Location: Diet, Genomics and Immunology Laboratory
2016 Annual Report
Objectives
Objective 1: Determine whether bioactive food components, such as vitamins A and D or lycopene, acting via vitamin A and D-receptor-mediated pathways and nuclear factor-kappaB signaling, exhibit dose-dependent inhibitory effects on macrophage-mediated remodeling of adipose tissue toward a pro-inflammatory phenotype in response to high fat diets in a swine model. [NP 107 Component 3, Problem Statement 3B].
Objective 2: Determine whether bioactives from food, including selenium, vitamin A, and thiol compounds, alter the immune response to model infectious organisms in mice through epigenetic changes, redox sensitive signaling pathways, and tissue remodeling by controlling cellular thiol levels, redox tone, and/or mitochondrial function. Compare responses of wild-type and genetically engineered mice with altered expression of one or more selenoproteins or proteins important for vitamin A or redox function, to identify specific proteins or pathways important for the effect of the nutrients under study on immune function and tissue remodeling. [NP 107 Component 3, Problem Statement 3B]
Approach
For Objective 1, in vitro and in vivo porcine models will be used to test the hypothesis that vitamin A or vitamin D or metabolites of dietary compounds that signal through retinoic acid receptor signaling pathways, such as lycopene, will promote an anti-inflammatory phenotype of adipose tissue macrophages and inhibit pro-inflammatory responses of adipose tissue macrophages to inflammatory ligands via inhibition of NF-kappaB signaling and epigenetic regulation of macrophage polarization.
For Objective 2, a mouse model will be used to test the hypothesis that bioactives from food, including selenium, vitamin A, and thiol compounds, alter the immune response to model infectious organisms in mice through epigenetic changes, redox-sensitive signaling pathways, and tissue remodeling by controlling cellular thiol levels, redox tone, and/or mitochondrial function. Responses of wild-type and genetically engineered mice with altered expression of one or more selenoproteins or proteins important for vitamin A or redox function, will be used to identify specific proteins or pathways important for the effect of the nutrients under study on immune function and tissue remodeling.
Progress Report
The central hypothesis being tested in objective 1 is that vitamin A (VA), either alone or in concert with cytokines like IL-4 or nutrients like vitamin D (VD), will foster the development of alternatively-activated macrophages (M2 macrophages) and/or inhibit the development of inflammatory macrophages (M1 macrophages) through epigenetic mechanisms. These macrophages (M1 or M2) differentially regulate the metabolic function of adipocytes and other cells; M1 macrophages are associated with insulin resistance and M2 macrophages with maintaining insulin sensitivity. To determine whether the bioactive VA metabolite, all-trans retinoic acid (ATRA) inhibits the development of M1 macrophages, primary porcine macrophages were treated with ATRA and then challenged in vitro, with an array of pro-inflammatory (lipopolysaccharide (LPS) + interferon-gamma (IFN-g), leptin, interleukin-1b (IL-1b), and nigericin (an inflammasome activator) or anti-inflammatory mediators (IL-4, IL-10 and IL-1B receptor antagonist). We found that ATRA reduced basal TNF protein, as well as TNF protein stimulated by LPS + IFN-g, IL-1b, and leptin but did not affect TNF protein stimulated by nigercin. We also found that ATRA decreased basal IL-1b protein, as well as IL-1b protein stimulated by LPS + IFN-g, but increased IL-1b protein stimulated by nigercin. These data suggest that ATRA acts to inhibit M1 macrophage polarization upstream of IL-1 b synthesis and inflammasome activation. These mechanistic data will allow us to focus our efforts. Studies to translate these findings to the human THP-1 human monocytic cell line are ongoing.
One epigenetic-based, regulatory mechanism whereby ATRA may mediate M1 macrophage polarization is thorough a small class of regulatory RNA known as microRNA. The exact number of porcine microRNAs is currently unknown. To address this uncertainty, we constructed a 1,033 gene porcine microRNAome by manually screening 1,956 human, mouse or bovine sequences for homology. Using our recently constructed non-redundant porcine RNA library and microRNA sequences as templates, we determined primary porcine macrophage microRNA induction or repression in response to M1- (LPS + IFN-g) or M2- (interleukin-4 (IL-4)), VA or VD) polarizing stimuli. We found 24 microRNA that were induced by LPS/IFN-g, including the microRNA, mir-155, that has previously been showed to be LPS induced in human and mouse macrophages and regulates responses to VD in human macrophages. Predicting miRNA-binding sites in mRNA is problematic because different methods often yield different results and cross-species conservation is somewhat limited. To address this, we screened 2,131 differentially regulated genes, using 3 separate software programs, for conserved mir-155 biding sites. We found that human-porcine conservation of Mir-155 targets sites occurs at five times the rate of human-mouse conservation. We also found 82 candidate genes with binding sites that were conserved among 3 species (pig, mouse and humans) including several that are associated with insulin resistance. These gene targets will be used to determine whether ATRA and/or VD regulate M1 macrophage differentiation through mir-155-mediated responses. These epigenetics-based integrated analysis tools will be useful for modeling human inflammatory and metabolic disease in pigs because of the close evolutionary and functional features of the two species. Studies to translate these findings to human THP-1 cells are being conducted.
