Advertisement

NOTICE: We are experiencing technical issues with Academy members trying to log into the JAND site using Academy member login credentials. We are working to resolve the issue as soon as possible. Alternatively, if you are an Academy member, you can access the JAND site by registering for an Elsevier account and claiming access using the links at the top of the JAND site. Email us at [email protected] for assistance. Thanks for your patience!

Microbiota Stability and Gastrointestinal Tolerance in Response to a High-Protein Diet with and without a Prebiotic, Probiotic, and Synbiotic: A Randomized, Double-Blind, Placebo-Controlled Trial in Older Women

Open AccessPublished:March 18, 2020DOI:https://doi.org/10.1016/j.jand.2019.12.009

      Abstract

      Background

      Higher protein intakes may help reduce sarcopenia and facilitate recovery from illness and injury in older adults. However, high-protein diets (HPDs) including animal-sourced foods may negatively perturb the microbiota, and provision of probiotics and prebiotics may mitigate these effects.

      Objective

      The aim of this study was to examine the effects of HPD, with and without a probiotic and/or prebiotic, on gut microbiota and wellness in older women.

      Design

      We conducted an 18-week, double-blind, placebo-controlled, crossover study.

      Participants/setting

      Participants were healthy, older women (mean age±standard deviation=73.7±5.6 years; n=26) recruited from Florida.

      Intervention

      Participants received a weight-maintenance HPD for 2-week periods and the following, in random order: HPD alone (1.5 to 2.2 g/kg/day protein); HPD plus multistrain probiotic formulation (1.54×109 Bifidobacterium bifidum HA-132, 4.62×109 Bifidobacterium breve HA-129, 4.62×109 Bifidobacterium longum HA-135, 4.62×109 Lactobacillus acidophilus HA-122, and 4.62×109 Lactobacillus plantarum HA-119), HPD plus prebiotic (5.6 g inulin), and HPD plus synbiotic (probiotic plus inulin), separated by 2-week washouts. Stools were collected per period for quantitative polymerase chain reaction (strain recovery) and 16S ribosomal RNA gene amplicon sequencing analyses (microbiota profile). Measures of gastrointestinal and general wellness were assessed.

      Main outcome measures

      Microbiota composition and probiotic strain recovery were measured.

      Statistical analyses

      Microbiota composition was analyzed by Wilcoxon signed-rank test and t test. Secondary outcomes were analyzing using generalized linear mixed models.

      Results

      The microbiota profile demonstrated relative stability with the HPD; representation of Lactobacillus, Lactococcus, and Streptococcus were enhanced, whereas butyrate producers, Roseburia and Anaerostipes, were suppressed. Lactococcus was suppressed with synbiotic vs other HPD periods. Recovery was confirmed for all probiotic strains. Indicators of wellness were unchanged, with the exception of a minimal increase in gastrointestinal distress with inulin. Fat-free mass increased from baseline to study end.

      Conclusions

      An HPD adhering to the recommended acceptable macronutrient distribution ranges maintains wellness in healthy older women and exerts minor perturbations to the microbiome profile, a group that may benefit from a higher protein intake. ClinicalTrials.gov ID: NCT #02445560.

      Keywords

      Research Question: What is the impact of a high-protein diet, with and without prebiotic and probiotic supplements, on microbiota composition and general wellness in older women?
      Key Findings: In this randomized, double-blind, placebo-controlled, crossover trial, a controlled high-protein diet alone, or when supplemented with a prebiotic, multistrain probiotic or synbiotic, did not affect probiotic recovery, bacterial diversity, or general wellness in older women. The microbiota of the older women demonstrated relative stability to the high-protein diet, with enhancement of Lactococcus and Streptococcus, and suppression of butyrate-producing genera, Roseburia and Anaerostipes. The high-protein diet, in keeping with the acceptable macronutrient distribution ranges and provided intermittently over 18 weeks, improved muscle mass in older women.
      The Recommended Dietary Allowance (RDA) for protein (0.8 g/kg) is considered suboptimal for older adults due to age-related metabolic and physiologic changes.
      • Landi F.
      • Calvani R.
      • Tosato M.
      • et al.
      Protein intake and muscle health in old age: From biological plausibility to clinical evidence.
      Higher protein intake is associated with reduced risk for sarcopenia,
      • Isanejad M.
      • Mursu J.
      • Sirola J.
      • et al.
      Dietary protein intake is associated with better physical function and muscle strength among elderly women.
      increased strength,
      • Mishra S.
      • Goldman J.D.
      • Sahyoun N.R.
      • Moshfegh A.J.
      Association between dietary protein intake and grip strength among adults aged 51 years and over: What We Eat in America, National Health and Nutrition Examination Survey 2011-2014.
      and possibly improved bone health.
      • Shams-White M.M.
      • Chung M.
      • Du M.
      • et al.
      Dietary protein and bone health: A systematic review and meta-analysis from the National Osteoporosis Foundation.
      ,
      • Wallace T.C.
      • Frankenfeld C.L.
      Dietary protein intake above the current RDA and bone health: A systematic review and meta-analysis.
      To maintain muscle mass and offset sarcopenia, a daily protein intake of 1.0 to 1.2 g/kg of body weight is considered essential, and an even higher protein intake may benefit older adults who are physically active and those experiencing acute or chronic disease.
      • Bauer J.M.
      • Diekmann R.
      Protein and older persons.
      ,
      • Bauer J.
      • Biolo G.
      • Cederholm T.
      • et al.
      Evidence-based recommendations for optimal dietary protein intake in older people: A position paper from the PROT-AGE Study Group.
      This high level of protein intake also seems required to preserve muscle mass during weight reduction.
      • Muscariello E.
      • Nasti G.
      • Siervo M.
      • et al.
      Dietary protein intake in sarcopenic obese older women.
      However, research suggests that 10% to 12% of community-dwelling older adults, and particularly women, have an inadequate protein intake level when considering the Estimated Average Requirement of 0.66 g/kg (equivalent to an RDA of 0.8 g/kg) as a reference point.
      • ter Borg S.
      • Verlaan S.
      • Mijnarends D.M.
      • Schols J.M.
      • De Groot L.C.
      • Luiking Y.C.
      Macronutrient intake and inadequacies of community-dwelling older adults, a systematic review.
      In addition, the number of older adults with inadequate protein intake levels rises to 23% to 27% when a proposed Estimated Average Requirement of 0.83 g/kg (corresponding to an RDA of 1.0 g/kg) is used as reference.
      • ter Borg S.
      • Verlaan S.
      • Mijnarends D.M.
      • Schols J.M.
      • De Groot L.C.
      • Luiking Y.C.
      Macronutrient intake and inadequacies of community-dwelling older adults, a systematic review.
      In addition, physically active older adults, for whom a minimum protein intake of 1.2 g/kg/day is recommended, may not meet their needs.
      • Ten Haaf D.S.M.
      • De Regt M.F.
      • Visser M.
      • et al.
      Insufficient protein intakes is highly prevalent among physically active elderly.
      For many adults, a protein intake corresponding to the RDA falls near or below the lower cut point of the acceptable macronutrient distribution range (AMDR) at 10% energy from protein. Conversely, in older women with lower energy requirements, higher protein diets (HPDs) may correspond closer to the higher cut point of the AMDR at 35% of energy from protein. Although the AMDR for protein is set to reduce the risk of disease while providing essential nutrients,
      Institute of Medicine
      Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids.
      little is known about the potential risks of consuming diets with >25% energy from protein. If higher-protein diets are to be broadly recommended to older adults, the potential adverse effects of such a dietary pattern warrant further study.
      The relationships among macronutrient diet composition, gut microbiota, and health and disease have not been elucidated. Few studies have examined the effects of an HPD (25% to 35% energy from protein) on microbiota composition.
      • David L.A.
      • Maurice C.F.
      • Carmody R.N.
      • et al.
      Diet rapidly and reproducibly alters the human gut microbiome.
      • Beaumont M.
      • Portune K.J.
      • Steuer N.
      • et al.
      Quantity and source of dietary protein influence metabolite production by gut microbiota and rectal mucosa gene expression: A randomized, parallel, double-blind trial in overweight humans.
      • Windey K.
      • De Preter V.
      • Louat T.
      • et al.
      Modulation of protein fermentation does not affect fecal water toxicity: A randomized cross-over study in healthy subjects.
      • Salonen A.
      • Lahti L.
      • Salojarvi J.
      • et al.
      Impact of diet and individual variation on intestinal microbiota composition and fermentation products in obese men.
      • Duncan S.H.
      • Belenguer A.
      • Holtrop G.
      • Johnstone A.M.
      • Flint H.J.
      • Lobley G.E.
      Reduced dietary intake of carbohydrates by obese subjects results in decreased concentrations of butyrate and butyrate-producing bacteria in feces.
      • Cotillard A.
      • Kennedy S.P.
      • Kong L.C.
      • et al.
      Dietary intervention impact on gut microbial gene richness.
      • Russell W.R.
      • Gratz S.W.
      • Duncan S.H.
      • et al.
      High-protein, reduced-carbohydrate weight-loss diets promote metabolite profiles likely to be detrimental to colonic health.
      David and colleagues
      • David L.A.
      • Maurice C.F.
      • Carmody R.N.
      • et al.
      Diet rapidly and reproducibly alters the human gut microbiome.
      compared a high-protein/very-high-fat, solely animal-sourced diet to a high-fiber, vegan diet. Predictably, the animal-sourced diet suppressed bacterial genera that hydrolyze plant polysaccharides, that is, Roseburia, Eubacterium rectale, and Ruminococcus bromii.
      • David L.A.
      • Maurice C.F.
      • Carmody R.N.
      • et al.
      Diet rapidly and reproducibly alters the human gut microbiome.
      Higher-protein diets intended for weight loss have also been shown to suppress Roseburia and E rectale,
      • Walker A.W.
      • Ince J.
      • Duncan S.H.
      • et al.
      Dominant and diet-responsive groups of bacteria within the human colonic microbiota.
      as well as Bifidobacterium,
      • Salonen A.
      • Lahti L.
      • Salojarvi J.
      • et al.
      Impact of diet and individual variation on intestinal microbiota composition and fermentation products in obese men.
      which are considered to be beneficial microbes. Although some have proposed that the negative impact of those diets may be mitigated by increasing the fiber content,
      • Cotillard A.
      • Kennedy S.P.
      • Kong L.C.
      • et al.
      Dietary intervention impact on gut microbial gene richness.
      others have noted that few studies have controlled for dietary variables known to significantly impact the microbiota,
      • Blachier F.
      • Beaumont M.
      • Portune K.J.
      • et al.
      High-protein diets for weight management: Interactions with the intestinal microbiota and consequences for gut health. A position paper by the my new gut study group.
      • Menni C.
      • Jackson M.A.
      • Pallister T.
      • Steves C.J.
      • Spector T.D.
      • Valdes A.M.
      Gut microbiome diversity and high-fibre intake are related to lower long-term weight gain.
      • Seganfredo F.B.
      • Blume C.A.
      • Moehlecke M.
      • et al.
      Weight-loss interventions and gut microbiota changes in overweight and obese patients: A systematic review.
      such as fiber content and energy intake, while manipulating protein intake.
      • Beaumont M.
      • Portune K.J.
      • Steuer N.
      • et al.
      Quantity and source of dietary protein influence metabolite production by gut microbiota and rectal mucosa gene expression: A randomized, parallel, double-blind trial in overweight humans.
      ,
      • Windey K.
      • De Preter V.
      • Louat T.
      • et al.
      Modulation of protein fermentation does not affect fecal water toxicity: A randomized cross-over study in healthy subjects.
      Based on prospective cohort studies, Pedersen and colleagues
      • Pedersen A.N.
      • Cederholm T.
      Health effects of protein intake in healthy elderly populations: A systematic literature review.
      suggested that an intake of 1.2 to 1.5 g/kg/day, representing approximately 15% to 20% energy from protein, could be considered safe for older adults.
      • Pedersen A.N.
      • Cederholm T.
      Health effects of protein intake in healthy elderly populations: A systematic literature review.
      However, additional research is required to determine the potential adverse effects of higher protein intake. One of these adverse effects may be to induce dysbiosis through disruption of the commensal microbiota, a common condition in older adults.
      • Claesson M.J.
      • Cusack S.
      • O'Sullivan O.
      • et al.
      Composition, variability, and temporal stability of the intestinal microbiota of the elderly.
      ,
      • Rehman T.
      Role of the gut microbiota in age-related chronic inflammation.
      Specifically, a decrease in carbohydrate fermentation and an increase in proteolytic fermentation has been found in older adults.
      • Bischoff S.C.
      Microbiota and aging.
      This fermentation shift may result in a decreased production of beneficial short-chain fatty acids (eg, butyrate), while increasing metabolites from protein fermentation, some of which are inflammatory in nature and are thought to increase the risk of renal and cardiovascular diseases.
      • Korpela K.
      Diet, microbiota, and metabolic health: Trade-off between saccharolytic and proteolytic fermentation.
      Increasing protein intake results in an increased amount of protein substrate reaching the colon,
      • Cummings J.H.
      • Hill M.J.
      • Bone E.S.
      • Branch W.J.
      • Jenkins D.J.
      The effect of meat protein and dietary fiber on colonic function and metabolism. II. Bacterial metabolites in feces and urine.
      ,
      • Macfarlane G.T.
      • Macfarlane S.
      Bacteria, colonic fermentation, and gastrointestinal health.
      which subsequently increases proteolytic fermentation.
      • Windey K.
      • De Preter V.
      • Louat T.
      • et al.
      Modulation of protein fermentation does not affect fecal water toxicity: A randomized cross-over study in healthy subjects.
      Particularly in older adults with lower overall energy intake, consuming more protein-rich foods may occur at the expense of fiber-containing foods, which may reduce fiber intake further below recommendations and worsen dysbiosis. As many older adults already present with decreased microbial diversity
      • Bischoff S.C.
      Microbiota and aging.
      and other deleterious, health-related changes in microbial balance,
      • Lakshminarayanan B.
      • Stanton C.
      • O'Toole P.W.
      • Ross R.P.
      Compositional dynamics of the human intestinal microbiota with aging: Implications for health.
      ,
      • O'Toole P.W.
      • Jeffery I.B.
      Gut microbiota and aging.
      the impact of higher protein intake on gut microbiota in this population requires investigation. In addition, the potential role of probiotics and prebiotics in mitigating the possible protein-induced suppression of certain beneficial genera, such as Bifidobacterium, requires elucidation.
      The primary aim of this study was to determine the effects of an HPD on fecal microbial communities in older women, as women are less likely to meet protein recommendations,
      • ter Borg S.
      • Verlaan S.
      • Mijnarends D.M.
      • Schols J.M.
      • De Groot L.C.
      • Luiking Y.C.
      Macronutrient intake and inadequacies of community-dwelling older adults, a systematic review.
      and thus may benefit from a higher-protein diet. In addition, we aimed to investigate the viability of a multistrain probiotic, with and without a prebiotic, in the context of an HPD, as prebiotics, particularly fructans, have been shown to enhance Bifidobacterium,
      • Vandeputte D.
      • Falony G.
      • Vieira-Silva S.
      • et al.
      Prebiotic inulin-type fructans induce specific changes in the human gut microbiota.
      and probiotics have demonstrated efficacy in restoring microbiota balance after perturbations.
      • Gagliardi A.
      • Totino V.
      • Cacciotti F.
      • et al.
      Rebuilding the gut microbiota ecosystem.
      It was hypothesized that consuming an HPD would produce undesirable changes in gut microbiota, including suppression of Bifidobacterium, whereas concomitant supplementation with a multistrain probiotic formulation and prebiotic may mitigate these effects. The impact of the HPD on the abundance of Akkermansia muciniphila and Faecalibacterium prausnitzii was also of interest, given they are inversely associated with disease.
      • Derrien M.
      • Belzer C.
      • de Vos W.M.
      Akkermansia muciniphila and its role in regulating host functions.
      ,
      • Ferreira-Halder C.V.
      • Faria A.V.S.
      • Andrade S.S.
      Action and function of Faecalibacterium prausnitzii in health and disease.
      Secondary aims were to determine the effects of the HPD on measures of wellness in older women.

