Although short-term feeding studies demonstrated effects of grains, fiber, and gluten on gut microbiome composition, the impact of habitual intake of these dietary factors is poorly understood. We examined whether habitual intakes of whole and refined grains, fiber, and gluten are associated with gut microbiota in a cross-sectional study. This study included 779 participants from the multi-ethnic Food and Microbiome Longitudinal Investigation study. Bacterial 16SV4 rRNA gene from baseline stool was amplified and sequenced using Illumina MiSeq. Read clustering and taxonomic assignment was performed using QIIME2. Usual dietary intake was assessed by a 137-item food frequency questionnaire. Association of diet with gut microbiota was assessed with respect to overall composition and specific taxon abundances. Whole grain intake was associated with overall composition, as measured by the Jensen–Shannon divergence (multivariable-adjusted Ptrend for quartiles = 0.03). The highest intake quartile was associated with higher abundance of Bacteroides plebeius, Faecalibacterium prausnitzii, Blautia producta, and Erysipelotrichaceae and lower abundance of Bacteroides uniformis. These bacteria also varied by dietary fiber intake. Higher refined grain and gluten intake was associated with lower Shannon diversity (Ptrend < 0.05). These findings suggest that whole grain and dietary fiber are associated with overall gut microbiome structure, largely fiber-fermenting microbiota. Higher refined grain and gluten intakes may be associated with lower microbial diversity.

Significance:

Regular consumption of whole grains and dietary fiber was associated with greater abundance of gut bacteria that may lower risk of colorectal cancer. Further research on the association of refined grains and gluten with gut microbial composition is needed to understand their roles in health and disease.

Whole grains are associated with lower risk of chronic diseases (1), including cancer and total cancer mortality, while refined grains do not have similar beneficial associations with cancer risk (2). Much of the evidence on cancer has focused on whole grains and colorectal cancer, and the World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) has concluded that there is “probable” evidence that whole grains reduce colorectal cancer risk (3). As a result, the U.S. Dietary Guidelines to promote health and prevent chronic diseases recommend a healthy dietary pattern that includes grains, at least half of which are whole grains (4). Research on components of overall dietary patterns, such as of grains, helps to build evidence on the building blocks of healthy dietary patterns and inform research and public health messages on healthy diet.

The beneficial association between whole grains and colorectal cancer risk may be at least partly attributable to fermentation of these foods by short-chain fatty acid (SCFA)-producing gut microbes. Whole grains, unlike refined grains, are rich in fermentable dietary fiber and consequently have a “prebiotic” effect on gut microbes to produce SCFAs, such as butyrate, which are suggested to facilitate growth and differentiation of normal colonocytes while inhibiting tumor cell growth (5). Thus, diets rich in whole rather than refined grains are hypothesized to promote growth of SCFA-producing microbes while decreasing abundance of proinflammatory species (6, 7).

Gluten is another component of both whole and refined grains that may also alter gut microbiome composition. In the human gastrointestinal tract, gluten, composed of grain storage proteins, is cleaved into proline- and glutamine-rich peptides by proteolytic gut microbiota (8). Incomplete digestion of gluten can lead to immune responses that are characteristic of celiac disease, wheat allergy, and non-celiac gluten sensitivity. Although the role of gut microbiota is unclear in these conditions, differences in microbial composition have been observed in patients with celiac disease when compared with healthy populations (9), suggesting that certain microbes are needed in the digestion of gluten-containing foods.

Although strong evidence exists on the beneficial effect of whole grain foods on colorectal cancer risk, randomized controlled trials of whole grain foods and the gut microbiome have shown inconsistent findings (10). In addition, limited evidence from dietary intervention studies of gluten suggests that gluten-free or low-gluten diets alter gut microbial composition in healthy populations by decreasing several beneficial species associated with carbohydrate metabolism (11–13), but the evidence is largely based on small, short-term trials, similar to whole grain feeding studies (14–20). Little is known regarding the impact of habitual intake of these dietary factors on gut microbiota, and given that interest in gluten has steadily grown in recent years with increasing adherence to gluten-free diets (21), understanding the association of gluten-containing foods and gut microbiota is also of public interest.

To contribute to the limited evidence on gluten, grains, and gut microbiota, we examined self-reported habitual consumption of whole and refined grains, and their major components dietary fiber and gluten, with gut microbial composition, among participants of the Food and Microbiome Longitudinal Investigation (FAMiLI; ref. 22).

Study Population

The FAMiLI is an ongoing prospective cohort study of racially and ethnically diverse residents in New York City and surrounding geographic areas. Briefly, men and women ages 40 years or older who were not currently pregnant or on long-term antibiotic therapy were invited to enroll through clinic, community, and web-based recruitment. Participants with recent antibiotic use were eligible if the last dose was at least 2 weeks prior to the enrollment date. To date, over 9,500 participants have enrolled, with the goal of 15,000, and continued recruitment and prospective follow-up for disease outcomes are planned. At baseline, participants completed a questionnaire containing demographic, lifestyle, and dietary questions, and submitted self-collected oral and stool samples. The study was approved by the New York University Langone Health Institutional Review Board (#s12-00855), and all participants provided written informed consent. For the current analysis, participants who completed both demographic and dietary questionnaires and provided a stool sample were included, as published elsewhere (22). In the first phase, 1,000 participants were recruited and of those, 873 stool samples were selected for microbiome sequencing and included in this analysis.

Dietary Assessment

Diet was assessed using a 137-item food frequency questionnaire (FFQ), known as the Dietary Questionnaire (DQX; ref. 23). The design of the DQX was based on two previously validated FFQs (24, 25) and is similar to the Diet History Questionnaire (DHQ), which was validated previously (26–29). The DQX was translated into Korean and Spanish for the FAMiLI participants. One additional page of culturally relevant foods was included with the DQX for Korean and Chinese participants, but because this page has not been validated, these food items were not included in the current study.

Study participants reported their frequency of consumption for 18 grain-containing items on the FFQ by selecting one of the 10 listed frequencies (never; less than once per month; one time per month; two to three times per month; one time per week; two times per week; three to four times per week; five to six times per week; one time per day; or two or more times per day) and one of the three listed serving sizes (small, medium, or large). The food items were categorized as whole or refined grains and as high- or low-gluten grain foods. Grain-based foods were classified as whole or refined based on the definition that whole grains retain 100% of the original kernel (30) and on extensive literature, including the WCRF/AICR systematic literature review on whole grains and colorectal cancer risk (3, 31). Whole grain foods in this analysis included high fiber cold cereals, dark wheat or rye breads, brown or wild rice, and other whole grains. Refined grain foods included cooked cereals or grits, other cold cereals, pancakes/waffles, white bread, corn bread, biscuits/muffins, white rice, pasta, pizza, and crackers. Five additional FFQ items were queried regarding refined grain-based sweets/dessert products; these items were not included in the primary analyses because they are frequently high in sugar and/or saturated fat, which may independently influence gut microbiota, but were included in sensitivity analyses.

To calculate the gluten content of whole and refined grain products, the amount of protein contained in wheat, barley, and rye food items was estimated from the USDA food composition database, FoodData Central (32). Then, the protein content was multiplied by a conversion factor of 0.75, based on the Osborne plant protein classification system (33). Previous estimates of the gluten fraction of proteins using this system applied conversion factors of 0.75 or 0.80 (34–36). Gluten estimation using the Osborne system has shown an acceptable correlation (r = 0.70; 95% confidence interval, 0.35–0.88) with gluten measurement using ELISAs (37).

The intakes of whole and refined grain, gluten, dietary fiber, and dietary fiber from grain-based foods were energy adjusted using the nutrient density method and then analyzed as continuous variables and sex-specific quartiles. For this analysis, participants who were missing ≥60 FFQ items, which also accounted for those missing >50% of the grain items (n = 48), and those who reported implausible total energy intake in the lowest or highest 1% (≤430 or ≥9,243 kcal/day for men and ≤412 or ≥6,703 kcal/day for women; n = 16) were excluded from this analysis. After exclusion of 64 subjects, 809 of 873 (93%) subjects remained.

