiotic-treated groups (Figure 4C). A. muciniphila was the most abundant species that was completely eliminated by antibiotics and completely restored to normal levels after co-housing. Reduction of A. muciniphila in the gut microbiota of antibiotic-treated mice was also observed by 16S PCR (Figure 4D). A. muciniphila is well known to degrade mucin, which results in production of acetate as a by-product.14 The second- and third most abundant bacterial species that were reduced by antibiotics and completely rescued by co-housing were Lactobacillus murinus and Lactobacillus johnsonii (Figure 4C). Notably, these 2 Lactobacillus species can also contribute to the production of acetate by the gut microbiota.42

F4
FIGURE 4.:
Antibiotic treatment early in life alters the abundance of bacterial species. Metagenomic whole genome shotgun sequencing was performed on fecal samples from 8- to 12-wk-old female mice. A, Difference in composition of bacteria in fecal samples is illustrated by NMDS plot of Bray-Curtis distances of bacterial taxa identified by MetaPlAn3. Covariance ellipses show 95% CI and stress = 0.048. B, Stacked bar chart of bacterial classes identified using MetaPlAn3. Individual mice are represented as separate columns. C, Bacterial species that are differentially abundant with Holm correction between UT, ABT, and CH mice. D, A. muciniphila specific and pan-16S PCR. Quantification is A. muciniphila product normalized to 16S product. *P < 0.05. ABT, antibiotic treated; CH, co-housed; NMDS, nonmetric multidimensional scaling; UT, untreated.

Genes that are related to metabolic pathways were then examined using HUMAnN3 and clustered into MetaCyc pathways. Of these, 103 were significantly different between untreated and antibiotic-treated mice (Figure S1E, SDC, https://links.lww.com/TP/C423). The NMDS plot of Bray-Curtis distances showed that mice treated with antibiotics have distinct genomic composition of metabolic genes as compared with untreated mice. This was normalized by co-housing (Figure 5A). Significantly different pathways were related to vitamin biosynthesis, amino acid biosynthesis, and nucleotide biosynthesis (Table S1, SDC, https://links.lww.com/TP/C423). Under current classifications, there are no pathways that specifically examine the metabolism of mucin and production of SCFAs from fiber.

F5
FIGURE 5.:
Antibiotic treatment early in life alters genes encoding enzymes involved in mucin metabolism. Genes encoding enzymes related to metabolic pathways were determined by whole genome shotgun sequencing of fecal samples in 8- to 12-wk-old female mice. A, Difference in the abundance of genes encoding metabolic pathways was determined by NMDS plot of Bray-Curtis distances. Covariance ellipses show 95% CI and stress = 0.009. B, The difference in abundance of genes encoding enzymes related to metabolic pathways was determined by NMDS plot of Bray-Curtis distances. Covariance ellipses show 95% CI and stress = 0.018. C, Genes encoding enzymes that metabolize mucin were quantified. D, Proportion of reads of β-N-acetylhexosaminidase from different bacterial species in the microbiome was determined. Individual mice are represented as separate columns.**P < 0.001. ABT, antibiotic treated; CH, co-housed; NMDS, nonmetric multidimensional scaling; UT, untreated.

The abundance of genes related to metabolically relevant enzymatic pathways was then assessed. Antibiotic-treated mice clustered separately from untreated and co-housed mice in the NDMS plot, with the latter groups clustering together (Figure 5B). There were 558 enzymatic reactions that were significantly different between untreated and antibiotic-treated mice that were related to enzyme cofactor biosynthesis, secondary metabolite degradation, and carbohydrate degradation (Figure S1F and Table S2, SDC, https://links.lww.com/TP/C423). When enzymes related to mucin degradation were specifically examined, genes related to 4 mucin-degrading enzyme reactions were detected. Similar to findings from inferred gene content using 16S analysis, β-N-acetylhexosaminidase was significantly decreased in antibiotic-treated mice as compared with untreated counterparts, and co-housing completely restored its abundance (Figure 5C). There was no detectable difference in abundance of the other 3 enzyme genes. β-N-acetylhexosaminidase catalyzes the cleavage of N-acetylglucosamine and N-acetylhexoseamine and is an initial step of mucin degradation.14 In untreated mice, a consistent proportion of identified reads from this gene in the microbiome was associated with the genome of A. muciniphila (Figure 5D). In antibiotic-treated mice, which lack A. muciniphila, the lower abundance of β-N-acetylhexosaminidase was disproportionally present from the genome of F. plautii. Co-housing mice restored the abundance of this gene from A. muciniphila. These findings indicate that the deficiency in A. muciniphila caused by antibiotic treatment reduces the metagenomic content of a mucin-degrading enzyme gene that may hamper mucin metabolism, and potentially the production of downstream SCFAs as a by-product. We were unable to identify enzymatic reactions that produce SCFAs specifically from degradation of resistant starches/fiber in the databases queried.