      Methods

      Study Design

      An 18-week randomized, double-blind, placebo-controlled, crossover study was conducted in Florida from May 2015 to December 2015. As suppression of Bifidobacterium by dietary means was unexplored, target sample size (n=20) was based on previous trials demonstrating significant microbiome profile changes with dietary interventions,
      • David L.A.
      • Maurice C.F.
      • Carmody R.N.
      • et al.
      Diet rapidly and reproducibly alters the human gut microbiome.
      ,
      • Walker A.W.
      • Ince J.
      • Duncan S.H.
      • et al.
      Dominant and diet-responsive groups of bacteria within the human colonic microbiota.
      with the expectation of a 25% dropout rate. Two-week periods of a controlled HPD were provided and, in random order, a multistrain probiotic, prebiotic, and synbiotic (probiotic and prebiotic) were administered, separated by 2-week washouts during which the participants consumed their usual diet (Figure 1). During baseline, dietary intake was assessed using 3 days of 24-hour recalls by the multipass method.
      • Guenther P.
      • DeMaio T.
      • Ingwersen L.
      • Berlin M.
      The multiple-pass approach for the 24-h recall in the Continuing Survey of Food Intakes by Individuals, 1994-1996.
      Portion-size aids were provided, including the US Department of Agriculture Food Model Booklets. Food intake was analyzed using ESHA Food Processor, version 11.0.117.

      ESHA Food Processor [computer program]. Version 11.0.117. Salem, OR: ESHA Research; 2015.

      To estimate energy expenditure and thus the requirement for weight maintenance, SenseWear Pro 2 armbands (Body Media multisensor devices) were worn over the triceps muscle
      • Andre D.
      • Pelletier R.
      • Farringdon J.
      • et al.
      The development of the SenseWear® armband, a revolutionary energy assessment device to assess physical activity and lifestyle. BodyMedia Inc. 2006.
      for a 7-day period, a method validated previously in community-dwelling older adults by comparison to the doubly labeled water method.
      • Mackey D.C.
      • Manini T.M.
      • Schoeller D.A.
      • et al.
      Validation of an armband to measure daily energy expenditure in older adults.
      During baseline and for the duration of the study, participants recorded their daily fasting weight (kg) using a portable scale (Eat Smart Precision Digital).
      Figure thumbnail gr1
      Figure 1Study design. An 18-week (randomized by Latin square), double-blinded, crossover study. Interventions include A=prebiotic, B=synbiotic, C=placebo, and D=probiotic.
      At randomization, weight (Seca 874 flat scale), height (Seca 217 portable stadiometer), hand-grip strength (JAMAR Plus+ digital dynamometer; Patterson Medical), and body composition by bioelectrical impedance analysis (HYDRA ECF/ICF Bio-Impedance Spectrum Analyzer, Model 4200; XITRON Technologies) were measured. Participants completed the Gastrointestinal Symptom Rating Scale (GSRS),
      • Kulich K.R.
      • Madisch A.
      • Pacini F.
      • et al.
      Reliability and validity of the Gastrointestinal Symptom Rating Scale (GSRS) and Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire in dyspepsia: A six-country study.
      Groningen Frailty Index,
      • Peters L.L.
      • Boter H.
      • Buskens E.
      • Slaets J.P.
      Measurement properties of the Groningen Frailty Indicator in home-dwelling and institutionalized elderly people.
      and SF-36v2 Health Survey.
      • Ware J.E.
      A 36-item short-form health survey (SF-36), I: Conceptual framework and item selection.
      These measures were repeated on the last day of each 2-week period for the duration of the study, with the exception of bioelectrical impedance analysis, which was measured at baseline and 18 weeks only. Fasting venous blood was drawn at baseline and at the end of each HPD period. Participants completed daily questionnaires assessing stool frequency, stool form, general health, and compliance with the diet and supplementation. Single stool and 24-hour urine samples were collected per period for microbiota profile analysis and future exploratory metabolomics, respectively. Three-day, 24-hour recalls were administered during washout periods. The study was approved by the Institutional Review Board (IRB-01) at the University of Florida and was performed according to the guidelines established by the Declaration of Helsinki. All participants provided written informed consent. Participants received a prorated honorarium for participation in the 18-week study.

      Randomization and Blinding

      Participants were randomized using a sealed envelope method (prepared by an individual unaffiliated with the study), following a Latin square design with four intervention sequences. Participants, investigators, study staff, and the statistician were blinded to the sequence until the completion of the statistical analyses.

      Participants

      Healthy women 65 years or older with a usual diet providing <15% of daily energy from protein and <20 g/day of fiber were recruited. Exclusion criteria included current intervention for immune-modulating diseases (eg, human immunodeficiency virus/acquired immunodeficiency syndrome, autoimmune, hepatitis, and cancer); diabetes or chronic kidney disease; vegetarianism; known food allergies or dietary restrictions; physician-diagnosed gastrointestinal disease or condition (such as ulcerative colitis, Crohn’s disease, gastroparesis, peptic ulcer disease, cancer, celiac disease, short bowel, and ileostomy); taking medication for constipation or diarrhea; antibiotic therapy during the preceding 2 months; current smoker; typically consume more than one alcoholic beverage per day; body mass index (calculated as kg/m2) >30; or plans to lose/gain weight during the next 6 months.

      HPD Intervention

      A controlled HPD composed of preportioned, commercially available foods was provided to participants during the four intervention periods. A 4-day rotation, 1,600-kcal base diet (all meals, snacks, and calorie-containing beverages) was provided, plus additional study foods to meet each participant’s energy requirement while maintaining the AMDR for carbohydrate, protein, and fat. The HPD included fresh meat (cut, weighed, and packaged); fresh fruits and vegetables; frozen meals; prepackaged foods, such as salads, protein drinks, breakfast foods; and an assortment of snacks. An example 1-day baseline menu is provided in Table 1 (available at www.jandonline.org). Shelf-stable foods were purchased in bulk before the study start to avoid changes in product formulation, and perishable products were ordered weekly. During the washout periods, participants were asked to return to their usual diet and were responsible for providing their own food. Participants were instructed to consume all food that was given to them during the HPD periods and to record any food that was not consumed in the daily questionnaire.

      Supplements

      The probiotic supplement contained Bifidobacterium bifidum HA-132 (1.54 billion), Bifidobacterium breve HA-129 (4.62 billion), Bifidobacterium longum HA-135 (4.62 billion), Lactobacillus acidophilus HA-122 (4.62 billion), and Lactobacillus plantarum HA-119 (4.62 billion) (Lallemand Health Solutions Inc). Placebo capsules were sensorially identical and consisted of encapsulated potato starch and magnesium stearate. Participants were instructed to consume one capsule daily with a meal.
      The prebiotic, inulin (Sensus FRUTAFIT IQ), and the placebo supplement composed of maltodextrin (PenNovo MD10) were packaged in coded, silver, opaque sachets (5.6 g/sachet). Codes were generated using a random-number generator and assigned by an individual unaffiliated with the study. Participants were instructed to consume one sachet daily by thoroughly mixing its contents into water or another beverage.

