Location: Diet, Genomics and Immunology Laboratory
2021 Annual Report
Objectives
Objective 1. Study the effect of resistant starch on the function of innate lymphoid cells, regulatory T cells, and regulatory macrophages in mucosal immunity and resistance to gastrointestinal infection. [NP107, C3, PS3B]
Objective 2. Examine the effect of cruciferous vegetables on the function of innate lymphoid cells, regulatory T cells, and regulatory macrophages in mucosal immunity and resistance to gastrointestinal infection. [NP107, C3, PS3B]
Objective 3. Define the effect of combining resistant starches with cruciferous vegetables on the function of innate lymphoid cells, regulatory T cells, and regulatory macrophages in mucosal immunity and resistance to gastrointestinal infection. [NP107, C3, PS3B]
Approach
The mucosal immune system is the first line of defense against a wide variety of bacterial, viral, and parasitic pathogens and must also regulate intestinal homeostasis. There is substantial cross-talk between the host immune system and the microbiome that modulates development of mucosal immunity and maintenance of intestinal homeostasis. Diet can affect the microbiome and, therefore, gut mucosal immunity and intestinal homeostasis. The composition of the microbiome can be altered by consumption of resistant starches (RS) or cruciferous vegetables (CV); but how this translates to changes in gut mucosal immunity and resistance to disease is largely unexplored. The goal of this project is to define how RS and CV affect the interaction between the gut microbiome and immune cells. This will be accomplished using rodent and porcine models to study the effect of feeding type 2 or 3 RS, or CV on activation of innate lymphoid cells (ILCs), as well as the activity/polarization of tissue macrophages (M's), and induction of T regulatory (Treg) cells at homeostasis and after challenge by enteric pathogens. This work will lead to the development of new biomarkers of immune status responsive to changes in nutrition, the microbiome, and identify nutrient-immune interactions potentially beneficial to human health. The studies will use a complementary approach to take advantage of the strengths of each animal model system. Mice will be used as a lower cost, high-throughput screening tool to evaluate the effect of RS and CV rich in dietary aryl hydrocarbon receptor (AhR) ligands on the microbiome and gut immune parameters. The results from these studies will be distilled into candidate foods to test mechanism-based effects in a pig model that are likely to yield data highly relevant to humans. The proposed mouse models in this project plan will provide flexibility to evaluate several classes of dietary RS and CV at various concentrations and combinations to evaluate mucosal responses to both bacterial and parasitic worm infections. Changes in mucosal cell populations of ILCs, Tregs and regulatory Mfs and their functional expression in explanted cells in vitro will provide a context for a diet-dependent mechanism in disease resistance. The effects RS and CV on the microbiome will be evaluated and correlated with changes to mucosal immunity. Subsequent studies in pigs using diet combinations optimized in mice and the use of a more human-like food matrix in pig feeding studies will inform recommendations for dietary RS and CV compositions predictive of improved intestinal health in humans. This will include challenge studies using infections in pigs caused by zoonotic E. coli and Trichuris suis (Ts) that are comparable to E. coli and whipworm infections in mice and humans. We have previously reported on the changes in metabolome and microbiome of Ts infected pigs affording us the opportunity to test the effects of dietary interventions on important diseases affecting humans.
Progress Report
Objective 1. Experiments were conducted using a Total Western Diet (TWD) based on National Health and Nutrition Examination Survey (NHANES) data that mimics the protein, fat, sugar, vitamin and mineral composition of a typical American diet. The effect of addition of a type 2 resistant starch, raw potato starch (RPS, 2, 5, 10% of diet), to the TWD on morphological and gene expression changes to the colon, cecum and microbiome were evaluated.
To evaluate the effects of RPS on the microbiome, cecal contents from mice fed the basal diet or the basal diet plus 2, 5, or 10% RPS were subjected to 16S sequencing. The principal component analysis (PCA) of the cecal 16S rRNA sequencing data showed 4 discreet groupings based on the dose of RPS in the diet. There were sex dependent differences in the relative abundance (RA) of various operational taxonomic units (OTUs) at different phylogenic levels but, in general, the effect of increasing amounts of RPS was similar, with either the RA of OTUs increasing or falling with different RPS doses. Of note, at the genus phylogenic level the Lachnospiraceae NK4A136 group went from 7% in mice fed the basal diet to 50% in mice fed the basal diet with 10% RPS in a dose dependent manner. This resulted in declines in other genera including Clostridium senso stricto 1, Rosburia, Bilophila, and Blauta. The Lachnospiraceae NK4A136 group are potential butyrate producers and the RPS dose dependent increase in the Lachnospiraceae NK4A136 group correlated with a RPS dose dependent increase in cecal butyrate production. In total, addition of RPS to the diet had a major impact on the microbiome composition and metabolites.