Normalizing the Gut Microbiota by Co-housing Prevents Neutrophil-mediated Allograft Vascular Injury in Antibiotic-treated Mice

We next examined whether dysbiosis of the gut microbiota caused by its early life disruption with antibiotics affects neutrophil-mediated allograft vascular injury by performing co-housing experiments. Aortic interposition allografts were performed into recipients at 8 to 12 wk of age. Antibiotic treatment markedly and significantly exacerbated neutrophil accumulation in both the arterial media and adventitia of allograft arteries (Figure 6A). This was related to increased medial injury as determined by a reduction in medial thickness that was accompanied by loss of cellularity and elastic laminae in the media of allograft arteries at day 30 posttransplantation (Figure 6B). Accumulation of other immune cells was not affected (Figure S2, SDC, https://links.lww.com/TP/C423). Notably, normalizing the microbiota of antibiotic-treated mice to resemble that of untreated counterparts via co-housing completely prevented neutrophil accumulation (Figure 6A). These findings directly establish that changes in the gut microbiota of adult graft recipients caused by early life disruption with antibiotics lead to dysregulation of neutrophils and increase medial injury after transplantation.

F6
FIGURE 6.:
Normalization of the gut microbiota by co-housing and administration of acetate reverses exacerbation of neutrophil accumulation caused by antibiotic treatment early in life. A, Aortic interposition allografts were transplanted into UT or ABT female mice. A group of UT mice were co-housed with ABT mice (UT CH) and of ABT mice were co-housed with UT mice (ABT CH). Also, some ABT mice were administered 100 mmol/L magnesium acetate in their drinking water for 2 wk before transplant and continuing until end point (ABT Ac). Arteries were recovered at 7 d posttransplantation and stained for neutrophils. Cell counts are normalized to surface area of media and adventitia. Inset is isotype control staining. B, Aortic interposition allografts were performed into UT, ABT, or ABT Ac mice. Arteries were recovered at day 30 posttransplanted, H&E stained, and medial thickness quantified. Magnification = ×400, scale bar = 50 μm. *P < 0.05, **P < 0.01. ABT, antibiotic treated; CH, co-housed; UT, untreated.

Acetate Prevents Neutrophil Accumulation and Related Injury in Vascular Allografts Caused by Early Life Disruption of the Gut Microbiota

We observed that early life disruption of the gut microbiota with antibiotics results in metagenomic changes, that relate to reduced acetate production, such as diminished abundance of the mucin-degrader A. muciniphila, reduction of genes encoding enzymes that metabolize mucin, and diminished abundance of bacterial species (L. murinus and L. johnsonii) that contribute to the production of acetate from dietary components. As such, the effect of acetate on regulating neutrophil responses in vascular allografts was examined by administering magnesium acetate (100 mmol/L) in the drinking water of some antibiotic-treated mice for 14 d before transplantation until end point. Administration of acetate in this manner has been shown to increase acetate levels and inhibit pathological neutrophil responses.43 Aortic interposition allografts were performed and artery segments collected at day 7 and 30 posttransplantation. Administration of acetate significantly and completely prevented the exacerbation of neutrophil accumulation caused by early life disruption of the gut microbiota by antibiotics (Figure 6A). Moreover, acetate also prevented medial injury assessed at day 30 posttransplantation, as determined by a preservation of medial thickness, cellularity, and elastic laminae integrity in the media of allograft arteries (Figure 6B). Acetate did not effect the accumulation of CD4 T cells, CD8 T cells, Tregs or macrophages (Figure S2, SDC, https://links.lww.com/TP/C423). This establishes that neutrophil-mediated allograft vascular injury caused by early life disruption of the gut microbiota with antibiotics is inhibited by acetate.