      Outcome Measures

      Fecal Microbiota Composition

      Fecal samples were collected during the last 3 days of each period. Coolers and commode collection kits (Fisher Scientific) were provided. Samples were mixed, aliquoted, and stored at –80°C within 6 hours of defecation. For DNA extraction, 5-mL aliquots of stool samples were thawed at 4°C. Genomic DNA was extracted using the QIAamp Fast DNA Stool Mini Kit (Qiagen) as per manufacturer’s protocol, with the following modifications: two washes with 50 mmol/L sodium phosphate buffer before the addition of InhibitEX, and a 0.1-mm zirconia/silica bead beating step (approximately 300 mg/tube, 4 m/s for 1 minute × 3) after incubation with InhibitEx. DNA concentrations were determined using a Nanodrop, and samples were stored at –20°C. Before quantitative polymerase chain reaction (qPCR) analysis, the samples were diluted fivefold in molecular biology–grade water.
      The template DNA for standard curve analysis was obtained by spiking 109 bacteria of the probiotic strain under investigation into approximately 250 mg of a clinical feces matrix, followed by DNA extraction, as described previously. Briefly, 1 g of lyophilized bacteria (B bifidum HA-132, B breve HA-129, B longum HA-135, L acidophilus HA-122. or L plantarum HA-119; Lallemand Health Solutions Inc) was dissolved in 99 mL phosphate-buffered saline (HyClone) and the total bacterial cell count of the suspension was determined by flow cytometry (BD BioSciences Accuri C6 Flow Cytometer) with Fluorescence Nucleic acid stain SYTO 24 labeling (ThermoFisher Scientific) to calculate the volume required to spike with 109 bacteria. Feces without the addition of bacterial suspension were used as negative control.
      Absolute quantification of B bifidum HA-132, B breve HA-129, B longum HA-135, L acidophilus HA-122, and L plantarum HA-119 in fecal samples was performed by qPCR using the ViiA 7 Real-Time PCR System (Thermo Fisher Scientific) and CFX384 Touch Real-Time PCR Detection System (Bio-Rad Laboratories). Standard curves for each strain were generated from feces spiked with 109 B bidifum HA-132, B breve HA-129, B longum HA-135, L acidophilus HA-122, and L plantarum HA-119, respectively. DNA recovered from spiked feces was serially diluted (10-fold) to generate templates for the standard curve ranging from 109 to 103 bacteria for L acidophilus HA-122 and L plantarum HA-119, and 109 to 104 for B bifidum HA-132, B breve HA-129, and B longum HA-135. The DNA samples to be quantified were diluted 5-fold in molecular biologygrade water before qPCR.
      The qPCR reaction mixture (25 μL) consisted of 300 nM of the appropriate primers, 1X SYBR Select Master Mix (Thermo Fisher Scientific, cat# 4472908) and 2.5 μL diluted DNA when assays were run on ViiA Real-Time PCR System in a 96-well format. Volumes were downscaled when assays were run on the CFX384 Touch Real-Time PCR Detection System to a final reaction volume of 10 μL, including 1 μL diluted DNA. Primer sequences are given in Table 2 (available at www.jandonline.org). Standard curve samples were analyzed in duplicate and unknown samples in triplicate. Cycling conditions consisted of initial incubations at 50°C and 95°C for 2 minutes each, followed by 40 cycles of 95°C for 15 seconds, 60°C for 30 seconds, and 72°C for 30 seconds. A dissociation curve analysis (60°C to 95°C) was also performed to ensure primer specificity.
      Relative quantification was carried out for A muciniphila, F prausnitzii, and total Bifidobacterium spp. DNA was normalized using 16S ribosomal DNA Universal Bacterial primers
      • Hartman A.L.
      • Lough D.M.
      • Barupal D.K.
      • et al.
      Human gut microbiome adopts an alternative state following small bowel transplantation.
      and the 2–ΔΔCT method to determine the relative fold gene expression when comparing baseline to each intervention period. Primer sequences and assay conditions for A muciniphila and F prausnitzii were obtained from Collado and colleagues
      • Collado M.C.
      • Derrien M.
      • Isolauri E.
      • de Vos W.M.
      • Salminen S.
      Intestinal integrity and Akkermansia muciniphila, a mucin-degrading member of the intestinal microbiota present in infants, adults, and the elderly.
      and Lopez-Siles and colleagues,
      • Lopez-Siles M.
      • Duncan S.H.
      • Garcia-Gil L.J.
      • Martinez-Medina M.
      Faecalibacterium prausnitzii: From microbiology to diagnostics and prognostics.
      respectively. The Bifidobacterium spp primers were designed in-house (forward: TGG AAG GTC TCG ATG GAG GT and reverse: CTG GAC AAG CCG TTC CTG AT) and use the same assay and cycling conditions as listed for the primers used in strain quantification. A dissociation curve analysis (60°C to 95°C) was also performed to ensure primer specificity for all assays. All qPCR reactions were prepared using the epMotion 5075tc liquid handling robot and Select SYBR Mastermix (Thermo Fisher Scientific) and analyzed on the CFX384 Touch Real-Time PCR Detection System (Bio-Rad Laboratories).
      DNA was amplified using gene-specific primers for the V4 hypervariable region of the 16S ribosomal RNA gene and tagged with unique identifiers. As described in MacPherson and colleagues,
      • MacPherson C.W.
      • Mathieu O.
      • Tremblay J.
      • et al.
      Gut bacterial microbiota and its resistome rapidly recover to basal state levels after short-term amoxicillin-clavulanic acid treatment in healthy adults.
      Illumina’s “16S Metagenomic Sequencing Library Preparation” guide (part #15044223 Rev. B)
      16S Metagenomic Sequencing Library Preparation: Preparing 16S Ribosomal RNA Gene Amplicons for the Illumina MiSeq System.
      was used to prepare the bacterial 16S ribosomal RNA gene libraries. The Qiagen HotStarTaq MasterMix was used for the first PCR (amplicon PCR) at 25 cycles with annealing temperatures of 55°C. For the second PCR (index PCR) only 50% of the reagent volumes were used. The template-specific primers, 515f (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′), flanked with appropriate overhang adapter sequences, were used. Before loading on an Illumina MiSeq, samples were diluted and pooled, and sequenced using a 500-cycle MiSeq Reagent Kit v3.
      The demultiplexed amplicons were processed with the quality filter software encapsulated in QIIME2 (Quantitative Insight Into Microbial Ecology–2).
      • Bolyen E.
      • Rideout J.R.
      • Dillon M.R.
      • et al.
      QIIME 2: Reproducible, interactive, scalable, and extensible microbiome data science.
      After visual inspection of the imported dataset and assessment of the high-quality reads throughout the length of the amplicon, the reads were trimmed at 240 bp and quality controlled with “quality-filter q-score” using the default parameters. The quality-controlled reads were clustered into amplicon sequence variants (ASVs), functionally equivalent to the more widely used operational taxonomic units, with the denoising algorithm Deblur, implemented in the QIIME2 suite. The resulting ASVs (and their relative abundances) were used to generate basic metrics (α-diversity per treatment),
      • Bolyen E.
      • Rideout J.R.
      • Dillon M.R.
      • et al.
      QIIME 2: Reproducible, interactive, scalable, and extensible microbiome data science.
      • Kruskal W.H.
      • Wallis W.
      Use of ranks in one-criterion variance analysis.

      McKinney W. Data structures for statistical computing in Python. In: van der Walt S, Millman J, eds. Proceedings of the 9th Python in Science Conference, Austin, TX, June 28-July 3, 2010:51-56.

      • McDonald D.
      • Clemente J.C.
      • Kuczynski J.
      • et al.
      The Biological Observation Matrix (BIOM) format or: How I learned to stop worrying and love the ome-ome.
      • Weiss S.
      • Xu Z.Z.
      • Peddada S.
      • et al.
      Normalization and microbial differential abundance strategies depend upon data characteristics.
      • Amir A.
      • McDonald D.
      • Navas-Molina J.A.
      • et al.
      Deblur rapidly resolves single-nucleotide community sequence patterns.
      • Bokulich N.A.
      • Subramanian S.
      • Faith J.J.
      • et al.
      Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing.
      • Price M.N.
      • Dehal P.S.
      • Arkin A.P.
      FastTree 2—Approximately maximum-likelihood trees for large alignments.
      • Lane D.J.
      16S/23S rRNA sequencing.
      • Katoh K.
      • Standley D.M.
      MAFFT multiple sequence alignment software version 7: Improvements in performance and usability.
      generate the MAFFT (encapsulated in QIIME2 framework) phylogeny and principal component analysis (with popular distance algorithms, like Bray-Curtis, Jaccard, Weighted UniFrac).
      • Bolyen E.
      • Rideout J.R.
      • Dillon M.R.
      • et al.
      QIIME 2: Reproducible, interactive, scalable, and extensible microbiome data science.
      ,
      • McDonald D.
      • Clemente J.C.
      • Kuczynski J.
      • et al.
      The Biological Observation Matrix (BIOM) format or: How I learned to stop worrying and love the ome-ome.
      • Weiss S.
      • Xu Z.Z.
      • Peddada S.
      • et al.
      Normalization and microbial differential abundance strategies depend upon data characteristics.
      • Amir A.
      • McDonald D.
      • Navas-Molina J.A.
      • et al.
      Deblur rapidly resolves single-nucleotide community sequence patterns.
      • Bokulich N.A.
      • Subramanian S.
      • Faith J.J.
      • et al.
      Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing.
      • Price M.N.
      • Dehal P.S.
      • Arkin A.P.
      FastTree 2—Approximately maximum-likelihood trees for large alignments.
      • Lane D.J.
      16S/23S rRNA sequencing.
      • Katoh K.
      • Standley D.M.
      MAFFT multiple sequence alignment software version 7: Improvements in performance and usability.
      • Lozupone C.
      • Knight R.
      UniFrac: A new phylogenetic method for comparing microbial communities.
      • Lozupone C.A.
      • Hamady M.
      • Kelley S.T.
      • Knight R.
      Quantitative and qualitative β diversity measures lead to different insights into factors that structure microbial communities.
      • Chang Q.
      • Luan Y.
      • Sun F.
      Variance adjusted weighted UniFrac: A powerful beta diversity measure for comparing communities based on phylogeny.
      • Chen J.
      • Bittinger K.
      • Charlson E.S.
      • et al.
      Associating microbiome composition with environmental covariates using generalized UniFrac distances.
      • McDonald D.
      • Vázquez-Baeza Y.
      • Koslicki D.
      • et al.
      Striped UniFrac: Enabling microbiome analysis at unprecedented scale.
      • Vázquez-Baeza Y.
      • Pirrung M.
      • Gonzalez A.
      • Knight R.
      EMPeror: A tool for visualizing high-throughput microbial community data.
      • Vázquez-Baeza Y.
      • Gonzalez A.
      • Smarr L.
      • et al.
      Bringing the dynamic microbiome to life with animations.
      The ASVs were then attributed to the closest known taxa with QIIME2’s “feature classifier,” classifying using the sk-learn machine learning classification module.
      The taxonomically attributed tables (tables containing known taxa and their relative abundance in each sample) were exported from the QIIME2 framework for further analysis (table collapsed at the genus level, L=6). The relative abundances profiles were compared between treatment groups using linear discriminant analysis effect size (LEfSe). This package is designed for the discovery of biomarkers from relative abundance microbiome tables.
      • Segata N.
      • Izard J.
      • Waldron L.
      • et al.
      Metagenomic biomarker discovery and explanation.
      For ease of use (because of its high number of software dependencies), the formatted tables have been uploaded to the Huttenhower galaxy online platform (http://huttenhower.sph.harvard.edu/galaxy/) and LEfSe was run from this implementation.