Gut Microbiome Assessment

Stool samples were self-collected by study participants upon enrollment using RNAlater collection kits. Participants were instructed to immediately ship the sample after collection, and average shipping time from the participant to the laboratory was 3–4 days. Because the samples were collected using RNAlater, shipments did not require cold or other special packaging. Samples were frozen immediately upon receipt until sequencing was performed. 16SV4 rRNA gene sequencing was conducted as described previously (22). DNA was extracted using the Mo Bio PowerSoil DNA isolation kit. The V4 region of the 16S rRNA gene was PCR amplified with the 515F/806R primer pair, which included sequencer adapter sequences used in the Illumina flowcell and sample-specific barcodes (38, 39). PCR products were quantified using PicoGreen (Invitrogen) and a plate reader (Infinite 200 PRO, Tecan). Sample PCR products were then pooled in equimolar amounts, purified using AMPure XP Beads (Beckman Coulter), and then quantified using a fluorometer (Qubit, Invitrogen). Amplicons were sequenced on a 151 bp × 12 bp × 151 bp MiSeq run (39).

Sequence reads were processed using QIIME2 (40). Briefly, sequence reads were demultiplexed and paired-end reads were joined, followed by quality filtering as described previously (41). The Deblur workflow was applied, which uses sequence error profiles to obtain putative error-free sequences, referred to as “sub” operational taxonomic units (s-OTU; ref. 42). s-OTUs were assigned taxonomy using a naïve Bayes classifier pretrained on the Greengenes (43) 13_8 99% OTUs, where the sequences have been trimmed to only include 250 bases from the 16S V4 region, bound by the 515F/806R primer pair. A phylogenetic tree was constructed via sequence alignment with MAFFT (44), filtering the alignment, and applying the FastTree algorithm (45) to generate the phylogenetic tree. The number of observed s-OTUs and Shannon diversity index were calculated in 100 iterations at 50 different rarefied sequencing depths (from 15 to 5,000 sequence reads per sample) and averaged for each subject at each depth (22).

In addition to the previously mentioned dietary exclusions, subjects for whom sequencing failed (n = 9) and subjects with sequencing depths <250 sequence reads per sample after the Deblur workflow (n = 21) were excluded, as described previously (22). After all exclusions, the final analytic cohort included 779 participants (291 men and 488 women). The data generated in this study are available upon request from the corresponding author.

Statistical Analysis

The associations of sex-specific quartiles of energy-adjusted whole and refined grain, gluten, and dietary and grain fiber intake with within-subject microbial diversity (alpha-diversity) were evaluated using the number of observed s-OTUs (richness) and the Shannon diversity index. The dietary exposures were modeled using multivariable-adjusted linear regression models with richness and the Shannon index as outcomes, and adjusted for age, sex, race/ethnicity, body mass index (BMI; categories of 18–<25 kg/m2, 25–<30 kg/m2, 30–<35 kg/m2, ≥35 kg/m2), smoking status (ever/never), alcohol intake (g/day), and total energy intake (kcal/day, continuous). We also tested whole grain models additionally adjusted for refined grains and vice versa, and whole and refined grain and gluten models additionally adjusted for dietary fiber and fruit and vegetable (servings/day) intake. Trend tests across quartiles of intake were calculated using the median quartile values, modeled as a continuous term. The association of quartiles of whole and refined grain, gluten, and dietary and grain fiber intake with overall gut microbiota composition was assessed using Jensen–Shannon divergence (JSD). Permutational multivariate analysis of variance was performed using R software (version 3.6.1; R Development Core Team, 2020), R package “vegan” (46), and adonis function to assess this association (47), and models were adjusted for the same covariates mentioned previously.

The analysis of composition of microbiomes (ANCOM) method (48) was used to identify s-OTUs and higher level taxa that were differentially abundant between the highest and lowest intakes of the various dietary exposures, adjusting for the previously mentioned covariates. For this analysis, taxa at the detection level ≥0.80 were considered significant.

In sensitivity analyses, analyses were repeated stratified by racial/ethnic group and after excluding participants who self-reported type 2 diabetes (n = 130) or inflammatory bowel disease (Crohn’s disease or ulcerative colitis; n = 21) at enrollment. Grain-based sweets and desserts were also included in the estimation of refined grain and gluten intake. Analyses were performed using R version 3.6.1 (R Core Team, 2020).

Data Availability

The 16S rRNA sequencing data that support the findings of this study have been deposited in the Sequence Read Archive (PRJNA559143), along with demographic metadata, to be released upon publication. Additional data on the study participants are available from the corresponding author upon reasonable request.

Participant Characteristics

In this study population, 36.7% of participants were White, 10.4% were Black, 15.0% were Hispanic, 30.2% were Korean, 3.1% were other Asian, and 1.0% were other race/ethnicities. Participants with the highest intakes of whole grain were older, had a lower BMI, and had higher total energy, dietary fiber, and fruit and vegetable intakes but lower calories from fat and intakes of alcohol, processed meat, and refined grains (Table 1). In addition, greater whole grain consumers were more likely to be Korean, more highly educated, regular aspirin users, and physically active. Conversely, greater refined grain consumers were generally younger, heavier, less physically active, and had lower intakes of dietary fiber and fruits and vegetables.

TABLE 1

Baseline characteristics of study participants by sex-specific quartiles of energy-adjusted whole and refined grain intake in the FAMiLI

