Associations of the Dietary Inflammatory Index with total adiposity and ectopic fat through the gut microbiota, LPS, and C-reactive protein in the Multiethnic Cohort–Adiposity Phenotype Study

Chloe P Lozano; Lynne R Wilkens; Yurii B Shvetsov; Gertraud Maskarinec; Song-Yi Park; John A Shepherd; Carol J Boushey; James R Hebert; Michael D Wirth; Thomas Ernst; Timothy Randolph; Unhee Lim; Johanna W Lampe; Loïc Le Marchand; Meredith A J Hullar

doi:10.1093/ajcn/nqab398

. 2021 Dec 6;115(5):1344–1356. doi: 10.1093/ajcn/nqab398

Associations of the Dietary Inflammatory Index with total adiposity and ectopic fat through the gut microbiota, LPS, and C-reactive protein in the Multiethnic Cohort–Adiposity Phenotype Study

Chloe P Lozano ^1,^✉, Lynne R Wilkens ², Yurii B Shvetsov ³, Gertraud Maskarinec ⁴, Song-Yi Park ⁵, John A Shepherd ⁶, Carol J Boushey ⁷, James R Hebert ⁸, Michael D Wirth ⁹, Thomas Ernst ¹⁰, Timothy Randolph ¹¹, Unhee Lim ¹², Johanna W Lampe ¹³, Loïc Le Marchand ¹⁴, Meredith A J Hullar ¹⁵

PMCID: PMC9071464 PMID: 34871345

ABSTRACT

Background

Mechanisms linking a proinflammatory diet to obesity remain under investigation. The ability of diet to influence the gut microbiome (GM) in creating chronic low-grade systemic inflammation provides a plausible connection to adiposity.

Objectives

Assess whether any associations seen between the Energy-Adjusted Dietary Inflammatory Index (E-DII score), total fat mass, visceral adipose tissue (VAT), or liver fat (percentage volume) operated through the GM or microbial related inflammatory factors, in a multiethnic cross-sectional study.

Methods

In the Multiethnic Cohort–Adiposity Phenotype Study (812 men, 843 women, aged 60–77 y) we tested whether associations between the E-DII and total adiposity, VAT, and liver fat function through the GM, LPS, and high-sensitivity C-reactive protein (hs-CRP). DXA-derived total fat mass, MRI-measured VAT, and MRI-based liver fat were measured. Participants provided stool and fasting blood samples and completed an FFQ. Stool bacterial DNA was amplified and the 16S rRNA gene was sequenced at the V1–V3 region. E-DII score was computed from FFQ data, with a higher E-DII representing a more proinflammatory diet. The associations between E-DII score, GM (10 phyla, 28 genera, α diversity), and adiposity phenotypes were examined using linear regression and mediation analyses, adjusting for confounders.

Results

There were positive total effects (c) between E-DII and total fat mass (c = 0.68; 95% CI: 0.47, 0.90), VAT (c = 4.61; 95% CI: 2.95, 6.27), and liver fat (c = 0.40; 95% CI: 0.27, 0.53). The association between E-DII score and total fat mass was mediated by LPS, Flavonifractor, [Ruminococcus] gnavus group, and Tyzzerella. The association between E-DII score and ectopic fat occurred indirectly through Fusobacteria, Christensenellaceae R-7 group, Coprococcus 2, Escherichia-Shigella, [Eubacterium] xylanophilum group, Flavonifractor, Lachnoclostridium, [Ruminococcus] gnavus group, Tyzzerella, [Ruminococcus] gnavus group (VAT only), and α diversity (liver fat only). There was no significant association between E-DII score and adiposity phenotype through hs-CRP.

Conclusions

Associations found between E-DII and adiposity phenotypes occurred through the GM and LPS.

Keywords: Dietary Inflammatory Index, gut microbiome, mediation, visceral adipose, liver fat, total adiposity, lipopolysaccharides, C-reactive protein, multiethnic, cohort

See corresponding editorial on page 1261.

Introduction

Obesity, which affects more than one-third of adults in the United States (1), induces a chronic systemic inflammatory state, contributing to type 2 diabetes, hypertension, and cancer, among other conditions (1, 2). A proinflammatory diet has been associated with obesity (3), including central adiposity (4, 5), and accompanying comorbidities in the liver (6). Although the mechanisms linking a proinflammatory diet to obesity are still under investigation, a plausible pathway is through the effect of diet on the gut microbiome (GM), which can cause chronic low-grade systemic inflammation and promote adiposity (7, 8). For example, a diet high in total fat, energy, or carbohydrates results in a shift in the microbiome often associated with reduced gut epithelial integrity that can cause increased LPS translocation. LPS can induce an innate immune response that leads to metabolic endotoxemia, chronic low-grade inflammation associated with obesity (7). Metabolic endotoxemia has been found in people with liver diseases, particularly those with cirrhosis (9). In mouse models, induced metabolic endotoxemia was found to increase total body weight, liver weight, and visceral adipose tissue (VAT) (5). In contrast, a diet high in fiber and polyphenol-rich plant foods can be fermented by the gut microbes to produce SCFAs, which decreases gut epithelial permeability, promotes an anti-inflammatory response, and results in a healthy body weight (7, 10).

Effects of specific dietary components are difficult to characterize and quantify because food components are consumed together or in lieu of others, and the resulting intakes are often correlated (11). The development of dietary indices has helped to facilitate research on the effect of whole dietary patterns on health outcomes (11). Previously, we reported that diet quality was strongly associated with microbial diversity and specific genera using 4 theoretically based diet quality indices: Healthy Eating Index–2010 (HEI-2010); Alternative Healthy Eating Index–2010; alternate Mediterranean Diet score; and Dietary Approaches to Stop Hypertension score (12). We also reported that a lower score for any of these 4 diet quality indices (indicating a lower diet quality), was associated with higher VAT and nonalcoholic fatty liver (12). The Dietary Inflammatory Index (DII) is unique among dietary indices because it is a validated measure of the pro- or anti-inflammatory potential of the diet (13, 14). The DII is scored based on intake of commonly consumed nutritional constituents that are associated with key blood inflammatory markers (14). Previous studies have explored the effect of the DII on the GM (15), obesity (16), and markers of liver fat (6). However, as far as we know, this is the first publication to examine the association between DII and adiposity through the indirect effect of the GM. Therefore, we aimed to assess whether any associations seen between DII score, DXA-derived total fat mass, MRI-measured VAT, or MRI-based liver fat operated through the GM or microbial related inflammatory factors, in a multiethnic cross-sectional study. The inflammatory factors included high-sensitivity C-reactive protein (hs-CRP) and lipopolysaccharide binding protein (LBP) (a proxy for serum LPS) (17).

Methods

Study population

The Multiethnic Cohort (MEC) is an ongoing longitudinal study focused on identifying risk factors for cancer and other chronic diseases, in particular diet, lifestyle, and genetic related risk factors (18). The MEC consists of >215,000 males and females from Hawaii and Los Angeles, California of mainly Japanese American, Native Hawaiian, white, African-American, or Latino ancestry, aged 45–75 y at recruitment between 1993 and 1996 (18). At baseline, MEC participants were mailed a self-administered 26-page questionnaire to obtain information on demographics, medical history, smoking history, medication use, family history of cancer, and physical activity. The questionnaire included a comprehensive quantitative FFQ (18).

The MEC Adiposity Phenotype Study (MEC-APS) is a subset of 1861 males and females from the MEC, re-recruited between 2013 and 2016 (Figure 1) (19). MEC-APS participants were aged 60–77 y and were enrolled using stratified sampling according to sex, self-identified race/ethnicity, and 6 categories of BMI to ensure similar distribution of the 5 main race/ethnic groups across BMI categories. As previously published (19), individuals were excluded if their self-reported BMI was outside the range of 18.5–40 kg/m²; if they smoked in the previous 2 y; had soft or metal body implants or any amputation; were taking thyroid or insulin medications; or had any serious medical concerns (e.g., kidney disease requiring dialysis, cancer diagnosis). Briefly, MEC-APS participants attended a clinic visit at the University of Hawaii (UH) or University of Southern California (USC), where they provided a fasting blood sample and repeated the FFQ from baseline. Participants also underwent anthropometric measurements, an abdominal MRI, and a whole-body DXA scan for total/regional fat measurement (19).

Study flow diagram. hs-CRP, high-sensitivity C-reactive protein; LBP, lipopolysaccharide binding protein; MEC, Multiethnic Cohort; MEC-APS, Multiethnic Cohort–Adiposity Phenotype Study.

For the current analysis, MEC-APS participants were excluded if they had implausible energy or macronutrient intakes based on the FFQ completed during the 2013–2016 clinic visit (n = 34), missing body imaging data (n = 104), missing gut microbial data (n = 78), or missing hs-CRP or LBP data (n = 6). Thus, a total of 812 males and 843 females were included in this analysis. Institutional Review Boards at UH and USC approved the protocol, and a written informed consent was provided by all participants.

MRI and DXA scanning

All MRI and DXA imaging and quality control procedures have been previously described (19). Whole-body DXA absorptiometry was conducted at UH or USC using a Hologic Discovery A fan-beam densitometer, from which total fat mass was estimated (19). All scans were centrally analyzed at the University of California San Francisco. Abdominal MRI scans were acquired using 3-tesla scanners (Siemens TIM Trio, software version VB13 at UH; General Electric HDx, software release 15M4 at USC) to quantify VAT in square centimeters as the average across 4 cross-sectional lumbar locations (L1–L2, L2–L3, L3–L4, L4–L5) using an axial gradient-echo sequence with breath holds and to calculate liver fat (percentage volume) using a series of axial triple gradient-echo Dixon-type scans. Percentage of liver fat was calculated in a circular region of interest in the lateral right lobe of the liver (15–20 cm²) that was manually selected to avoid intrahepatic vessels and bile ducts (19).

DII

The DII was developed to measure the inflammatory potential of the diet. The DII is based on ∼2000 peer-reviewed journal articles that examined the association between diet and 6 blood inflammatory markers (IL-1β, IL-4, IL-6, IL-10, TNF-α, CRP) (14). Among these articles, 45 food components were identified as being associated with ≥1 of the 6 inflammatory markers. Given that overall consumption of energy in the diet is associated with inflammation, and many studies showed strong inverse associations between DII scores and total energy intake, a method for accounting for energy intake, the Energy-Adjusted DII (E-DII) score, was later developed (13). A total of 27 of a possible 45 food parameters were available from the MEC FFQ. These included: carbohydrate; protein; total fat; saturated, monounsaturated, and polyunsaturated fats; omega-3 (ω-3) and ω-6 fatty acids; alcohol; fiber; cholesterol; vitamins A, B-6, B-12, C, D, and E; thiamin; riboflavin; niacin; iron; magnesium; zinc; selenium; folate; β-carotene; and caffeine. Only foods, and not supplements, contributed to the E-DII scores for MEC. Components not included were eugenol, garlic, ginger, onion, trans fat, turmeric, saffron, green tea/black tea, falan-3-ol, flavones, flavonols, flavonones, isoflavones, anthocyanins, pepper, thyme/oregano, and rosemary.

The E-DII was constructed based on relating the intakes reported on the FFQ to dietary intake data from surveys or studies conducted in 11 countries (United Sates, Australia, Bahrain, Denmark, India, Japan, New Zealand, Taiwan, South Korea, Mexico, and the United Kingdom) (14). Dietary data from these countries were compiled to provide a mean ± SD intake for each component score. A z-score was created for each food component for each participant and converted to a centered proportion score. Each E-DII component is scored based on intake per 1000 kcal (13). E-DII component scores are summed to create the overall E-DII score. A higher E-DII score indicates a more proinflammatory diet, and a lower score indicates a more anti-inflammatory diet. For this study, E-DII scores were calculated from the FFQ completed during the 2013–2015 clinic visit.

Blood biomarkers analysis

hs-CRP and LBP were measured in the MEC-APS blood samples at the UH Cancer Center's Analytical Biochemistry Shared Resource in serum and plasma following the manufacturer's protocol (20–23). Briefly, hs-CRP was assessed in serum using a Cobas MiraPlus clinical chemistry analyzer (Roche Diagnostics) and a latex particle enhanced immunoturbidimetry–based kit from Pointe Scientific. Circulating concentration of LBP was measured in heparinized plasma using a commercial ELISA kit (Cell Sciences Inc). Samples were diluted 1:1000 and the assay was conducted according to the kit protocol with a standard curve of 1.5–100 ng/mL (23).