      Wellness

      Participants were asked to record in a daily questionnaire, their fasting weight, hours of sleep, level of hunger, supplement and diet compliance, antibiotic use, doctor visits, changes in physical activity, medication usage, stool frequency, and stool form using the Bristol Stool Form Scale (1 to 2=slow transit, 3 to 5=normal transit, 6 to 7=fast transit).
      • Lewis S.J.
      • Heaton K.W.
      Stool form scale as a useful guide to intestinal transit time.
      Daily symptoms were monitored using a 7-point Likert-like scale (0=none, 3=moderate, 6=very severe) and included constipation, diarrhea, and fatigue. Syndromes included gastrointestinal distress (bloating, flatulence, cramping, and abdominal noises); psychological (feeling anxious, depressed, or stressed); cephalic (headache and dizziness); and emetic (nausea and vomiting) using the same 7-point scale.
      • Hanifi A.
      • Culpepper T.
      • Mai V.
      • et al.
      Evaluation of Bacillus subtilis R0179 on gastrointestinal viability and general wellness: A randomised, double-blind, placebo-controlled trial in healthy adults.
      Participants completed the GSRS,
      • Kulich K.R.
      • Madisch A.
      • Pacini F.
      • et al.
      Reliability and validity of the Gastrointestinal Symptom Rating Scale (GSRS) and Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire in dyspepsia: A six-country study.
      SF-36v2 Health Survey,
      • Ware J.E.
      A 36-item short-form health survey (SF-36), I: Conceptual framework and item selection.
      and Groningen Frailty Indicator
      • Peters L.L.
      • Boter H.
      • Buskens E.
      • Slaets J.P.
      Measurement properties of the Groningen Frailty Indicator in home-dwelling and institutionalized elderly people.
      every 2 weeks at study visits. Scores of the SF-36v2 Health Survey (originally reported on a 0 to 100 scale) were normalized to the 2009 US population using a linear transformation with a mean±standard deviation (SD) score=50±10 (QualityMetric Health Outcomes Scoring software, version 4.5).
      The Groningen Frailty Indicator was scored using guidelines outlined by Peters and colleagues.
      • Peters L.L.
      • Boter H.
      • Buskens E.
      • Slaets J.P.
      Measurement properties of the Groningen Frailty Indicator in home-dwelling and institutionalized elderly people.
      Hand-grip strength was assessed during each period using the Jamar Plus+ digital dynamometer (Patterson Medical) as described by Flood and colleagues.
      • Flood A.
      • Chung A.
      • Parker H.
      • Kearns V.
      • O'Sullivan T.A.
      The use of hand grip strength as a predictor of nutrition status in hospital patients.
      Fasting blood was collected at baseline and during each HPD period. Lipid panel, comprehensive metabolic profile, and C-reactive protein were analyzed by accepted clinical laboratory methods. Insulin-like growth factor-1 was measured in serum samples using Human IGF-1 Milliplex Map Kit (cat# HIGFMAG-52K; Millipore Corp) according to manufacturer’s directions. The beads were analyzed on a Luminex 200 instrument (Luminex Corp) with xPONENT software, version 3.1.

      Compliance

      Two-week supplies of the sachets and capsules were provided during each intervention period. Participants were asked to return any unconsumed supplements at the end of each period and to record their intake compliance of the probiotic, prebiotic, and diet in their daily questionnaire.

      Statistical Analyses of Wellness Outcomes

      All wellness data were analyzed on an intent-to-treat basis (n=26). Unless noted otherwise, data represent the least squares mean±standard error of mean and significance at a P value <0.05. For body weight, equivalence testing was performed using a tolerance level set at 1 kg. Bioelectrical impedance analysis data were analyzed using a paired Student’s t test. Sleep was compared using a quasi-binomial distribution of daily category modeling using mean daily proportion of people with <7 hours of sleep. Nonparametric analysis, estimates, and least squares mean difference Tukey’s honestly significant difference were used to analyze symptoms and syndromes recorded in the daily questionnaires. Biweekly means of stool frequency and Bristol Stool Form Scale (slow transit, types 1 and 2; normal transit, types 3 to 5; and fast transit, types 6 and 7) and GSRS were log-transformed and analyzed using generalized linear mixed models with a random effect of subject and a fixed effect of intervention. As no sequence, week, or washout/carryover effects were seen, all washouts were combined. For log constipation, an effect of sequence and its interaction with intervention was added. A pairwise comparison of the sequences was then conducted. For Groningen Frailty Indicator, a binomial model with a random effect of subject was used, and for the SF-36v2, the scored norm-based data were analyzed with a generalized linear mixed model fitted with intervention as a fixed effect and a random-subject effect. Serum markers were analyzed using a general linear mixed model with a random subject effect and a fixed effect of treatment or intervention and log transformation where appropriate. Kruskal-Wallis one-way analysis of variance was used to compare qPCR data and Shannon α-diversity index with α set a .05.
      • Kruskal W.H.
      • Wallis W.
      Use of ranks in one-criterion variance analysis.

      Results

      The study flow diagram is presented in Figure 2. Participant characteristics at baseline (n=26; mean±SD=age 73.7±5.6 years) are shown in Table 3. Twenty-one participants completed the entire study. No study-related adverse events were reported during the study. Estimated energy requirement for participants at baseline was 1,839±39 kcal/day (mean±SD). HPD provided was 1,792±36 kcal/day (mean±SD), individualized within ±50 kcal/day of each participant’s estimated energy requirement. The base HPD provided 30% energy from protein, 46% from carbohydrate, and 27% from fat. With adjustments for energy (primarily carbohydrate), protein provision was 1.8 g/kg/day (range=1.5 to 2.2 g/kg/day) and 27% of energy. Fiber was provided at 15.5±0.4 g/day (mean±SD). Energy intake assessed by 24-hour recall during baseline was 1,455±506 kcal/day (mean±SD) and washouts were 1,667±615 kcal/day (mean±SD). Compliance of participants was 99.5% for the stool protocol, 96.6% for probiotic consumption based on capsule counts, 95.0% for the prebiotic based on packet use, 99.8% for completion of the daily questionnaires, and most study foods were consumed.
      Figure thumbnail gr2
      Figure 2Flow diagram of participant recruitment, allocation, and analysis. Interventions include A=prebiotic, B=synbiotic, C=placebo, and D=probiotic.
      Table 3Baseline characteristics of older women participating in an evaluation of a high-protein diet, with and without prebiotic and probiotic supplements, on microbiota composition and general wellness (n=26)
      CharacteristicsData
      Sex, male/female, n0/26
      Age, y, mean±SD
      SD=standard deviation.
      73.7±5.6
      Race/ethnicity, n (%)
      White23 (88.5)
      Black2 (7.7)
      American Indian or Alaska Native1 (3.8)
      Not Hispanic or Latino26 (100)
      Body mass index, mean±SD (range)25.7±3.1 (20.0 to 29.9)
      Normal weight, n (%)12 (46.2)
      Overweight, n (%)14 (53.8)
      Groningen Frailty Indicator score,
      Score >4 indicates frailty.
      mean±SD
      1.6±1.7
      SF-36v2, mean±SD
      Mental Component Summary57.8±10.0
      Physical Component Summary55.6±5.8
      Hand-grip strength, kg, mean±SD20.3±1.0
      Maximum strength of dominant hand.
      Laboratory values, mean±SD
      Total cholesterol, mg/dL (mmol/L)197.9±7.4 (51.3±1.9)
      Triglycerides, mg/dL (mmol/L)92.3±6.9 (1.04±0.08)
      High-density lipoprotein cholesterol, mg/dL (mmol/L)68.9±3.3 (17.8±0.9)
      Low-density lipoprotein cholesterol, mg/dL (mmol/L)110.6±6.6 (28.6±1.7)
      Very-low-density lipoprotein cholesterol, mg/dL (mmol/L)18.5±1.4 (4.8±0.4)
      Fasting glucose, mg/dL (mmol/L)92.1±1.8 (5.1±0.1)
      Blood urea nitrogen, mg/dL (mmol/L)16.7±1.2 (6.0±0.4)
      Albumin, g/dL (g/L)4.3±0.1 (43.1±1)
      Total protein, g/dL (g/L)7.0±0.1 (70±1)
      Creatinine, mg/dL (μmol/L)0.73±0.02 (65±2)
      C-reactive protein, mg/dL (mg/L)0.50±0.13 (5.0±1.3)
      Values reported as <0.3 mg/dL were assumed to be 0.3 mg/dL.
      Alanine transaminase, U/L16.6±1.3
      Alanine aminotransferase, U/L21.7±1.0
      Alkaline phosphatase, U/L72.1±3.5
      a SD=standard deviation.
      b Score >4 indicates frailty.
      c Maximum strength of dominant hand.
      d Values reported as <0.3 mg/dL were assumed to be 0.3 mg/dL.

      Strain Recovery and Microbiota Composition

      Absolute quantification by qPCR of five probiotic strains from stools, from participants with complete collections, confirmed strain recovery in the context of the HPD (n=20) (Figure 3). Probiotic strains were recovered in fecal samples after probiotic ingestion during the probiotic and synbiotic periods, with minimal strain detection during the prebiotic and placebo periods. All five strains were recovered during probiotic and synbiotic periods for 18 of 20 participants. One participant showed no detection of B breve HA-129 after both probiotic and synbiotic interventions and another showed no detection of B breve HA-129, B longum HA-135, and L acidophilus HA-122 during the synbiotic intervention. As washouts before the probiotic interventions did not detect any of the strains, detection of the remaining four of five and two of five strains, respectively, confirmed supplement compliance in the 20 participants. No significant differences were seen between interventions for the relative quantification by real-time PCR of A muciniphila, F prausnitzii, and Bifidobacterium spp (Figure 4). Shannon α-diversity index was not statistically different among HPD periods compared to baseline and washouts (Figure 5; available at www.jandonline.org) or when comparing pooled HPD periods to baseline and washout (Figure 6; available at www.jandonline.org). Using principal coordinate analysis based on the UniFrac metric, no clustering by treatment was detected (Figure 7; available at www.jandonline.org). Figure 8 (available at www.jandonline.org) shows the UniFrac analysis for the individual participants and the four intervention orders of the Latin design. Although there were some differentiated ASVs when separated out by intervention, the differences were small and not meaningful, given the small sample sizes of the subgroups.
      Figure thumbnail gr3
      Figure 3Absolute quantification by real-time polymerase chain reaction of the five probiotic strains (Bifidobacterium bifidum HA-132, Bifidobacterium breve HA-129, Bifidobacterium longum HA-135, Lactobacillus acidophilus HA-122, and Lactobacillus plantarum HA-119) (n=20). Assay’s limit of detection is indicated by the dotted line. Solid black lines indicate the median absolute quantification value for each strain during each intervention.
      Figure thumbnail gr4
      Figure 4Relative quantification by real-time polymerase chain reaction of Akkermansia muciniphila, Faecalibacterium prausnitzii, and Bifidobacterium spp (n=20). Mean LOG2 fold-change from baseline are represented by solid black lines (P<0.05).
      Figure 9 depicts the microbiome profile (relative abundance of bacteria observed) at three different taxonomic levels, at baseline, and during each HPD and washout period for each participant. The same profiles are presented at three different taxonomic levels, with detail level increasing from left to right. Using LEfSe to compare the relative bacterial proportions between the HPD vs washouts demonstrated enrichment of Roseburia and Anaerostipes during washouts and enrichment of Lactobacillus, Lactococcus, and Streptococcus in the HPD periods (pooled) (Figure 10A). Relative abundance by period of Roseburia and Anaerostipes are shown in Figure 11 (available at www.jandonline.org) and Lactococcus and Streptococcus in Figure 12 (available at www.jandonline.org). Lactobacillus was also higher during baseline vs washouts (Figure 13; available at www.jandonline.org). There were no significant differences between placebo vs probiotic or placebo vs prebiotic interventions. However, for the placebo vs synbiotic comparison, the placebo group was higher in Lactococcus or inversely, these taxa were suppressed in the synbiotic intervention (Figure 10B). Washouts were compared to baseline to determine whether the intermittent provision of the HPD produced any carryover effect with time. Pooled washouts vs baseline exhibited enhanced Actinomyces at baseline (Figure 14A; available at www.jandonline.org). There was no evidence of a difference between baseline and washout 1. However, washout 2 showed minor changes in unknown taxa (Figure 14B; available at www.jandonline.org), washout 3 showed increased Bacteroidetes and Actinobacteria at all taxonomic levels (and is most similar to baseline, probably responsible for most of the difference in the “pooled washouts vs baseline” comparison) (Figure 14C; available at www.jandonline.org), and washout 4 was enriched in Fusobacteria compared with baseline (Figure 14D; available at www.jandonline.org).
      Figure thumbnail gr5
      Figure 9Fecal microbiota profile. Phyla, family, and genus levels for the four intervention periods (placebo, probiotic, prebiotic, and synbiotic) and washouts. The same profiles are presented at three different taxonomic levels, with detail level increasing from left to right.
      Figure thumbnail gr6
      Figure 10Linear discriminant analysis (LDA) and cladogram analyses of fecal bacteria during high-protein diet (HPD) interventions. (A) LDA score represents log changes in relative bacterial representation between pooled HPD interventions and washout periods. The cladogram shows the phylogenetic relationship among the bacterial genus, families, and orders that were different between the pooled HPD diet interventions and washout periods. (B) LDA score represents log changes in relative bacterial representation between the placebo and synbiotic HPD periods. The cladogram shows the phylogenetic relationship between the bacterial genus, families, and orders that were different between the placebo and synbiotic HPD periods (threshold=2.0, α=.05).