Whole grain quartilesa,bRefined grain quartilesa,b
Q1Q2Q3Q4Q1Q2Q3Q4
N = 196N = 195N = 193N = 196N = 195N = 196N = 193N = 196
Range: Males0–8.0>8.0–20.2>20.2–46.5>46.5–2990–47.7>47.7–80.5>80.5–120>120–343
Females0–10.3>10.3–27.5>27.5–63.8>63.8–4480–41.9>41.9–74.4>74.4–119>119–370
Age (years) 57.6 (11.2) 57.2 (10.5) 60.2 (11.4) 62.8 (11.0) 64.5 (10.9) 59.7 (10.7) 58.2 (10.2) 55.4 (11.1) 
Female, % 62.6 62.6 62.9 62.6 62.6 62.6 62.9 62.6 
BMI (kg/m229.1 (7.0) 28.5 (6.3) 27.6 (5.9) 24.4 (4.9) 25.9 (5.3) 27.5 (6.8) 28.4 (6.3) 27.9 (6.5) 
Race/ethnicity, % 
 White 37.9 52.3 48.5 13.8 26.7 43.1 44.8 37.9 
 Black 15.4 12.3 9.8 5.6 5.1 14.9 12.9 10.3 
 Hispanic 28.2 18.5 13.4 2.1 7.7 11.3 19.6 23.6 
 Korean 14.4 11.3 23.7 75.9 59.0 25.6 19.6 21.0 
 Asian, other than Korean 3.6 3.6 3.6 2.1 1.0 4.1 2.1 5.6 
 Otherc 0.5 2.1 1.0 0.5 0.5 1.0 1.0 1.5 
College graduate or higher, % 44.1 59.5 64.9 53.3 49.7 61.5 58.8 51.8 
Physical activity, % 
 None 21.0 13.3 7.7 11.8 9.2 10.8 13.4 20.5 
 <4 hours/week 56.4 60.0 54.1 49.2 44.1 61.0 62.4 52.3 
 4 or more hours/week 22.6 26.7 38.1 39.0 46.7 28.2 24.2 27.2 
Ever smoker, % 33.3 41.0 31.4 24.6 30.8 36.9 30.9 31.8 
Alcohol consumption (g/day) 8.0 (22.2) 14.2 (31.9) 8.0 (22.5) 4.0 (15.6) 9.4 (29.5) 9.6 (23.7) 9.6 (24.2) 5.6 (16.8) 
Family history of cancer, % 55.9 55.9 53.1 39.0 45.6 57.9 53.1 47.2 
Regular aspirin use, % 32.3 33.3 33.5 41.5 34.9 34.4 36.6 34.9 
History of ulcerative colitis, % 3.1 1.0 2.1 2.6 1.5 2.6 2.6 2.1 
History of Crohn’s disease, % 0.5 1.0 0.5 0.0 0.0 1.0 0.5 0.5 
Total energy (kcal/day) 1,889 (1,047) 2,015 (1,212) 2,052 (1,050) 2,122 (1,036) 2,106 (1,017) 1,905 (935) 2,059 (1,064) 2,007 (1,306) 
Total fat (% kcal/day) 31.2 (8.2) 30.9 (6.6) 29.4 (5.4) 25.8 (4.6) 28.0 (7.7) 29.9 (7.0) 30.2 (6.2) 29.2 (5.6) 
Total carbohydrates (% kcal/day) 51.7 (11.2) 50.4 (9.4) 54.1 (8.1) 58.3 (7.0) 54.3 (11.2) 53.0 (10.3) 52.6 (8.3) 54.5 (7.9) 
Red meat (servings/day) 0.6 (0.7) 0.7 (0.8) 0.6 (0.6) 0.5 (0.7) 0.6 (0.8) 0.5 (0.5) 0.6 (0.7) 0.6 (0.9) 
Processed meat (servings/day) 0.7 (1.0) 0.6 (0.8) 0.4 (0.7) 0.2 (0.5) 0.4 (0.8) 0.5 (0.8) 0.6 (0.8) 0.5 (0.8) 
Fruits and vegetables (servings/day) 5.9 (4.6) 7.2 (4.5) 8.1 (4.7) 9.6 (5.3) 9.8 (5.9) 7.2 (4.4) 7.4 (4.5) 6.4 (4.2) 
Total dairy (servings/day) 2.1 (2.2) 2.2 (2.0) 2.0 (2.0) 1.3 (1.5) 1.7 (1.9) 2.0 (1.9) 2.2 (2.2) 1.7 (1.8) 
Whole grains (energy-adj; g/day) 3.5 (3.1) 16.4 (5.0) 38.1 (10.2) 126.6 (74.9) 79.0 (88.5) 41.3 (51.6) 33.6 (38.4) 30.7 (39.3) 
Refined grains (energy-adj; g/day) 105.2 (76.3) 98.0 (58.8) 92.9 (58.7) 62.8 (55.7) 23.1 (13.9) 60.9 (9.8) 95.2 (12.0) 179.7 (54.6) 
Gluten (energy-adj; g/day), 2.1 (1.4) 2.7 (1.3) 3.2 (1.8) 2.8 (2.1) 1.6 (1.5) 2.3 (1.2) 3.0 (1.3) 3.9 (1.9) 
Dietary fiber (energy-adj; g/day) 18.0 (10.2) 22.9 (13.2) 28.1 (15.8) 35.1 (17.8) 31.6 (18.3) 23.2 (13.0) 25.2 (14.0) 24.1 (16.3) 
Whole grain quartilesa,bRefined grain quartilesa,b
Q1Q2Q3Q4Q1Q2Q3Q4
N = 196N = 195N = 193N = 196N = 195N = 196N = 193N = 196
Range: Males0–8.0>8.0–20.2>20.2–46.5>46.5–2990–47.7>47.7–80.5>80.5–120>120–343
Females0–10.3>10.3–27.5>27.5–63.8>63.8–4480–41.9>41.9–74.4>74.4–119>119–370
Age (years) 57.6 (11.2) 57.2 (10.5) 60.2 (11.4) 62.8 (11.0) 64.5 (10.9) 59.7 (10.7) 58.2 (10.2) 55.4 (11.1) 
Female, % 62.6 62.6 62.9 62.6 62.6 62.6 62.9 62.6 
BMI (kg/m229.1 (7.0) 28.5 (6.3) 27.6 (5.9) 24.4 (4.9) 25.9 (5.3) 27.5 (6.8) 28.4 (6.3) 27.9 (6.5) 
Race/ethnicity, % 
 White 37.9 52.3 48.5 13.8 26.7 43.1 44.8 37.9 
 Black 15.4 12.3 9.8 5.6 5.1 14.9 12.9 10.3 
 Hispanic 28.2 18.5 13.4 2.1 7.7 11.3 19.6 23.6 
 Korean 14.4 11.3 23.7 75.9 59.0 25.6 19.6 21.0 
 Asian, other than Korean 3.6 3.6 3.6 2.1 1.0 4.1 2.1 5.6 
 Otherc 0.5 2.1 1.0 0.5 0.5 1.0 1.0 1.5 
College graduate or higher, % 44.1 59.5 64.9 53.3 49.7 61.5 58.8 51.8 
Physical activity, % 
 None 21.0 13.3 7.7 11.8 9.2 10.8 13.4 20.5 
 <4 hours/week 56.4 60.0 54.1 49.2 44.1 61.0 62.4 52.3 
 4 or more hours/week 22.6 26.7 38.1 39.0 46.7 28.2 24.2 27.2 
Ever smoker, % 33.3 41.0 31.4 24.6 30.8 36.9 30.9 31.8 
Alcohol consumption (g/day) 8.0 (22.2) 14.2 (31.9) 8.0 (22.5) 4.0 (15.6) 9.4 (29.5) 9.6 (23.7) 9.6 (24.2) 5.6 (16.8) 
Family history of cancer, % 55.9 55.9 53.1 39.0 45.6 57.9 53.1 47.2 
Regular aspirin use, % 32.3 33.3 33.5 41.5 34.9 34.4 36.6 34.9 
History of ulcerative colitis, % 3.1 1.0 2.1 2.6 1.5 2.6 2.6 2.1 
History of Crohn’s disease, % 0.5 1.0 0.5 0.0 0.0 1.0 0.5 0.5 
Total energy (kcal/day) 1,889 (1,047) 2,015 (1,212) 2,052 (1,050) 2,122 (1,036) 2,106 (1,017) 1,905 (935) 2,059 (1,064) 2,007 (1,306) 
Total fat (% kcal/day) 31.2 (8.2) 30.9 (6.6) 29.4 (5.4) 25.8 (4.6) 28.0 (7.7) 29.9 (7.0) 30.2 (6.2) 29.2 (5.6) 
Total carbohydrates (% kcal/day) 51.7 (11.2) 50.4 (9.4) 54.1 (8.1) 58.3 (7.0) 54.3 (11.2) 53.0 (10.3) 52.6 (8.3) 54.5 (7.9) 
Red meat (servings/day) 0.6 (0.7) 0.7 (0.8) 0.6 (0.6) 0.5 (0.7) 0.6 (0.8) 0.5 (0.5) 0.6 (0.7) 0.6 (0.9) 
Processed meat (servings/day) 0.7 (1.0) 0.6 (0.8) 0.4 (0.7) 0.2 (0.5) 0.4 (0.8) 0.5 (0.8) 0.6 (0.8) 0.5 (0.8) 
Fruits and vegetables (servings/day) 5.9 (4.6) 7.2 (4.5) 8.1 (4.7) 9.6 (5.3) 9.8 (5.9) 7.2 (4.4) 7.4 (4.5) 6.4 (4.2) 
Total dairy (servings/day) 2.1 (2.2) 2.2 (2.0) 2.0 (2.0) 1.3 (1.5) 1.7 (1.9) 2.0 (1.9) 2.2 (2.2) 1.7 (1.8) 
Whole grains (energy-adj; g/day) 3.5 (3.1) 16.4 (5.0) 38.1 (10.2) 126.6 (74.9) 79.0 (88.5) 41.3 (51.6) 33.6 (38.4) 30.7 (39.3) 
Refined grains (energy-adj; g/day) 105.2 (76.3) 98.0 (58.8) 92.9 (58.7) 62.8 (55.7) 23.1 (13.9) 60.9 (9.8) 95.2 (12.0) 179.7 (54.6) 
Gluten (energy-adj; g/day), 2.1 (1.4) 2.7 (1.3) 3.2 (1.8) 2.8 (2.1) 1.6 (1.5) 2.3 (1.2) 3.0 (1.3) 3.9 (1.9) 
Dietary fiber (energy-adj; g/day) 18.0 (10.2) 22.9 (13.2) 28.1 (15.8) 35.1 (17.8) 31.6 (18.3) 23.2 (13.0) 25.2 (14.0) 24.1 (16.3) 

Abbreviations: BMI, body mass index; energy-adj, energy-adjusted.

aWhole and refined grain quartiles are energy-adjusted using the nutrient density method.

bValues presented are mean (SD) unless otherwise indicated.

c“Other” race/ethnicity included participants who reported more than one or mixed race/ethnicity.