To understand the effect of dietary RPS on gene expression, RNASeq analyses was performed on cecum tissue. A PCA of cecum gene expression showed a RPS dose-dependent segregation into 4 groups analogous to the PCA results obtained from 16S sequencing. Animals fed 2% RPS had 9 genes that were differentially up (5) or downregulated (4) more than 1.5-fold. Animals fed 5% RPS had 294 genes that were differentially up (153) or downregulated (140) more than 1.5- fold. Animals fed 10% had 923 genes that were differentially up (601) or downregulated (342) more than 1.5-fold. We functionally annotated differentially expressed genes using our Porcine Translational Research Database. The database serves as a manually to translate data found in rodents or pigs to human. Classes of genes that are overrepresented in upregulated genes in the 10% RPS group include; immunoglobulin kappa and lamba variable region genes, genes involved in oxygen- or nitric oxide dependent intracellular killing, integrins, structural proteins and Type 1 IFN-induced. Classes of genes that are overrepresented in genes that are downregulated in the 10% RPS group include; vitamin A and vitamin D, lipid metabolism, glycolysis and bile acid metabolism. These data indicate that RPS has wide-ranging effects on genes involved in immunity and metabolism.
We examined the potential beneficial or deleterious effects of butyrate or proprionate (at levels at or below the cecal contents of our RPS-treated mice), on the response of pig intestinal epithelial cells to E. coli-derived ligands. These cells have been shown to reproduce the effect of ex vivo-derived cells in several of our previous experiments. We measured cells exposed to a comprehensive panel of immunostimulants from Escherichia coli including outer membrane vesicles (OMVs), muramyldipeptide (MDP, a NOD2 ligand), Pam3CysSerLys4, (Pam3CSK4, a TLR1/2 ligand), nigericin (an activator of the NLRP3 inflammasome) and ultrapure butyrate lipopolysaccharide (LPS), a TLR4 ligand. Our results suggest that undifferentiated IPEC-J2 cells exposed to butyrate, but not proprionate, synthesize more IL-8 in response to OMVs and Pam-3CSK4, but not to LPS, nigericin or MDP. These data indicate a role of butyrate in selectively modulating TLR2 signaling in intestinal epithelial cells.
Contrary to our hypothesis that RPS should improve the outcome of Citrobacter rodentium (Cr)-induced colitis, we found that mice fed the 10% RPS diet led to an increased colonization of the colon, enlarged spleens, and increased colon pathology. Additional studies indicate this effect is restricted to the 10% RPS level. In a preliminary study using the parasite Trichuris muris, a higher worm burden was observed only in mice fed the 10% RPS diet. Future research will focus on understanding how consumption of 10% RPS diet interferes with resistance to these two model organisms.
Continued progress was also made on Objective 2. Initiation of pig studies is delayed due to difficulties conducting large animal studies during the pandemic. We continued to investigate the effect of cruciferous-derived compounds on Citrobacter rodentium (Cr) infection model using Cr-susceptible C3H/HeN mice. Dietary indole-3-carbinol (I3C) significantly inhibited body weight loss and the increase in spleen size in Cr infected mice but colonization of the colon was not affected. In addition, I3C treatment reduced the inflammatory response to Cr infection by maintaining anti-inflammatory cytokine IL-22 mRNA levels while reducing expression of other pro-inflammatory cytokines including IL17A, IL6, IL1ß, TNF-a, and IFN. Serum cytokine levels of IL17, TNF-a, IL12p70, and G-CSF also were down-regulated by I3C in Cr-infected mice. Dietary I3C specifically enhanced the Cr-specific IgG response to Cr infection. In general, dietary I3C reduced the Cr-induced pro-inflammatory response in susceptible C3H/HeN mice and alleviated the physiological changes and tissue damage induced by Cr infection but not Cr colonization.
We also studied the relationship between diet, the gut microbiome and Cr infection. A Principal Coordinate Analysis (PCA, Bray-Curtis distance) of uninfected or Cr infected mice fed an AIN-93M diet ± I3C showed significant changes in beta-diversity resulting from a Cr infection that was not further altered by dietary I3C. Importantly, I3C alone did not affect beta diversity. Alpha diversity was not affected by Cr infection or I3C treatment. Cr decreased the abundance of Bacteroidetes but increased the abundance of Firmicutes and Actinobacteria. The relationship between specific Cr-induced changes to the microbiome and Cr-induced pathology remain unclear but data analysis is continuing to look for specific changes at various phylogenic levels.