DISCUSSION

In this study, we provide new information on how early life disruption of the gut microbiota affects its metagenome and related immunoregulatory properties and how the resulting changes influence immune injury of allograft blood vessels. Specifically, dysbiosis of the gut microbiota that exacerbates neutrophil-mediated vascular injury is characterized by a reduction in metabolic genes related to the production of SCFAs, and acetate prevents neutrophil-mediated allograft vascular injury. These findings advance our knowledge of how the gut microbiota controls immune responses that injure organ transplants by establishing the role of acetate as an immune regulator.

We observe a sex-dependent effect on the influence of the gut microbiota on neutrophil responses, which is consistent with the differential susceptibility of females to some immunopathologies through effects of the gut microbiota.44,45 This differential susceptibility may relate specifically to neutrophils because females are more susceptible than males to systemic lupus erythamatosus and the etiology of this autoimmune disease involves the development of immunopathological responses toward autoantigens from neutrophils.46 Recently, RNA sequencing has identified an enhanced responsiveness of female neutrophils to type I interferon stimulation as compared with male counterparts. This is related to altered bioenergetics of female and male neutrophils that are driven by estradiol exposure.47,48 With regard to vascular effects, eliminating the gut microbiota increases outward hypertrophic remodeling of resistance arteries in female mice to a greater extent than males, so the gut microbiota may contribute to sex differences in certain vascular conditions.49

In our studies, co-housing mice that were treated with antibiotics early in life with untreated counterparts prevented neutrophil-mediated allograft vascular injury, indicating that changes in the composition of the gut microbiota were responsible for exacerbating neutrophil responses. Treatment of mice with antibiotics in our experiments reduces the amount of bacteria to nearly undetectable levels, so dysbiosis of the gut microbiota in adults is likely to be a result of differential population of the intestinal tract by bacterial species caused by the delay in exposure to bacteria until after weaning.28 Indeed, altering the timing of exposure of animals to gut colonizing bacteria around the time of weaning leads to differences in the establishment of gut microbiota populations later in life.50 This may be a result of the differential availability of intestinal niches available for population by bacteria at different times during development.51

The anaerobic bacterium A. muciniphila contributes to the production of acetate as a by-product of mucin degradation and was the most abundant bacterial species reduced by early life disruption of the gut microbiota with antibiotics. In humans, A. muciniphila is highly abundant in the colonic mucosa, and it is being investigated as a probiotic for several immune-mediated conditions.14,52-55 The reduction in A. muciniphila caused by antibiotics early in life was related to a reduced abundance of genes that encode mucin-degrading enzymes, such as β-N-acetylhexosaminidase, which catalyzes the cleavage of terminal β-D-N-acetylglucosamine and β-D-GalNAc residues of oligosaccharides that are important for bacterial taxa to derive resources from mucin.56,57 The second- and thirdmost abundant bacterial species that were reduced by disruption of the gut microbiota with antibiotics were L. johnsonii and L. murinus. Lactobacillus bacteria, which are implicated in many aspects of health and disease.58 One of the defining features of bacteria within the Lactobacillus genus is the production of lactic acid through the fermentation of indigestible carbohydrates such as dietary fiber. Lactic acid is then converted to SCFAs, including acetate, by other bacteria in the community.42L. johnsonii may also be capable of directly producing acetate.59,60 In patients infected with HIV, reduction in Lactobacillus species in the gut microbiota increases neutrophil accumulation in the intestinal mucosa, but the roles of L. murinus and L. johnsonii were not specifically examined.61

Our observation that directly providing acetate to mice reduces neutrophil-mediated injury of allografts indicates that this SCFA is immunoregulatory. Although the mechanism by which neutrophils exacerbate medial injury in our experiments was not specifically examined, these innate immune cells induce vascular smooth muscle apoptosis in allograft arteries through a reactive oxygen species-mediated pathway.27 Neutrophils can also produce high levels of nitric oxide, and inhibition of inducible nitric oxide synthase protects against early ischemic injury of liver allografts.62 Acetate inhibits neutrophil responses by activating the G-protein coupled receptor 43.43 The cellular mechanisms by which this SCFA inhibits neutrophils are poorly understood. Neutrophils can migrate toward a chemotactic gradient of acetate in vitro.63 Although this suggests a proinflammatory effect of this SCFA, elevated levels of acetate in the blood could inhibit the extravasation of neutrophils to sites of inflammation that are outside the blood vessel lumen. Acetate also induces neutrophil apoptosis in vitro.43 Administration of acetate in the drinking water or a high fiber diet is efficacious for attenuating neutrophil pathologies such as gout.10 In transplantation, Wu et al64 recently demonstrated that acetate from the gut microbiota reduces rejection of renal transplants through an effect on Treg. We were not able to quantify circulating levels of acetate because of technical challenges, and this is a limitation of our studies.