Whole Grain and Fiber Intake

Greater whole grain consumption was not associated with the number of observed s-OTUs (richness; Ptrend = 0.96) or the Shannon diversity index (Ptrend = 0.79; Fig. 1A; alpha-diversity) after adjustment for age, sex, race/ethnicity, BMI, smoking status, alcohol intake, and total energy intake. However, higher whole grain intake was associated with overall gut microbiome composition, as measured by JSD (Fig. 1B; beta-diversity, Q4 vs. Q1 R2 = 0.23%, P = 0.01; R2 for trend = 0.22%, Ptrend = 0.02), with adjustment for age, sex, and race/ethnicity. The amount of variation remained consistent after further adjustment for BMI, smoking status, alcohol intake, and total energy intake (Q4 vs. Q1 R2 = 0.23%, P = 0.01; R2 for trend = 0.22%, Ptrend = 0.02; Fig. 1C). We identified bacterial taxa associated with greater whole grain consumption (Fig. 1D): s-OTUs of Bacteroides plebeius, Faecalibacterium prausnitzii, and Blautia producta were enriched, while B. uniformis was depleted. Additional significantly differentially abundant s-OTUs belonged to the following higher-order groups: Erysipelotrichaceae and Roseburia were enriched at detection levels of 0.90 and 0.80, respectively, Oscillospira and Rikenellaceae were depleted, and Lachnospira, Ruminococcus, and Ruminococcaceae were both enriched and depleted. Dietary fiber, a major whole grain nutrient, was positively correlated with whole grain intake (r = 0.42) and was similarly associated with beta-diversity (JSD, Q4 vs. Q1 R2 = 0.21%, P = 0.02; R2 for trend = 0.20%, Ptrend = 0.03) but not with alpha-diversity (Ptrend for richness = 0.30, Ptrend for Shannon diversity index = 0.38; Supplementary Fig. S1). Abundant taxa associated with greater dietary fiber intake (Supplementary Fig. S1C) were similar to those associated with whole grains (Fig. 1E). When whole grain models were additionally adjusted for dietary fiber or fruit and vegetable intake, the results for alpha-diversity (Ptrend for richness = 0.59 and 0.84, respectively; Ptrend for Shannon diversity index = 0.43 and 0.60, respectively) and for differentially abundant taxa did not materially change; however, whole grain intake was no longer significantly associated with microbial community diversity as measured by JSD after adjustment for dietary fiber or fruit and vegetable intake (R2 for trend = 0.19% and 0.15%, respectively; Ptrend = 0.06 and 0.21, respectively). Grain fiber was similarly positively correlated with whole grain intake (r = 0.48), but higher intake was not significantly associated with alpha- (Ptrend for richness = 0.54, Ptrend for Shannon diversity index = 0.53) or beta-diversity (R2 for trend = 0.15%, Ptrend = 0.19) or any specific microbial taxa.

FIGURE 1

Gut microbiome alpha- and beta-diversity according to energy-adjusted quartiles of whole grain intake in the FAMiLI. A, Boxplot of Shannon diversity index by whole grain quartiles. B, Principal coordinate analysis of the JSD by whole grain quartiles. C, Bar plot illustrating the R2 for model covariates derived from JSD. D, Volcano plots of differentially abundant s-OTUs as detected by ANCOM (model adjusted for age, sex, race, BMI, smoking status, alcohol intake, and total energy) between quartile 4 (Q4) and quartile 1 (Q1) of whole grain intake. The x-axis represents the difference in mean centered log ratio (clr)-transformed abundance between Q4 and Q1, and the y-axis represents the ANCOM W Statistic. s-OTU points are colored according to level of ANCOM significance, with 0.90 being the highest level and grey points indicating s-OTUs that were not significant. E, Cladograms of phylum through species level taxa; color represents clr mean difference between Q4 and Q1 of whole grain intake.

FIGURE 1

Gut microbiome alpha- and beta-diversity according to energy-adjusted quartiles of whole grain intake in the FAMiLI. A, Boxplot of Shannon diversity index by whole grain quartiles. B, Principal coordinate analysis of the JSD by whole grain quartiles. C, Bar plot illustrating the R2 for model covariates derived from JSD. D, Volcano plots of differentially abundant s-OTUs as detected by ANCOM (model adjusted for age, sex, race, BMI, smoking status, alcohol intake, and total energy) between quartile 4 (Q4) and quartile 1 (Q1) of whole grain intake. The x-axis represents the difference in mean centered log ratio (clr)-transformed abundance between Q4 and Q1, and the y-axis represents the ANCOM W Statistic. s-OTU points are colored according to level of ANCOM significance, with 0.90 being the highest level and grey points indicating s-OTUs that were not significant. E, Cladograms of phylum through species level taxa; color represents clr mean difference between Q4 and Q1 of whole grain intake.

Close modal

Refined Grain and Gluten Intake

Refined grain consumers were less likely to consume whole grains in their diet (r = −0.27). Refined grain consumption was associated with a lower number of observed s-OTUs (richness, Ptrend = 0.04) and lower Shannon diversity index (Ptrend = 0.03; Fig. 2A), but not related to differences in overall microbiome composition (Ptrend = 0.05; Fig. 2B). Greater refined grain consumption was not significantly associated with any specific taxa (Fig. 2C). Most refined grain food items contributed to gluten intake. Thus, refined grain was moderately positively correlated with gluten intake (r = 0.46) and weakly negatively correlated with dietary fiber (r = −0.18). Higher consumption of gluten was similarly associated with lower Shannon diversity (Ptrend = 0.03), not associated with beta-diversity (JSD Ptrend = 0.19), and not associated with specific gut microbial taxa, other than depletion of s-OTUs of Blautia obeum (Fig. 3). Refined grain consumption was associated with lower alpha-diversity after additional adjustment for energy-adjusted dietary fiber and fruit and vegetable intake, although P for trend values were no longer statistically significant in most models—(Ptrend for richness = 0.09 and 0.06, respectively; Ptrend for Shannon diversity index = 0.07 and 0.04, respectively), but there remained no differences in overall microbial composition (R2 for trend = 0.17% and 0.18%, respectively; Ptrend = 0.11 and 0.08, respectively) or any specific taxa. Additional adjustment for energy-adjusted dietary fiber and fruit and vegetable intake in gluten models did not materially change the results for alpha- (Ptrend for richness = 0.08 and 0.02, respectively; Ptrend for Shannon diversity index = 0.04 and 0.01, respectively) or beta-diversity (R2 for trend = 0.16% and 0.15%, respectively; Ptrend = 0.17 and 0.20, respectively) or for taxonomic abundance.

FIGURE 2

Gut microbiome alpha- and beta-diversity according to energy-adjusted quartiles of refined grain intake in the FAMiLI. A, Boxplot of Shannon diversity index by refined grain quartiles. B, Principal coordinate analysis of the JSD by refined grain quartiles. C, Volcano plots of differentially abundant s-OTUs as detected by ANCOM (model adjusted for age, sex, race, BMI, smoking status, alcohol intake, and total energy) between quartile 4 (Q4) and quartile 1 (Q1) of refined grain intake. The x-axis represents the difference in mean centered log ratio (clr)-transformed abundance between Q4 and Q1, and the y-axis represents the ANCOM W Statistic. s-OTU points are colored according to level of ANCOM significance, with 0.90 being the highest level and gray points indicating s-OTUs that were not significant.

FIGURE 2

Gut microbiome alpha- and beta-diversity according to energy-adjusted quartiles of refined grain intake in the FAMiLI. A, Boxplot of Shannon diversity index by refined grain quartiles. B, Principal coordinate analysis of the JSD by refined grain quartiles. C, Volcano plots of differentially abundant s-OTUs as detected by ANCOM (model adjusted for age, sex, race, BMI, smoking status, alcohol intake, and total energy) between quartile 4 (Q4) and quartile 1 (Q1) of refined grain intake. The x-axis represents the difference in mean centered log ratio (clr)-transformed abundance between Q4 and Q1, and the y-axis represents the ANCOM W Statistic. s-OTU points are colored according to level of ANCOM significance, with 0.90 being the highest level and gray points indicating s-OTUs that were not significant.

Close modal
FIGURE 3

Gut microbiome alpha- and beta-diversity according to energy-adjusted quartiles of gluten intake in the FAMiLI. A, Boxplot of Shannon diversity index by gluten quartiles. B, Principal coordinate analysis of the JSD by gluten quartiles. C, Volcano plots of differentially abundant s-OTUs as detected by ANCOM (model adjusted for age, sex, race, BMI, smoking status, alcohol intake, and total energy) between quartile 4 (Q4) and quartile 1 (Q1) of gluten intake. The x-axis represents the difference in mean centered log ratio (clr)-transformed abundance between Q4 and Q1, and the y-axis represents the ANCOM W Statistic. s-OTU points are colored according to level of ANCOM significance, with 0.90 being the highest level and gray points indicating s-OTUs that were not significant.