We then examined whether a whole food rich in glucosinolates, such as red cabbage cruciferous microgreens (RCMG) would elicit more robust changes to the gut microbiome than I3C in mice fed either a low-fat (LFD) or high-fat diet (HFD) without or with added RCMG. There was significant difference in beta diversity between the LFD and HFD and between mice fed the LFD and LFD supplemented with RCMG. However, there were no significant differences between HFD and HFD supplemented with RCMG. Alpha diversity indices were also analyzed. In mice fed a LFD, all four indices of alpha diversity (Chao1, ACE, Shannon and Simpson’s indexes) showed significant increases in alpha diversity for RCMG supplementation. In mice fed a high fat diet with RCMG only the Simpson index did not show changes in alpha diversity. These results suggest RCMG may modulate the gut microbiome but additional factors such as the type and quantity of dietary fat may also play a role in the effects of MG on the microbiome. Additional analyses are underway. In summary, Cr infection can alter the gut microbiome and these infection-induced changes can be modified by diet. The cruciferous glucosinolates-derived compound, I3C, provided protection against a Cr infection, but the protective effect of I3C did not appear to occur through modulation of the microbiome. RCMG, as with other cruciferous vegetables, are rich sources of glucosinolates, protect against high fat diet-induced risk factors such as inflammation and can alter the gut microbiome.
Recently there have been numerous attempts to improve the assembly, annotation and analysis of the porcine genome. Despite these efforts, several recent studies indicate that there is a significant amount of work that needs to be done towards a “finished version.” To that end, we have documented 6,820 genes that have at least one error in the 3 latest builds of the genome; NCBI and Ensembl builds 11.1 and MARC build 1.0. Our analysis of 7,455 Ensembl loci, reveals that only 48.1% of genes are correctly assembled and annotated. We have recently extended this analysis to the 4,398 genes in the Ensembl assembly of MARC 1.0 and have found a similar error rate. In contrast analysis of 9,087 genes in NCBI build 11.1, reveals that 70% of genes are correctly assembled and annotated. A comprehensive examination of these errors revealed 7 systematic sources. Correcting these errors will lead to dramatic improvements in the assembly and annotation of the porcine genome.
Using our improved build of the genome, we conducted an expanded analysis of 4,038 genes involved in the immune response of humans, mice or pigs, reveals that 81% of genes are conserved between the 3 species. When a gene is missing from one of the 3 genomes, pigs are 4.5X more likely to have the human gene than mice (3.8% vs 0.8%). In the first analysis of 3,480 genes involved in nutrition or metabolism of humans, mice or pigs, 72.2 % of genes are conserved between the 3 species. When a gene is missing from one of the 3 genomes, pigs are 6.5X more likely to have the human gene than mice (5.9% vs 0.9%). The analysis also illuminates pathways where mice may be a more appropriate model for humans. For example, pigs lack the enzymes that initiate starch and lipid metabolism in the upper gastrointestinal tract. These data strongly indicate that overall, swine are a scientifically acceptable intermediate species (rodent-human) for conducting scientific research on certain relationships between nutrition and immunity.
Accomplishments
1. Gut immune cells release alarming molecules to activate nerve signaling. The luminal surface of the intestine contains enterochromaffin (EC) cells that function as chemosensors to translate cues from intestinal contents into nerve-activating chemicals like serotonin to modify intestinal physiology. ARS scientists in Beltsville, Maryland, worked with National Institute of Health scientists from the National Cancer Institute in Bethesda, Maryland, to show that mice infected with the common whipworm trigger innate immune cells to produce a protein cytokine messenger molecule called IL-33 which, in turn, activates EC in the intestine to produce serotonin that activates enteric nerves to stimulate intestinal gut motility. Because serotonin and cytokine production are partly dependent on diet and nutrition, there is the potential to target mechanisms that control parasite infection of the gut and neuroendocrine disorders such as irritable bowel syndrome through diet. This work is important to those interested in the maintenance of healthy homeostasis in the intestine.
Review Publications
Elolimy, A., Washam, C., Byrum, S., Chen, C.T., Dawson, H.D., Bowlin, A.K., Randolph, C., Saraf, M.K., Yeruva, L. 2020. Formula diet alters ileal metagenome and transcriptome at weaning and during the postweaning period in a porcine model. mSystems. 5:e00457-20. https://doi.org/10.1128/mSystems.00457-20.
Ye, S., Matthan, N., Lamon-Fava, S., Solano Aguilar, G., Turner, J., Walker, M.E., Chai, Z., Lakshman, S., Urban Jr, J.F., Lichtenstein, A.H. 2021. Western and heart healthy dietary patterns differentially affect the expression of genes associated with lipid metabolism, interferon signaling and inflammation in the jejunum of Ossabaw pigs. Journal of Nutritional Biochemistry. 90:108577. https://doi.org/doi: 10.1016/j.jnutbio.2020.108577.
Chen, Z., Luo, J., Li, J., Kim, G., Stewart, A., Urban Jr, J.F., Huang, Y., Chen, S., Wu, L., Chester, A., Trinchieri, G., Li, W., Wu, C. 2020. Interleukin-33 promotes seratonin release from enterochromaffin cells for intestional homeostasis. Immunity. 54(1):151-163.e6. https://doi.org/doi:10.1016/j.immuni.2020.10.014.
