Yogurt Supplementation to Alter Bone Biomarkers, the Gut Microbiota and Inflammation in Older Adults

Identifying Mechanisms Underpinning the Impact of Daily Yogurt Supplementation on the Gut Microbiome, Inflammation and Bone Biomarkers in Older Adults

University of Massachusetts, Worcester, The Lawrence Senior Center, Danone Institute International, American Society for Nutrition, The International Osteoporosis Foundation

Caribbean Latino adults are a vulnerable population at risk for developing osteoporosis. In addition to lifestyle factors, such as dietary intake, the investigators hypothesize that a unhealthy gut microbial environment coupled with high inflammation contributes to the risk of developing osteoporosis. There has been little research conducted on mechanisms underpinning how a low-intensity dietary intervention, including supplementation with daily yogurt, can affect the gut health of Caribbean Latino adults. The study objectives are to: 1) determine whether daily yogurt supplementation reduces bone turnover (biomarkers of overall bone health) and inflammation compared to a control group that maintains their usual diet (void of yogurt); and 2) collect preliminary data on the effects of daily yogurt supplementation on the gut microbiome compared to diet control group.

Recruitment: A total of 40 Caribbean Latino men and women ≥50 y will be recruited and enrolled from Greater Lawrence, MA, USA. Subjects will be recruited using flyers distributed throughout the communities including Senior Centers and other community organizations and agencies. The study team will also set up tables/booths at local festivals and community events. Interested participants will be directed to call a study-center call line to learn more about the study and to be screened for inclusion, if interested. For those who are eligible, the study team member will schedule the first interview, which will include obtaining consent. Intervention Protocol: 40 study participants will undergo 12 study visits at the Lawrence Senior Center. They will first undergo a 4-week wash in baseline period. Participants will then undergo a two-month intervention (yogurt supplementation or control [maintenance of usual diet]). The study period will include weekly interviews (12 visits in total). Baseline Month First interview (visit -3) in week 1 of the baseline month participants will be provided written informed consent. A trained bilingual interviewer will describe the study and answer any questions participants may have. The interviewer will also be available to read the consent form to participants in the event that he/she is unable to read. The consent forms will be stored in a locked file cabinet inside of a locked office at UMass Lowell. During the first visit, a trained bilingual interviewer will complete a baseline questionnaire, a 24-hour dietary recall where the interviewer will record all foods eaten by participants the previous day, and complete physical assessment measures, such as height and body weight. Participants will be provided fecal collection materials and the interviewer will instruct participants on the proper aseptic technique for fecal collection (see collection methods section). The interview will collect the fecal tube at the next visit (week 2 of the baseline month). The first interview will take approximately 2 hours to complete. Interviews in week 2 and 3 of the baseline month (visits -2, -1): At each interview, the trained interviewer will collect the fecal sample and provide a new collection kit for the following visit. A 24-hour dietary recall will be completed and a physical assessment measures will be obtained. These interviews will take approximately 1 hour to complete. Interview in week 4 of the baseline month (visit 0): During this visit, the trained interviewer will collect the fecal sample and provide a new collection kit for the following visit. A 24-hour dietary recall will be completed and physical assessment measures will be obtained. A certified phlebotomist will collect 23 milliliters of blood for baseline data on inflammation and bone turnover. This interview will take approximately 1½ hours to complete. Intervention Months 1 and 2 (8 weeks total): Upon completion of 4-week baseline period, participants will be randomly assigned to intervention and control groups by a project manager not associated with the project (study manager of the Center for Population Health, University of Massachusetts, Lowell). This study manager will randomize de-identified participants to either intervention or control at random to ensure an even number of men and women in each group. They will then participate