Location: Immunity and Disease Prevention Research2018 Annual Report
Objective 1: Determine how diet quality (assessed using the Healthy Eating Index), nutritional status (assessed using biomarkers in a cross-sectional study) and adherence to a diet following Dietary Guidelines recommendations for intake of fat and fat-soluble vitamins affect immune function and inflammation. 1A: In the cross-sectional WHNRC Phenotyping Study (CSPS) determine if diet quality and intestinal dysbiosis are independently associated with systemic immune activation. 1B: In the WHNRC DGA Intervention Trial (IT) of adults with indicators of metabolic syndrome, determine if following the DGA diet improves markers of systemic and intestinal inflammation relative to a Typical American (TA) diet. Objective 2: Determine the degree of modulation and the mechanism of activation or inhibition of blood monocytes by different types of dietary fatty acids (including saturated fatty acids and docosahexaenoic acid [DHA]) and by fruit-derived dietary polyphenols or their metabolites. 2A: Determine (1) whether the high fat/sugar challenge meal administered during the CSPS induces postprandial monocyte activation; (2) whether this activation is mediated by saturated fatty acids; (3) whether and how the challenge meal-induced monocyte activation is suppressed by docosahexaenoic acid; and (4) in the DGA IT whether the diets affect challenge meal-induced monocyte activation. 2B: In subjects from the CSPS determine whether addition of DHA to the high fat/sugar challenge meal inhibits monocyte activation. 2C: In cell culture studies determine whether bioactive phytochemicals known to inhibit signaling pathways in monocytes, or their metabolites, also suppress SFA-induced monocyte activation. Objective 3: Removed per the PDRAM. Objective 4: Determine how diets enriched with polyphenol-rich fruits such as strawberries and grapes affect monocyte/macrophage function in obesity, determine possible chemical components of the fruits responsible for changes in function, and determine the mechanisms involved in changes in function. 4A: Determine if dietary strawberries and grapes affect monocyte/macrophage function, bacterial burden, morbidity and mortality in diet-induced obese mice infected with gram-negative bacteria. 4B: Determine if the polyphenols of strawberries and grapes are responsible for modulating monocytes in diet-induced obese mice infected with gram-negative bacteria. 4C: Determine mechanisms by which components of strawberries and grapes may modulate the function of monocytes isolated from diet-induced obese mice.
Objective 1 will utilize samples exclusively from the two human studies, the Western Human Nutrition Research Center (WHNRC) Cross-Sectional Phenotyping Study and the WHNRC Dietary Guidelines for Americans (DGA) Intervention Trial. Thus the designs of these studies are described under Objective 1 and the sample size calculations given relate to the goals of Objective 1. 1A: Such activation takes several forms and we will differentiate among pathways defined by the activity of pro-inflammatory T-helper (Th) cells (Th1, Th2 and Th17) and T-regulatory (Treg) cells. We hypothesize that those with low diet quality (including high solid fat and added sugar [SOFA] and low n-3 polyunsaturated fatty acids [PUFA]), or low intake (or status) of key nutrients (including vitamin D) will have greater immune activation after adjustment for appropriate covariates (e.g., age, BMI and sex). In addition, we hypothesize that dysbiosis of the gut microbiota (e.g., high levels of Proteobacteria) will be associated with gut inflammation that, in turn, will be associated with systemic immune activation. Microbiota will be assessed in stool using 16S rRNA gene sequence and inflammation by stool calprotectin and neopterin levels. 1B: DGA diet is optimized to minimize inflammation by decreasing SOFA, and increasing vitamin D, n-3 PUFA, fruit and vegetable intake. Objectives 2 will also utilize samples from both of these studies. In addition, Objectives will utilize cell culture methods to examine effects of dietary components on regulating cellular functions, including the effects of DHA (Objective 2B) and phytochemicals (Objectives 2C) on monocyte activation. Objective 4 will utilize a mouse model to examine the effect of diets rich in strawberry and grape preparations (freeze-dried whole fruit or fruit extracts) on monocyte/macrophage function in mice fed standard and high-fat diets and infected with gram-negative bacteria. Cell culture studies will also be used to examine the effect of fruit-derived phytochemicals on monocyte/macrophage function.