      Wellness

      Participants’ metabolic panel and liver function tests were normal at baseline (Table 3) and remained unchanged throughout the study, with the exception of blood urea nitrogen. Mean±SD blood urea nitrogen increased during all HPD periods as follows: prebiotic 22.0±1.3 mg/dL (7.9±0.5 mmol/L); synbiotic 23.8±1.3 mg/dL (8.5±0.5 mmol/L); placebo 22.6±1.3 mg/dL (8.1±0.5 mmol/L); and probiotic 21.9±1.3 mg/dL (7.8±0.5 mmol/L), compared to the baseline mean±SD of 16.7±1.2 mg/dL (6.0±0.4 mmol/L) (P<0.001). A significant increase of 0.21 ng/mL in serum insulin-like growth factor-1 (baseline vs all HPD periods) was seen, with no differences among HPD periods (mean±SD=4.24±0.08, 4.25±0.08, 4.25±0.08, 4.25±0.08 ng/mL; not significant). Assessment of body composition showed an increase of fat-free mass from mean±SD of 41.1±1.1 kg at baseline to 43.2±1.2 kg at study end (P=0.03). In addition, a trend (defined as P<0.10) was observed for decreased fat mass (P=0.05) and percent fat (P=0.05). Intracellular (P=0.03) and total body water (P=0.03) were also significantly increased compared to baseline. SF-36v2 scores remained unchanged with interventions, with four participants exhibiting frailty (Figure 15; available at www.jandonline.org).
      As the dataset of the daily questionnaire contained a large proportion of “no reported symptoms” for diarrhea, constipation, and for the psychological, cephalic, and emetic syndromes scores, statistical analyses were precluded. No differences were seen for fatigue or sleep; however, level of hunger was lower during HPD and washout periods compared to baseline (P<0.001) (Table 4). Daily symptom reporting indicated that gastrointestinal distress was very low, but higher for the prebiotic and synbiotic periods compared to baseline (P<0.001) (Table 4). Mean±SD stool frequency at baseline (1.38±0.10 stools/day) and Bristol Stool Form Scale (3.84±0.21; normal transit) were unchanged with interventions.
      Table 4Participant wellness outcomes reported by the study participants during baseline, each of the four 2-week high-protein diet intervention periods, and washout periods
      VariableFatigue
      Severity of symptoms from 0 (none) to 6 (very severe).
      Fatigue=mean of nonzero reports; individuals with zero reporting removed (n=18) (not significant).
      Hunger
      Severity of symptoms from 0 (none) to 6 (very severe).
      Hunger=mean of nonzero reports (n=26) (P<0.001).
      Gastrointestinal distress
      Gastrointestinal distress=mean of bloating, flatulence, abdominal cramping, and abdominal noises scored on a scale of 0 (no symptoms present) to 6 (very severe) (n=26) (P<0.001).
      Sleep
      Sleep categories: <5 h ,5 to 6 h, 6 to 7 h, 7 to 8 h, 8 to 9 h, >9 h. A quasi-binomial distribution of daily category (<7 h or ≥7 h) modeling mean daily proportion of participants with sleep <7 h (n=26) (not significant).
      Stool Frequency
      Stool frequency as mean number of stools per day (n=26) (not significant).
      Bristol Stool Scale Score
      Bristol Stool Form Scale responses grouped as slow transit (types 1 and 2), normal transit (types 3 to 5), and fast transit (types 6 and 7) (n=26) (not significant).
      least squares mean±standard error of mean
      Baseline1.39±0.201.76±0.220.31±0.070.40±0.171.38±0.103.84±0.21
      High-protein diet
      Prebiotic+placebo1.40±0.201.20±0.22
      P<0.05.
      0.58±0.08
      P<0.05.
      0.32±0.161.38±0.103.71±0.21
      Prebiotic+probiotic1.73±0.211.23±0.23
      P<0.05.
      0.62±0.08
      P<0.05.
      0.36±0.161.43±0.103.79±0.22
      Control+placebo1.56±0.201.20±0.22
      P<0.05.
      0.36±0.080.33±0.161.39±0.103.78±0.21
      Control+probiotic1.34±0.211.01±0.22
      P<0.05.
      0.39±0.080.34±0.161.39±0.103.58±0.21
      Washout period
      11.60±0.211.31±0.22
      P<0.05.
      0.31±0.080.40±0.171.42±0.103.75±0.21
      21.54±0.211.32±0.22
      P<0.05.
      0.32±0.080.32±0.161.38±0.103.74±0.21
      31.41±0.221.10±0.23
      P<0.05.
      0.29±0.080.31±0.151.36±0.113.77±0.22
      a Severity of symptoms from 0 (none) to 6 (very severe).
      b Fatigue=mean of nonzero reports; individuals with zero reporting removed (n=18) (not significant).
      c Hunger=mean of nonzero reports (n=26) (P<0.001).
      d Gastrointestinal distress=mean of bloating, flatulence, abdominal cramping, and abdominal noises scored on a scale of 0 (no symptoms present) to 6 (very severe) (n=26) (P<0.001).
      e Sleep categories: <5 h ,5 to 6 h, 6 to 7 h, 7 to 8 h, 8 to 9 h, >9 h. A quasi-binomial distribution of daily category (<7 h or ≥7 h) modeling mean daily proportion of participants with sleep <7 h (n=26) (not significant).
      f Stool frequency as mean number of stools per day (n=26) (not significant).
      g Bristol Stool Form Scale responses grouped as slow transit (types 1 and 2), normal transit (types 3 to 5), and fast transit (types 6 and 7) (n=26) (not significant).
      P<0.05.
      For the GSRS, mean syndrome scores were all <2 (slight discomfort). Abdominal pain was higher with the prebiotic (P<0.05), synbiotic (P<0.01), and placebo (P<0.01), but not the probiotic when compared to baseline plus washout. However, when compared to baseline alone, there were no differences in abdominal pain. Indigestion syndrome was higher when participants received the prebiotic (P<0.001), synbiotic (P<0.001), and probiotic (P<0.05), but not the placebo compared to baseline plus washout. However, when compared to baseline only, indigestion syndrome was significantly higher only for the prebiotic (P<0.01) and synbiotic (P<0.001). Results of the GSRS questionnaire are shown in Table 5.
      Table 5Gastrointestinal Symptom Rating Scale syndrome
      Syndromes are means of individual symptom ratings of 1 (no discomfort at all) to 7 (very severe discomfort).
      reported weekly by the study participants during baseline and each of the 2-week high-protein diet intervention periods
      PeriodAbdominal pain
      Abdominal pain scale (abdominal pain, hunger pains and nausea symptoms) (P=0.01).
      Constipation
      Constipation scale (constipation, hard stools, and feeling of incomplete evacuation symptoms) (P=0.12).
      Diarrhea
      Diarrhea (diarrhea, loose stools, and urgent need for defecation symptoms) (P=0.11).
      Indigestion
      Indigestion (stomach rumbling, bloating, burping, and increased flatus symptoms) (P<0.001).
      Reflux
      Reflux scale (heartburn and acid regurgitation symptoms) (P=0.15).
      least squares mean±standard error of mean
      Baseline1.17±0.041.38±0.071.16±0.031.37±0.051.19±0.06
      High-protein diet
      Prebiotic+placebo1.24±0.061.52±0.151.24±0.111.63±0.10
      P<0.01.
      1.18±0.06
      Prebiotic+probiotic1.33±0.111.54±0.151.19±0.061.78±0.14
      P<0.001.
      1.27±0.11
      Control+placebo1.29±0.061.64±0.161.07±0.031.47±0.081.46±0.16
      Control+probiotic1.15±0.041.78±0.241.09±0.041.45±0.061.24±0.07
      a Syndromes are means of individual symptom ratings of 1 (no discomfort at all) to 7 (very severe discomfort).
      b Abdominal pain scale (abdominal pain, hunger pains and nausea symptoms) (P=0.01).
      c Constipation scale (constipation, hard stools, and feeling of incomplete evacuation symptoms) (P=0.12).
      d Diarrhea (diarrhea, loose stools, and urgent need for defecation symptoms) (P=0.11).
      e Indigestion (stomach rumbling, bloating, burping, and increased flatus symptoms) (P<0.001).
      f Reflux scale (heartburn and acid regurgitation symptoms) (P=0.15).
      ∗∗ P<0.01.
      ∗∗∗ P<0.001.