Fecal microbiome analysis

A detailed description of sample collection and fecal microbiome analysis is available elsewhere (24). Briefly, participants collected stool samples in RNAlater (Ambion) and samples were stored in their home freezer until delivered to the clinic (25). Samples were shipped on dry ice from Honolulu or Los Angeles to Seattle, where they were stored at −80°C and processed (26). DNA was extracted and amplified for the V1–V3 region of the 16S rRNA genes, and amplicons were sequenced on the MiSeq platform (Illumina) (26). An in-house bioinformatic pipeline was used to generate operational taxonomic units, sequences were processed using Quantitative Insights Into Microbial Ecology (QIIME) v.1.8. The α diversity (Shannon) index (25, 27) was calculated in QIIME based on the mean of 10 subsamples with rarefaction to 10,000 sequences per sample (12).

Statistical methods

The final dataset included 1655 MEC-APS participants. To compare characteristics of individuals with the most anti-inflammatory diet with those with more proinflammatory diets, E-DII scores were ranked and split into approximately equal tertiles, with tertile 1 representing the most anti-inflammatory diets and tertile 3 the most proinflammatory. The associations of E-DII scores with the adiposity outcomes were examined using linear regression models of total fat mass, VAT, and liver fat, with adjustment for age, education (<college education, ≥college education), physical activity [<1.65 MET·(metabolic equivalent of task) h/d, ≥1.65 MET·h/d), race/ethnicity (African American, Native Hawaiian, Japanese American, Latino, white), sex, smoking status (never, ever), and antibiotic use in the last year (yes, no). VAT and liver fat models were additionally adjusted for total fat mass. A separate exposure category was created for missing data for physical activity (n = 45), education (n = 11), and antibiotic use (n = 13). Mediation analyses of hs-CRP, LBP, and GM [α diversity (Shannon), phyla, or genera], for the association of E-DII with adiposity phenotypes (Figure 2) were conducted through linear regression using the PROCESS macro (v3.5 by Andrew F Hayes) for SPSS. A bootstrap method using iterations of computed samples (5000) was used to determine the significance of the indirect effects (ab) (Figure 2) of the GM (α diversity, phyla, or genera), LBP, or hs-CRP (28) for the association of E-DII and adiposity phenotypes. Centered log ratio transformations of genera were used to account for the compositional nature of microbial sequencing data (12). The current mediation analysis included all 10 phyla and 28 genera that were previously found to be associated with diet quality (HEI-2010 scores) in the MEC (29). The same approach was applied to analyze the data stratified by sex and by race/ethnicity to assess if associations seen in the whole sample were present within subgroups. To examine effect modification between E-DII and adiposity phenotypes by sex and ethnicity, interaction terms were included in the primary E-DII models.

Model: the relation between E-DII and adiposity phenotype through the indirect effect of hs-CRP, LBP, α diversity, genera, and phyla. E-DII, Energy-Adjusted Dietary Inflammatory Index; hs-CRP, high-sensitivity C-reactive protein; LBP, lipopolysaccharide binding protein; VAT, visceral adipose tissue.

To assess whether any E-DII component was independently associated with the 3 adiposity phenotypes, we assessed the association between each E-DII component and adiposity phenotypes, controlling in the regression models for a modified total E-DII score, which did not include the respective E-DII component being assessed. For example, to assess whether the E-DII component carbohydrate was not solely driving any associations found between total E-DII and adiposity phenotypes, we assessed the association between the E-DII carbohydrate component and adiposity phenotypes, controlling in the model for a modified total E-DII score (modified total E-DII score = total E-DII score − E-DII carbohydrate component).

To correct for multiple comparisons in the mediation analysis, the Bonferroni-adjusted significance level was set as <0.0013 (0.05/38 models run for each adiposity phenotype; including models for 28 genera and 10 phyla). To understand the dependence among adiposity phenotypes we assessed the correlations between adiposity phenotypes. To assess our decision for a priori selecting the 28 genera previously associated with HEI-2010 scores in MEC for analyses, we tested the correlation between E-DII and HEI-2010 scores. All data were analyzed using IBM SPSS Statistics version 26 software (IBM Corp).

Results

The current analysis included 812 males and 843 females (Table 1). Participants with the most anti-inflammatory diets (E-DII tertile 1) had a lower total energy intake and total fat mass compared with participants with the most proinflammatory diets (E-DII tertile 3). Compared with participants in tertile 3, higher proportions of participants in tertile 1 conducted above the median per day of physical activity, were female, white, never smokers, or had a BMI <25, whereas a lower proportion had a BMI ≥30. The participants’ mean age was similar across E-DII tertiles, as were education level, antibiotic use, the proportions of participants who were African-American, Native Hawaiian, Japanese American, and Latino, and the proportion of participants with a BMI between 25 and <30.

TABLE 1.

Characteristics of participants (n = 1655) in the Multiethnic Cohort–Adiposity Phenotype Study at the 2013–2015 clinic visit, by E-DII score tertiles¹

	Tertile 1	Tertile 2	Tertile 3
Participants (n)	551	552	552
E-DII score	−3.68 ± 0.52	−2.12 ± 0.49	0.16 ± 1.01
Age, y	69.3 ± 2.7	69.1 ± 2.6	69.1 ± 2.8
Total energy intake, kcal/d	1641 (1295–2121)	1672 (1250–2194)	1752 (1312–2384)
Total fat mass, kg	23.9 ± 8.3	25.4 ± 8.8	26.2 ± 8.4
Visceral adipose tissue, cm²	141.5 ± 78.5	168.3 ± 83.2	191.8 ± 81.3
Liver fat, % volume	4.8 ± 3.9	5.6 ± 4.4	6.7 ± 5.3
Physical activity, MET·h/d
<1.65	227 (28.3%)	260 (32.4%)	315 (39.3%)
≥1.65	302 (37.4%)	276 (34.2%)	230 (28.5%)
Missing	22 (48.9%)	16 (35.6%)	7 (15.6%)
Sex
Men	218 (26.8%)	258 (31.8%)	336 (41.4%)
Women	333 (39.5%)	294 (34.9%)	216 (25.6%)
Race/ethnicity
African American	83 (32.3%)	86 (33.5%)	88 (34.2%)
Native Hawaiian	88 (34.0%)	79 (30.5%)	92 (35.5%)
Japanese American	123 (29.1%)	158 (37.4%)	141 (33.4%)
Latino	110 (32.9%)	106 (31.7%)	118 (35.3%)
White	147 (38.4%)	123 (32.1%)	113 (29.5%)
Education
<College education	248 (30.4%)	260 (31.9%)	308 (37.7%0
≥College education	296 (35.7%)	289 (34.9%)	243 (29.3%)
Missing	7 (63.6%)	3 (27.3%)	1 (9.1%)
Smoking
Never	354 (34.9%)	350 (34.5%)	311 (30.6%)
Ever	197 (30.8%)	202 (31.6%)	241 (37.7%)
Antibiotic use in the previous year
Yes	124 (22.5%)	122 (22.1%)	111 (20.1%)
No	425 (77.1%)	76.8 (76.8%)	436 (79.0%)
Missing	2 (0.4%)	6 (1.1%)	5 (0.9%)
BMI, kg/m²
<25	210 (42.4%)	165 (33.3%)	120 (24.2%)
25 to <30	215 (32.1%)	225 (33.6%)	229 (34.2%)
≥30	126 (25.7%)	162 (33.0%)	203 (41.3%)

Open in a new tab

Tertile 1 represents the lowest E-DII scores and most anti-inflammatory diets and tertile 3 represents the highest E-DII scores and most proinflammatory diets. Data presented as mean ± SD, median (IQR), or n (%). E-DII, Energy-Adjusted Dietary Inflammatory Index.

The E-DII score was associated [total effect (c)] with each adiposity phenotype. A more proinflammatory diet was associated with higher total fat mass (c = 0.68; 95% CI: 0.47, 0.90), VAT (c = 4.61; 95% CI: 2.95, 6.27), and liver fat (c = 0.40; 95% CI: 0.27, 0.53) (Table 2). In assessing interdependence between adiposity phenotypes, total fat mass was only weakly correlated with liver fat (r = 0.20), and moderately correlated with VAT (r = 0.44), and VAT was moderately correlated with liver fat (r = 0.40) (all P values <0.001; data not shown). In contrast, E-DII and HEI-2010 scores were strongly correlated (r = −0.78, P < 0.001).

TABLE 2.

Association between E-DII score and adiposity phenotype (total effect) in Multiethnic Cohort–Adiposity Phenotype Study participants (n = 1655)¹

	Total effect
Adiposity phenotype	β	SE	95% CI
Total fat mass, kg	0.683	0.111	0.465, 0.901
Visceral adipose tissue, cm²	4.610	0.848	2.947, 6.272
Liver fat, %	0.401	0.066	0.272, 0.530

Open in a new tab

Associations of E-DII scores with the adiposity outcomes were examined using linear regression models with adjustment for age, education, physical activity, race/ethnicity, sex, smoking status, and antibiotic use in the last year. Visceral and liver fat models were additionally adjusted for total fat mass. Mediation analyses were conducted using the PROCESS macro for SPSS. E-DII, Energy-Adjusted Dietary Inflammatory Index.

We present all results; however, regarding mediators where both Path a (a) and Path b (b) met the Bonferroni adjusted P value of <0.0013, the relation between E-DII and total fat mass significantly operated indirectly through LBP (ab = 0.095; 95% CI: 0.038, 0.159), Flavonifractor (ab = −0.038; 95% CI: −0.069, −0.013), [Ruminococcus] gnavus group (ab = −0.072; 95% CI: −0.116, −0.036), and Tyzzerella (ab = −0.059; 95% CI: −0.098, −0.027) (Table 3). For the total fat mass model, a higher E-DII score was associated with higher LBP (a = 389.150; 95% CI: 153.487, 624.811) (Supplemental Table 1), and there was a positive association between LBP and total fat mass (b = 0.0002; 95% CI: 0.0002, 0.0003). At the genera level, a higher E-DII score was associated with higher Flavonifractor (a = 0.123; 95% CI: 0.084, 0.161), [Ruminococcus] gnavus group (a = 0.170; 95% CI: 0.125, 0.216), and Tyzzerella (a = 0.112; 95% CI: 0.064, 0.161). Higher Flavonifractor (b = −0.391; 95% CI: −0.645, −0.128), [Ruminococcus] gnavus group (b = −0.364; 95% CI: −0.587, −0.141), and Tyzzerella (b = −0.388; 95% CI: −0.597, −0.180) were associated with lower total fat mass.

TABLE 3.

Indirect effect and direct effect for the association of E-DII with total fat mass through hs-CRP, LBP, and gut microbiome, in MEC-APS participants (n = 1655)¹