We have obtained and are breeding multiple selenoprotein KO strains. Several KO strains, selenoprotein P, GPX1 and GPX2, were Helicobacter positive. A substantial effort was made to rid the strains of Helicobacter prior to their use in our models since Helicobacter spp. are known to influence gut immunity and inflammation, as thus, would be a confounding variable in our experiments. We successfully eradicated Helicobacter spp. from our mice and are currently evaluating the effect of KO of selenoprotein P in our two model systems. We are also currently evaluating the effect of KO of selenoprotein 15 in our two model systems. Additional experiments testing GPX1 and GPX2 mice will be initiated during the latter half of year 3. Experiments to identify sulfhydryl compounds which can reverse the effects of selenium deficiency including N-acetyl cysteine, 2-ME, lipoic acid, CoQ10 and its reduced form, and vitamin E in the Citrobacter rodentium (Cr) and/or Hpb models have not yielded any antioxidants that can substitute for the role of selenium. Due to the negative results obtained thus far and to allow additional time and resources to be devoted to use of the KO mouse models, this line of research has been terminated. Preliminary evidence, however, indicates that the GPX mimetic, ebselen, was able to reverse the effect of Se deficiency in the Cr model. This suggests that one of the four selenoprotein GPXs found in mice may be critical for controlling inflammation in response to a gastrointestinal bacterial infection.
We found that VA deficiency exacerbated Cr infections leading to higher rates of mortality, increased colonic colonization and pathology, and spread to the systemic compartment. Unfortunately, these initial observations were also observed and published by another research group immediately prior to our publication effort. We have some additional unique observations on goblet cell dysfunction that we are following up on in hopes of still being able to publish our results. To further explore the role of VA in mucosal immunity, we fed pregnant mice control (1X), or VAS (5X) AIN-93G-based diet beginning on day 14 of gestation but this had no effect on the course of Cr infections. A one-time high bolus dose (20,000 U/Kg) of VA administered just prior to infection had no effect, so these studies will no longer be pursued. To determine how quickly various aspects of immunity to Cr can be recovered, we fed VAD mice, a VAS diet starting the day of, or 7 days after infection. We found that administering VAS diet at the time of infection reversed many adverse effects of VAD on Cr infection while starting 7 days post-infection was only partially effective. These observations will be part of the eventual manuscript. A study to examine the effect of VAD on the microbiome was just completed and a principle component analysis of the data indicates the VAD does alter the microbiome. Additional analyses are being performed to identify specific families of bacteria that are altered by VAD.
In collaboration with ARS scientists from Environmental Microbial and Food Safety (EMFSL) and the Invasive Insect Biocontrol and Behavior Laboratory (IIBBL), we tested the effect of pomegranate peel extracts on pathogenic bacteria, both in vitro and in vivo. The results suggest that the pomegranate peel extract (POM) alters swarming properties of Cr in vitro and reduced the pathogenicity of Cr infections in vivo. To gain further insight into the mechanisms by which POM affected swarming, RNAseq data were generated for Cr grown under swarming conditions in the presence or absence of POM. Significant differences in gene expression have been identified and the analysis is continuing. A study to determine the effect of feeding POM on the microbiome was just completed and preliminary analysis indicates major POM-induced changes to the microbiome, including an increase in Akkermansia muciniphila. The abundance of this bacterial strain has been reported to inversely correlate with inflammation and metabolic disorders, associated with obesity, and is less abundant in patients with inflammatory bowel disease (IBD). These results raise the possibility that POM may have beneficial effects for IBD patients and could impact the development of metabolic syndrome.
Accomplishments
1. Conducted an in-depth analysis of the porcine, murine and human inflammasomes. Herein ARS scientists at the Beltsville Human Nutrition Center, Beltsville, Maryland, provide an expanded in silico analysis using an improved assembly of the porcine transcriptome that provides an in depth analysis of genes that are related to inflammasomes. Comparisons of the expansion or contraction of orthologous gene families indicated more similar rates and classes of genes in humans and pigs than in mice; however, several novel porcine or artiodactyl-specific paralogs or pseudogenes were identified. Despite these similarities, two out of four canonical inflammsome pathways that included the Absent in melanoma 2 (AIM2) and NLR family and CARD domain containing 4 (NLRC4) pathways were found to be missing in pigs. This work supports using swine to model both human immunological and inflammatory responses to infection. However, caution must be exercised as pigs differ from humans in several fundamental pathways.
2. Defined the host-adapted state of Citrobacter rodentium (Cr) by transcriptomic analysis. Citrobacter rodentium is a mouse pathogen that mimics many aspects of enteropathogenic Escherichia coli infections. Host-adapted (HA) Cr cells that are shed at the peak of infection have been reported to be hyperinfective. The exact mechanism underlying this phenomenon has remained elusive since the pathogen loses it’s HA ‘status’ immediately upon sub-culturing in laboratory media. By sequencing the entire transcriptome from the feces of infected mouse and media grown Cr, ARS scientists at the Beltsville Human Nutrition Center, Beltsville, Maryland, observed that the entire transcriptional machinery, as well as several transcriptional regulators to be differentially expressed, including fis, a global regulator that affects growth, replication, and expression of the type 3 secretion system, which is critical for inducing disease. Notably absent among the HA expressed genes were 19 fimbrial operons and non-fimbrial adhesions and several non-LEE encoded effectors. This work provides basic mechanistic observations related to Citrobacter rodentium infectivity that relate to human and animal health.