      Discussion

      The primary aim of this study was to determine whether an HPD exerted undesirable changes on fecal microbial communities, focusing on diversity and suppression of abundance of health-enhancing taxa Bifidobacterium spp, A muciniphila, and F prausnitzii. There was no evidence that diversity was affected by study interventions. These findings are in agreement with Beaumont and colleagues,
      • Beaumont M.
      • Portune K.J.
      • Steuer N.
      • et al.
      Quantity and source of dietary protein influence metabolite production by gut microbiota and rectal mucosa gene expression: A randomized, parallel, double-blind trial in overweight humans.
      who found no significant changes in diversity with a higher-protein diet in overweight adults. In the present study, there was no evidence that quantitative Bifidobacterium abundance was affected by HPD interventions. Although an age-related decline in Bifidobacterium has been reported in frail older adults,
      • Hébuterne X.
      Gut changes attributed to ageing: Effects on intestinal microflora.
      only four participants were assessed as frail based on the Groningen Frailty Indicator. Bifidobacterium levels observed were similar to what has been reported in an adult population.
      • Odamaki T.
      • Kato K.
      • Sugahara H.
      • et al.
      Age-related changes in gut microbiota composition from newborn to centenarian: A cross-sectional study.
      In contrast, Zhu and colleagues
      • Zhu Q.
      • Huang L.
      • Zhu J.
      • Lu J.
      • Qin H.
      Analysis of gut microbiota in long-lived older adults and their relatives: A gradual change with ageing.
      found a low level of abundance for Bifidobacterium in 375 older adults residing in China. Similarly, there was no evidence that quantitative abundance of A muciniphila or F prausnitzii, a noted butyrate producer,
      • Lopez-Siles M.
      • Duncan S.H.
      • Garcia-Gil L.J.
      • Martinez-Medina M.
      Faecalibacterium prausnitzii: From microbiology to diagnostics and prognostics.
      was affected by the HPD interventions. Abundance of F prausnitzii is reported to be inversely correlated with age
      • Zhu Q.
      • Huang L.
      • Zhu J.
      • Lu J.
      • Qin H.
      Analysis of gut microbiota in long-lived older adults and their relatives: A gradual change with ageing.
      ; however, the sample size of the present study precludes such analysis.
      An HPD consisting solely of animal-sourced foods was shown to suppress the bacterial genera Roseburia, E rectale, and R bromii.
      • David L.A.
      • Maurice C.F.
      • Carmody R.N.
      • et al.
      Diet rapidly and reproducibly alters the human gut microbiome.
      In the present study, the relative representation of Roseburia and Anaerostipes were suppressed with the HPD, but no changes were seen in Ruminococcus and Eubacterium. The provision of fiber from plant-sourced foods may have maintained Ruminococcus and Eubacterium, as these genera respond to the consumption of vegetarian diets.
      • Tomova A.
      • Bukovsky I.
      • Rembert E.
      • et al.
      The effects of vegetarian and vegan diets on gut microbiota.
      The butyrate producers Roseburia and Anaerostipes are considered beneficial constituents of the microbiota, as butyrate is essential for the maintenance of intestinal barrier function, among other critical roles as a mediator of inflammation.
      • Brahe L.K.
      • Astrup A.
      • Larsen L.H.
      Is butyrate the link between diet, intestinal microbiota and obesity-related metabolic diseases?.
      Suppression of these taxa may have negatively affected colonic butyrate production; however, this was not measured. Inulin provided in the prebiotic and synbiotic periods did not help to sustain the relative abundance of Anaerostipes, which has been shown to utilize this substrate.
      • Gagliardi A.
      • Totino V.
      • Cacciotti F.
      • et al.
      Rebuilding the gut microbiota ecosystem.
      Of note, higher abundance of Roseburia is associated with whole-grain intake
      • Martinez I.
      • Lattimer J.M.
      • Hubach K.L.
      • et al.
      Gut microbiome composition is linked to whole grain-induced immunological improvements.
      and higher diet quality,
      • Liu Y.
      • Ajami N.J.
      • El-Serag H.B.
      • et al.
      Dietary quality and the colonic mucosa-associated gut microbiome in humans.
      whereas lower abundance has been associated with inflammation in kidney disease
      • Jiang S.
      • Xie S.
      • Lv D.
      • et al.
      A reduction in the butyrate producing species Roseburia spp. and Faecalibacterium prausnitzii is associated with chronic kidney disease progression.
      and inflammatory bowel disease,
      • Machiels K.
      • Joossens M.
      • Sabino J.
      • et al.
      A decrease of the butyrate-producing species Roseburia hominis and Faecalibacterium prausnitzii defines dysbiosis in patients with ulcerative colitis.
      further evidence that suppression may be unfavorable.
      As demonstrated by the linear discriminant analysis–LEfSe analysis, increases in relative abundance of Lactobacillus, Lactococcus, and Streptococcus were seen with the HPD. This is in contrast to the study by David and colleagues,
      • David L.A.
      • Maurice C.F.
      • Carmody R.N.
      • et al.
      Diet rapidly and reproducibly alters the human gut microbiome.
      which demonstrated increased abundances of Alistipes, Bilophila, and Bacteroides with an animal-sourced diet providing similar energy from protein. As these genera are considered resistant to the bactericidal effects of bile acids, their increase was logically due to the very-high-fat provision in the experimental animal-based diet provided vs the higher protein. In the present study, the higher relative abundance of Lactobacillus in the HPD vs washout period may been due to the provision of yogurt containing Lactobacillus delbrueckii subsp bulgaricus during the controlled HPD; however, levels were only higher in the placebo and prebiotic HPD periods. Not surprisingly, the linear discriminant analysis–LEfSe analysis did not demonstrate enrichment in Bifidobacterium or Lactobacillus with the provision of probiotic strains. This may be due to the stability of the microbiome in the absence of dysbiosis, as has been reported previously,
      • Kristensen N.B.
      • Bryrup T.
      • Allin K.H.
      • Nielsen T.
      • Hansen T.H.
      • Pedersen O.
      Alterations in fecal microbiota composition by probiotic supplementation in healthy adults: A systematic review of randomized controlled trials.
      ,
      • Sommer F.
      • Anderson J.M.
      • Bharti R.
      • Raes J.
      • Rosenstiel P.
      The resilience of the intestinal microbiota influences health and disease.
      the dose amount in relation to the microbiome, and the possibility of displacement. The increase in relative abundance of Streptococcus may be unfavorable. Higher relative abundance of Streptococcus has been associated with major depressive disorder
      • Cheung S.G.
      • Goldenthal A.R.
      • Uhlemann A.C.
      • Mann J.J.
      • Miller J.M.
      • Sublette M.E.
      Systematic review of gut microbiota and major depression.
      and atherosclerotic cardiovascular disease
      • Jie Z.
      • Xia H.
      • Zhong S.L.
      • et al.
      The gut microbiome in atherosclerotic cardiovascular disease.
      ; however, the microbiome–metabolite–disease interrelationships and potential causation have yet to be elucidated. Although human research is lacking, a recent animal study similarly showed significantly increased Streptococcus with an HPD.
      • Mu C.
      • Yang Y.
      • Luo Z.
      • Guan L.
      • Zhu W.
      The colonic microbiome and epithelial transcriptome are altered in rats fed a high-protein diet compared with a normal-protein diet.
      Interestingly, the relative abundance of Streptococcus during the HPD periods was lowest during the synbiotic period, suggesting mitigation of the HPD effect. Ho and colleagues
      • Ho J.
      • Nicolucci A.C.
      • Virtanen H.
      • et al.
      Effect of prebiotic on microbiota, intestinal permeability and glycemic control in children with type 1 diabetes.
      recently noted that Streptococcus was lower with prebiotic treatment vs placebo in children with type 1 diabetes, providing support for this premise. Increased abundance of other proteolytic taxa, predominantly from the phylum Bacteroidetes,
      • Korpela K.
      Diet, microbiota, and metabolic health: Trade-off between saccharolytic and proteolytic fermentation.
      expected to respond to the HPD, did not change. This may be due to the provision of fiber from a variety of plant sources, supporting saccharolytic vs proteolytic fermentation.
      • Korpela K.
      Diet, microbiota, and metabolic health: Trade-off between saccharolytic and proteolytic fermentation.
      In addition, although the HPD contained daily meat, plant proteins contributed a significant amount of the total protein. It has previously been suggested that the microbiota response to plant-sourced proteins may differ from that of meat
      • Korpela K.
      Diet, microbiota, and metabolic health: Trade-off between saccharolytic and proteolytic fermentation.
      and thus, may have modulated the impact of the HPD.
      We examined the effects of a multistrain probiotic supplement, prebiotic, and synbiotic on strain recovery in the context of a higher-protein diet. The presence of the five strains was consistently detected during administration and strain recovery was not impacted by the HPD or prebiotic administration. Provision of the multistrain probiotic did not increase the abundance of Bifidobacterium or Lactobacillus spp, as is most often seen when probiotics are provided to individuals without dysbiosis.
      • McFarland L.V.
      Use of probiotics to correct dysbiosis of normal microbiota following disease or disruptive events: A systematic review.
      The strains administered may have temporarily replaced resident strains. It is also possible that HPD may have blunted the expected increase in Bifidobacterium. Supplementation with 5 g of fiber from inulin did not enhance Bifidobacterium spp abundance, although this dose has been previously shown to exert such an effect.
      • So D.
      • Whelan K.
      • Rossi M.
      • et al.
      Dietary fiber intervention on gut microbiota composition in healthy adults: A systematic review and meta-analysis.
      Alternatively, as both naturally occurring and added fructans and other oligosaccharides are widespread in the food supply, the dose of inulin administered may not have been sufficient to elicit an additional increase in Bifidobacterium, especially considering that dose–response to inulin is thought to be limited.
      • So D.
      • Whelan K.
      • Rossi M.
      • et al.
      Dietary fiber intervention on gut microbiota composition in healthy adults: A systematic review and meta-analysis.
      An increase in flatulence, which drove the change in gastrointestinal distress during the prebiotic and synbiotic periods, confirmed fermentation of this substrate, suggesting that the inulin was fermented by resident gas-producing bacteria, such as from Clostridium cluster IV, which includes F prausnitzii.
      • Moens F.
      • De Vuyst L.
      Inulin-type fructan degradation capacity of Clostridium cluster IV and XIVa butyrate-producing colon bacteria and their associated metabolic outcomes.
      However, there was no evidence that the absolute abundance of F prausnitzii was affected by the HPD intervention or the supplements. The inulin may have induced changes in fermentation and metabolite production without impacting relative abundance of specific genera.
      A secondary aim of this research was to explore the effects of HPD on wellness, with and without probiotics and prebiotics. An increase in abdominal pain was observed with the HPD and mitigated by the probiotic. In comparison, Skov and colleagues
      • Skov A.R.
      • Toubro S.
      • Ronn B.
      • Holm L.
      • Astrup A.
      Randomized trial on protein vs carbohydrate in ad libitum fat reduced diet for the treatment of obesity.
      reported no difference in gastrointestinal discomfort or gastrointestinal symptoms with ad libitum consumption of a low-fat HPD (25% energy) for weight loss. While abdominal pain and indigestion were reported with the prebiotic, the average scoring level was below clinical significance. This finding is in accordance with Sairanen and colleagues,
      • Sairanen U.
      • Piirainen L.
      • Grasten S.
      • et al.
      The effect of probiotic fermented milk and inulin on the functions and microecology of the intestine.
      who reported increased gastrointestinal symptoms with a probiotic-supplemented fermented milk with 4 g/day inulin vs without. In young women, daily consumption of 8 g oligofructose resulted in an increased probability of reporting at least one gastrointestinal symptom (primarily flatulence), whereas a 16-g dose produced an even higher probability of reporting gastrointestinal symptoms in young men and women.
      • Dahl W.J.
      • Wright A.R.
      • Specht G.J.
      • et al.
      Consuming foods with added oligofructose improves stool frequency: A randomised trial in healthy young adults.
      Pedersen and colleagues
      • Pedersen A.
      • Sandstrom B.
      • Van Amelsvoort J.M.
      The effect of ingestion of inulin on blood lipids and gastrointestinal symptoms in healthy females.
      supplemented a low-fat diet with 14 g inulin in healthy young women and reported that self-reported gastrointestinal symptoms were higher with inulin, findings that are in accordance with the present study. Adaptation to the inulin, a decrease in symptom frequency and intensity over time, was not observed, and this is in agreement with other trials.
      • Dahl W.J.
      • Wright A.R.
      • Specht G.J.
      • et al.
      Consuming foods with added oligofructose improves stool frequency: A randomised trial in healthy young adults.
      ,
      • Pedersen A.
      • Sandstrom B.
      • Van Amelsvoort J.M.
      The effect of ingestion of inulin on blood lipids and gastrointestinal symptoms in healthy females.
      Suppression of hunger level was observed with the HPD and persisted during all washout periods, although participants resumed their usual protein intake. However, energy intake was lower than the participants’ average requirement during baseline, and thus may explain this finding. Hunger levels and energy intakes were similar in the HPD and washout periods, suggesting no specific effect of the HPD. Previous research on higher-protein diets and their enhanced effect on satiety or hunger suppression have been primarily limited to weight-loss studies,
      • Nickols-Richardson S.M.
      • Coleman M.D.
      • Volpe J.J.
      • Hosig K.W.
      Perceived hunger is lower and weight loss is greater in overweight premenopausal women consuming a low-carbohydrate/high-protein vs high-carbohydrate/low-fat diet.
      ,
      • Halton T.L.
      • Hu F.B.
      The effects of high protein diets on thermogenesis, satiety and weight loss: A critical review.
      and thus this effect of protein may be more apparent with energy restriction.
      Aging is associated with a decline in lean mass and an increase in fat mass, and high levels of mortality and impairment are associated with sarcopenia in older adults.
      • Litchford M.D.
      Counteracting the trajectory of frailty and sarcopenia in older adults.
      Data from the present study suggest that older women on an intermittent HPD (every other 2-week period) over 18 weeks may exhibit improved body composition, specifically by increasing lean body mass. This was an unexpected result, given that resistance exercises, which have been deemed required for a short-term effect, were not included in the intervention.
      • Tieland M.
      • van de Rest O.
      • Dirks M.L.
      • et al.
      Protein supplementation improves physical performance in frail elderly people: A randomized, double-blind, placebo-controlled trial.
      Instead, participants were asked to maintain their usual physical activity level. Weight stability suggested that participants were compliant. However, dietary protein intake has been associated with decreased loss of lean mass over longer time frames.
      • Houston D.K.
      • Nicklas B.J.
      • Ding J.
      • et al.
      Dietary protein intake is associated with lean mass change in older, community-dwelling adults: The Health, Aging, and Body Composition (Health ABC) Study.
      Notable limitations of this trial were the lengths of the intervention and washout periods. It is possible that if an HPD is administered for longer periods of time, further modulation of the microbiota may occur. The qPCR data suggest that after probiotic intake, strains were detected in a few individuals during washout periods. Carryover effects were seen for L plantarum HA-119 (four subjects), B bifidum HA-132 (seven subjects), and L acidophilus HA-122 (one subject), indicating that a 2-week washout period was not adequate for all participants, and strains may have persisted into the following treatment period for some participants. Typical crossover studies with probiotics include a washout period of variable duration, lasting 2 to 5 weeks between intervention periods.
      • Culpepper T.
      • Rowe C.C.
      • Rusch C.T.
      • et al.
      Three probiotic strains exert different effects on plasma bile acid profiles in healthy obese adults: Randomised, double-blind placebo-controlled crossover study.
      • Giannetti E.
      • Maglione M.
      • Alessandrella A.
      • et al.
      A mixture of 3 bifidobacteria decreases abdominal pain and improves the quality of life in children with irritable bowel syndrome: A multicenter, randomized, double-blind, placebo-controlled, crossover trial.
      • Spaiser S.J.
      • Culpepper T.
      • Nieves Jr., C.
      • et al.
      Lactobacillus gasseri KS-13, Bifidobacterium bifidum G9-1, and Bifidobacterium longum MM-2 ingestion induces a less inflammatory cytokine profile and a potentially beneficial shift in gut microbiota in older adults: A randomized, double-blind, placebo-controlled, crossover study.
      • Macfarlane S.
      • Cleary S.
      • Bahrami B.
      • Reynolds N.
      • Macfarlane G.T.
      Synbiotic consumption changes the metabolism and composition of the gut microbiota in older people and modifies inflammatory processes: A randomised, double-blind, placebo-controlled crossover study.
      • Asemi Z.
      • Khorrami-Rad A.
      • Alizadeh S.A.
      • Shakeri H.
      • Esmaillzadeh A.
      Effects of synbiotic food consumption on metabolic status of diabetic patients: A double-blind randomized cross-over controlled clinical trial.
      • Murakami K.
      • Habukawa C.
      • Nobuta Y.
      • Moriguchi N.
      • Takemura T.
      The effect of Lactobacillus brevis KB290 against irritable bowel syndrome: A placebo-controlled double-blind crossover trial.
      • Guandalini S.
      • Magazzu G.
      • Chiaro A.
      • et al.
      VSL#3 improves symptoms in children with irritable bowel syndrome: A multicenter, randomized, placebo-controlled, double-blind, crossover study.
      However, a recent study found that probiotic persistence after end of intervention appears to be strain-specific,
      • Taverniti V.
      • Koirala R.
      • Dalla Via A.
      • et al.
      Effect of cell concentration on the persistence in the human intestine of four probiotic strains administered through a multispecies formulation.
      suggesting that the adequate washout period for crossover trials will likely vary according to the strains administered. Therefore, more studies on probiotic strain-specific persistence behaviors in human populations are required. Alternatively, adaptive designs could be useful in that regard; qPCR could be performed after the 2-week washout period and negative results would validate passing onto the next period, or prolongation of the washout in the case of positive results. In addition, in future studies, participant stratification based on their baseline microbiome profiles may help decrease interindividual variation among older adults, which in turn may allow identification of characteristics distinguishing dietary or probiotic intervention responders from nonresponders. Examining a sample of only 20 women was an additional limitation. Findings may differ in older men and younger cohorts and thus, the generalizability of the results may be limited. Further, given intraindividual variation, microbiota findings may differ with geographical region and subgroup selection.