	Indirect effect²			Direct effect
Mediator	β	SE	95% CI	β	SE	95% CI
High sensitivity C-reactive protein, mg/L	0.092	0.034	0.031, 0.165	0.591	0.106	0.383, 0.800
Lipopolysaccharide binding protein, μg/mL	0.095	0.031	0.038, 0.159³	0.588	0.108	0.377, 0.799
α Diversity (Shannon)	−0.022	0.012	−0.049, −0.003	0.705	0.111	0.486, 0.923
Phyla⁴
Actinobacteria	−0.014	0.001	−0.040, 0.005	0.698	0.111	0.479, 0.916
Bacteroidetes	−0.001	0.004	−0.010, 0.005	0.684	0.111	0.466, 0.902
Cyanobacteria	0.007	0.007	−0.005, 0.024	0.676	0.111	0.458, 0.895
Firmicutes	0.009	0.009	−0.007, 0.030	0.674	0.111	0.455, 0.892
Fusobacteria	−0.017	0.011	−0.041, 0.001	0.700	0.111	0.481, 0.918
Lentisphaerae	−0.006	0.006	−0.020, 0.003	0.689	0.111	0.471, 0.907
Proteobacteria	−0.004	0.006	−0.018, 0.006	0.687	0.111	0.469, 0.905
Synergistetes	−0.001	0.004	−0.010, 0.009	0.684	0.111	0.466, 0.902
Tenericutes	−0.012	0.009	−0.031, 0.002	0.695	0.111	0.477, 0.913
Verrucomicrobia	−0.008	0.007	−0.025, 0.004	0.691	0.111	0.473, 0.909
Genera⁴
Anaerostipes	−0.009	0.015	−0.039, 0.021	0.692	0.111	0.472, 0.912
Bacteroides	−0.007	0.009	−0.027, 0.009	0.690	0.111	0.472, 0.909
Christensenellaceae R-7 group	−0.035	0.015	−0.067, −0.010	0.718	0.111	0.499, 0.937
Collinsella	−0.016	0.012	−0.043, 0.006	0.699	0.111	0.480, 0.918
Coprococcus 2	0.021	0.018	−0.013, 0.059	0.662	0.111	0.442, 0.882
Erysipelotrichaceae UCG-003	0.022	0.011	0.003, 0.047	0.662	0.111	0.443, 0.880
Escherichia-Shigella	−0.009	0.011	−0.034, 0.011	0.692	0.111	0.473, 0.911
[Eubacterium] coprostanoligenes group	0.006	0.008	−0.009, 0.024	0.677	0.111	0.460, 0.895
[Eubacterium] eligens group	0.020	0.022	−0.023, 0.064	0.663	0.111	0.441, 0.886
[Eubacterium] xylanophilum group	−0.040	0.019	−0.081, −0.005	0.723	0.111	0.503, 0.944
Faecalibacterium	0.022	0.013	−0.002, 0.049	0.662	0.111	0.443, 0.880
Flavonifractor	−0.038	0.014	−0.069, −0.013³	0.721	0.111	0.503, 0.939
Lachnoclostridium	0.000	0.011	−0.022, 0.022	0.683	0.111	0.464, 0.902
Lachnospira	0.003	0.014	−0.025, 0.032	0.680	0.111	0.460, 0.899
Lachnospiraceae NC2004 group	−0.011	0.009	−0.032, 0.002	0.694	0.111	0.476, 0.912
Lachnospiraceae ND3007 group	−0.026	0.013	−0.053, −0.005	0.709	0.111	0.490, 0.927
Lachnospiraceae UCG-001	−0.014	0.009	−0.034, −0.001	0.697	0.111	0.479, 0.915
Lachnospiraceae; other	−0.001	0.010	−0.023, 0.020	0.684	0.111	0.465, 0.903
Odoribacter	−0.002	0.012	−0.027, 0.020	0.686	0.111	0.467, 0.905
Oscillibacter	−0.008	0.008	−0.027, 0.007	0.691	0.111	0.473, 0.909
Parabacteroides	−0.005	0.010	−0.026, 0.015	0.688	0.111	0.469, 0.906
Ruminiclostridium 5	−0.009	0.015	−0.038, 0.020	0.692	0.111	0.472, 0.912
Ruminococcaceae UCG-013	0.014	0.013	−0.009, 0.041	0.670	0.111	0.451, 0.889
Ruminococcaceae UCG-014	−0.016	0.013	−0.043, 0.007	0.699	0.111	0.480, 0.918
Ruminococcaceae uncultured	0.000	0.003	−0.008, 0.007	0.684	0.111	0.466, 0.901
Ruminococcus 1	−0.007	0.017	−0.040, 0.025	0.690	0.111	0.470, 0.911
[Ruminococcus] gnavus group	−0.072	0.021	−0.116, −0.036³	0.756	0.111	0.536, 0.975
Tyzzerella	−0.059	0.018	−0.098, −0.027³	0.742	0.111	0.524, 0.960

Open in a new tab

Mediation analyses conducted through linear regression using the PROCESS macro for SPSS adjusting for age, education, physical activity, race/ethnicity, sex, smoking status, and antibiotic use. E-DII, Energy-Adjusted Dietary Inflammatory Index; hs-CRP, high-sensitivity C-reactive protein; LBP, lipopolysaccharide binding protein; MEC-APS, Multiethnic Cohort–Adiposity Phenotype Study.

A bootstrap method using iterations of computed samples (5000) was used to determine the significance of the indirect effects.

Significant after Bonferroni adjustment (P < 0.0013).

⁴

Centered log ratio transformation for skewed distributions.

We found that the association between E-DII score and VAT arose through the phylum Fusobacteria (ab = 0.450; 95% CI: 0.181, 0.787), and through the genera Christensenellaceae R-7 group (ab = 0.532; 95% CI: 0.274, 0.843), Coprococcus 2 (ab = 0.464; 95% CI: 0.215, 0.764), Escherichia-Shigella (ab = 0.455; 95% CI: 0.202, 0.761), [Eubacterium] xylanophilum group (ab = 0.666; 95% CI: 0.357, 1.021), Flavonifractor (ab = 0.483; 95% CI: 0.231, 0.803), Lachnoclostridium (ab = 0.437; 95% CI: 0.197, 0.733), [Ruminococcus] gnavus group (ab = 0.786; 95% CI: 0.460, 1.196), and Tyzzerella (ab = 0.651; 95% CI: 0.353, 1.027) (Table 4). For the VAT model, there was an inverse association between E-DII score and Christensenellaceae R-7 group (a = −0.146; 95% CI: −0.193, −0.099), Coprococcus 2 (a = −0.107; 95% CI: −0.141, −0.074), and [Eubacterium] xylanophilum group (a = −0.149; 95% CI: −0.190, −0.109). Lower Christensenellaceae R-7 group (b = −4.139; 95% CI: −5.759, −2.519), Coprococcus 2 (b = −4.298; 95% CI: −6.540, −2.057), and [Eubacterium] xylanophilum group (b = −4.273; 95% CI: −6.134, −2.411) was associated with higher VAT (Supplemental Table 2). Conversely, for the VAT model, we found positive associations between E-DII and the phylum Fusobacteria (a = 0.073; 95% CI: 0.033, 0.114), and between the genera Escherichia-Shigella (a = 0.089; 95% CI: 0.034, 0.144), Flavonifractor (a = 0.124; 95% CI: 0.085, 0.163), Lachnoclostridium (a = 0.052; 95% CI: 0.031, 0.072), [Ruminococcus] gnavus group (a = 0.172; 95% CI: 0.127, 0.218), and Tyzzerella (a = 0.115; 95% CI: 0.066, 0.163). There was a positive association between Fusobacteria (b = 5.625; 95% CI: 3.766, 7.485), Escherichia- Shigella (b = 3.906; 95% CI: 2.522, 5.290), Flavonifractor (b = 4.824; 95% CI: 2.865, 6.782), Lachnoclostridium (b = 9.716; 95% CI: 6.078, 13.355), [Ruminococcus] gnavus group (b = 4.438; 95% CI: 2.777, 6.098), and Tyzzerella (b = 4.553; 95% CI: 3.001, 6.104) and VAT.

TABLE 4.

Indirect effect and direct effect for the association of E-DII with visceral adipose tissue through hs-CRP, LBP, and gut microbiome, in MEC-APS participants (n = 1655)¹

	Indirect effect²			Direct effect
Mediator	β	SE	95% CI	β	SE	95% CI
High sensitivity C-reactive protein, mg/L	−0.002	0.035	−0.088, 0.059	4.612	0.848	2.949, 6.275
Lipopolysaccharide binding protein, μg/mL	0.073	0.060	−0.012, 0.216	4.536	0.847	2.875, 6.198
α Diversity (Shannon)	0.171	0.095	0.015, 0.382	4.439	0.850	2.772, 6.106
Phyla³
Actinobacteria	0.131	0.088	−0.025, 0.327	4.479	0.851	2.810, 6.147
Bacteroidetes	0.051	0.078	−0.099, 0.218	4.559	0.844	2.903, 6.215
Cyanobacteria	0.097	0.064	−0.03, 0.241	4.513	0.847	2.851, 6.174
Firmicutes	−0.125	0.078	−0.306, −0.001	4.734	0.849	3.069, 6.399
Fusobacteria	0.450	0.155	0.181, 0.787³	4.160	0.843	2.507, 5.812
Lentisphaerae	0.076	0.060	−0.016, 0.215	4.533	0.847	2.872, 6.195
Proteobacteria	0.168	0.095	0.004, 0.366	4.442	0.845	2.785, 6.099
Synergistetes	−0.013	0.041	−0.001, 0.054	4.622	0.847	2.960, 6.284
Tenericutes	0.326	0.130	0.082, 0.602	4.284	0.841	2.634, 5.934
Verrucomicrobia	0.081	0.064	−0.020, 0.228	4.529	0.847	2.868, 6.190
Genera³
Anaerostipes	0.060	0.112	−0.158, 0.297	4.549	0.855	2.872, 6.226
Bacteroides	0.270	0.109	0.080, 0.510	4.340	0.846	2.681, 5.998
Christensenellaceae R-7 group	0.532	0.146	0.274, 0.843³	4.078	0.847	2.416, 5.739
Collinsella	0.145	0.101	−0.029, 0.363	4.464	0.852	2.794, 6.135
Coprococcus 2	0.464	0.140	0.215, 0.764³	4.146	0.853	2.474, 5.818
Erysipelotrichaceae UCG-003	0.213	0.096	0.058, 0.432	4.396	0.847	2.735, 6.057
Escherichia-Shigella	0.455	0.145	0.202, 0.761³	4.154	0.844	2.499, 5.810
[Eubacterium] coprostanoligenes group	0.002	0.024	−0.046, 0.056	4.608	0.848	2.945, 6.271
[Eubacterium] eligens group	0.125	0.157	−0.173, 0.450	4.484	0.864	2.790, 6.179
[Eubacterium] xylanophilum group	0.666	0.167	0.357, 1.021³	3.943	0.853	2.270, 5.617
Faecalibacterium	0.011	0.090	−0.172, 0.187	4.598	0.852	2.927, 6.269
Flavonifractor	0.483	0.144	0.231, 0.803³	4.127	0.846	2.468, 5.786
Lachnoclostridium	0.437	0.136	0.197, 0.733³	4.173	0.845	2.516, 5.829
Lachnospira	0.059	0.108	−0.142, 0.289	4.550	0.854	2.875, 6.226
Lachnospiraceae NC2004 group	0.181	0.095	0.026, 0.397	4.428	0.846	2.768, 6.088
Lachnospiraceae ND3007 group	0.118	0.092	−0.048, 0.316	4.492	0.852	2.821, 6.163
Lachnospiraceae UCG-001	0.076	0.066	−0.034, 0.231	4.534	0.849	2.868, 6.199
Lachnospiraceae; other	−0.108	0.086	−0.301, 0.034	4.718	0.851	3.049, 6.387
Odoribacter	0.226	0.100	0.054, 0.446	4.383	0.851	2.715, 6.052
Oscillibacter	0.163	0.088	0.022, 0.362	4.447	0.848	2.784, 6.109
Parabacteroides	0.112	0.081	−0.020, 0.293	4.497	0.850	2.829, 6.165
Ruminiclostridium 5	0.562	0.155	0.286, 0892	4.048	0.849	2.383, 5.713
Ruminococcaceae UCG-013	0.251	0.107	0.066, 0.488	4.358	0.850	2.691, 6.025
Ruminococcaceae UCG-014	0.459	0.136	0.229, 0.754	4.151	0.848	2.487, 5.814
Ruminococcaceae uncultured	−0.031	0.050	−0.143, 0.059	4.641	0.847	2.980, 6.302
Ruminococcus 1	0.396	0.145	0.136, 0.710	4.214	0.855	2.536, 5.891
[Ruminococcus] gnavus group	0.786	0.188	0.460, 1.196³	3.824	0.854	2.149, 5.500
Tyzzerella	0.651	0.172	0.353, 1.027³	3.959	0.848	2.300, 5.621

Open in a new tab

Mediation analyses conducted through linear regression using the PROCESS macro for SPSS adjusting for age, education, physical activity, race/ethnicity, sex, smoking status, antibiotic use, and total fat mass. E-DII, Energy-Adjusted Dietary Inflammatory Index; hs-CRP, high-sensitivity C-reactive protein; LBP, lipopolysaccharide binding protein; MEC-APS, Multiethnic Cohort–Adiposity Phenotype Study

A bootstrap method using iterations of computed samples (5000) was used to determine the significance of the indirect effects.

Centered log ratio transformation for skewed distributions.

⁴

Significant after Bonferroni adjustment (P < 0.0013).

The association between E-DII score and liver fat indirectly occurred through α diversity (ab = 0.037; 95% CI: 0.015, 0.063), the phylum Fusobacteria (ab = 0.038; 95% CI: 0.016, 0.066), and the genera Christensenellaceae R-7 group, (ab = 0.050; 95% CI: 0.026, 0.073), Coprococcus 2 (ab = 0.041; 95% CI: 0.020, 0.065), Escherichia-Shigella (ab = 0.036; 95% CI: 0.016, 0.062), [Eubacterium] xylanophilum group (ab = 0.052; 95% CI: 0.029, 0.079), Flavonifractor (ab = 0.043; 95% CI: 0.021, 0.070), Lachnoclostridium (ab = 0.053; 95% CI: 0.025, 0.085), Ruminiclostridium 5 (ab = 0.024; 95% CI: 0.002, 0.050), [Ruminococcus] gnavus group (ab = 0.085; 95% CI: 0.052, 0.122), and Tyzzerella (ab = 0.035; 95% CI: 0.015, 0.061) (Table 5).

TABLE 5.