In summary, we show that early life disruption of the gut microbiota can enhance immune responses that injure transplant blood vessels through effects on the immunoregulatory SCFA acetate. These findings provide a basis for further study of immunoregulatory effects of the gut microbiota in transplantation and of immune regulatory properties of A. muciniphila, Lactobacillus spp., and acetate.

ACKNOWLEDGMENTS

We are grateful to the SFU Big Data Hub, the Cedar supercluster, and Compute Canada for maintenance of bioinformatics data and software. Expert animal care and technical assistance were provided by the Animal Research Center at SFU.

References

1. Rooks MG, Garrett WS. Gut microbiota, metabolites and host immunity. Nat Rev Immunol. 2016;16:341–352.
2. Arrieta MC, Stiemsma LT, Dimitriu PA, et al.; CHILD Study Investigators. Early infancy microbial and metabolic alterations affect risk of childhood asthma. Sci Transl Med. 2015;7:307ra152.
3. Vangay P, Ward T, Gerber JS, et al. Antibiotics, pediatric dysbiosis, and disease. Cell Host Microbe. 2015;17:553–564.
4. den Besten G, van Eunen K, Groen AK, et al. The role of short-chain fatty acids in the interplay between diet, gut microbiota, and host energy metabolism. J Lipid Res. 2013;54:2325–2340.
5. Arpaia N, Campbell C, Fan X, et al. Metabolites produced by commensal bacteria promote peripheral regulatory T-cell generation. Nature. 2013;504:451–455.
6. Vieira AT, Galvão I, Macia LM, et al. Dietary fiber and the short-chain fatty acid acetate promote resolution of neutrophilic inflammation in a model of gout in mice. J Leukoc Biol. 2017;101:275–284.
7. Vieira AT, Macia L, Galvão I, et al. A role for gut microbiota and the metabolite-sensing receptor GPR43 in a murine model of gout. Arthritis Rheumatol. 2015;67:1646–1656.
8. Furusawa Y, Obata Y, Fukuda S, et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells. Nature. 2013;504:446–450.
9. Garrett WS. Immune recognition of microbial metabolites. Nat Rev Immunol. 2020;20:91–92.
10. Ottman N, Davids M, Suarez-Diez M, et al. Genome-scale model and omics analysis of metabolic capacities of akkermansia muciniphila reveal a preferential mucin-degrading lifestyle. Appl Environ Microbiol. 2017;83:e01014–e01017.
11. Le Leu RK, Hu Y, Brown IL, et al. Effect of high amylose maize starches on colonic fermentation and apoptotic response to DNA-damage in the colon of rats. Nutr Metab (Lond). 2009;6:11.
12. Holscher HD. Dietary fiber and prebiotics and the gastrointestinal microbiota. Gut Microbes. 2017;8:172–184.
13. Tabibian JH, Kenderian SS. The microbiome and immune regulation after transplantation. Transplantation. 2017;101:56–62.
14. Lu H, He J, Wu Z, et al. Assessment of microbiome variation during the perioperative period in liver transplant patients: a retrospective analysis. Microb Ecol. 2013;65:781–791.
15. Ling Q, Xu X, Wang B, et al. The origin of new-onset diabetes after liver transplantation: liver, islets, or gut? Transplantation. 2016;100:808–813.
16. Fricke WF, Maddox C, Song Y, et al. Human microbiota characterization in the course of renal transplantation. Am J Transplant. 2014;14:416–427.
17. Lee JR, Muthukumar T, Dadhania D, et al. Gut microbial community structure and complications after kidney transplantation: a pilot study. Transplantation. 2014;98:697–705.
18. Lee JR, Magruder M, Zhang L, et al. Gut microbiota dysbiosis and diarrhea in kidney transplant recipients. Am J Transplant. 2019;19:488–500.
19. Lecronier M, Tashk P, Tamzali Y, et al. Gut microbiota composition alterations are associated with the onset of diabetes in kidney transplant recipients. PLoS One. 2020;15:e0227373.
20. Lei YM, Chen L, Wang Y, et al. The composition of the microbiota modulates allograft rejection. J Clin Invest. 2016;126:2736–2744.