      Conclusions

      Provision of a higher-protein diet, compliant with the AMDRs for macronutrient contents, elicited no change in self-reported wellness or interference with probiotic strains recovery in older women. Modulation of microbiota profile by the HPD, specifically the suppression of butyrate-producing organisms, suggests a need to consider dietary quality, such as inclusion of whole grains, to help maintain these taxa and their beneficial metabolic activities. Supplementing the HPD with a multistrain probiotic may benefit gastrointestinal wellness even in healthy individuals reporting a high level of wellness. Intermittent HPD provision over 18 weeks suggests benefit for lean mass. However, the controlled diet made it possible to demonstrate mild gastrointestinal symptoms with as little as 5 g of inulin per day, confirming fermentation without an increase in Bifidobacterium and suggesting that the HPD may have suppressed this effect. Given the high interindividual variation observed among older women, future research should consider stratification by microbiome phenotype when testing effects of dietary interventions.

      What Is the Current Knowledge on This Topic?

      Studies investigating gut microbiome, health, and disease often disregard diet, although plant-based foods, diet quality, and macronutrient intakes modulate the microbiota and its activity. Few studies have explored the effects of an HPD on microbiota profile and evaluation of tolerance to an HPD has been underexplored.

      How Does This Research Add to Knowledge on This Topic?

      Provision of an HPD, AMDR-compliant for macronutrient contents, maintained general wellness and increased lean body mass in older women. The HPD suppressed butyrate-producing bacteria, but did not affect probiotic recovery, and thus their efficacy.

      How Might This Knowledge Impact Current Dietetics Practice?

      The findings of this study suggest that attention to dietary quality, such as the inclusion of whole grains, may be needed to help maintain levels of certain microbes and their beneficial metabolic activities while consuming an HPD.

      Acknowledgements

      The authors would like to thank the women who participated in the study and the dedicated team of undergraduate student volunteers for their invaluable assistance with the study.

      Author Contributions

      A. L. Ford, W. J. Dahl, T.A. Tompkins, and S.-A. Girard designed the trial. A. L. Ford and W. J. Dahl conducted the clinical trial. M. Christman carried out the statistical analysis of wellness outcomes and J. Auger, the microbiota. V. Nagulesapillai, A. Piano, and A. L. Ford carried out the microbiota and persistence analyses. A. L. Ford, W. J. Dahl, V. Nagulesapillai, A. Piano, J. Auger, and M. Christman wrote the manuscript. All authors reviewed the final manuscript.

      Supplementary Materials

      Table 1An example 1-day menu of the 1,600-kcal/day base high-protein diet provided to the study participants during the four 2-week high-protein diet interventions
      MealItemEnergy, kcalProtein, gCarbohydrate, gFiber, gFat, gSugar, gSodium, mg
      BreakfastEgg Beaters Smart Cups
      Conagra Brands.
      60112000220
      Thomas' Double Protein English Muffins
      Bimbo Bakeries.
      150726112210
      Smucker’s Concord Grape Jelly
      The J.M. Smucker Company.
      35090085
      SnackFriendship Fit to Go 1% low-fat cottage cheese,
      Saputo Dairy Foods USA Inc.
      5 oz
      110205024440
      Kellogg’s Special K Protein Meal Bar
      Kellogg NA Co.
      (honey almond)
      18010225612160
      LunchArtisan Bistro Wild Alaskan Salmon Bake
      Ruiz Food Products.
      2401630171280
      Dannon Light and Fit Greek Raspberry Chocolate nonfat yogurt
      Danone North America.
      8012900750
      DinnerBeef cubed steak, 5 oz2032800100116
      Libby's Vegetable Cups
      Libby’s Brand Holding.
      (cut green beans, 4 oz)
      1012101140
      Potato, flesh and skin, 170 g13133040110
      SnackJustin’s Peanut Butter
      Justin’s.
      (1.15 oz squeeze pack)
      19077316370
      Saltines Original Fresh Stacks
      Mondelēz Global LLC.
      22043915>1440
      Total1,609120181<1646392,141
      a Conagra Brands.
      b Bimbo Bakeries.
      c The J.M. Smucker Company.
      d Saputo Dairy Foods USA Inc.
      e Kellogg NA Co.
      f Ruiz Food Products.
      g Danone North America.
      h Libby’s Brand Holding.
      i Justin’s.
      j Mondelēz Global LLC.
      Table 2Primer sequences used to quantify, by quantitative polymerase chain reaction (qPCR), the strains provided in the probiotic formulation during two of the high-protein diet periods
      StrainPrimer nameSequence (5' to 3')Gene targetAnnealing temperature, °CAmplicon size, bp
      Bifidobacterium bifidum HA-132HA-132_GB_NC2_FAAGTGTGAGCCGGTGATAGCNoncoding sequence6078
      HA-132_GB_NC2_RCAGTACGTCGGCCGTTACAT
      Bifidobacterium breve HA-129HA-129_225-F2CGACCCTAATGACGTGGAGGHypothetical protein60195
      HA-129_225-R2CATTTCAGCCAGTACGTGCG
      Bifidobacterium longum HA-135HA-135_AP_HP10_FGTCGCCACATTTCATCGCAAHypothetical protein6099
      HA-135_AP_HP10_RGAGAGCTTCGATTGGCGAAC
      Lactobacillus acidophilus HA-122HA-122- F1AGAATCAAGCAGAGACTGGCTACGAn ORF
      ORF=open reading frame.
      in plasmid
      60150
      HA-122- R1GGACCGGATTTGAGTAGAGGTA
      Lactobacillus plantarum HA-119HA119_GB_PP_F2AGGTTGACATGGACACTGCTPhage protein6074
      HA119_GB_PP_R2CCGTTTGTCCAACAACCGAA
      a ORF=open reading frame.
      Figure thumbnail fx1
      Figure 5Pirateplot showing the distribution of Shannon α-diversity index for participants (intent-to-treat) during baseline, each high-protein diet period (placebo, prebiotic, probiotic, and synbiotic) and washouts (P=0.36) (bar/line=median, band/box=highest-density interval).
      Figure thumbnail fx2
      Figure 6Pirateplot showing the distribution of Shannon α-diversity index for participants (intent-to-treat) during baseline, high-protein diet intervention periods combined, and washouts (P=0.34) (bar/line=median; band/box=highest-density interval).
      Figure thumbnail fx3
      Figure 7Principal coordinate analysis based on the weighted UniFrac metric from fecal samples of all participants (intent-to-treat) during baseline, high-protein diet interventions (prebiotic, synbiotic, placebo, and probiotic), and washouts (pooled), each indicated by a different color.
      Figure thumbnail fx4
      Figure 8Principal coordinate analysis based on the weighted UniFrac metric from fecal samples of all participants (intent-to-treat) at all intervention periods. (A) Each individual is indicated by a different color and (B) each intervention order of the Latin design (A=prebiotic, B=synbiotic, C=placebo, D=probiotic).
      Figure thumbnail fx5
      Figure 11Relative abundance of (A) Roseburia and (B) Anaerostipes for participants (n=20) at baseline and during the high-protein diet periods and washouts.
      Figure thumbnail fx6
      Figure 12Relative abundance of genera (A) Lactococcus (B) Streptococcus for participants (n=20) at baseline and during the high-protein diet periods and washouts.
      Figure thumbnail fx7
      Figure 13Relative abundance of Lactobacillus for participants (n=20) at baseline and during the high-protein diet periods and washouts.
      Figure thumbnail fx8
      Figure 14Linear discriminant analysis (LDA) of fecal bacteria of participants (n=20) during baseline vs washout periods. (A) LDA score represents log changes in relative bacterial representation between baseline and washouts. (B) LDA score represents log changes in relative bacterial representation between baseline and washout 2. (Note: Taxonomy is not determined at the genus level). (C) LDA score represents log changes in relative bacterial representation between baseline and washout 3. (D) LDA score represents log changes in relative bacterial representation between baseline and washout 4 (threshold=2.0, α=.05).
      Figure thumbnail fx9
      Figure 15Radar chart of the SF-36v2 health domains, Mental Component Summary (MCS) and Physical Component Summary (PCS) of participants (n=26) for baseline and each high-protein diet treatment period. No differences were observed for SF-36v2 for eight health domains: physical functioning (PF), role-physical (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role-emotional (RE), and mental health (MH), MCS or PCS scores. A=prebiotic; B=synbiotic; C=placebo; D=probiotic; NBS=norm-based scoring.