Indirect and direct effect for the association of E-DII with liver fat through hs-CRP, LBP, and gut microbiome, in MEC-APS participants (n = 1655)¹

	Indirect effect²			Direct effect
Mediator	β	SE	95% CI	β	SE	95% CI
High sensitivity C-reactive protein, mg/L	0.002	0.003	−0.002, 0.011	0.399	0.066	0.269, 0.528
Lipopolysaccharide binding protein, μg/mL	0.008	0.006	−0.001, 0.023	0.393	0.066	0.264, 0.522
α Diversity (Shannon)	0.037	0.012	0.015, 0.063³	0.364	0.066	0.236, 0.493
Phyla⁴
Actinobacteria	0.000	0.007	−0.014, 0.015	0.401	0.066	0.271, 0.531
Bacteroidetes	0.006	0.009	−0.011, 0.023	0.395	0.065	0.267, 0.524
Cyanobacteria	0.003	0.004	−0.004, 0.012	0.399	0.066	0.269, 0.528
Firmicutes	−0.014	0.007	−0.030, −0.003	0.415	0.066	0.286, 0.544
Fusobacteria	0.038	0.013	0.016, 0.066³	0.363	0.065	0.235, 0.491
Lentisphaerae	0.008	0.005	−0.002, 0.020	0.393	0.066	0.264, 0.523
Proteobacteria	0.017	0.009	0.001, 0.037	0.384	0.066	0.256, 0.513
Synergistetes	−0.002	0.006	−0.015, 0.008	0.403	0.066	0.274, 0.532
Tenericutes	0.024	0.009	0.006, 0.043	0.378	0.066	0.249, 0.506
Verrucomicrobia	0.009	0.007	−0.002, 0.024	0.392	0.066	0.263, 0.521
Genera⁴
Anaerostipes	0.004	0.010	−0.015, 0.025	0.398	0.067	0.267, 0.528
Bacteroides	0.035	0.012	0.012, 0.060	0.366	0.065	0.238, 0.494
Christensenellaceae R-7 group	0.050	0.012	0.026, 0.073³	0.353	0.066	0.224, 0.482
Collinsella	−0.003	0.008	−0.022, 0.013	0.405	0.066	0.274, 0.535
Coprococcus 2	0.041	0.012	0.020, 0.065³	0.360	0.066	0.231, 0.490
Erysipelotrichaceae UCG-003	0.004	0.005	−0.005, 0.016	0.397	0.066	0.267, 0.527
Escherichia-Shigella	0.036	0.012	0.016, 0.062³	0.365	0.066	0.236, 0.494
[Eubacterium] coprostanoligenes group	−0.003	0.008	−0.020, 0.013	0.404	0.066	0.276, 0.533
[Eubacterium] eligens group	0.003	0.013	−0.003, 0.051	0.378	0.067	0.246, 0.510
[Eubacterium] xylanophilum group	0.052	0.013	0.029, 0.079³	0.349	0.066	0.219, 0.479
Faecalibacterium	0.006	0.008	−0.009, 0.022	0.395	0.066	0.265, 0.525
Flavonifractor	0.043	0.013	0.021, 0.070³	0.358	0.066	0.230, 0.487
Lachnoclostridium	0.053	0.015	0.025, 0.085³	0.348	0.065	0.220, 0.475
Lachnospira	0.009	0.009	−0.007, 0.027	0.392	0.067	0.262, 0.523
Lachnospiraceae NC2004 group	0.024	0.011	0.004, 0.048	0.378	0.065	0.249, 0.506
Lachnospiraceae ND3007 group	0.016	0.009	0.002, 0.035	0.385	0.066	0.255, 0.515
Lachnospiraceae UCG-001	0.002	0.004	−0.006, 0.012	0.399	0.066	0.269, 0.528
Lachnospiraceae; other	−0.017	0.008	−0.034, −0.004	0.418	0.066	0.288, 0.548
Odoribacter	0.017	0.009	0.002, 0.036	0.385	0.066	0.255, 0.514
Oscillibacter	0.008	0.006	−0.002, 0.023	0.393	0.066	0.264, 0.523
Parabacteroides	0.011	0.008	−0.002, 0.029	0.390	0.066	0.260, 0.520
Ruminiclostridium 5	0.024	0.012	0.002, 0.050³	0.377	0.066	0.247, 0.508
Ruminococcaceae UCG-013	0.021	0.009	0.005, 0.040	0.381	0.066	0.251, 0.510
Ruminococcaceae UCG-014	0.044	0.011	0.025, 0.068	0.357	0.066	0.228, 0.846
Ruminococcaceae uncultured	−0.005	0.007	−0.019, 0.009	0.406	0.066	0.277, 0.534
Ruminococcus 1	0.018	0.013	−0.006, 0.045	0.383	0.067	0.252, 0.514
[Ruminococcus] gnavus group	0.085	0.018	0.052, 0.122³	0.317	0.066	0.187, 0.446
Tyzzerella	0.035	0.011	0.015, 0.061³	0.366	0.066	0.236, 0.496

Open in a new tab

Mediation analyses conducted through linear regression using the PROCESS macro for SPSS adjusting for age, education, physical activity, race/ethnicity, sex, smoking status, antibiotic use, and total fat mass. E-DII, Energy-Density Dietary Inflammatory Index; hs-CRP, high-sensitivity C-reactive protein; LBP, lipopolysaccharide binding protein; MEC-APS, Multiethnic Cohort–Adiposity Phenotype Study; NC.

A bootstrap method using iterations of computed samples (5000) was used to determine the significance of the indirect effects.

Significant after Bonferroni adjustment (P < 0.0013).

⁴

Centered log ratio transformation for skewed distributions.

For the liver fat model, there was an inverse association between E-DII score and α diversity (a = −0.041; 95% CI: −0.062, −0.019), Christensenellaceae R-7 group (a = −0.146; 95% CI: −0.193, −0.099), Coprococcus 2 (a = −0.107; 95% CI: −0.141, −0.074), and [Eubacterium] xylanophilum group (a = −0.149; 95% CI: −0.190, −0.109) (Supplemental Table 3). We found an inverse association between α diversity (b = −0.900; 95% CI: −1.187, −0.614), Christensenellaceae R-7 group (b = −0.348; 95% CI: −0.473, −0.223), Coprococcus 2 (b = −0.350; 95% CI: −0.524, −0.177), and [Eubacterium] xylanophilum group (b = −0.336; 95% CI: −0.480, −0.192) and liver fat. Conversely, for the liver fat model, there was a positive association between E-DII score and the phylum Fusobacteria (a = 0.073; 95% CI: 0.033, 0.114), and the genera Escherichia-Shigella (a = 0.089; 95% CI: 0.034, 0.144), Flavonifractor (a = 0.124; 95% CI: 0.085, 0.163), Lachnoclostridium (a = 0.052; 95% CI: 0.031, 0.072), Ruminiclostridium (a = 0.172; 95% CI: 0.127, 0.218), and Tyzzerella (a = 0.115; 95% CI: 0.066, 0.163). We found a positive association between Fusobacteria (b = 0.490; 95% CI: 0.346, 0.633), Escherichia-Shigella (b = 0.304; 95% CI: 0.196, 0.411), Flavonifractor (b = 0.410; 95% CI: 0.259,0.561), Lachnoclostridium (b = 1.129; 95% CI: 0.850, 1.407), Ruminiclostridium (b = 0.347; 95% CI: 0.082, 0.611), [Ruminococcus] gnavus group (b = 0.458; 95% CI: 0.330, 0.586), and Tyzzerella (b = 0.223; 95% CI: 0.102, 0.344) and liver fat.

There was no significant modifying effect of sex on the association of E-DII and the adiposity phenotypes. In subgroup analyses by sex, the total effect of E-DII on total fat mass and ectopic fat was significant for both males and females (P values <0.05) (data not shown). The interaction between E-DII and adiposity phenotype by ethnicity was only significant for the liver fat model. In subgroup analyses by race/ethnicity, the total effect of E-DII was significant for Japanese Americans and whites for total fat mass, significant for whites for VAT, and significant for Latinos and whites for liver fat (P values <0.05) (data not shown).

Although multiple E-DII components were found to be significantly associated with the adiposity phenotypes (Supplemental Table 4), only the E-DII components cholesterol and magnesium were found to be significantly associated with all 3 adiposity phenotypes after adjusting for the modified E-DII total score. The cholesterol E-DII component score was positively associated with total fat mass (β = 9.61; 95% CI: 5.45, 13.77), VAT (β = 101.49; 95% CI: 58.85, 144.13), and liver fat (β = 4.33; 95% CI: 1.71, 6.96), where a higher cholesterol E-DII score reflects higher cholesterol intake. The magnesium E-DII component score also was positively associated with total fat mass (β = 6.67; 95% CI: 3.96, 9.37), VAT (β = 80.37; 95% CI: 52.74, 108.01), and liver fat (β = 3.28, 95% CI: 1.58, 4.98), where a higher magnesium E-DII score reflects a lower magnesium intake. When assessing the association between the E-DII component magnesium and each adiposity phenotype, the modified E-DII total score was no longer significant in the model (data not shown). This finding was unique only to the magnesium component score.

Discussion

In this multiethnic sample we found that consuming a more proinflammatory diet, as assessed with the E-DII score, was associated with a higher total fat mass, VAT, and liver fat, compared with following a more anti-inflammatory diet. LBP, a proxy for LPS, and the genera Flavonifractor, [Ruminococcus] gnavus group, and Tyzzerella significantly mediated the association between E-DII and total fat mass. The relation between E-DII and both VAT and liver fat occurred indirectly through the phylum Fusobacteria, and through the genera Christensenellaceae R-7 group, Coprococcus 2, Escherichia-Shigella, [Eubacterium] xylanophilum group, Flavonifractor, Lachnoclostridium, [Ruminococcus] gnavus group, and Tyzzerella. In addition, the association between E-DII and VAT existed through the [Ruminococcus] gnavus group, and the relation between E-DII and liver fat operated through α diversity.

Consistent with our findings, a meta-analysis of 22 observational studies (cross-sectional, case-control, and cohort studies) in 33,219 adults with obesity reported that following a proinflammatory diet (higher DII score) was associated with higher BMI (3). Within this meta-analysis, separate analyses were conducted by study design, and the strongest association between DII and BMI was observed in prospective cohort studies (3). Also, a meta-analysis of 32 studies in 103,071 adults with obesity or obesity-related diseases found that a 1-unit increase in DII score was associated with a 1.81-cm increase in waist circumference (30). The only study examining DII and liver fat, a cross-sectional analysis of 794 overweight or obese participants in the PREDIMED (PREvencion con DIeta MEDiterranea) trial, found that a proinflammatory diet (higher DII score) could contribute to obesity and fatty liver disease (6). These previous studies used proxy measures for obesity, VAT, and liver fat, including BMI (3), waist circumference (30), waist-to-hip ratio (30), noninvasive blood markers of liver status (alanine aminotransferase, aspartate aminotransferase, γ-glutamyltransferase) (6) or a Fatty Liver Index (6) score to study the association between DII and adiposity phenotypes. In comparison, our study used direct measures of total fat mass through DXA, and of VAT and liver fat through the gold standard assessment method of MRI.

We also explored the independent associations of the E-DII components and the adiposity phenotypes. We found that a higher E-DII component score for magnesium (lower magnesium intake) was an independent driver of the association between E-DII score, total fat mass, VAT, and liver fat. Magnesium, which is an important part of the photosynthetic machinery of plants, functions both as a proxy for plant intake and as a potential regulator of inflammatory potential (31, 32). Low magnesium concentrations have previously been associated with higher CRP (33, 34), inflammatory cytokines (33, 35), insulin resistance (36), and obesity (33). The mechanism between magnesium intake and obesity is still unclear (33); however, in mouse models it has been reported that low dietary magnesium is associated with a higher capacity for gut microbiota to harvest energy from dietary intake (37). This finding was unique to the E-DII component for magnesium, with no other E-DII component independently driving the associations between an inflammatory diet and adiposity phenotypes. The absence of effect for other individual components is not surprising because all remaining E-DII components have a small effect on adiposity, and only collectively is this effect significant. Of course, the overarching purpose of developing the DII/E-DII was to provide a summary measure for overall inflammatory effect of the diet.

Only the association between E-DII and total fat mass was explained by higher LBP and, therefore, higher LPS concentrations. LPS is found in the cell wall of Gram-negative bacteria (38) and is known to contribute to low-grade inflammation and obesity when translocated across the gut epithelial membrane (5, 39). The significant mediation by LPS and not by hs-CRP in the current analysis might be because LPS is a part of Gram-negative bacteria; therefore, LPS is a more direct measure of microbially mediated inflammation (5).