21. Qiu F, Lu W, Ye S, et al. Berberine promotes induction of immunological tolerance to an allograft via downregulating memory CD8+ T-cells through altering the gut microbiota. Front Immunol. 2021;12:646831.
22. Guo Y, Wang Q, Li D, et al. Vendor-specific microbiome controls both acute and chronic murine lung allograft rejection by altering CD4+ Foxp3+ regulatory T cell levels. Am J Transplant. 2019;19:2705–2718.
23. Alhabbab R, Blair P, Elgueta R, et al. Diversity of gut microflora is required for the generation of B cell with regulatory properties in a skin graft model. Sci Rep. 2015;5:11554.
24. Zhang Z, Liu L, Tang H, et al. Immunosuppressive effect of the gut microbiome altered by high-dose tacrolimus in mice. Am J Transplant. 2018;18:1646–1656.
25. Healy DG, Watson RW, O’Keane C, et al. Neutrophil transendothelial migration potential predicts rejection severity in human cardiac transplantation. Eur J Cardiothorac Surg. 2006;29:760–766.
26. von Rossum A, Laher I, Choy JC. Immune-mediated vascular injury and dysfunction in transplant arteriosclerosis. Front Immunol. 2014;5:684.
27. So M, Lee TD, Hancock Friesen CL. Neutrophils are responsible for impaired medial smooth muscle cell recovery and exaggerated allograft vasculopathy in aortic allografts exposed to prolonged cold ischemia. J Heart Lung Transplant. 2013;32:360–367.
28. Rey K, Manku S, Enns W, et al. Disruption of the gut microbiota with antibiotics exacerbates acute vascular rejection. Transplantation. 2018;102:1085–1095.
29. von Rossum A, Rey K, Enns W, et al. Graft-derived IL-6 amplifies proliferation and survival of effector T cells that drive alloimmune-mediated vascular rejection. Transplantation. 2016;100:2332–2341.
30. Bolyen E, Rideout JR, Dillon MR, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol. 2019;37:852–857.
31. Callahan BJ, McMurdie PJ, Rosen MJ, et al. DADA2: high-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–583.
32. Katoh K, Misawa K, Kuma K, et al. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 2002;30:3059–3066.
33. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30:1312–1313.
34. Quast C, Pruesse E, Yilmaz P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41:D590–D596.
35. Douglas GM, Maffei VJ, Zaneveld JR, et al. PICRUSt2 for prediction of metagenome functions. Nat Biotechnol. 2020;38:685–688.
36. Caspi R, Altman T, Billington R, et al. The MetaCyc database of metabolic pathways and enzymes and the BioCyc collection of pathway/genome databases. Nucleic Acids Res. 2014;42:D459–D471.
37. Bairoch A. The ENZYME database in 2000. Nucleic Acids Res. 2000;28:304–305.
38. Beghini F, McIver LJ, Blanco-Míguez A, et al. Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3. Elife. 2021;10:e65088.
39. Segata N, Waldron L, Ballarini A, et al. Metagenomic microbial community profiling using unique clade-specific marker genes. Nat Methods. 2012;9:811–814.
40. Franzosa EA, McIver LJ, Rahnavard G, et al. Species-level functional profiling of metagenomes and metatranscriptomes. Nat Methods. 2018;15:962–968.
41. Shin J, Noh JR, Chang DH, et al. Elucidation of Akkermansia muciniphila probiotic traits driven by mucin depletion. Front Microbiol. 2019;10:1137.
42. Markowiak-Kopeć P, Śliżewska K. The effect of probiotics on the production of short-chain fatty acids by human intestinal microbiome. Nutrients. 2020;12:1107.
43. Maslowski KM, Vieira AT, Ng A, et al. Regulation of inflammatory responses by gut microbiota and chemoattractant receptor GPR43. Nature. 2009;461:1282–1286.
44. Yurkovetskiy L, Burrows M, Khan AA, et al. Gender bias in autoimmunity is influenced by microbiota. Immunity. 2013;39:400–412.