in 8 weekly interviews with a trained bilingual interviewer. The intervention group will be provided their yogurt weekly (7, 5 ounce containers per week) and will be asked to bring back empty yogurt containers to their next visit. Interview in week 1 of Intervention Month 1 (visit #1): During the week 1 visit, participants will complete a 24-hour dietary recall performed by a trained bilingual interviewer and will also undergo a physical assessment of their weight and height. Participants will be provided with their yogurt for the week and a fecal collection kit and instructions. The fecal sample will be picked up the following week by the study interviewer. This interview will take approximately 1 hour to complete. Interview in week 2-7 of Intervention Months 1 & 2 (visits 2-7) Participants will complete 6 weekly interviews. The trained interviewer will collect the fecal sample, provide new fecal sample kit, administer a 24-hour dietary recall and obtain weight and height measures. If in the yogurt intervention group, participants will receive an additional week of yogurt. At each follow-up interview, the participants will be asked by the trained interviewer if they have gas, bloating or any feelings of gastrointestinal discomfort or signs of allergy after eating the yogurt. If the participant responds "yes", the trained interviewer will document the participant's symptoms on the Adverse Event sheet (attached) and will then notify the PI. The PI (Registered Dietitian) will call the participant to inquire further about the symptoms and decide with the participant if they should remain in the study. All AEs will be reported to the IRB within 24h of participant report to the PI. Participants in the control group will be reminded to continue their usual diet and abstain from yogurt intake. These interviews will take approximately 1 hour each to complete. Interviews week 4 of Intervention Month 2 (final visit 8): During this interview, the trained interviewer will collect the fecal sample. A 24-hour dietary recall will be completed and physical assessment measures will be obtained. A certified phlebotomist will collect 23 milliliters of blood for baseline data on inflammation and bone turnover. This interview will take approximately 1 ½ hours to complete. Additional Information on Data Collection Measures: Study Questionnaire (collected during week 1 of the wash-in period): This questionnaire will ask participants to report their age, sex, education, income, marital status, employment status and place of birth, questions on health history (e.g., told by a health care provider if they have health conditions such as diabetes, osteoporosis, and cancer), health behaviors (smoking, alcohol consumption, physical activity, and medication use) will be obtained and intakes of vitamin D and calcium will also be assessed. Twelve, 24-hour Dietary Recalls: The investigators will obtain dietary data using an automated multiple-pass interview method in conjunction with the University of Minnesota Nutrition Data System (NDS) software. The USDA automated multiple-pass method is a computerized method of collecting 24-hour dietary recalls. This method is completed by a trained interviewer either in person or by telephone and is designed to reduce respondent burden. This approach uses 5 steps to collect information about food and beverages consumed over the past 24 hours, as follows: Step 1: Collect a list of foods and beverages consumed the previous day (quick list). Step 2: Probe for foods and beverages forgotten during the quick list. Step 3: Collect time and eating occasion for each food reported Step 4: For each food, collect a detailed description, amount consumed and any additions to the food such as milk on cereal or cream in coffee. Step 5: Conduct a final probe for anything else consumed. This information will be entered directly into the computer system by the trained interviewer. The trained bilingual interviewer will administer the 24-h recall during the interviews and will enter dietary data directly into a computer. Food models will be used to help participants in estimating portion sizes. Blood sample: 23 milliliters of blood will be taken at the end of the baseline period (week 4, baseline, visit 0), and again at the end of the 8-week intervention (week 4, Interview Month 2, visit 8). Blood samples will be obtained at the Lawrence Senior Center by a certified phlebotomist. Samples will be transported in temperature controlled containers immediately after blood draw, and stored at the UML in a freezer at -20℃ until laboratory analysis. Fecal Samples: Participants will be asked to collect a fecal sample at