Despite there being a positive association between E-DII and total fat mass, through higher LBP concentrations, there was no significant mediation by Gram-negative genera. Activation of the Toll-like receptor 4 (TLR4) pathways can vary in a strain-dependent manner (38) or via other bacterial pathways (40). Therefore, we might have seen a significant positive association between E-DII and total fat mass through the gut microbiota if data at the strain level, ascertained by metagenomic sequencing, were available. Also, the effect of Gram-negative bacteria on LBP concentrations could be through a cumulative dose (41), and not due to a single genus.

Flavonifractor could impact inflammation and adiposity through several mechanisms. The negative association between E-DII and total fat mass that existed through Flavonifractor was unique to this adiposity phenotype because there was a positive association between E-DII and ectopic fat through Flavonifractor. More specifically there was an inverse association with the E-DII and Flavonifractor (Path a) in the total fat mass mediation analysis. Both in vitro and in vivo studies have shown that either live or heat-killed Flavonifractor elicit a blunted inflammatory response to LPS stimulation or a high-fat diet in mice (42). Ogita et al. (43) showed that an oral dose of F. plautii suppressed activation of T-helper-2 cell immune responses and the immune regulation of regulatory T cells in mice. In contrast, the literature also supports an inverse association between Flavonifractor and obesity and nonalcoholic fatty liver disease (NAFLD) (44, 45) but an increased association with colorectal cancer (CRC) and the potential for inflammatory bowel disease (46). Flavonifractor is responsible for the breakdown of the flavonoid, quercetin, a prominent dietary antioxidant (47). A recent study identified increased flavonoid metabolism by Flavonifractor through the catechol ortho-cleavage pathway associated with increased 4-hydroxyphenyl metabolites in CRC patients (48) as a potential biomarker associated with increased cancer risk. Further research is needed to understand the conflicting associations found between Flavonifractor, total fat mass, and ectopic fat, and whether these associations are confounded by other metabolic factors.

We found negative associations between E-DII score and total fat mass through the [Ruminococcus] gnavus group and Tyzzerella, and positive associations between E-DII and ectopic fat through the [Ruminococcus] gnavus group and Tyzzerella. The [Ruminococcus] gnavus group can use mucin as a carbon source and produce secondary bile acids, which both can directly contribute to the breakdown of gut barrier function and increase translocation of metabolites into systemic circulation (49–51). Recent studies have shown that glucorhamnan excreted by R. gnavus induces inflammatory cytokine secretion (TNF-α), which is also dependent on binding to TLR4 receptors (52). Additionally, the production of secondary acids impacts lipid metabolism through the farnesoid X receptor and could increase lipid build-up in the liver (48, 53, 54). The [Ruminococcus] gnavus group has also previously been associated with NAFLD in the MEC (24). Tyzzerella is a potentially pathogenic bacterial genus that has been linked to cardiovascular disease (55). Therefore, we would expect a positive association between the [Ruminococcus] gnavus group, Tyzzerella, and total fat mass. Ectopic fat is more metabolically active than subcutaneous fat (56), which could explain differences in the associations of E-DII with total fat mass and ectopic fat through the [Ruminococcus] gnavus group and Tyzzerella; however, further investigation is needed.

Previous evidence supports a positive association between gut permeability, VAT, and liver fat (39). We did find the suggestion of an association between LBP and VAT (P = 0.035) and liver fat (P = 0.002), which did not reach statistical significance after Bonferroni adjustment. Also, previous studies have reported a positive relation between following a proinflammatory diet (higher DII score) and CRP (57), and between CRP and adiposity (58, 59). For the total fat mass model, our results suggested a relation between E-DII and hs-CRP (P = 0.005) which did not reach statistical significance after Bonferroni adjustment. A significant relation was found between hs-CRP and total fat mass (P < 0.001).

Consistently, we saw a positive association between E-DII score, VAT, and liver fat through lower abundances of Christensenellaceae R-7 group, Coprococcus 2, and [Eubacterium] xylanophilum group. The association between Christensenellaceae and BMI has been reported as one of the strongest links between the microbiome and metabolic disease and is inversely related to BMI (60). Coprococcus 2 has been found to be higher in women with obesity and polycystic ovary syndrome, compared with healthy but obese women (61). [Eubacterium] xylanophilum ferments complex phytochemicals to produce SCFAs including butyrate (62, 63). Butyrate has been found to be antiobesogenic in human studies (64); therefore, the production of this SCFA by the [Eubacterium] xylanophilum group might help explain the mediating role of these bacteria between E-DII and ectopic fat.

The positive association between E-DII, VAT, and liver fat operated through a higher abundance of the phylum Fusobacteria and through the genera Escherichia-Shigella and Lachnoclostridium. Fusobacteria is pathogenic, and in a systematic review evaluating the differences in gut microbiota between people with obesity or of a normal weight, Fusobacteria were more abundant in adults with obesity (65) and have been associated with NAFLD in the MEC (24). Both Escherichia-Shigella and Lachnoclostridium are involved in converting primary bile acids to secondary bile acids (66) and have been positively associated with NAFLD (24).

Lastly, we found that the association between E-DII score and liver fat occurred indirectly through lower α diversity. There are no known mediation studies by which to compare these results; however, in NHANES participants it was reported that enterolignans, biomarkers of microbiota diversity (67), were negatively associated with the E-DII score (36). Moreover, proinflammatory diets, such as the Western diet, are known to be associated with lower microbial diversity (36, 68), and more anti-inflammatory diets, such as the Mediterranean diet, with higher diversity (36, 68).

The strengths of the current study include the large multiethnic population, the use of a validated FFQ, a validated stool sample collection and analysis protocols, and key lifestyle covariates. Another strength is the use of the validated E-DII for scoring the inflammatory potential of the diet, and the use of DXA and MRI to directly measure total fat mass, VAT, and liver fat. Limitations include the single stool sample and blood measurements. Repeated measures of stool samples, and DXA and MRI imaging over time will enable a more precise examination of the temporality of these relations. Participants’ ages ranged from 60 to 77 y; therefore, results might not be generalized to other age groups. Although participants in MEC-APS were relatively healthy, we did not control for comorbidities or use of nonantibiotic drugs, which might have impacted the GM (69). Further research is needed on drug–microbe interactions, where comorbidities co-occur in the MEC. Another limitation was the limited number of participants within each race/ethnic and sex group. Finally, given the observational nature of our data, any conclusion about causality will await the conduct of intervention studies.

In conclusion, we found that the GM and LPS indirectly accounted for the observed associations between E-DII and adiposity phenotypes. The results of this study suggest that following a more anti-inflammatory diet can help to minimize low-grade inflammation and the accumulation of total fat mass and could promote a healthy GM and prevent the build-up of ectopic fat. Large prospective studies with repeat measurements are needed to confirm our findings.

Supplementary Material

nqab398_Supplemental_Tables

Click here for additional data file.^{(80.3KB, docx)}

Acknowledgments

We acknowledge the contribution of the following study staff whose excellent performance made this research possible: Recruitment and Data Collection Core staff at the University of Southern California (USC), including Dr Kristine Monroe for her leadership in recruitment and data collection, as well as her scientific contributions to the Multiethnic Cohort–Adiposity Phenotype Study; Recruitment and Data Collection Core staff at the University of Hawaii Cancer Center (UHCC); Data Management and Analysis Core staff at USC and the UHCC, Project Administrative Core staff at the UHCC; and Dr Adrian Franke and the Analytical Biochemistry Shared Resource at the UHCC for analyzing the blood samples.

The authors’ responsibilities were as follows—CPL, LLM, YBS, GM, S-YP, UL, JWL, MAJH: designed the research; CPL: conducted the research; MAJH, LRW, CJB, JRH, MDW, TE, TR, JAS: analyzed the specimens and data; CPL, MAJH: wrote the paper; all coauthors: provided critical review; CPL, LLM, MAJH: had primary responsibility for final content; and all authors: read and approved the final manuscript. JRH owns controlling interest in Connecting Health Innovations LLC (CHI), a company that has licensed the right to his invention of the dietary inflammatory index (DII^®) from the University of South Carolina in order to develop computer and smart phone applications for patient counseling and dietary intervention in clinical settings. CHI owns exclusive right to the E-DII^TM. The subject matter of this paper will not have any direct bearing on that work, nor has that activity exerted any influence on this project. All other authors report no conflicts of interest.

Notes

Supported by P01 CA168530/CA/NCI NIH HHS/United States, T32 CA229110/CA/NCI NIH HHS/United States, U01 CA164973/CA/NCI NIH HHS/United States.

Supplemental Tables 1–4 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/ajcn/.

Abbreviations used: a, Path a; ab, indirect effect; b, Path b, c, total effect; CRC, colorectal cancer; DII, Dietary Inflammatory Index; E-DII, Energy-Adjusted Dietary Inflammatory Index; GM, gut microbiome; HEI-2010, Healthy Eating Index–2010; hs-CRP, high-sensitivity C-reactive protein; LBP, lipopolysaccharide-binding protein; MEC, Multiethnic Cohort; MEC-APS, Multiethnic Cohort–Adiposity Phenotype Study; MET, ·metabolic equivalent of task; NAFLD, nonalcoholic fatty liver disease; QIIME, Quantitative Insights Into Microbial Ecology; TLR4, Toll-like receptor 4,; UH, University of Hawaii; USC, University of Southern California; VAT, visceral adipose tissue.

Contributor Information

Chloe P Lozano, University of Hawaii Cancer Center, Honolulu, HI, USA.

Lynne R Wilkens, University of Hawaii Cancer Center, Honolulu, HI, USA.

Yurii B Shvetsov, University of Hawaii Cancer Center, Honolulu, HI, USA.

Gertraud Maskarinec, University of Hawaii Cancer Center, Honolulu, HI, USA.

Song-Yi Park, University of Hawaii Cancer Center, Honolulu, HI, USA.

John A Shepherd, University of Hawaii Cancer Center, Honolulu, HI, USA.

Carol J Boushey, University of Hawaii Cancer Center, Honolulu, HI, USA.

James R Hebert, University of South Carolina, Cancer Prevention and Control Program, Department of Epidemiology and Biostatistics, Arnold School of Public Health, Columbia, SC, USA.

Michael D Wirth, University of South Carolina, Cancer Prevention and Control Program, Department of Epidemiology and Biostatistics, Arnold School of Public Health, Columbia, SC, USA.

Thomas Ernst, University of Maryland, Department of Diagnostic Radiology and Nuclear Medicine, Baltimore, MD, USA.

Timothy Randolph, Fred Hutchinson Cancer Research Center, Public Health Sciences Division, Seattle, WA, USA.

Unhee Lim, University of Hawaii Cancer Center, Honolulu, HI, USA.

Johanna W Lampe, Fred Hutchinson Cancer Research Center, Public Health Sciences Division, Seattle, WA, USA.

Loïc Le Marchand, University of Hawaii Cancer Center, Honolulu, HI, USA.

Meredith A J Hullar, Fred Hutchinson Cancer Research Center, Public Health Sciences Division, Seattle, WA, USA.

Data Availability

Data described in the manuscript, code book, and analytic code will be made available upon request pending application and approval.