45. Johnson BM, Gaudreau MC, Gudi R, et al. Gut microbiota differently contributes to intestinal immune phenotype and systemic autoimmune progression in female and male lupus-prone mice. J Autoimmun. 2020;108:102420.
46. Darrah E, Andrade F. NETs: the missing link between cell death and systemic autoimmune diseases? Front Immunol. 2012;3:428.
47. Blazkova J, Gupta S, Liu Y, et al. Multicenter systems analysis of human blood reveals immature neutrophils in males and during pregnancy. J Immunol. 2017;198:2479–2488.
48. Gupta S, Nakabo S, Blanco LP, et al. Sex differences in neutrophil biology modulate response to type I interferons and immunometabolism. Proc Natl Acad Sci U S A. 2020;117:16481–16491.
49. Edwards JM, Roy S, Tomcho JC, et al. Microbiota are critical for vascular physiology: germ-free status weakens contractility and induces sex-specific vascular remodeling in mice. Vascul Pharmacol. 2020;125-126:106633.
50. Schloss PD, Schubert AM, Zackular JP, et al. Stabilization of the murine gut microbiome following weaning. Gut Microbes. 2012;3:383–393.
51. Berry D, Stecher B, Schintlmeister A, et al. Host-compound foraging by intestinal microbiota revealed by single-cell stable isotope probing. Proc Natl Acad Sci U S A. 2013;110:4720–4725.
52. Collado MC, Derrien M, Isolauri E, et al. Intestinal integrity and Akkermansia muciniphila, a mucin-degrading member of the intestinal microbiota present in infants, adults, and the elderly. Appl Environ Microbiol. 2007;73:7767–7770.
53. Derrien M, Vaughan EE, Plugge CM, et al. Akkermansia muciniphila gen. nov., sp. nov., a human intestinal mucin-degrading bacterium. Int J Syst Evol Microbiol. 2004;54:1469–1476.
54. Earley H, Lennon G, Balfe Á, et al. The abundance of Akkermansia muciniphila and its relationship with sulphated colonic mucins in health and ulcerative colitis. Sci Rep. 2019;9:15683.
55. Depommier C, Everard A, Druart C, et al. Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: a proof-of-concept exploratory study. Nat Med. 2019;25:1096–1103.
56. Chen X, Xu L, Jin L, et al. Efficient and regioselective synthesis of β-GalNAc/GlcNAc-lactose by a bifunctional transglycosylating β-N-Acetylhexosaminidase from bifidobacterium bifidum. Appl Environ Microbiol. 2016;82:5642–5652.
57. Xu W, Yang W, Wang Y, et al. Structural and biochemical analyses of β-N-acetylhexosaminidase Am0868 from Akkermansia muciniphila involved in mucin degradation. Biochem Biophys Res Commun. 2020;529:876–881.
58. Fine RL, Mubiru DL, Kriegel MA. Friend or foe? Lactobacillus in the context of autoimmune disease. Adv Immunol. 2020;146:29–56.
59. He T, Zhu YH, Yu J, et al. Lactobacillus johnsonii L531 reduces pathogen load and helps maintain short-chain fatty acid levels in the intestines of pigs challenged with Salmonella enterica Infantis. Vet Microbiol. 2019;230:187–194.
60. Date Y, Nakanishi Y, Fukuda S, et al. In vitro evaluation method for screening of candidate prebiotic foods. Food Chem. 2014;152:251–260.
61. Hensley-McBain T, Wu MC, Manuzak JA, et al. Increased mucosal neutrophil survival is associated with altered microbiota in HIV infection. Plos Pathog. 2019;15:e1007672.
62. Tsuchihashi S, Kaldas F, Chida N, et al. FK330, a novel inducible nitric oxide synthase inhibitor, prevents ischemia and reperfusion injury in rat liver transplantation. Am J Transplant. 2006;6:2013–2022.
63. Vinolo MA, Ferguson GJ, Kulkarni S, et al. SCFAs induce mouse neutrophil chemotaxis through the GPR43 receptor. PLoS One. 2011;6:e21205.
64. Wu H, Singer J, Kwan TK, et al. Gut microbial metabolites induce donor-specific tolerance of kidney allografts through induction of T regulatory cells by short-chain fatty acids. J Am Soc Nephrol. 2020;31:1445–1461.

Supplemental Digital Content

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.