In support of Objective 1, approximately 300 study volunteers will complete the Phenotyping Study this year. The study design calls for recruiting equal numbers of male and female volunteers in specific age categories (18-33 years (y); 34-49y; and 50-65y) and specific body mass index (BMI) categories (18.5-24.9; 25-29.9; and 30-39.9 kilograms/square meters). Recruitment has been low among men and in the highest BMI category and advertising was adjusted this year to address these problems. A second issue that arose this year was a lack of sufficient time among ARS scientists to keep pace with analysis of dietary intake data. To address this issue, a collaboration began with a dietary analysis expert from the University of California, Davis. Laboratory work under Sub-objective 1A continued this year. Immune activation and inflammation markers were measured by flow cytometry and plasma cytokine analysis, respectively. Gut health markers include stool consistency, fecal pH, gut inflammation (e.g. fecal calprotectin) and indirect markers of endotoxin exposure (e.g. plasma LPS-binding protein). A non-invasive assay of colonocyte RNA expression-preserved stool is also being tested. The characterization of gut microbiota from Phenotyping Study subjects continued under Sub-objective 1A. Last year, methods for extracting bacterial DNA from stool were evaluated, and this year, laboratory and bioinformatic methods for characterizing the gut microbiome by DNA sequence analysis were compared. A manuscript describing these results was completed and shared via bioRxiv, a site for biologists to share preliminary findings. Using these methods, stool DNA was prepared from the first 140 study volunteers. The DNA sequence of variable region 4 of the 16S bacterial ribosome gene was determined and used to characterize the intestinal bacterial community. A preliminary analysis (using the DADA2 and phyloseq packages in the R statistical software platform) found that the gut microbial communities from individuals who habitually consumed a high level of soluble fiber (above 3.9 grams per 1000 calories (g/1000kcal)) were significantly different from those who consumed low levels of soluble fiber (below 2.5g/1000kcal per day). Those consuming high fiber had higher levels of Coprococcus and Lachnospria species and lower levels of Ruminococcus species from the family Lachnosipriaceae. These findings are of interest because Coprococcus species ferment dietary fiber in the colon, and may be beneficial to human health, while Ruminococcus species may degrade mucus that protects the intestinal epithelium, which may cause intestinal damage and inflammation, thus impairing gut health. In addition to the characterization of the gut microbiome via DNA sequence analysis under Objective 1, bacterial relationships with food components and host immunity can be directly measured through in vitro experimentation. In preparation for such experiments, a pilot study was conducted in which four types of bacterial culture media were inoculated with fresh stool samples from five individuals and cultured anaerobically. The medium that produced a bacterial community most closely resembling that of the corresponding stool sample was chosen for use in ongoing experiments to select storage conditions for clinical samples that maximize the survival of anaerobic bacteria. A principal goal of Objective 2 is to determine the main cause of postprandial (PP) inflammation seen after a high-fat meal. Two causes are likely. First, absorption of bacterial endotoxin (ET) from the gut may stimulate immune cells in the blood via Toll-like receptor 4 (TLR4) to cause cytokine production, which promotes inflammation. Second, free fatty acids (FFA) in PP blood that derive from the recently consumed meal may also directly stimulate the TLR4 receptor. Results from the Phenotyping Study show that plasma ET concentrations in healthy adults were below the sensitivity of the standard ET assay, suggesting that any ET must be present at very low levels. To increase the sensitivity of ET analysis, ARS scientists developed an assay to measure ET indirectly. The new assay measured cytokine production in PP whole blood, which contains immune cells, and measured the ability of Polymyxin B (PMB), an inhibitor of ET activity, to block this activity. Results with the first approach revealed that cytokine production in PP blood from Phenotyping Study volunteers after a meal with 60% of energy from fat was not significantly higher than seen in fasting blood. However, the production of cytokines was partially diminished by PMB, suggesting that the blood samples did contain ET at a very low level. The preliminary results from a second study funded by the National Institute of Food and Agriculture (2032-53000-001-14I) found that cytokine production in PP blood after a meal with less fat (36% of energy) was not affected by PMB, suggesting no role for ET with a lower fat meal. However, results from PP plasma (as compared to whole blood) from the same studies showed temporal changes in cytokine concentration in plasma that followed a pattern similar to that of plasma FFA. To assess the role of FFA, whole blood samples were incubated with lipoprotein lipase to release FFA from cell membranes, showing a significant increase in cytokine production. These results suggest that plasma FFA, independent of ET, is one factor triggering PP inflammation. Thus, decreasing the consumption of foods that increase PP levels of FFA will diminish inflammation. A second approach to examining PP inflammation, characterizing temporal changes in gene expression in PP blood, was also examined. A pilot study of 20 volunteers measured inflammatory gene expression in whole blood by targeted polymerase