References

1. National Institute of Diabetes and Digestive and Kidney Diseases . Overweight and obesity statistics. [Internet]. [cited March 24, 2021]. Available from: https://www.niddk.nih.gov/health-information/health-statistics/overweight-obesity [Google Scholar]
2. Heber D. An integrative view of obesity. Am J Clin Nutr. 2010;91(1):280S–3S..] [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Kord Varkaneh H, Fatahi S, Tajik S, Rahmani J, Zarezadeh M, Shab-Bidar S. Dietary inflammatory index in relation to obesity and body mass index: a meta-analysis. Nutr Food Sci. 2018;48(5):702–21. [Google Scholar]
4. Ruiz-Canela M, Zazpe I, Shivappa N, Hébert JR, Sánchez-Tainta A, Corella D, Salas-Salvadó J, Fitó M, Lamuela-Raventós RM, Rekondo Jet al. Dietary inflammatory index and anthropometric measures of obesity in a population sample at high cardiovascular risk from the PREDIMED (PREvención con DIeta MEDiterránea) trial. Br J Nutr. 2015;113(6):984–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, Neyrinck AM, Fava F, Tuohy KM, Chabo Cet al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes. 2007;56(7):1761–72. [DOI] [PubMed] [Google Scholar]
6. Cantero I, Abete I, Babio N, Arós F, Corella D, Estruch R, Fitó M, Hebert JR, Martínez-González MÁ, Pintó Xet al. Dietary inflammatory index and liver status in subjects with different adiposity levels within the PREDIMED trial. Clin Nutr. 2018;37(5):1736–43. [DOI] [PubMed] [Google Scholar]
7. Cuevas-Sierra A, Ramos-Lopez O, Riezu-Boj JI, Milagro FI, Martinez JA. Diet, gut microbiota, and obesity: links with host genetics and epigenetics and potential applications. Adv Nutr. 2019;10(Suppl 1):S17–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Citronberg JS, Curtis KR, White E, Newcomb PA, Newton K, Atkinson C, Song X, Lampe JW, Hullar MA.. Association of gut microbial communities with plasma lipopolysaccharide-binding protein (LBP) in premenopausal women. ISME J. 2018;12(7):1631–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Trebicka J, Bork P, Krag A, Arumugam M. Utilizing the gut microbiome in decompensated cirrhosis and acute-on-chronic liver failure. Nat Rev Gastroenterol Hepatol. 2021;18(3):167–80. [DOI] [PubMed] [Google Scholar]
10. Chapkin RS, Navarro SL, Hullar MAJ, Lampe JW. Diet and gut microbes act coordinately to enhance programmed cell death and reduce colorectal cancer risk. Dig Dis Sci. 2020;65(3):840–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol. 2002;13(1):3–9. [DOI] [PubMed] [Google Scholar]
12. Maskarinec G, Hullar MAJ, Monroe KR, Shepherd JA, Hunt J, Randolph TW, Wilkens LR, Boushey CJ, Le Marchand L, Lim Uet al. Fecal microbial diversity and structure are associated with diet quality in the multiethnic cohort adiposity phenotype study. J Nutr. 2019;149(9):1575–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Hébert JR, Shivappa N, Wirth MD, Hussey JR, Hurley TG. Perspective: the dietary inflammatory index (DII)—lessons learned, improvements made, and future directions. Adv Nutr. 2019;10(2):185–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Shivappa N, Steck SE, Hurley TG, Hussey JR, Hébert JR. Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr. 2014;17(8):1689–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Zheng J, Hoffman KL, Chen J-S, Shivappa N, Sood A, Browman GJ, Dirba DD, Hanash S, Wei P, Hebert JRet al. Dietary inflammatory potential in relation to the gut microbiome: results from a cross-sectional study. Br J Nutr. 2020;124(9):931–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Aslani Z, Sadeghi O, Heidari-Beni M, Zahedi H, Baygi F, Shivappa N, Hébert JR, Moradi S, Sotoudeh G, Asayesh Het al. Association of dietary inflammatory potential with cardiometabolic risk factors and diseases: a systematic review and dose–response meta-analysis of observational studies. Diabetol Metab Syndr. 2020;12(1):86. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Mirmajidi S, Izadi A, Saghafi-Asl M, Vahid F, Karamzad N, Amiri P, Shivappa N, Hébert JR. Inflammatory potential of diet: association with chemerin, omentin, lipopolysaccharide-binding protein, and insulin resistance in the apparently healthy obese. J Am Coll Nutr. 2019;38(4):302–10. [DOI] [PubMed] [Google Scholar]
18. Kolonel LN, Henderson BE, Hankin JH, Nomura AM, Wilkens LR, Pike MC, Stram DO, Monroe KR, Earle ME, Nagamine FS. A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol. 2000;151(4):346–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Lim U, Monroe KR, Buchthal S, Fan B, Cheng I, Kristal BS, Lampe JW, Hullar MA, Franke AA, Stram DOet al. Propensity for intra-abdominal and hepatic adiposity varies among ethnic groups. Gastroenterology. 2019;156(4):966–75.e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Conroy SM, Maskarinec G, Morimoto Y, Franke AA, Cooney RV, Wilkens LR, Goodman MT, Hernadez BY, Le Marchand L, Henderson BEet al. Non-Hodgkin lymphoma and circulating markers of inflammation and adiposity in a nested case-control study: the multiethnic cohort. Cancer Epidemiol Biomarkers Prev. 2013;22(3):337–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Ollberding NJ, Kim Y, Shvetsov YB, Wilkens LR, Franke AA, Cooney RV, Maskarinec G, Hernandez BY, Henderson BE, Marchand LLet al. Prediagnostic leptin, adiponectin, C-reactive protein, and the risk of postmenopausal breast cancer. Cancer Prev Res (Phila Pa). 2013;6(3):188–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Chai W, Morimoto Y, Cooney RV, Franke AA, Shvetsov YB, Le Marchand L, Haiman CA, Kolonel LN, Goodman MT, Maskarinec G. Dietary red and processed meat intake and markers of adiposity and inflammation: the multiethnic cohort study. J Am Coll Nutr. 2017;36(5):378–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Citronberg JS, Wilkens LR, Le Marchand L, Lim U, Monroe KR, Hullar MAJ, White E, Newcomb PA, Lampe JW. Plasma lipopolysaccharide-binding protein and colorectal cancer risk: a nested case–control study in the multiethnic cohort. Cancer Causes Control. 2018;29(1):115–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Hullar MAJ, Jenkins IC, Randolph TW, Curtis KR, Monroe KR, Ernst T, Shepherd JA, Stram DO, Cheng I, Kristal BSet al. Associations of the gut microbiome with hepatic adiposity in the multiethnic cohort adiposity phenotype study. Gut Microbes. 2021;13(1):1965463. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Fu BC, Randolph TW, Lim U, Monroe KR, Cheng I, Wilkens LR, Le Marchand L, Hullar MAJ, Lampe JW. Characterization of the gut microbiome in epidemiologic studies: the multiethnic cohort experience. Ann Epidemiol. 2016;26(5):373–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Fu BC, Randolph TW, Lim U, Monroe KR, Cheng I, Wilkens LR, Le Marchand L, Lampe JW, Hullar MAJ. Temporal variability and stability of the fecal microbiome: the multiethnic cohort study. Cancer Epidemiol Biomarkers Prev. 2019;28(1):154–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Shannon CE. A mathematical theory of communication. Bell Syst Tech J. 1948; 27(3):379–423. [Google Scholar]
28. Hayes AF. Introduction to mediation, moderation, and conditional process analysis. A regression-based approach. 2nd ed. New York (NY): Guilford Publications; 2017. [Google Scholar]
29. Gloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ. Microbiome datasets are compositional: and this is not optional. Front Microbiol. 2017;8:2224. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Farhangi MA, Vajdi M. The association between dietary inflammatory index and risk of central obesity in adults: an updated systematic review and meta-analysis. Int J Vitam Nutr Res. 2020;90(5–6):535–52. [DOI] [PubMed] [Google Scholar]
31. Dominguez LJ, Veronese N, Guerrero-Romero F, Barbagallo M. Magnesium in infectious diseases in older people. Nutrients. 2021;13(1):180. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. King DE. Inflammation and elevation of C-reactive protein: does magnesium play a key role?. Magnes Res. 2009;22(2):57–9. [PubMed] [Google Scholar]
33. Kokot T, Malczyk E, Wilemska-Kucharzewska K, Muc-Wierzgoń M, Fatyga E. Serum magnesium and abdominal obesity and its consequences. In: Watson RR, editor.Nutrition in the prevention and treatment of abdominal obesity. 2nd ed. Elsevier; 2018.. p. 383–91. [Google Scholar]
34. King DE, III AGM, G ME, W RF. Dietary magnesium and C-reactive protein levels. J Am Coll Nutr. 2005;24(3):166–71. [DOI] [PubMed] [Google Scholar]
35. Rodríguez-Hernández H, Gonzalez JL, Rodríguez-Morán M, Guerrero-Romero F. Hypomagnesemia, insulin resistance, and non-alcoholic steatohepatitis in obese subjects. Arch Med Res. 2005;36(4):362–6. [DOI] [PubMed] [Google Scholar]
36. Shivappa N, Wirth MD, Murphy EA, Hurley TG, Hébert JR. Association between the dietary inflammatory index (DII) and urinary enterolignans and C-reactive protein from the National Health and Nutrition Examination Survey-2003–2008. Eur J Nutr. 2019;58(2):797–805. [DOI] [PubMed] [Google Scholar]
37. García-Legorreta A, Soriano-Pérez LA, Flores-Buendía AM, Media-Campos ON, Noriega LG, Granados-Portilo O, Nambo-Venegas R, Tovar AR, Mendoza-Vargas A, Barrera-Oviedo D. Effect of dietary magnesium content on intestinal microbiota of rats. Nutrients. 2020;12(9):2889. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Farhana A, Khan YS. Biochemistry, lipopolysaccharide. StatPearls. [Internet]. Treasure Island (FL): StatPearls Publishing; 2021[cited May 18, 2021]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK554414/. [PubMed] [Google Scholar]
39. Gummesson A, Carlsson LMS, Storlien LH, Bäckhed F, Lundin P, Löfgren L, Stenlöf K, Lam YY, Fagerberg B, Carlsson B. Intestinal permeability is associated with visceral adiposity in healthy women. Obesity. 2011;19(11):2280–2. [DOI] [PubMed] [Google Scholar]
40. Yan C-G, Chen X, Li L, Castellanos FX, Bai T-J, Bo Q-J, Lu X, Sun J, Zhang L, Guo Het al. Reduced default mode network functional connectivity in patients with recurrent major depressive disorder. Proc Natl Acad Sci U S A. 2019;116(18):9078–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Gutsmann T, Müller M, Carroll SF, MacKenzie RC, Wiese A, Seydel U. Dual role of lipopolysaccharide (LPS)-binding protein in neutralization of LPS and enhancement of LPS-induced activation of mononuclear cells. Infect Immun. 2001;69(11):6942–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Mikami A, Ogita T, Namai F, Shigemori S, Sato T, Shimosato T. Oral administration of Flavonifractor plautii attenuates inflammatory responses in obese adipose tissue. Mol Biol Rep. 2020;47(9):6717–25. [DOI] [PubMed] [Google Scholar]
43. Ogita T, Yamamoto Y, Mikami A, Shigemori S, Sato T, Shimosato T. Oral administration of Flavonifractor plautii strongly suppresses Th2 immune responses in mice. Front Immunol. 2020;11:379. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Kasai C, Sugimoto K, Moritani I, Tanaka J, Oya Y, Inoue H, Tameda M, Shiraki K, Ito M, Takei Y. Comparison of the gut microbiota composition between obese and non-obese individuals in a Japanese population, as analyzed by terminal restriction fragment length polymorphism and next-generation sequencing. BMC Gastroenterol. 2015;15(1):100. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Pan X, Wen SW, Kaminga AC, Liu A. Gut metabolites and inflammation factors in non-alcoholic fatty liver disease: a systematic review and meta-analysis. Sci Rep. 2020;10(1):8848. [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Li X, Zeng F, Huang Y, Liu B. The positive effects of Grifola frondosa heteropolysaccharide on NAFLD and regulation of the gut microbiota. Int J Mol Sci. 2019;20(21):5302. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Boots AW, Haenen G, Bast A. Health effects of quercetin: from antioxidant to nutraceutical. Eur J Pharmacol. 2008;585(2-3):325–37. [DOI] [PubMed] [Google Scholar]
48. Gupta A, Dhakan DB, Maji A, Saxena R, Prasoodanan PK V, Mahajan S, Pulikkan J, Kurian J, Gomez AM, Scaria Jet al. Association of Flavonifractor plautii, a flavonoid-degrading bacterium, with the gut microbiome of colorectal cancer patients in India. mSystems. 2019;4(6):e00438–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Png CW, Lindén SK, Gilshenan KS, Zoetendal EG, McSweeney CS, Sly LI, McGuckin MA, Florin THJ. Mucolytic bacteria with increased prevalence in IBD mucosa augment in vitro utilization of mucin by other bacteria. Am J Gastroenterol. 2010;105(11):2420–8. [DOI] [PubMed] [Google Scholar]
50. Crost EH, Tailford LE, Le Gall G, Fons M, Henrissat B, Juge N. Utilisation of mucin glycans by the human gut symbiont Ruminococcus gnavus is strain-dependent. PLoS One. 2013;8(10):e76341. [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Tailford LE, Owen CD, Walshaw J, Crost EH, Hardy-Goddard J, Le Gall G, de Vos WM, Taylor GL, Juge N. Discovery of intramolecular trans-sialidases in human gut microbiota suggests novel mechanisms of mucosal adaptation. Nat Commun. 2015;6(1):7624. [DOI] [PMC free article] [PubMed] [Google Scholar]
52. Henke MT, Kenny DJ, Cassilly CD, Vlamakis H, Xavier RJ, Clardy J. Ruminococcus gnavus, a member of the human gut microbiome associated with Crohn's disease, produces an inflammatory polysaccharide. Proc Natl Acad Sci U S A. 2019;116(26):12672–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
53. Gupta VK, Kim M, Bakshi U, Cunningham KY, Davis JM, Lazaridis KN, Nelson H, Sung J. A predictive index for health status using species-level gut microbiome profiling. Nat Commun. 2020;11(1):2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
54. Ginos BNR, Navarro SL, Schwarz Y, Gu H, Wang D, Randolph TW, Shojaie A, Hullar MAJ, Lampe PD, Kratz M. Circulating bile acids in healthy adults respond differently to a dietary pattern characterized by whole grains, legumes and fruits and vegetables compared to a diet high in refined grains and added sugars: a randomized, controlled, crossover feeding study. Metabolism. 2018;83:197–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
55. Kelly TN, Bazzano LA, Ajami NJ, He H, Zhao J, Petrosino JF, Correa A, He J. Gut microbiome associates with lifetime cardiovascular disease risk profile among Bogalusa Heart Study participants. Circ Res. 2016;119(8):956–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
56. Shuster A, Patlas M, Pinthus JH, Mourtzakis M. The clinical importance of visceral adiposity: a critical review of methods for visceral adipose tissue analysis. Br J Radiol. 2012;85(1009):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
57. Shivappa N, Steck SE, Hurley TG, Hussey JR, Ma Y, Ockene IS, Tabung F, Hébert JR. A population-based dietary inflammatory index predicts levels of C-reactive protein in the seasonal variation of blood cholesterol study (SEASONS). Public Health Nutr. 2014;17(8):1825–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
58. Maki KC, Rains TM, Bell M, Reeves MS, Farmer MV, Yasunaga K. Fat mass, abdominal fat distribution, and C-reactive protein concentrations in overweight and obese men and women. Metab Syndr Relat Disord. 2011;9(4):291–6. [DOI] [PubMed] [Google Scholar]
59. Festa A, D'Agostino R Jr, Williams K, Karter A, Mayer-Davis E, Tracy R, Haffner S. The relation of body fat mass and distribution to markers of chronic inflammation. Int J Obes. 2001;25(10):1407–15. [DOI] [PubMed] [Google Scholar]
60. Waters JL, Ley RE. The human gut bacteria Christensenellaceae are widespread, heritable, and associated with health. BMC Biol. 2019;17(1):83. [DOI] [PMC free article] [PubMed] [Google Scholar]
61. Zhou L, Ni Z, Cheng W, Yu J, Sun S, Zhai D, Yu C, Cai Z. Characteristic gut microbiota and predicted metabolic functions in women with PCOS. Endocr Connect. 2019;9(1):63–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
62. Duncan SH, Russell WR, Quartieri A, Rossi M, Parkhill J, Walker AW, Flint HJ. Wheat bran promotes enrichment within the human colonic microbiota of butyrate-producing bacteria that release ferulic acid. Environ Microbiol. 2016;18(7):2214–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
63. Van Gylswyk NO, Van der Torn JJTK. Eubacterium uniforme sp. nov. and Eubacterium xylanophilum sp. nov., fiber-digesting bacteria from the rumina of sheep fed corn stover. Int J Syst Evol Microbiol. 1985;35:323–326. [Google Scholar]
64. Chakraborti CK. New-found link between microbiota and obesity. World J Gastrointest Pathophysiol. 2015;6(4):110. [DOI] [PMC free article] [PubMed] [Google Scholar]
65. Crovesy L, Masterson D, Rosado EL. Profile of the gut microbiota of adults with obesity: a systematic review. Eur J Clin Nutr. 2020;74(9):1251–62. [DOI] [PubMed] [Google Scholar]
66. Lee J-Y, Arai H, Nakamura Y, Fukiya S, Wada M, Yokota A. Contribution of the 7β-hydroxysteroid dehydrogenase from Ruminococcus gnavus N53 to ursodeoxycholic acid formation in the human colon. J Lipid Res. 2013;54(11):3062–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
67. Hullar MAJ, Lancaster SM, Li F, Tseng E, Beer K, Atkinson C, Wähälä K, Copeland WK, Randolph TW, Newton KMet al. Enterolignan producing phenotypes are associated with increased gut microbial diversity and altered composition in premenopausal women in the United States. Cancer Epidemiol Biomarkers Prev. 2015;24(3):546–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
68. Singh RK, Chang H-W, Yan D, Lee KM, Ucmak D, Wong K, Abrouk M, Farahnik B, Nakamura M, Zhu THet al. Influence of diet on the gut microbiome and implications for human health. J Transl Med. 2017;15(1):73. [DOI] [PMC free article] [PubMed] [Google Scholar]
69. Vich Vila A, Collij V, Sanna S, Sinha T, Imhann F, Bourgonje AR, Mujagic Z, Jonkers D, Masclee AAM, Fu J. Impact of commonly used drugs on the composition and metabolic function of the gut microbiota. Nat Commun. 2020;11:362. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