FIGURE 3

Gut microbiome alpha- and beta-diversity according to energy-adjusted quartiles of gluten intake in the FAMiLI. A, Boxplot of Shannon diversity index by gluten quartiles. B, Principal coordinate analysis of the JSD by gluten quartiles. C, Volcano plots of differentially abundant s-OTUs as detected by ANCOM (model adjusted for age, sex, race, BMI, smoking status, alcohol intake, and total energy) between quartile 4 (Q4) and quartile 1 (Q1) of gluten intake. The x-axis represents the difference in mean centered log ratio (clr)-transformed abundance between Q4 and Q1, and the y-axis represents the ANCOM W Statistic. s-OTU points are colored according to level of ANCOM significance, with 0.90 being the highest level and gray points indicating s-OTUs that were not significant.

Close modal

Sensitivity Analyses

In sensitivity analyses, associations by racial/ethnic group did not significantly differ, but our limited sample size severely limited our ability to examine differences (Supplementary Fig. S2). Similarly, findings remained largely unchanged when we excluded participants who reported a history of type 2 diabetes (energy-adjusted whole grains: Ptrend for richness = 0.84 and Ptrend for Shannon diversity index = 0.96; JSD R2 for trend = 0.26% and Ptrend = 0.02) or inflammatory bowel disease (energy-adjusted whole grains: Ptrend for richness = 0.92 and Ptrend for Shannon diversity index = 0.76; JSD R2 for trend = 0.21% and Ptrend = 0.02), because grain-based, gluten-containing foods may be avoided with these conditions (Supplementary Figs. S3 and S4). When we included grain-based sweets and dessert items with refined grain (Ptrend for richness = 0.04 and Ptrend for Shannon diversity index = 0.02; JSD R2 for trend = 0.20% and Ptrend = 0.03) and gluten intakes (Ptrend for richness = 0.07 and Ptrend for Shannon diversity index = 0.03; JSD R2 for trend = 0.15% and Ptrend = 0.18), findings also remained unchanged (Supplementary Figs. S5 and S6).

In this study, we found that whole grain intake influences the overall gut microbial composition, which was largely explained by the higher abundance of F. prausnitzii, B. plebeius, and Erysipelotrichaceae and lower abundance of B. uniformis, Oscillospira, and Rikenellaceae. These bacteria were also associated with higher dietary fiber intakes. We further found that refined grain and gluten consumers have lower Shannon diversity but do not have any significant variation in specific bacterial taxa, even with the inclusion of grain-based sweets and desserts.

In contrast to refined grains, whole grains are a rich source of dietary fiber, nutrients, and various bioactive compounds, such as phytochemicals. These compounds may reduce risk of colorectal cancer and other cancers through various antiproliferative and inhibitory effects against cancer cells (49) that may be at least partially modulated by gut microbiota (50). Through gut microbial fermentation of dietary fibers, the production of the SCFA butyrate is thought to suppress colonic inflammation and carcinogenesis by protecting DNA damage in colonocytes induced by oxidative stress (51), facilitating normal colonocyte growth (52), facilitating assembly of tight junction proteins to maintain the intestinal barrier (53), and inhibiting tumor cell growth (54). However, unlike whole grains, grain fiber was not associated with alpha- or beta-diversity or abundant taxa in this study, which raises the question whether other components of whole grains beyond dietary fiber may be associated with gut microbiota. Whole grains are more complex than refined grains and contain additional nutrients, including unsaturated fatty acids and various phytochemicals, which have unknown associations with the gut microbiome. Future studies that utilize whole genome shotgun sequencing techniques and examine the fecal metabolome will help elucidate food-microbiome relationships.

We found that greater whole grain intake was associated with increased beta-diversity and increased abundance of various SCFA-producing species, including s-OTUs from F. prausnitzii and family Erysipelotrichaceae. These findings provide further evidence that whole grain consumption is associated with lower colorectal cancer risk due to an associated healthier gut microbial phenotype. Previous intervention trials of whole grains have yielded mixed results related to F. prausnitzii (14–20), which may be partly due to differences in study type (crossover vs. parallel dietary interventions), study length, type of whole grain or dietary fiber, dose, and baseline gut microbiota of different study populations (e.g., healthy vs. obese; ref. 55). Evidence from animal models also suggest that high-fat or Western dietary patterns, which are low in whole grains, may increase abundance of Erysipelotrichaceae (56), but additional studies report higher (57, 58) and lower (59–61) abundances of Erysipelotrichaceae in inflammatory bowel disease and colorectal cancer. This suggests that differences may exist between species. We found that s-OTUs of Erysipelotrichaceae were still enriched with greater whole grain intake after exclusion of participants with self-reported inflammatory bowel disease. Additional studies that utilize whole genome shotgun sequencing are warranted to clarify the relationship between whole grains, SCFA-producing species, and risk of inflammatory bowel disease and colorectal cancer.

Conversely, greater whole grain consumption was associated with lower abundance of s-OTUs of B. uniformis and family Rikenellaceae from phylum Bacteroidetes. Although these propionate-producing microbes have not been associated with whole grain or dietary fiber consumption, evidence suggests potential associations with animal protein and dietary fat consumption (62–64). In our study, participants with the highest whole grain intakes consumed less total fat and processed meat and more fruits and vegetables, suggesting that B. uniformis and Rikenellaceae may be associated with dietary patterns that are less healthy and commonly low in whole grains and dietary fiber (65).

Higher whole grain intake was associated with higher abundance of s-OTUs of B. plebeius, also from phylum Bacteroidetes, but the significance of this finding is uncertain. Although B. plebeius was previously reported to have decreased among participants who consumed a 12-week high-fiber rye bread intervention (66), it was also identified to contain an enzyme capable of digesting nori seaweed among native Japanese populations (67). This evidence may at least partially explain our finding because participants with the highest whole grain intake were more likely to be of Korean descent; however, seaweed-containing food items were not included on the original FFQ and therefore, were unable to be assessed.

Current evidence on the relationship between gluten and gut microbiota in healthy adult populations is limited. To our knowledge, three dietary intervention trials of gluten-free (11, 12) or low-gluten (13) diets were conducted in adult non-celiac disease populations. Among 10 Spanish adults ages 23–40 years maintained on a gluten-free diet for 1 month, statistically significant decreases in Bifidobacterium and Lactobacillus and increases in Escherichia coli and Enterobacteriaceae compared with baseline were observed (11). A second trial of 21 Dutch men and women ages 16–61 years reported alterations in eight taxa, with the greatest decrease in family Veillonellaceae, after following a 4-week gluten-free diet (12). A third crossover trial of 54 Danish adults ages 22–65 years reported that an 8-week low-gluten diet altered abundance of 14 bacterial species in comparison with a high-gluten diet of equal dietary fiber content, including decreased abundance of four Bifidobacterium species (13). Given that Bifidobacterium and Veillonellaceae species mediate or are directly involved in SCFA production, respectively (68, 69), this evidence suggests that gluten-free dietary patterns may cause undesirable shifts in gut microbial composition. We did not observe similar findings as previous studies, but we did observe a depletion of s-OTUs of B. obeum with gluten intake. This, along with our finding that s-OTUs of B. producta were enriched with higher whole grain intake, may reflect the association of Blautia genus with visceral fat accumulation and obesity (70). This finding suggests that greater consumption of whole grains, rather than refined grains, which was correlated with gluten intake, may be associated with enrichment of Blautia species. However, additional studies are needed to understand the significance of different Blautia species and to understand whether gluten has an undesirable effect on gut microbial composition in healthy populations.