their home using standardized procedures with Omnigene Gut kits. a. The stool collection kit will contain: i. Gloves ii. Marker iii. Stool collection hat iv. One stool collection OMNIgene•Gut (DNA genotek, Canada). The kit includes the collection device showed in Figure 1. v. Paper instructions with images vi. Brown paper bags vii. Plastic container with tight seal Collection Methods Specimen Labeling: The collection tube provided in the kit will have a customized easy to fill up label that will be use to provide specific information about the sample: date, time of the day, and consistency of the samples. The kit will be inspected for integrity and cleanliness prior to sample collection before labeling. In case of noticing lack of integrity, the device should be discarded and another collection device should be used. Participants will fill the label with the appropriate information: month, day, time and fecal consistency. Specimen collection: If needed the participant should urinate before the fecal collection to avoid sample contamination with urine. Then, the stool collection hat should be placed on the toilet as shown in Figure 3. Following patients should deposit the stools into the hat. Once finish, the purple cap will be unscrew for the OMNIgene•Gut collection. The collection device will be kept straight up in one hand, and with the other hand, the samples will be collected from the stool collection using the spatula provided in the kit. The purple top of the collection device will be filled completely. Any excess will be scraped off with the spatula, and the purple cap will be then screwed back tightly. The tubes will be shaken for at least 30 seconds until the samples do not contain large particles. Then samples will be bagged (brown paper bags will be provided for privacy) and place it in the plastic container also provided. Samples will be kept in the participant's refrigerator (or any cool, dark, dry space) up to 7 days. Participants will bring their samples to each subsequent interview. The samples will be stored at UML in a -80 degree freezer until all samples are collected. Samples will be placed on dry ice and driven by an investigator to the Center for Microbiome Research laboratory at UMass Medical School for analysis using 16S rRNA sequencing. Translation: The screening form, recruitment flyer, questionnaire on sociodemographic, health and health behaviors and informed consent form will be translated to Spanish by a bilingual staff member in our Center for Population Health and Health Disparities. All materials will be back translated by bilingual staff at the Lawrence Senior Center. A signed copy of the "Certificate of Translation" will be submitted to the IRB once completed. Describe the participant population: 40 Caribbean Latino adults age ≥50y will be recruited by flyers displayed in the Lawrence Multi-Purpose Senior Center. The senior center serves approximately 200-400 adults per day (over 5000 yearly). 75% of those served are considered Hispanic with the majority of Hispanics coming from Caribbean descent. Recruitment Information Participants will be recruited posting flyers around the Lawrence Multi-Purpose Senior Center and other community organizations and agencies. The study team will also set up tables/ booths at local festivals and community events. Sample size calculation: Based on data from Hooshman and colleagues 2011 published in the British Journal of Nutrition, sample size was calculated for absolute difference in osteocalcin between two dietary supplemented groups (OC after intervention with dried plums 17.19 +- 0.7 vs. OC after intervention with dried apples 15.6 +- 0.6) as >0.99 with 20 people per group. In addition, effect sizes were used from the same study for CRP in the dried plum group (20.5 +- 3.0) and the dried apple group (14.9 +- 3.0) and provided a power >0.99 with a sample size of 20 per group. Collaborators: i. The investigators are collaborating with the Lawrence Multipurpose Senior Center and City of Lawrence Mayor's Health Task Force team for research effort support. The Senior Center will serve as the study site, as well as provide support for recruitment and will provide a study staff member for interviews and blood & fecal collection. ii. The investigators will also collaborate with the UMass Medical School, Center for Microbiome Research. Dr. Ana Luis Maldonado is our primary contact at the Center and will provide her expertise on the microbiome for fecal sample collection, analysis, and interpretation of results. All fecal samples will be sent to the UMass Medical School for analysis. Data will be de-identified and coded by study ID number.