nqab398_Supplemental_Tables

Click here for additional data file.^{(80.3KB, docx)}

Data Availability Statement

Data described in the manuscript, code book, and analytic code will be made available upon request pending application and approval.

[bib1] 1. National Institute of Diabetes and Digestive and Kidney Diseases . Overweight and obesity statistics. [Internet]. [cited March 24, 2021]. Available from: https://www.niddk.nih.gov/health-information/health-statistics/overweight-obesity [Google Scholar]

[bib2] 2. Heber D. An integrative view of obesity. Am J Clin Nutr. 2010;91(1):280S–3S..] [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib3] 3. Kord Varkaneh H, Fatahi S, Tajik S, Rahmani J, Zarezadeh M, Shab-Bidar S. Dietary inflammatory index in relation to obesity and body mass index: a meta-analysis. Nutr Food Sci. 2018;48(5):702–21. [Google Scholar]

[bib4] 4. Ruiz-Canela M, Zazpe I, Shivappa N, Hébert JR, Sánchez-Tainta A, Corella D, Salas-Salvadó J, Fitó M, Lamuela-Raventós RM, Rekondo Jet al. Dietary inflammatory index and anthropometric measures of obesity in a population sample at high cardiovascular risk from the PREDIMED (PREvención con DIeta MEDiterránea) trial. Br J Nutr. 2015;113(6):984–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] 5. Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, Neyrinck AM, Fava F, Tuohy KM, Chabo Cet al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes. 2007;56(7):1761–72. [DOI] [PubMed] [Google Scholar]

[bib6] 6. Cantero I, Abete I, Babio N, Arós F, Corella D, Estruch R, Fitó M, Hebert JR, Martínez-González MÁ, Pintó Xet al. Dietary inflammatory index and liver status in subjects with different adiposity levels within the PREDIMED trial. Clin Nutr. 2018;37(5):1736–43. [DOI] [PubMed] [Google Scholar]

[bib7] 7. Cuevas-Sierra A, Ramos-Lopez O, Riezu-Boj JI, Milagro FI, Martinez JA. Diet, gut microbiota, and obesity: links with host genetics and epigenetics and potential applications. Adv Nutr. 2019;10(Suppl 1):S17–30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib8] 8. Citronberg JS, Curtis KR, White E, Newcomb PA, Newton K, Atkinson C, Song X, Lampe JW, Hullar MA.. Association of gut microbial communities with plasma lipopolysaccharide-binding protein (LBP) in premenopausal women. ISME J. 2018;12(7):1631–41. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib9] 9. Trebicka J, Bork P, Krag A, Arumugam M. Utilizing the gut microbiome in decompensated cirrhosis and acute-on-chronic liver failure. Nat Rev Gastroenterol Hepatol. 2021;18(3):167–80. [DOI] [PubMed] [Google Scholar]

[bib10] 10. Chapkin RS, Navarro SL, Hullar MAJ, Lampe JW. Diet and gut microbes act coordinately to enhance programmed cell death and reduce colorectal cancer risk. Dig Dis Sci. 2020;65(3):840–51. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib11] 11. Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol. 2002;13(1):3–9. [DOI] [PubMed] [Google Scholar]

[bib12] 12. Maskarinec G, Hullar MAJ, Monroe KR, Shepherd JA, Hunt J, Randolph TW, Wilkens LR, Boushey CJ, Le Marchand L, Lim Uet al. Fecal microbial diversity and structure are associated with diet quality in the multiethnic cohort adiposity phenotype study. J Nutr. 2019;149(9):1575–84. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib13] 13. Hébert JR, Shivappa N, Wirth MD, Hussey JR, Hurley TG. Perspective: the dietary inflammatory index (DII)—lessons learned, improvements made, and future directions. Adv Nutr. 2019;10(2):185–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib14] 14. Shivappa N, Steck SE, Hurley TG, Hussey JR, Hébert JR. Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr. 2014;17(8):1689–96. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib15] 15. Zheng J, Hoffman KL, Chen J-S, Shivappa N, Sood A, Browman GJ, Dirba DD, Hanash S, Wei P, Hebert JRet al. Dietary inflammatory potential in relation to the gut microbiome: results from a cross-sectional study. Br J Nutr. 2020;124(9):931–42. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib16] 16. Aslani Z, Sadeghi O, Heidari-Beni M, Zahedi H, Baygi F, Shivappa N, Hébert JR, Moradi S, Sotoudeh G, Asayesh Het al. Association of dietary inflammatory potential with cardiometabolic risk factors and diseases: a systematic review and dose–response meta-analysis of observational studies. Diabetol Metab Syndr. 2020;12(1):86. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] 17. Mirmajidi S, Izadi A, Saghafi-Asl M, Vahid F, Karamzad N, Amiri P, Shivappa N, Hébert JR. Inflammatory potential of diet: association with chemerin, omentin, lipopolysaccharide-binding protein, and insulin resistance in the apparently healthy obese. J Am Coll Nutr. 2019;38(4):302–10. [DOI] [PubMed] [Google Scholar]

[bib18] 18. Kolonel LN, Henderson BE, Hankin JH, Nomura AM, Wilkens LR, Pike MC, Stram DO, Monroe KR, Earle ME, Nagamine FS. A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol. 2000;151(4):346–57. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib19] 19. Lim U, Monroe KR, Buchthal S, Fan B, Cheng I, Kristal BS, Lampe JW, Hullar MA, Franke AA, Stram DOet al. Propensity for intra-abdominal and hepatic adiposity varies among ethnic groups. Gastroenterology. 2019;156(4):966–75.e10. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib20] 20. Conroy SM, Maskarinec G, Morimoto Y, Franke AA, Cooney RV, Wilkens LR, Goodman MT, Hernadez BY, Le Marchand L, Henderson BEet al. Non-Hodgkin lymphoma and circulating markers of inflammation and adiposity in a nested case-control study: the multiethnic cohort. Cancer Epidemiol Biomarkers Prev. 2013;22(3):337–47. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib21] 21. Ollberding NJ, Kim Y, Shvetsov YB, Wilkens LR, Franke AA, Cooney RV, Maskarinec G, Hernandez BY, Henderson BE, Marchand LLet al. Prediagnostic leptin, adiponectin, C-reactive protein, and the risk of postmenopausal breast cancer. Cancer Prev Res (Phila Pa). 2013;6(3):188–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib22] 22. Chai W, Morimoto Y, Cooney RV, Franke AA, Shvetsov YB, Le Marchand L, Haiman CA, Kolonel LN, Goodman MT, Maskarinec G. Dietary red and processed meat intake and markers of adiposity and inflammation: the multiethnic cohort study. J Am Coll Nutr. 2017;36(5):378–85. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib23] 23. Citronberg JS, Wilkens LR, Le Marchand L, Lim U, Monroe KR, Hullar MAJ, White E, Newcomb PA, Lampe JW. Plasma lipopolysaccharide-binding protein and colorectal cancer risk: a nested case–control study in the multiethnic cohort. Cancer Causes Control. 2018;29(1):115–23. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] 24. Hullar MAJ, Jenkins IC, Randolph TW, Curtis KR, Monroe KR, Ernst T, Shepherd JA, Stram DO, Cheng I, Kristal BSet al. Associations of the gut microbiome with hepatic adiposity in the multiethnic cohort adiposity phenotype study. Gut Microbes. 2021;13(1):1965463. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib25] 25. Fu BC, Randolph TW, Lim U, Monroe KR, Cheng I, Wilkens LR, Le Marchand L, Hullar MAJ, Lampe JW. Characterization of the gut microbiome in epidemiologic studies: the multiethnic cohort experience. Ann Epidemiol. 2016;26(5):373–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] 26. Fu BC, Randolph TW, Lim U, Monroe KR, Cheng I, Wilkens LR, Le Marchand L, Lampe JW, Hullar MAJ. Temporal variability and stability of the fecal microbiome: the multiethnic cohort study. Cancer Epidemiol Biomarkers Prev. 2019;28(1):154–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib27] 27. Shannon CE. A mathematical theory of communication. Bell Syst Tech J. 1948; 27(3):379–423. [Google Scholar]

[bib28] 28. Hayes AF. Introduction to mediation, moderation, and conditional process analysis. A regression-based approach. 2nd ed. New York (NY): Guilford Publications; 2017. [Google Scholar]