Additional components of diet, beyond whole grains and dietary fiber, may influence gut microbial composition, such as dietary fat (71) and fruits and vegetables (72), which correlated with whole grain consumption in this population. Other dietary components, such as artificial sweeteners (73), were not assessed in this study. High whole grain consumption may also be indicative of healthier dietary patterns. Dietary fiber is a major component of healthy dietary patterns, and intake was associated with similar microbial taxa as whole grain intake in this study. Grain fiber was not associated with these taxa, suggesting that whole grain components, other than fiber, may influence gut microbiota. Adjustment for fruit and vegetable intake in models of whole grain and dietary fiber did not change associations with microbial taxa, suggesting that fruit and vegetable intake was not responsible for the gut microbial composition associations observed. Similarly, the addition of grain-based sweets and desserts in refined grain and gluten models did not change associations, suggesting that these foods, which are often low in dietary fiber and high in sugar and/or saturated fat, were also not associated with gut microbial composition. Additional studies are needed to elucidate the associations between individual foods and nutrients, as well as overall dietary patterns, on the gut microbiome.

Strengths of this study include a large study population with collection of stool samples and the assessment of habitual, rather than shorter-term, dietary intakes using the FFQ. This allowed us to examine a wide range of whole and refined grain, dietary fiber, and gluten intakes in relation to gut microbial composition. A limitation of this study was the limited sample size when stratifying by race/ethnicity, likely hindering our ability to observe significant differences between groups. Furthermore, Chinese- and Korean-specific grain foods were not available for analysis, which may have contributed to measurement error of the exposures. Because the DQX and DHQ have not been validated in Chinese-American or Korean-American populations (27), we acknowledge that consumption of certain cultural foods may have been underestimated. There is very limited data on dietary patterns of the Korean American population, but commonly consumed Korean foods include kimchi, rice, garlic, green onions, Korean soups and stews, and various Korean condiments (74). Although none of these foods are considered whole grains, kimchi and other vegetables may contribute to vegetable and dietary fiber intake. Thus, underestimation of vegetable dietary fiber intake by the DQX may have contributed to residual confounding. Similarly, refined grain consumption may have been underestimated among Korean American participants, which may have contributed to measurement error and attenuation of the refined grain results. However, rice consumption in the native Korean diet has steadily declined (75) so it is unclear whether or to what degree rice consumption was underestimated by the DQX in this study. Acculturation based on length of residence in the United States may also influence dietary patterns of Korean Americans to be more “Americanized” versus “Korean,” but this information was unavailable in this study. Additional studies are needed to validate the DQX and DHQ among Korean American and other Asian American subgroups. Additional limitations of this study include the cross-sectional design, which did not allow us to examine within-person changes in grain and gluten intakes, gut microbiota, and other lifestyle factors over time, as well as any potential health-related consequences of long-term intake of grain- and gluten-containing foods and related changes in the gut microbiome. With continued enrollment, future analyses in the FAMiLI cohort will have greater statistical power to investigate associations by racial/ethnic group and will be able to examine longitudinal changes in dietary intakes and gut microbiome as repeated stool samples and dietary intakes are collected over time.

In summary, our findings from this cross-sectional study of U.S. adults suggest that higher refined grain and gluten consumption may be associated with lower gut microbial alpha-diversity, and that higher whole grain consumption may be associated with altered gut microbial composition. Although these findings suggest that whole grains, independent of their grain fiber and gluten content, may contribute to a healthier gut microbial profile, additional studies are needed to confirm our findings and provide additional evidence as to how these dietary exposures influence gut microbiota and subsequent risk for chronic diseases, including colorectal cancer.

R.B. Hayes reports grants from NCI during the conduct of the study. No disclosures were reported by the other authors.

C.Y. Um: Conceptualization, formal analysis, methodology, writing-original draft, writing-review and editing. B.A. Peters: Formal analysis, methodology, writing-review and editing. H.S. Choi: Resources, investigation. P. Oberstein: Resources, investigation. D.B. Beggs: Resources, investigation. M. Usyk: Formal analysis, writing-review and editing. F. Wu: Resources, investigation. R.B. Hayes: Project administration, writing-review and editing. S.M. Gapstur: Conceptualization, writing-review and editing. M.L. McCullough: Conceptualization, supervision, writing-review and editing. J. Ahn: Resources, supervision, project administration, writing-review and editing.

We are grateful to the FAMiLI participants for contributing their time to participate in this study. We thank our partners at Korean Community Services for facilitating recruitment in the Korean community.

This work was supported by the NCI of the NIH P30 CA016087 (J. Ahn), P20 CA252728 (J. Ahn), R01 CA159036 (R.B. Hayes), and U01 CA250186 (J. Ahn).

Note: Supplementary data for this article are available at Cancer Research Communications Online (https://aacrjournals.org/cancerrescommun/).