chain reaction (PCR) analysis and by global messenger RNA (mRNA) sequencing. Data analysis is in progress and preliminary results show distinct patterns of expression of inflammatory genes among study participants. The ARS scientist responsible for Objective 4 retired in December, 2017. This objective focused on how diet-induced obesity in a mouse model might impair certain immune responses and increase the risk of developing infections. To determine possible defects in monocyte function induced by obesity, C57BL/6 mice (10 per group) were fed a low fat or a high fat diet and monocytes were isolated from their spleens. Phagocytic activity of the monocytes was determined using E.coli Bioparticles® that fluoresce after uptake by monocytes. The expression of TLR2 and TLR4, which signal bacterial recognition at the cell surface of monocytes, was measured by flow cytometry. The levels of several cytokines secreted by monocytes in response to bacteria were measured from isolated immune cells treated with bacterial lipopolysaccharide (LPS). A 9% increase in splenic monocytes was observed from the obese compared to the lean mice, suggesting greater immune-activation, though there was no difference in phagocytic activity or TLR2 and TLR4 expression between the two groups. Increases in four pro-inflammatory cytokines (TNF-Alpha, IL-6, MIP-1-Beta, and MIP-1-Alpha), and a decrease on one chemoattractant for inflammatory cells (CXCL1), were observed in monocytes from obese compared to lean mice after stimulation with LPS. These data suggest that diet-induced obesity produces an inflammatory phenotype in the circulating monocyte population and a potential defect in the production of the neutrophil chemotactic chemokine CXCL1. A manuscript documenting these findings was submitted for publication. Three subordinate projects have examined the effect of specific nutrients on immune function. One project in collaboration with the Thrasher Foundation (2032-53000-001-11T) is to evaluate the effect of vitamin A supplementation at birth on intestinal microbiota composition of Bangladeshi infants because such changes may affect vaccine responses. One manuscript has been published and two more have been submitted for publication. A second study (2032-53000-001-13T) with Micronutrients Initiative is examining the effect of zinc supplementation on immune function of children at risk of zinc deficiency. Enrollment and follow-up of infants was completed last year, laboratory work was completed this year, and data analysis is in progress. In a subordinate project on milk, health, and genetics with University of California, Davis (UCD), and funded by the California Dairy Research Foundation, the objective is to determine how lactase persistence genotypes and dairy consumption interact to impact human health. The first phase of the project is a meta-analysis of genome-wide association studies (GWAS) to determine the impact of the genetic neighborhood of the lactase persistence gene on markers of disease risk. In FY18, the CARDIoGRAMplusC4D 1000 Genomes-based GWAS dataset was analyzed; this data set includes 9.4 million variants with 60,801 coronary artery disease (CAD) cases analyzed under an additive model and 123,504 controls. The objective of the second phase of the project is to investigate the interaction between genotype, dairy consumption, and numerous health outcomes in the Phenotyping Study cohort (Sub-objective 1A). In FY18, additional participants were genotyped for lactase persistence (now n=172) and the lactose contents of individual food items reported by these 172 subjects in 24-hour recalls have been expert-curated. Data analysis is in progress. In another subordinate project with UCD, publicly available infant and adult stool metagenomes and metagenomes simulated from a custom database of known enzymes were used for computational experiments to determine sequencing format and depth needed to accurately quantitate functional annotation of enzymes in stool metagenomes. This information is needed to inform metagenomic sequencing of stool from participants in the Phenotyping Cohort (Objective 1A). A manuscript is in progress.
1. Development of software to analyze metatranscriptomes. The effect of a dietary component on a gut microbiome can be studied via a metatranscriptome - gene expression of all microbes in a sample. However, one barrier to the use of metatranscriptomics has been the availability of software to analyze the results. ARS scientists in Davis, California, with collaborators at the University of California, Davis, published such a software package, Simple Analysis of Metatranscriptomes through Sequence Annotation version 2 (SAMSA2; https://github.com/transcript/samsa2). SAMSA2 is designed for standalone use on a supercomputing cluster or large workstation, or to work with additional reference databases. This represents a significant advance over the original SAMSA software developed by this same team, which was dependent on third party resources for the computationally-expensive mapping step. SAMSA2 can thus be utilized by a broader group of biologists than the original SAMSA, increasing the breadth and potential impact of metatranscriptome analysis in multiple disciplines in the biological sciences.
Huda, M., Ahmad, S.D., Alam, J., Khanam, A., Alam, N., Raqib, R., Laugero, K.D., Stephensen, C.B. 2018. Infant cortisol stress-response is associated with thymic function and vaccine response. Stress: The International Journal on Biology of Stress. https://doi.org/10.1080/10253890.2018.1484445.
Westreich, S.T., Treiber, M.L., Mills, D.A., Lemay, D.G. 2018. SAMSA2: a standalone metatranscriptome analysis pipeline. BMC Bioinformatics. 19:175. https://doi.org/10.1186/s12859-018-2189-z.