Participants assigned to the yogurt group were provided with a biweekly supply of yogurt at every other visit and were directed to consume one serving daily (5-ounce, 141 grams) and to store the remaining yogurt at 40˚ F or lower until consumed. Additionally, all participants were instructed to keep a log of any yogurt consumption time and changes to bowel health. The ingredients of the yogurt include Pasteurized Grade A Milk, Cane Sugar, Yogurt Cultures (L. bulgaricus, S. thermophilus), and Vanilla Extract. The intervention meets the Codex Alimentarius definition of yogurt. In addition, the yogurt lacked preservatives, added fruit or pectin, was not enriched with added prebiotics or probiotics.

The control group was asked to abstain from yogurt consumption. Diet (including yogurt consumption) was monitored by weekly 24-hour dietary recalls throughout the study period.

Participants will be asked to consume one serving of yogurt per day. Nutrient composition of yogurt provided to participants in the yogurt group: Nutrients per serving Serving Size: 141 g Energy (kcal) 170 Total Fat (g) 8 Total Carbohydrate (g) 16 Sugars (g) 15 Protein (g) 5 Sodium (mg) 85 Calcium (%) 15 Vitamin A (%) 4 Ingredients: Pasteurized Grade A Milk, Cane Sugar, Yogurt Cultures (S. thermophilus, L. bulgaricus), Vanilla Extract.

To evaluate bone health at baseline and at study completion, bone formation markers, pro-collagen I alpha 1 (P1NP) pg/mL and osteocalcin (OC) pg/mL will be measured in serum using ELISA kits.

To evaluate bone health at baseline and at study completion, bone resorption marker, tartrate-resistant acid phosphatase subunit 5b (TRAP5b) in units/L will be measured in serum using ELISA kits.

To evaluate the phylogeny and taxonomy of the gut microbiota at baseline and throughout the intervention, bacterial DNA in the feces will be amplified at the phylogenetically informative V4 region of 16S rRNA gene using universal primers 515F/806R. Libraries will be sequenced using the Illumina MiSeq platform, following a 2 × 300 base pair paired-end protocol.

To evaluate metabolic function of the gut microbiota at baseline and at study completion, ten individual short chain fatty acid (SCFA) isomers [acetate, propionate, butyrate, isobutyrate, valerate, isovalerate, 2-methylbutyrate, 3-methylvalerate, 4-methylvalerate, hexanoate] will be quantified via liquid chromatography-tandem mass spectrometry. Individual concentrations of SCFA will be presented as umol/g.

To evaluate systemic inflammation at baseline and at study completion, C-Reactive Protein (CRP) mg/L will be measured by Clinical Chemistry Analyzer.

To evaluate inflammation and immune function at baseline and at study completion, eight key inflammatory cytokines [TNF-α, IL-6, IL-10, IFN-γ, IL-1β, IL-8, IL-12p70, MCP-1] will be quantified from human plasma using the FirePlex Human Inflammation - Immunoassay Panel (ab243550, Abcam). A pro-inflammatory cytokine score (cytokine score, ICS) will be calculated for each participant by ranking the inflammatory cytokines by tertile. For all cytokines, except IL-10, the top tertile of data (≥ 67%) received a rank of 2, the middle tertile data (33 - 66%) received a rank of 1, and the bottom tertile of data (< 33%) received a rank of 0. Due to the anti-inflammatory properties of IL-10, the ranking system was opposite, where the top tertile of IL-10 values received a rank of 0. All the rankings were summed to produce a maximum score of 16, indicating highest inflammation, and a minimum score of 0, indicating lowest inflammation.

As a measure of intestinal integrity at baseline and at study completion, the concentrations of lipopolysaccharide binding protein (LBP) pg/mL and intestinal fatty acid binding protein (FABP2) pg/mL will be measured in plasma using Abcam ELISA kits (ab213805, ab234566).

Inclusion Criteria: self-reported Caribbean Latino descent at least 50 years of age Exclusion Criteria: usual yogurt intake (> 2 servings per week) use of antibiotics in the past 6 months regular weekly laxative use self-reported osteoporosis history of osteoporosis-related fracture self-reported diagnosis of cancer self-reported diagnosis of gastrointestinal diseases (Crohn's disease, ulcerative colitis, Clostridium difficile infection) history of gastrointestinal alteration procedures (appendectomy, gastric bypass surgery) use of osteoporosis drugs use of steroids use of chemotherapeutic agents

De-identified participant data available upon request to the study PI (Kelsey Mangano, University of Massachusetts, Lowell). Gut microbiota 16S rRNA sequencing data will be deposited to a public data repository.