1.
Jonnalagadda
SS
,
Harnack
L
,
Liu
RH
,
McKeown
N
,
Seal
C
,
Liu
S
, et al
.
Putting the whole grain puzzle together: health benefits associated with whole grains–summary of American Society for Nutrition 2010 Satellite Symposium
.
J Nutr
2011
;
141
:
1011S
22S
.
2.
Gaesser
GA
.
Whole grains, refined grains, and cancer risk: a systematic review of meta-analyses of observational studies
.
Nutrients
2020
;
12
:
3756
.
3.
World Cancer Research Fund International/American Institute for Cancer Research
.
Diet, nutrition, physical activity and colorectal cancer. Continuous Update Project Expert Report
;
2018
. Available from: dietandcancerreport.org.
4.
U.S. Department of Health and Human Services and U.S. Department of Agriculture
.
2015–2020 Dietary Guidelines for Americans
. 8th Edition. December
2015
. Available from: http://health.gov/dietaryguidelines/2015/guidelines/.
5.
McNabney
SM
,
Henagan
TM
.
Short chain fatty acids in the colon and peripheral tissues: a focus on butyrate, colon cancer, obesity and insulin resistance
.
Nutrients
2017
;
9
:
1348
.
6.
Martinez
I
,
Lattimer
JM
,
Hubach
KL
,
Case
JA
,
Yang
J
,
Weber
CG
, et al
.
Gut microbiome composition is linked to whole grain-induced immunological improvements
.
ISME J
2013
;
7
:
269
80
.
7.
Singh
RK
,
Chang
HW
,
Yan
D
,
Lee
KM
,
Ucmak
D
,
Wong
K
, et al
.
Influence of diet on the gut microbiome and implications for human health
.
J Transl Med
2017
;
15
:
73
.
8.
Koiv
V
,
Tenson
T
.
Gluten-degrading bacteria: availability and applications
.
Appl Microbiol Biotechnol
2021
;
105
:
3045
59
.
9.
Marasco
G
,
Di Biase
AR
,
Schiumerini
R
,
Eusebi
LH
,
Iughetti
L
,
Ravaioli
F
, et al
.
Gut microbiota and celiac disease
.
Dig Dis Sci
2016
;
61
:
1461
72
.
10.
Koecher
KJ
,
McKeown
NM
,
Sawicki
CM
,
Menon
RS
,
Slavin
JL
.
Effect of whole-grain consumption on changes in fecal microbiota: a review of human intervention trials
.
Nutr Rev
2019
;
77
:
487
97
.
11.
De Palma
G
,
Nadal
I
,
Collado
MC
,
Sanz
Y
.
Effects of a gluten-free diet on gut microbiota and immune function in healthy adult human subjects
.
Br J Nutr
2009
;
102
:
1154
60
.
12.
Bonder
MJ
,
Tigchelaar
EF
,
Cai
X
,
Trynka
G
,
Cenit
MC
,
Hrdlickova
B
, et al
.
The influence of a short-term gluten-free diet on the human gut microbiome
.
Genome Med
2016
;
8
:
45
.
13.
Hansen
LBS
,
Roager
HM
,
Sondertoft
NB
,
Gobel
RJ
,
Kristensen
M
,
Valles-Colomer
M
, et al
.
A low-gluten diet induces changes in the intestinal microbiome of healthy Danish adults
.
Nat Commun
2018
;
9
:
4630
.
14.
Costabile
A
,
Klinder
A
,
Fava
F
,
Napolitano
A
,
Fogliano
V
,
Leonard
C
, et al
.
Whole-grain wheat breakfast cereal has a prebiotic effect on the human gut microbiota: a double-blind, placebo-controlled, crossover study
.
Br J Nutr
2008
;
99
:
110
20
.
15.
Christensen
EG
,
Licht
TR
,
Kristensen
M
,
Bahl
MI
.
Bifidogenic effect of whole-grain wheat during a 12-week energy-restricted dietary intervention in postmenopausal women
.
Eur J Clin Nutr
2013
;
67
:
1316
21
.
16.
Foerster
J
,
Maskarinec
G
,
Reichardt
N
,
Tett
A
,
Narbad
A
,
Blaut
M
, et al
.
The influence of whole grain products and red meat on intestinal microbiota composition in normal weight adults: a randomized crossover intervention trial
.
PLoS One
2014
;
9
:
e109606
.
17.
Roager
HM
,
Vogt
JK
,
Kristensen
M
,
Hansen
LBS
,
Ibrugger
S
,
Maerkedahl
RB
, et al
.
Whole grain-rich diet reduces body weight and systemic low-grade inflammation without inducing major changes of the gut microbiome: a randomised cross-over trial
.
Gut
2019
;
68
:
83
93
.
18.
Vuholm
S
,
Nielsen
DS
,
Iversen
KN
,
Suhr
J
,
Westermann
P
,
Krych
L
, et al
.
Whole-grain rye and wheat affect some markers of gut health without altering the fecal microbiota in healthy overweight adults: a 6-week randomized trial
.
J Nutr
2017
;
147
:
2067
75
.
19.
Cooper
DN
,
Kable
ME
,
Marco
ML
,
De Leon
A
,
Rust
B
,
Baker
JE
, et al
.
The effects of moderate whole grain consumption on fasting glucose and lipids, gastrointestinal symptoms, and microbiota
.
Nutrients
2017
;
9
:
173
.
20.
Vanegas
SM
,
Meydani
M
,
Barnett
JB
,
Goldin
B
,
Kane
A
,
Rasmussen
H
, et al
.
Substituting whole grains for refined grains in a 6-wk randomized trial has a modest effect on gut microbiota and immune and inflammatory markers of healthy adults
.
Am J Clin Nutr
2017
;
105
:
635
50
.
21.
Kim
HS
,
Patel
KG
,
Orosz
E
,
Kothari
N
,
Demyen
MF
,
Pyrsopoulos
N
, et al
.
Time trends in the prevalence of celiac disease and gluten-free diet in the US population: results from the national health and nutrition examination surveys 2009–2014
.
JAMA Intern Med
2016
;
176
:
1716
7
.
22.
Peters
BA
,
Yi
SS
,
Beasley
JM
,
Cobbs
EN
,
Choi
HS
,
Beggs
DB
, et al
.
US nativity and dietary acculturation impact the gut microbiome in a diverse US population
.
ISME J
2020
;
14
:
1639
50
.
23.
National Cancer Institute Cancer Data Access System
.
Dietary questionnaire (DQX) datasets
. Available from: https://biometry.nci.nih.gov/cdas/datasets/plco/97/.
24.
Willett
WC
,
Sampson
L
,
Stampfer
MJ
,
Rosner
B
,
Bain
C
,
Witschi
J
, et al
.
Reproducibility and validity of a semiquantitative food frequency questionnaire
.
Am J Epidemiol
1985
;
122
:
51
65
.
25.
Block
G
,
Hartman
AM
,
Dresser
CM
,
Carroll
MD
,
Gannon
J
,
Gardner
L
.
A data-based approach to diet questionnaire design and testing
.
Am J Epidemiol
1986
;
124
:
453
69
.
26.
Thompson
FE
,
Subar
AF
,
Brown
CC
,
Smith
AF
,
Sharbaugh
CO
,
Jobe
JB
, et al
.
Cognitive research enhances accuracy of food frequency questionnaire reports: results of an experimental validation study
.
J Am Diet Assoc
2002
;
102
:
212
25
.
27.
Subar
AF
,
Thompson
FE
,
Kipnis
V
,
Midthune
D
,
Hurwitz
P
,
McNutt
S
, et al
.
Comparative validation of the Block, Willett, and National Cancer Institute food frequency questionnaires: the Eating at America's Supplementary Table Study
.
Am J Epidemiol
2001
;
154
:
1089
99
.
28.
Subar
AF
,
Kipnis
V
,
Troiano
RP
,
Midthune
D
,
Schoeller
DA
,
Bingham
S
, et al
.
Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study
.
Am J Epidemiol
2003
;
158
:
1
13
.
29.
Kipnis
V
,
Subar
AF
,
Midthune
D
,
Freedman
LS
,
Ballard-Barbash
R
,
Troiano
RP
, et al
.
Structure of dietary measurement error: results of the OPEN biomarker study
.
Am J Epidemiol
2003
;
158
:
14
21
.
30.
van der Kamp
JW
,
Poutanen
K
,
Seal
CJ
,
Richardson
DP
.
The HEALTHGRAIN definition of 'whole grain'
.
Food Nutr Res
2014
;
58
.
31.
Schwingshackl
L
,
Schwedhelm
C
,
Hoffmann
G
,
Knuppel
S
,
Laure Preterre
A
,
Iqbal
K
, et al
.
Food groups and risk of colorectal cancer
.
Int J Cancer
2018
;
142
:
1748
58
.
32.
U.S. Department of Agriculture, Agricultural Research Service
.
FoodData Central, 2019
. Available from: fdc.nal.usda.gov.
33.
Shewry
PR
,
Halford
NG
.
Cereal seed storage proteins: structures, properties and role in grain utilization
.
J Exp Bot
2002
;
53
:
947
58
.
34.
van Overbeek
FM
,
Uil-Dieterman
IG
,
Mol
IW
,
Kohler-Brands
L
,
Heymans
HS
,
Mulder
CJ
.
The daily gluten intake in relatives of patients with coeliac disease compared with that of the general Dutch population
.
Eur J Gastroenterol Hepatol
1997
;
9
:
1097
9
.
35.
Lebwohl
B
,
Cao
Y
,
Zong
G
,
Hu
FB
,
Green
PHR
,
Neugut
AI
, et al
.
Long term gluten consumption in adults without celiac disease and risk of coronary heart disease: prospective cohort study
.
BMJ
2017
;
357
:
j1892
.
36.
Jamnik
J
,
Garcia-Bailo
B
,
Borchers
CH
,
El-Sohemy
A
.
Gluten intake is positively associated with plasma α2-macroglobulin in young adults
.
J Nutr
2015
;
145
:
1256
62
.
37.
Assor
E
,
Davies-Shaw
J
,
Marcon
MA
,
Mahmud
FH
.
Estimation of dietary gluten content using total protein in relation to gold standard testing in a variety of foods
.
J Nutr Food Sci
2014
;
04
:
296
.
38.
Caporaso
JG
,
Lauber
CL
,
Walters
WA
,
Berg-Lyons
D
,
Lozupone
CA
,
Turnbaugh
PJ
, et al
.
Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample
.
Proc Nat Acad Sci U S A
2011
;
108
:
4516
22
.
39.
Caporaso
JG
,
Lauber
CL
,
Walters
WA
,
Berg-Lyons
D
,
Huntley
J
,
Fierer
N
, et al
.
Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms
.
ISME J
2012
;
6
:
1621
4
.
40.
Caporaso
JG
,
Kuczynski
J
,
Stombaugh
J
,
Bittinger
K
,
Bushman
FD
,
Costello
EK
, et al
.
QIIME allows analysis of high-throughput community sequencing data
.
Nat Methods
2010
;
7
:
335
6
.
41.
Bokulich
NA
,
Subramanian
S
,
Faith
JJ
,
Gevers
D
,
Gordon
JI
,
Knight
R
, et al
.
Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing
.
Nat Methods
2013
;
10
:
57
9
.
42.
Amir
A
,
McDonald
D
,
Navas-Molina
JA
,
Kopylova
E
,
Morton
JT
,
Zech Xu
Z
, et al
.
Deblur rapidly resolves single-nucleotide community sequence patterns
.
mSystems
2017
;
2
:
e00191
16
.
43.
DeSantis
TZ
,
Hugenholtz
P
,
Larsen
N
,
Rojas
M
,
Brodie
EL
,
Keller
K
, et al
.
Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB
.
Appl Environ Microbiol
2006
;
72
:
5069
72
.
44.
Katoh
K
,
Misawa
K
,
Kuma
K
,
Miyata
T
.
MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform
.
Nucleic Acids Res
2002
;
30
:
3059
66
.
45.
Price
MN
,
Dehal
PS
,
Arkin
AP
.
FastTree: computing large minimum evolution trees with profiles instead of a distance matrix
.
Mol Biol Evol
2009
;
26
:
1641
50
.
46.
Oksanen
J
,
Blanchet
FG
,
Friendly
M
,
Kindt
R
,
Legendre
P
,
McGlinn
D
, et al
.
vegan community ecology package version 2.5–7 November 2020
;
2020
.
47.
Anderson
MJ
.
A new method for non-parametric multivariate analysis of variance
.
2001
;
26
:
32
46
.
48.
Mandal
S
,
Van Treuren
W
,
White
RA
,
Eggesbo
M
,
Knight
R
,
Peddada
SD
.
Analysis of composition of microbiomes: a novel method for studying microbial composition
.
Microb Ecol Health Dis
2015
;
26
:
27663
.
49.
Zhu
Y
,
Sang
S
.
Phytochemicals in whole grain wheat and their health-promoting effects
.
Mol Nutr Food Res
2017
;
61
.
50.
Jefferson
A
,
Adolphus
K
.
The effects of intact cereal grain fibers, including wheat bran on the gut microbiota composition of healthy adults: a systematic review
.
Front Nutr
2019
;
6
:
33
.
51.
Rosignoli
P
,
Fabiani
R
,
De Bartolomeo
A
,
Spinozzi
F
,
Agea
E
,
Pelli
MA
, et al
.
Protective activity of butyrate on hydrogen peroxide-induced DNA damage in isolated human colonocytes and HT29 tumour cells
.
Carcinogenesis
2001
;
22
:
1675
80
.
52.
Nepelska
M
,
de Wouters
T
,
Jacouton
E
,
Beguet-Crespel
F
,
Lapaque
N
,
Dore
J
, et al
.
Commensal gut bacteria modulate phosphorylation-dependent PPARgamma transcriptional activity in human intestinal epithelial cells
.
Sci Rep
2017
;
7
:
43199
.
53.
Peng
L
,
Li
ZR
,
Green
RS
,
Holzman
IR
,
Lin
J
.
Butyrate enhances the intestinal barrier by facilitating tight junction assembly via activation of AMP-activated protein kinase in Caco-2 cell monolayers
.
J Nutr
2009
;
139
:
1619
25
.
54.
Fung
KY
,
Cosgrove
L
,
Lockett
T
,
Head
R
,
Topping
DL
.
A review of the potential mechanisms for the lowering of colorectal oncogenesis by butyrate
.
Br J Nutr
2012
;
108
:
820
31
.
55.
Sawicki
CM
,
Livingston
KA
,
Ross
AB
,
Jacques
PF
,
Koecher
K
,
McKeown
NM
.
Evaluating whole grain intervention study designs and reporting practices using evidence mapping methodology
.
Nutrients
2018
;
10
:
1052
.
56.
Fleissner
CK
,
Huebel
N
,
Abd El-Bary
MM
,
Loh
G
,
Klaus
S
,
Blaut
M
.
Absence of intestinal microbiota does not protect mice from diet-induced obesity
.
Br J Nutr
2010
;
104
:
919
29
.
57.
Chen
W
,
Liu
F
,
Ling
Z
,
Tong
X
,
Xiang
C
.
Human intestinal lumen and mucosa-associated microbiota in patients with colorectal cancer
.
PLoS One
2012
;
7
:
e39743
.
58.
Zhu
Q
,
Jin
Z
,
Wu
W
,
Gao
R
,
Guo
B
,
Gao
Z
, et al
.
Analysis of the intestinal lumen microbiota in an animal model of colorectal cancer
.
PLoS One
2014
;
9
:
e90849
.
59.
Dey
N
,
Soergel
DA
,
Repo
S
,
Brenner
SE
.
Association of gut microbiota with post-operative clinical course in Crohn's disease
.
BMC Gastroenterol
2013
;
13
:
131
.
60.
Gevers
D
,
Kugathasan
S
,
Denson
LA
,
Vazquez-Baeza
Y
,
Van Treuren
W
,
Ren
B
, et al
.
The treatment-naive microbiome in new-onset Crohn's disease
.
Cell Host Microbe
2014
;
15
:
382
92
.
61.
Labbe
A
,
Ganopolsky
JG
,
Martoni
CJ
,
Prakash
S
,
Jones
ML
.
Bacterial bile metabolising gene abundance in Crohn's, ulcerative colitis and type 2 diabetes metagenomes
.
PLoS One
2014
;
9
:
e115175
.
62.
Wu
GD
,
Chen
J
,
Hoffmann
C
,
Bittinger
K
,
Chen
YY
,
Keilbaugh
SA
, et al
.
Linking long-term dietary patterns with gut microbial enterotypes
.
Science
2011
;
334
:
105
8
.
63.
Daniel
H
,
Gholami
AM
,
Berry
D
,
Desmarchelier
C
,
Hahne
H
,
Loh
G
, et al
.
High-fat diet alters gut microbiota physiology in mice
.
ISME J
2014
;
8
:
295
308
.
64.
Wang
B
,
Kong
Q
,
Li
X
,
Zhao
J
,
Zhang
H
,
Chen
W
, et al
.
A high-fat diet increases gut microbiota biodiversity and energy expenditure due to nutrient difference
.
Nutrients
2020
;
12
:
3197
.
65.
Cordain
L
,
Eaton
SB
,
Sebastian
A
,
Mann
N
,
Lindeberg
S
,
Watkins
BA
, et al
.
Origins and evolution of the Western diet: health implications for the 21st century
.
Am J Clin Nutr
2005
;
81
:
341
54
.
66.
Lappi
J
,
Salojarvi
J
,
Kolehmainen
M
,
Mykkanen
H
,
Poutanen
K
,
de Vos
WM
, et al
.
Intake of whole-grain and fiber-rich rye bread versus refined wheat bread does not differentiate intestinal microbiota composition in Finnish adults with metabolic syndrome
.
J Nutr
2013
;
143
:
648
55
.
67.
Hehemann
JH
,
Correc
G
,
Barbeyron
T
,
Helbert
W
,
Czjzek
M
,
Michel
G
.
Transfer of carbohydrate-active enzymes from marine bacteria to Japanese gut microbiota
.
Nature
2010
;
464
:
908
12
.
68.
Riviere
A
,
Selak
M
,
Lantin
D
,
Leroy
F
,
De Vuyst
L
.
Bifidobacteria and butyrate-producing colon bacteria: importance and strategies for their stimulation in the human gut
.
Front Microbiol
2016
;
7
:
979
.
69.
Flint
HJ
,
Duncan
SH
,
Scott
KP
,
Louis
P
.
Links between diet, gut microbiota composition and gut metabolism
.
Proc Nutr Soc
2015
;
74
:
13
22
.
70.
Ozato
N
,
Saito
S
,
Yamaguchi
T
,
Katashima
M
,
Tokuda
I
,
Sawada
K
, et al
.
Blautia genus associated with visceral fat accumulation in adults 20–76 years of age
.
NPJ Biofilms Microbiomes
2019
;
5
:
28
.
71.
Wan
Y
,
Wang
F
,
Yuan
J
,
Li
J
,
Jiang
D
,
Zhang
J
, et al
.
Effects of dietary fat on gut microbiota and faecal metabolites, and their relationship with cardiometabolic risk factors: a 6-month randomised controlled-feeding trial
.
Gut
2019
;
68
:
1417
29
.
72.
Frankenfeld
CL
,
Hullar
MAJ
,
Maskarinec
G
,
Monroe
KR
,
Shepherd
JA
,
Franke
AA
, et al
.
The Gut microbiome is associated with circulating dietary biomarkers of fruit and vegetable intake in a multiethnic cohort
.
J Acad Nutr Diet
2022
;
122
:
78
98
.
73.
Ruiz-Ojeda
FJ
,
Plaza-Diaz
J
,
Saez-Lara
MJ
,
Gil
A
.
Effects of sweeteners on the gut microbiota: a review of experimental studies and clinical trials
.
Adv Nutr
2019
;
10
:
S31
48
.
74.
Lee
SK
,
Sobal
J
,
Frongillo
EA
Jr
.
Acculturation and dietary practices among Korean Americans
.
J Am Diet Assoc
1999
;
99
:
1084
9
.
75.
Kim
SH
.
Cultural perspectives and current consumption changes of cooked rice in Korean diet
.
Nutr Res Pract
2007
;
1
:
8
13
.