3. Demonstrated the selenoproteins regulate the activity of inflammatory cells that expel parasitic worms. The micronutrient selenium (Se) induces an alternatively activated anti-inflammatory macrophage or M2 cell that is dependent upon prostaglandin J2 (15d-PGJ2). ARS scientists at the Beltsville Human Nutrition Center, Beltsville, Maryland and colleagues at Pennsylvania State University, University Park, Pennsylvania showed that mice with an adequate level of Se in the diet expressed normal levels of M2 cells after infection with a parasitic nematode worm Nippostrongylus brasiliensis (Nb). In contrast, Nb-infected mice on a Se deficient diet had reduced M2 cells, and the worms produced more infective eggs and were more slowly expelled from the intestine. Nb Infection of genetically modified mice that did not express selenoproteins in the macrophage had fewer M2 cells and showed reduced worm clearance and higher egg production. Notably, blocking prostaglandin production in Nb-infected mice using the nonsteroidal anti-inflammatory drug indomethacin inhibited worm expulsion, but the protective response was restored by giving 15d-PGJ2. This work is important to those in animal and human health where parasitic worms are a problem as well as understanding the importance of adequate Se in the diet.
Review Publications
Dawson, H.D., Smith, A.D., Chen, C.T., Urban Jr, J.F. 2016. An in-depth comparison of the porcine, murine and human inflammasomes; lessons from the porcine genome and transcriptome. Veterinary Microbiology. pii:S0378-1135(16)30137-7. doi: 10.1016/j.vetmic.2016.05.013.
Vannella, K.M., Ramalingam, T.R., De Queiroz, P.R., Sciurba, J., Barron, L., Borthwick, L., Smith, A.D., Mentink-Kane, M., White, S., Thompson, R.W., Cheever, A.W., Bock, K., Moore, I., Fitz, L.J., Urban Jr, J.F., Wynn, T.A. 2016. Acidic Chitinase Limits Allergic Inflammation and Promotes Intestinal Nematode Expulsion. Nature Immunology. 17(5):538-44 doi: 101038/ni.3417.
Sun, R., Notari, L., Vanuytsel, T., Madden, K., Bohl, J., Ramalingam, T.R., Wynn, T.A., Urban Jr, J.F., Zhao, A., Shea-Donohue, T. 2016. Interleukin-13 receptor a1-dependent responses in the intestine are critical to parasite clearance. Infection and Immunity. 84(4):1032-1044. doi: 10.1128IAI.00990-009915.
Everts, B., Tussiwand, R., Dreesen, L., Fairfax, K.C., Huang, S.C., Smith, A.M., Oneil, C.M., Lam, W.Y., Edelson, B.T., Urban Jr, J.F., Murphy, K.M., Pearce, E.J. 2016. Migratory CD103+ dendritic cells suppress Helminth-driven Type 2 immunity through constitutive expression of IL-12. Journal of Experimental Medicine. 213(1):35-51. doi: 10.1084/jem.20150235..
Nelson, S.M., Shay, A.E., James, J.L., Carlson, B., Urban Jr, J.F., Prabhu, K.S. 2016. Selenoprotein expression in macrophages is critical for optimal clearance of parasitic helminth Helminth Nippostrongylus brasiliensis. Journal of Biological Chemistry. 291(6):2787-2798. doi: 10.1074/jbc.M115.684738.
Guo, L., Huang, Y., Chen, X., Hu-Li, J., Urban Jr, J.F., Paul, W.E. 2015. Innate immunological function of TH2 cells in vivo. Nature Immunology. 16:1051-1059.
Huang, Y., Guo, L., Qiu, J., Chen, X., Hu-Li, J., Siebenlist, U., Williamson, P.R., Paul, W.E., Urban Jr, J.F., Paul, W.E. 2015. IL-25-responsive, lineage-negative, KLRG1(hi) cells are multipotential “inflammatory” type-2 innate lymphoid cells. Nature Immunology. 16(2):161-169. doi: 10.1038/ni.3078
Chen, F., Wu, W., Millman, A., Craft, J.F., Chen, E., Patel, N., Boucher, J.L., Urban Jr, J.F., Kim, C.C., Gause, W.C. 2014. Neutrophils prime a long-lived effector macrophage phenotype that mediates accelerated helminth expulsion. Nature Immunology. 15(10):938-946. doi: 10.1038/ni.2984.
Smith, A.D., Yan, X., Chen, C.T., Dawson, H.D., Bhagwat, A.A. 2016. Understanding the host-adapted state of Citrobacter rodentium by transcriptomic analysis. Archives Of Microbiology. 198:353-362.