[bib29] 29. Gloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ. Microbiome datasets are compositional: and this is not optional. Front Microbiol. 2017;8:2224. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib30] 30. Farhangi MA, Vajdi M. The association between dietary inflammatory index and risk of central obesity in adults: an updated systematic review and meta-analysis. Int J Vitam Nutr Res. 2020;90(5–6):535–52. [DOI] [PubMed] [Google Scholar]

[bib31] 31. Dominguez LJ, Veronese N, Guerrero-Romero F, Barbagallo M. Magnesium in infectious diseases in older people. Nutrients. 2021;13(1):180. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib32] 32. King DE. Inflammation and elevation of C-reactive protein: does magnesium play a key role?. Magnes Res. 2009;22(2):57–9. [PubMed] [Google Scholar]

[bib33] 33. Kokot T, Malczyk E, Wilemska-Kucharzewska K, Muc-Wierzgoń M, Fatyga E. Serum magnesium and abdominal obesity and its consequences. In: Watson RR, editor.Nutrition in the prevention and treatment of abdominal obesity. 2nd ed. Elsevier; 2018.. p. 383–91. [Google Scholar]

[bib34] 34. King DE, III AGM, G ME, W RF. Dietary magnesium and C-reactive protein levels. J Am Coll Nutr. 2005;24(3):166–71. [DOI] [PubMed] [Google Scholar]

[bib35] 35. Rodríguez-Hernández H, Gonzalez JL, Rodríguez-Morán M, Guerrero-Romero F. Hypomagnesemia, insulin resistance, and non-alcoholic steatohepatitis in obese subjects. Arch Med Res. 2005;36(4):362–6. [DOI] [PubMed] [Google Scholar]

[bib36] 36. Shivappa N, Wirth MD, Murphy EA, Hurley TG, Hébert JR. Association between the dietary inflammatory index (DII) and urinary enterolignans and C-reactive protein from the National Health and Nutrition Examination Survey-2003–2008. Eur J Nutr. 2019;58(2):797–805. [DOI] [PubMed] [Google Scholar]

[bib37] 37. García-Legorreta A, Soriano-Pérez LA, Flores-Buendía AM, Media-Campos ON, Noriega LG, Granados-Portilo O, Nambo-Venegas R, Tovar AR, Mendoza-Vargas A, Barrera-Oviedo D. Effect of dietary magnesium content on intestinal microbiota of rats. Nutrients. 2020;12(9):2889. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib38] 38. Farhana A, Khan YS. Biochemistry, lipopolysaccharide. StatPearls. [Internet]. Treasure Island (FL): StatPearls Publishing; 2021[cited May 18, 2021]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK554414/. [PubMed] [Google Scholar]

[bib39] 39. Gummesson A, Carlsson LMS, Storlien LH, Bäckhed F, Lundin P, Löfgren L, Stenlöf K, Lam YY, Fagerberg B, Carlsson B. Intestinal permeability is associated with visceral adiposity in healthy women. Obesity. 2011;19(11):2280–2. [DOI] [PubMed] [Google Scholar]

[bib40] 40. Yan C-G, Chen X, Li L, Castellanos FX, Bai T-J, Bo Q-J, Lu X, Sun J, Zhang L, Guo Het al. Reduced default mode network functional connectivity in patients with recurrent major depressive disorder. Proc Natl Acad Sci U S A. 2019;116(18):9078–83. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib41] 41. Gutsmann T, Müller M, Carroll SF, MacKenzie RC, Wiese A, Seydel U. Dual role of lipopolysaccharide (LPS)-binding protein in neutralization of LPS and enhancement of LPS-induced activation of mononuclear cells. Infect Immun. 2001;69(11):6942–50. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib42] 42. Mikami A, Ogita T, Namai F, Shigemori S, Sato T, Shimosato T. Oral administration of Flavonifractor plautii attenuates inflammatory responses in obese adipose tissue. Mol Biol Rep. 2020;47(9):6717–25. [DOI] [PubMed] [Google Scholar]

[bib43] 43. Ogita T, Yamamoto Y, Mikami A, Shigemori S, Sato T, Shimosato T. Oral administration of Flavonifractor plautii strongly suppresses Th2 immune responses in mice. Front Immunol. 2020;11:379. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib44] 44. Kasai C, Sugimoto K, Moritani I, Tanaka J, Oya Y, Inoue H, Tameda M, Shiraki K, Ito M, Takei Y. Comparison of the gut microbiota composition between obese and non-obese individuals in a Japanese population, as analyzed by terminal restriction fragment length polymorphism and next-generation sequencing. BMC Gastroenterol. 2015;15(1):100. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib45] 45. Pan X, Wen SW, Kaminga AC, Liu A. Gut metabolites and inflammation factors in non-alcoholic fatty liver disease: a systematic review and meta-analysis. Sci Rep. 2020;10(1):8848. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib46] 46. Li X, Zeng F, Huang Y, Liu B. The positive effects of Grifola frondosa heteropolysaccharide on NAFLD and regulation of the gut microbiota. Int J Mol Sci. 2019;20(21):5302. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib47] 47. Boots AW, Haenen G, Bast A. Health effects of quercetin: from antioxidant to nutraceutical. Eur J Pharmacol. 2008;585(2-3):325–37. [DOI] [PubMed] [Google Scholar]

[bib48] 48. Gupta A, Dhakan DB, Maji A, Saxena R, Prasoodanan PK V, Mahajan S, Pulikkan J, Kurian J, Gomez AM, Scaria Jet al. Association of Flavonifractor plautii, a flavonoid-degrading bacterium, with the gut microbiome of colorectal cancer patients in India. mSystems. 2019;4(6):e00438–19. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib49] 49. Png CW, Lindén SK, Gilshenan KS, Zoetendal EG, McSweeney CS, Sly LI, McGuckin MA, Florin THJ. Mucolytic bacteria with increased prevalence in IBD mucosa augment in vitro utilization of mucin by other bacteria. Am J Gastroenterol. 2010;105(11):2420–8. [DOI] [PubMed] [Google Scholar]

[bib50] 50. Crost EH, Tailford LE, Le Gall G, Fons M, Henrissat B, Juge N. Utilisation of mucin glycans by the human gut symbiont Ruminococcus gnavus is strain-dependent. PLoS One. 2013;8(10):e76341. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib51] 51. Tailford LE, Owen CD, Walshaw J, Crost EH, Hardy-Goddard J, Le Gall G, de Vos WM, Taylor GL, Juge N. Discovery of intramolecular trans-sialidases in human gut microbiota suggests novel mechanisms of mucosal adaptation. Nat Commun. 2015;6(1):7624. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib52] 52. Henke MT, Kenny DJ, Cassilly CD, Vlamakis H, Xavier RJ, Clardy J. Ruminococcus gnavus, a member of the human gut microbiome associated with Crohn's disease, produces an inflammatory polysaccharide. Proc Natl Acad Sci U S A. 2019;116(26):12672–7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib53] 53. Gupta VK, Kim M, Bakshi U, Cunningham KY, Davis JM, Lazaridis KN, Nelson H, Sung J. A predictive index for health status using species-level gut microbiome profiling. Nat Commun. 2020;11(1):2020. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib54] 54. Ginos BNR, Navarro SL, Schwarz Y, Gu H, Wang D, Randolph TW, Shojaie A, Hullar MAJ, Lampe PD, Kratz M. Circulating bile acids in healthy adults respond differently to a dietary pattern characterized by whole grains, legumes and fruits and vegetables compared to a diet high in refined grains and added sugars: a randomized, controlled, crossover feeding study. Metabolism. 2018;83:197–204. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib55] 55. Kelly TN, Bazzano LA, Ajami NJ, He H, Zhao J, Petrosino JF, Correa A, He J. Gut microbiome associates with lifetime cardiovascular disease risk profile among Bogalusa Heart Study participants. Circ Res. 2016;119(8):956–64. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib56] 56. Shuster A, Patlas M, Pinthus JH, Mourtzakis M. The clinical importance of visceral adiposity: a critical review of methods for visceral adipose tissue analysis. Br J Radiol. 2012;85(1009):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib57] 57. Shivappa N, Steck SE, Hurley TG, Hussey JR, Ma Y, Ockene IS, Tabung F, Hébert JR. A population-based dietary inflammatory index predicts levels of C-reactive protein in the seasonal variation of blood cholesterol study (SEASONS). Public Health Nutr. 2014;17(8):1825–33. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib58] 58. Maki KC, Rains TM, Bell M, Reeves MS, Farmer MV, Yasunaga K. Fat mass, abdominal fat distribution, and C-reactive protein concentrations in overweight and obese men and women. Metab Syndr Relat Disord. 2011;9(4):291–6. [DOI] [PubMed] [Google Scholar]

[bib59] 59. Festa A, D'Agostino R Jr, Williams K, Karter A, Mayer-Davis E, Tracy R, Haffner S. The relation of body fat mass and distribution to markers of chronic inflammation. Int J Obes. 2001;25(10):1407–15. [DOI] [PubMed] [Google Scholar]

[bib60] 60. Waters JL, Ley RE. The human gut bacteria Christensenellaceae are widespread, heritable, and associated with health. BMC Biol. 2019;17(1):83. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib61] 61. Zhou L, Ni Z, Cheng W, Yu J, Sun S, Zhai D, Yu C, Cai Z. Characteristic gut microbiota and predicted metabolic functions in women with PCOS. Endocr Connect. 2019;9(1):63–73. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib62] 62. Duncan SH, Russell WR, Quartieri A, Rossi M, Parkhill J, Walker AW, Flint HJ. Wheat bran promotes enrichment within the human colonic microbiota of butyrate-producing bacteria that release ferulic acid. Environ Microbiol. 2016;18(7):2214–25. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib63] 63. Van Gylswyk NO, Van der Torn JJTK. Eubacterium uniforme sp. nov. and Eubacterium xylanophilum sp. nov., fiber-digesting bacteria from the rumina of sheep fed corn stover. Int J Syst Evol Microbiol. 1985;35:323–326. [Google Scholar]

[bib64] 64. Chakraborti CK. New-found link between microbiota and obesity. World J Gastrointest Pathophysiol. 2015;6(4):110. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib65] 65. Crovesy L, Masterson D, Rosado EL. Profile of the gut microbiota of adults with obesity: a systematic review. Eur J Clin Nutr. 2020;74(9):1251–62. [DOI] [PubMed] [Google Scholar]

[bib66] 66. Lee J-Y, Arai H, Nakamura Y, Fukiya S, Wada M, Yokota A. Contribution of the 7β-hydroxysteroid dehydrogenase from Ruminococcus gnavus N53 to ursodeoxycholic acid formation in the human colon. J Lipid Res. 2013;54(11):3062–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib67] 67. Hullar MAJ, Lancaster SM, Li F, Tseng E, Beer K, Atkinson C, Wähälä K, Copeland WK, Randolph TW, Newton KMet al. Enterolignan producing phenotypes are associated with increased gut microbial diversity and altered composition in premenopausal women in the United States. Cancer Epidemiol Biomarkers Prev. 2015;24(3):546–54. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib68] 68. Singh RK, Chang H-W, Yan D, Lee KM, Ucmak D, Wong K, Abrouk M, Farahnik B, Nakamura M, Zhu THet al. Influence of diet on the gut microbiome and implications for human health. J Transl Med. 2017;15(1):73. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib69] 69. Vich Vila A, Collij V, Sanna S, Sinha T, Imhann F, Bourgonje AR, Mujagic Z, Jonkers D, Masclee AAM, Fu J. Impact of commonly used drugs on the composition and metabolic function of the gut microbiota. Nat Commun. 2020;11:362. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Associations of the Dietary Inflammatory Index with total adiposity and ectopic fat through the gut microbiota, LPS, and C-reactive protein in the Multiethnic Cohort–Adiposity Phenotype Study

Chloe P Lozano

Lynne R Wilkens

Yurii B Shvetsov

Gertraud Maskarinec

Song-Yi Park

John A Shepherd

Carol J Boushey

James R Hebert

Michael D Wirth

Thomas Ernst

Timothy Randolph

Unhee Lim

Johanna W Lampe

Loïc Le Marchand

Meredith A J Hullar

ABSTRACT

Background

Objectives

Methods

Results

Conclusions

Introduction

Methods

Study population

FIGURE 1.

MRI and DXA scanning

DII

Blood biomarkers analysis

Fecal microbiome analysis

Statistical methods

FIGURE 2.

Results

TABLE 1.

TABLE 2.

TABLE 3.

TABLE 4.

TABLE 5.

Discussion

Supplementary Material

Acknowledgments

Notes

Contributor Information

Data Availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases