Mediterranean Diet and Weight Loss: Targeting the Bile Acid/Gut Microbiome Axis to Reduce Colorectal Cancer (Bridge CRC)

Mediterranean Diet and Weight Loss: Targeting the Bile Acid/Gut Microbiome Axis to Reduce Colorectal Cancer

Mediterranean Diet and Weight Loss: Targeting the Bile Acid/Gut Microbiome Axis to Reduce Colorectal Cancer

A Mediterranean Diet (MedDiet), a largely plant-based dietary pattern, is relevant to CRC prevention and microbial production of anti-cancer metabolites in observational studies. A MedDiet can shift BA metabolism as shown in primates and when combined with calorie restriction, shows superior adherence and weight control in humans, given its palatability. To date, no studies have tested in an RCT the effects of a MedDiet alone (MedA), WL through lifestyle intervention (WL-A) or a calorie-restricted MedDiet for WL (WL-Med) on the BA-gut microbiome axis and its relevance to CRC prevention among AAs. A multidisciplinary team combining expertise in psychology, nutrition, microbiology, molecular cell biology, computational biology, medicine and biostatistics, proposes to conduct a four-arm RCT in which 232 obese AAs, 45-75 years old complete one of the following 6-month interventions: Med-A, weight stable; WL-A, calorie restriction with no diet pattern change; WLMed; or Control. The investigators will use samples and data collected at baseline, mid-study (month-3) and post-intervention to compare the effects of the interventions on 1) Concentration and composition of circulating and fecal BAs; 2) Gut microbiota and metabolic function; and 3) Gene expression profiles of exfoliated intestinal epithelial cells.

Colorectal cancer (CRC) is associated with multiple risk factors including, obesity, low fiber diets, and diets high in animal protein and saturated fat (SFat). African Americans (AAs) have a higher prevalence of these risk factors and they have the highest incidence of CRC and related mortality. These multiple risk factors are also linked to higher circulating and fecal bile acids (BA) and a shift in BA amino acid conjugation from glycine to taurine. These BA-related changes can alter the composition, structure, and metabolic activity of the gut microbiota, fostering conditions for gut bacteria to expand and metabolize taurine-conjugated BAs to genotoxic hydrogen sulfide (H2S) and the tumor promoter, deoxycholic acid (DCA); a colonic milieu conducive to the formation of CRC. The investigators have shown that the abundance of H2S-producing bacteria is significantly higher in the colon of AAs compared to non-Hispanic whites (NHWs) and is a defining feature among AA CRC cases implicating these bacteria as contributors to CRC development in a race-dependent manner. Moreover, the microbial difference is associated with higher intake of SFat and animal protein in AAs, providing a pivotal intervention target. The investigators hypothesize that targeting the BA-gut microbiome axis to suppress abundance, growth and metabolic activity of H2S and DCA producing bacteria through diet and weight loss (WL) may reduce CRC risk, especially among AAs. A Mediterranean Diet (MedDiet), a largely plant-based dietary pattern, is relevant to CRC prevention and microbial production of anti-cancer metabolites in observational studies. A MedDiet can shift BA metabolism as shown in primates and when combined with calorie restriction, shows superior adherence and weight control in humans, given its palatability. To date, no studies have tested in an RCT the effects of a MedDiet alone (MedA), WL through lifestyle intervention (WL-A) or a calorie-restricted MedDiet for WL (WL-Med) on the BA-gut microbiome axis and its relevance to CRC prevention among AAs. Our multidisciplinary team combining expertise in psychology, nutrition, microbiology, molecular cell biology, computational biology, medicine and biostatistics, propose to conduct a four-arm RCT in which 232 obese AAs, 45-75 years old complete one of the following 6-month interventions: Med-A, weight stable; WL-A, calorie restriction with no diet pattern change; WLMed; or Control. The investigators will use samples and data collected at baseline, mid-study (month-3) and post-intervention to compare the effects of the interventions on 1) Concentration and composition of circulating and fecal BAs; 2) Gut microbiota and metabolic function; and 3) Gene expression profiles of exfoliated intestinal epithelial cells. The investigators approach is strong given the multidisciplinary team, use of evidence-based lifestyle interventions, and sophisticated -omics analyses to examine crosstalk between diet/WL, gut microbiome, and host intestinal physiology. If successful, this study could have profound public health impact on CRC risk among AAs and other high-risk populations, that would translate into timely dissemination opportunities.

This study is a four-arm RCT designed to test the effect of a weight-stable MedDiet alone (Med-A) lifestyle intervention, calorie restriction for WL with no diet composition change (WL-A) intervention, a MedDiet WL lifestyle intervention (WL-Med) compared to Control on BA metabolism, the gut microbiome, and gene expression in exfoliated intestinal epithelial cells. The study includes 1) screening potential subjects; 2) baseline assessment of diet, lifestyle behaviors (e.g., physical activity), anthropometrics, and blood and fecal sampling; 3) a four-arm RCT; and 4) mid-(month 3) and post-intervention (month 6) assessments.

Med-A will attend a one-hour, in-person individual session with a registered dietitian (RD) during the two weeks prior to the start of the intervention. For subjects randomized to Med-A the study RD will instruct on adoption of an eating pattern consistent with a MedDiet using an individualized MedDiet exchange list and companion guide. Recommended daily exchanges are based on individual caloric needs to maintain weight. We will not ask subjects to abstain from alcohol during the trial despite its known effects on BA metabolism, but we will recommend that only 5% of calories come from alcohol taken with meals. Following the initial session, subjects will meet for 24 individual sessions (1-hour, held approximately weekly) in-person or virtually over the remaining 6 months. Additional asynchronous learning materials will be distributed weekly through a private Facebook group. The Med-A group will be asked to maintain their usual physical activity.

WL-A will attend a one-hour, in-person individual session with a registered dietitian (RD) during the two weeks prior to the intervention. For WL-A, the focus will be on daily calorie restriction (-500-750 kcal/day) to achieve a 1-2 lb. per week WL and 5% WL from baseline at 6 months in the context of the subject's typical diet pattern. We will not ask subjects to abstain from alcohol during the trial despite its known effects on BA metabolism, but we will recommend that only 5% of calories come from alcohol taken with meals. Following the initial session, subjects will meet for 24 individual, virtual or in-person sessions (1-hour, held approximately weekly) over the remaining 6 months. Additional asynchronous learning materials will be distributed weekly through a private Facebook group. The WL-A group will be prescribed an activity program. Physical activity will be monitored via FitBit.

WL-Med will attend a one-hour, in-person session with a registered dietitian (RD) prior to the intervention. The RD will instruct on an eating pattern consistent with a MedDiet using an individualized exchange list. Exchanges are based on individual caloric needs to lose weight (WL-Med, calorie restriction to achieve a 1-2 lb. per week WL and 5% WL from baseline at 6 months). We will not ask subjects to abstain from alcohol despite its known effects on BA metabolism, but we will recommend that only 5% of calories come from alcohol taken with meals. Following the initial session, subjects will meet remotely or in-person for 24 individual sessions (1-hour, held approximately weekly). Additional asynchronous learning materials will be distributed weekly through a private Facebook group. The WL-Med group will be prescribed an activity program. Some asynchronous lessons will contain information about physical activity. Physical activity will be monitored via FitBit.

The study RD will meet individually with the Control group subjects in-person for 1-hour at the start of the 6-month intervention. Control participants will be instructed to maintain current eating and activity patterns and weight over the next 6 months. No dietary recommendations are provided, and they will receive weekly health newsletters that include non-diet related health topics (e.g., flu prevention). Contact will be made again at month-3 and post-intervention (month-6) research visits and during monthly phone calls to collect data pertaining to recent diet intake. At the month-3 assessment, weight will be checked and those with >2.5% WL from baseline will receive additional instruction from the RD to maintain lifestyle patterns. All WL-Med materials are offered to the group in a self-guided format following the 6-month intervention.

Absolute measurement of BAs in stool and serum obtained at baseline will be performed under the direction of Co-I Ridlon. Samples will be extracted, and supernatants will be dried and resuspended for LC/MS analysis following validated and published methods. We will quantify all major primary and secondary BAs (e.g., DCA) and glycine and taurine conjugates and their ratios, as well as total BAs and total unconjugated BAs. Authentic reference BAs will be purchased from Sigma-Aldrich and Steraloids. Blind duplicate samples will be used to assess inter- and intra-batch precision.

Absolute measurement of BAs in stool and serum obtained at mid-study (3 month follow up) will be performed under the direction of Co-I Ridlon. Samples will be extracted, and supernatants will be dried and resuspended for LC/MS analysis following validated and published methods. We will quantify all major primary and secondary BAs (e.g., DCA) and glycine and taurine conjugates and their ratios, as well as total BAs and total unconjugated BAs. Authentic reference BAs will be purchased from Sigma-Aldrich and Steraloids. Blind duplicate samples will be used to assess inter- and intra-batch precision.

Absolute measurement of BAs in stool and serum obtained at post-intervention (6 month follow up) will be performed under the direction of Co-I Ridlon. Samples will be extracted, and supernatants will be dried and resuspended for LC/MS analysis following validated and published methods. We will quantify all major primary and secondary BAs (e.g., DCA) and glycine and taurine conjugates and their ratios, as well as total BAs and total unconjugated BAs. Authentic reference BAs will be purchased from Sigma-Aldrich and Steraloids. Blind duplicate samples will be used to assess inter- and intra-batch precision.

The UIC Genomics core will PCR amplify genomic DNA with primers CS1_515F and CS2_806R (modified from the set used by the Earth Microbiome Project) targeting the V4 region of microbial small subunit ribosomal RNA genes. Amplicons will be generated using a two-stage PCR protocol. The V4 region of the 16S rRNA gene will be sequenced with the Illumina MiSeq platform to generate 2x250 bp paired end reads per sample. Environmental controls will be included in the sequences to distinguish from any contaminants in reagents or the lab environment.

The UIC Genomics core will PCR amplify genomic DNA with primers CS1_515F and CS2_806R (modified from the set used by the Earth Microbiome Project) targeting the V4 region of microbial small subunit ribosomal RNA genes. Amplicons will be generated using a two-stage PCR protocol. The V4 region of the 16S rRNA gene will be sequenced with the Illumina MiSeq platform to generate 2x250 bp paired end reads per sample. Environmental controls will be included in the sequences to distinguish from any contaminants in reagents or the lab environment.

The UIC Genomics core will PCR amplify genomic DNA with primers CS1_515F and CS2_806R (modified from the set used by the Earth Microbiome Project) targeting the V4 region of microbial small subunit ribosomal RNA genes. Amplicons will be generated using a two-stage PCR protocol. The V4 region of the 16S rRNA gene will be sequenced with the Illumina MiSeq platform to generate 2x250 bp paired end reads per sample. Environmental controls will be included in the sequences to distinguish from any contaminants in reagents or the lab environment.

From stool preserved in Ambion Denaturation Solution, eukaryotic polyA+ RNA will be isolated using the Active Motif mTRAP Maxi kit followed by DNA removal with DNAFree (Invitrogen). Libraries will be quantified using the Library Quantification kit (Kapa Biosystems), and sequencing will be performed on an Illumina HiSeq 2500 platform using standard Illumina protocols. RNA reads will be mapped with the STAR aligner using the default parameters to the Ensembl GRCh38 human reference. Reads will be examined for quality control using FastQC and quantified using HTSeq-count. Sequencing reads will be filtered to remove genes present in low abundance. For stool exfoliated cells, the RNA-seq gene count matrix is very sparse, with most entries corresponding to zero transcripts; thus, genes in stool will be removed if >33% of the samples contain only 0 or 1 read.

From stool preserved in Ambion Denaturation Solution, eukaryotic polyA+ RNA will be isolated using the Active Motif mTRAP Maxi kit followed by DNA removal with DNAFree (Invitrogen). Libraries will be quantified using the Library Quantification kit (Kapa Biosystems), and sequencing will be performed on an Illumina HiSeq 2500 platform using standard Illumina protocols. RNA reads will be mapped with the STAR aligner using the default parameters to the Ensembl GRCh38 human reference. Reads will be examined for quality control using FastQC and quantified using HTSeq-count. Sequencing reads will be filtered to remove genes present in low abundance. For stool exfoliated cells, the RNA-seq gene count matrix is very sparse, with most entries corresponding to zero transcripts; thus, genes in stool will be removed if >33% of the samples contain only 0 or 1 read.

From stool preserved in Ambion Denaturation Solution, eukaryotic polyA+ RNA will be isolated using the Active Motif mTRAP Maxi kit followed by DNA removal with DNAFree (Invitrogen). Libraries will be quantified using the Library Quantification kit (Kapa Biosystems), and sequencing will be performed on an Illumina HiSeq 2500 platform using standard Illumina protocols. RNA reads will be mapped with the STAR aligner using the default parameters to the Ensembl GRCh38 human reference. Reads will be examined for quality control using FastQC and quantified using HTSeq-count. Sequencing reads will be filtered to remove genes present in low abundance. For stool exfoliated cells, the RNA-seq gene count matrix is very sparse, with most entries corresponding to zero transcripts; thus, genes in stool will be removed if >33% of the samples contain only 0 or 1 read.

Measured with a food frequency questionnaire, 24-hour diet recalls, and screener which will be aggregated to evaluate a total adherence score

Measured with a food frequency questionnaire, 24-hour diet recalls, and screener which will be aggregated to evaluate a total adherence score

Measured with a food frequency questionnaire, 24-hour diet recalls, and screener which will be aggregated to evaluate a total adherence score

Measured with a food frequency questionnaire, 24-hour diet recalls, and screener which will be aggregated to evaluate a total adherence score

Measured with a food frequency questionnaire, 24-hour diet recalls, and screener which will be aggregated to evaluate a total adherence score

Measured with a food frequency questionnaire, 24-hour diet recalls, and screener which will be aggregated to evaluate a total adherence score

Measured with a food frequency questionnaire, 24-hour diet recalls, and screener which will be aggregated to evaluate a total adherence score

Inclusion Criteria: Men and women 45-75 years of age Self-identify as AA BMI 30-50 kg/m2 Willingness to participate in all procedures including maintaining weight/current physical activity if randomized to Med-A/Control Willingness and ability to provide informed consent Willingness to be randomized Understands English Has access to a phone Plans to reside in Chicago for the next 8-10 months. Exclusion Criteria: renal disease autoimmune disorders immunodeficiency malabsorptive disorders significant gastrointestinal and/or hepatic diseases severe ischemic heart disease severe pulmonary disease history of bariatric surgery alcohol abuse (> 50 grams/day) illicit drug abuse (other than marijuana based on self-report) combustible tobacco use uncontrolled diabetes based on HbA1c>9.0% eating disorder cancer treatment within the past 12 months history of CRC genetic predisposition to CRC (e.g., Lynch syndrome) weight > 450 lbs. (weight limitation of the DXA scanner) currently adhering to a MedDiet based on a diet screener self-reported WL > 3% in the past 12 months currently on a WL diet or actively involved in a formal WL program (e.g., Weight Watchers) food allergies that would interfere with adopting a MedDiet antibiotic use in the past 3 months night-shift work regular use (i.e., ≥ 3 times per week) of prebiotics/probiotics/synbiotics, dietary fiber supplements, or laxatives, Gait disorder currently pregnant active Covid-19 infection within 6 weeks of recruitment/data collection.