      References

        • Landi F.
        • Calvani R.
        • Tosato M.
        • et al.
        Protein intake and muscle health in old age: From biological plausibility to clinical evidence.
        Nutrients. 2016; 8
        • Isanejad M.
        • Mursu J.
        • Sirola J.
        • et al.
        Dietary protein intake is associated with better physical function and muscle strength among elderly women.
        Br J Nutr. 2016; : 1281-1291
        • Mishra S.
        • Goldman J.D.
        • Sahyoun N.R.
        • Moshfegh A.J.
        Association between dietary protein intake and grip strength among adults aged 51 years and over: What We Eat in America, National Health and Nutrition Examination Survey 2011-2014.
        PLoS One. 2018; 13e0191368
        • Shams-White M.M.
        • Chung M.
        • Du M.
        • et al.
        Dietary protein and bone health: A systematic review and meta-analysis from the National Osteoporosis Foundation.
        Am J Clin Nutr. 2017; 105: 1528-1543
        • Wallace T.C.
        • Frankenfeld C.L.
        Dietary protein intake above the current RDA and bone health: A systematic review and meta-analysis.
        J Am Coll Nutr. 2017; 36: 481-496
        • Bauer J.M.
        • Diekmann R.
        Protein and older persons.
        Clin Geriatr Med. 2015; 31: 327-338
        • Bauer J.
        • Biolo G.
        • Cederholm T.
        • et al.
        Evidence-based recommendations for optimal dietary protein intake in older people: A position paper from the PROT-AGE Study Group.
        J Am Med Dir Assoc. 2013; 14: 542-559
        • Muscariello E.
        • Nasti G.
        • Siervo M.
        • et al.
        Dietary protein intake in sarcopenic obese older women.
        Clin Interv Aging. 2016; 11: 133-140
        • ter Borg S.
        • Verlaan S.
        • Mijnarends D.M.
        • Schols J.M.
        • De Groot L.C.
        • Luiking Y.C.
        Macronutrient intake and inadequacies of community-dwelling older adults, a systematic review.
        Ann Nutr Metab. 2015; 66: 242-255
        • Ten Haaf D.S.M.
        • De Regt M.F.
        • Visser M.
        • et al.
        Insufficient protein intakes is highly prevalent among physically active elderly.
        J Nutr Health Aging. 2018; 22: 1112-1114
        • Institute of Medicine
        Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids.
        The National Academies Press, Washington, DC2005
        • David L.A.
        • Maurice C.F.
        • Carmody R.N.
        • et al.
        Diet rapidly and reproducibly alters the human gut microbiome.
        Nature. 2014; 505: 559-563
        • Beaumont M.
        • Portune K.J.
        • Steuer N.
        • et al.
        Quantity and source of dietary protein influence metabolite production by gut microbiota and rectal mucosa gene expression: A randomized, parallel, double-blind trial in overweight humans.
        Am J Clin Nutr. 2017; 106: 1005-1019
        • Windey K.
        • De Preter V.
        • Louat T.
        • et al.
        Modulation of protein fermentation does not affect fecal water toxicity: A randomized cross-over study in healthy subjects.
        PLoS One. 2012; 7e52387
        • Salonen A.
        • Lahti L.
        • Salojarvi J.
        • et al.
        Impact of diet and individual variation on intestinal microbiota composition and fermentation products in obese men.
        ISME J. 2014; 8: 2218-2230
        • Duncan S.H.
        • Belenguer A.
        • Holtrop G.
        • Johnstone A.M.
        • Flint H.J.
        • Lobley G.E.
        Reduced dietary intake of carbohydrates by obese subjects results in decreased concentrations of butyrate and butyrate-producing bacteria in feces.
        Appl Environ Microbiol. 2007; 73: 1073-1078
        • Cotillard A.
        • Kennedy S.P.
        • Kong L.C.
        • et al.
        Dietary intervention impact on gut microbial gene richness.
        Nature. 2013; 500: 585-588
        • Russell W.R.
        • Gratz S.W.
        • Duncan S.H.
        • et al.
        High-protein, reduced-carbohydrate weight-loss diets promote metabolite profiles likely to be detrimental to colonic health.
        Am J Clin Nutr. 2011; 93: 1062-1072
        • Walker A.W.
        • Ince J.
        • Duncan S.H.
        • et al.
        Dominant and diet-responsive groups of bacteria within the human colonic microbiota.
        ISME J. 2010; 5: 220
        • Blachier F.
        • Beaumont M.
        • Portune K.J.
        • et al.
        High-protein diets for weight management: Interactions with the intestinal microbiota and consequences for gut health. A position paper by the my new gut study group.
        Clin Nutr. 2019; 38: 1012-1022
        • Menni C.
        • Jackson M.A.
        • Pallister T.
        • Steves C.J.
        • Spector T.D.
        • Valdes A.M.
        Gut microbiome diversity and high-fibre intake are related to lower long-term weight gain.
        Int J Obes (Lond). 2017; 41: 1099-1105
        • Seganfredo F.B.
        • Blume C.A.
        • Moehlecke M.
        • et al.
        Weight-loss interventions and gut microbiota changes in overweight and obese patients: A systematic review.
        Obes Rev. 2017; 18: 832-851
        • Pedersen A.N.
        • Cederholm T.
        Health effects of protein intake in healthy elderly populations: A systematic literature review.
        Food Nutr Res. 2014; 58
        • Claesson M.J.
        • Cusack S.
        • O'Sullivan O.
        • et al.
        Composition, variability, and temporal stability of the intestinal microbiota of the elderly.
        Proc Natl Acad Sci U S A. 2011; 108: 4586-4591
        • Rehman T.
        Role of the gut microbiota in age-related chronic inflammation.
        Endocr Metab Immune Disord Drug Targets. 2012; 12: 361-367
        • Bischoff S.C.
        Microbiota and aging.
        Curr Opin Clin Nutr Metab Care. 2016; 19: 26-30
        • Korpela K.
        Diet, microbiota, and metabolic health: Trade-off between saccharolytic and proteolytic fermentation.
        Annu Rev Food Sci Technol. 2018; 9: 65-84
        • Cummings J.H.
        • Hill M.J.
        • Bone E.S.
        • Branch W.J.
        • Jenkins D.J.
        The effect of meat protein and dietary fiber on colonic function and metabolism. II. Bacterial metabolites in feces and urine.
        Am J Clin Nutr. 1979; 32: 2094-2101
        • Macfarlane G.T.
        • Macfarlane S.
        Bacteria, colonic fermentation, and gastrointestinal health.
        J AOAC Int. 2012; 95: 50-60
        • Lakshminarayanan B.
        • Stanton C.
        • O'Toole P.W.
        • Ross R.P.
        Compositional dynamics of the human intestinal microbiota with aging: Implications for health.
        J Nutr Health Aging. 2014; 18: 773-786
        • O'Toole P.W.
        • Jeffery I.B.
        Gut microbiota and aging.
        Science. 2015; 350: 1214-1215
        • Vandeputte D.
        • Falony G.
        • Vieira-Silva S.
        • et al.
        Prebiotic inulin-type fructans induce specific changes in the human gut microbiota.
        Gut. 2017; 66: 1968-1974
        • Gagliardi A.
        • Totino V.
        • Cacciotti F.
        • et al.
        Rebuilding the gut microbiota ecosystem.
        Int J Environ Res Public Health. 2018; 15
        • Derrien M.
        • Belzer C.
        • de Vos W.M.
        Akkermansia muciniphila and its role in regulating host functions.
        Microb Pathog. 2017; 106: 171-181
        • Ferreira-Halder C.V.
        • Faria A.V.S.
        • Andrade S.S.
        Action and function of Faecalibacterium prausnitzii in health and disease.
        Best Pract Res Clin Gastroenterol. 2017; 31: 643-648
        • Guenther P.
        • DeMaio T.
        • Ingwersen L.
        • Berlin M.
        The multiple-pass approach for the 24-h recall in the Continuing Survey of Food Intakes by Individuals, 1994-1996.
        Am J Clin Nutr. 1997; 65 (S1316-S1316)
      1. ESHA Food Processor [computer program]. Version 11.0.117. Salem, OR: ESHA Research; 2015.

        • Andre D.
        • Pelletier R.
        • Farringdon J.
        • et al.
        The development of the SenseWear® armband, a revolutionary energy assessment device to assess physical activity and lifestyle. BodyMedia Inc. 2006.
        • Mackey D.C.
        • Manini T.M.
        • Schoeller D.A.
        • et al.
        Validation of an armband to measure daily energy expenditure in older adults.
        J Gerontol A Biol Sci Med Sci. 2011; 66: 1108-1113
        • Kulich K.R.
        • Madisch A.
        • Pacini F.
        • et al.
        Reliability and validity of the Gastrointestinal Symptom Rating Scale (GSRS) and Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire in dyspepsia: A six-country study.
        Health Qual Life Outcomes. 2008; 6: 12
        • Peters L.L.
        • Boter H.
        • Buskens E.
        • Slaets J.P.
        Measurement properties of the Groningen Frailty Indicator in home-dwelling and institutionalized elderly people.
        J Am Med Dir Assoc. 2012; 13: 546-551
        • Ware J.E.
        A 36-item short-form health survey (SF-36), I: Conceptual framework and item selection.
        Med Care. 1992; 30: 476-483
        • Hartman A.L.
        • Lough D.M.
        • Barupal D.K.
        • et al.
        Human gut microbiome adopts an alternative state following small bowel transplantation.
        Proc Natl Acad Sci U S A. 2009; 106: 17187-17192
        • Collado M.C.
        • Derrien M.
        • Isolauri E.
        • de Vos W.M.
        • Salminen S.
        Intestinal integrity and Akkermansia muciniphila, a mucin-degrading member of the intestinal microbiota present in infants, adults, and the elderly.
        Appl Environ Microbiol. 2007; 73: 7767-7770
        • Lopez-Siles M.
        • Duncan S.H.
        • Garcia-Gil L.J.
        • Martinez-Medina M.
        Faecalibacterium prausnitzii: From microbiology to diagnostics and prognostics.
        ISME J. 2017; 11: 841-852
        • MacPherson C.W.
        • Mathieu O.
        • Tremblay J.
        • et al.
        Gut bacterial microbiota and its resistome rapidly recover to basal state levels after short-term amoxicillin-clavulanic acid treatment in healthy adults.
        Sci Rep. 2018; 8: 11192
      2. 16S Metagenomic Sequencing Library Preparation: Preparing 16S Ribosomal RNA Gene Amplicons for the Illumina MiSeq System.
        • Bolyen E.
        • Rideout J.R.
        • Dillon M.R.
        • et al.
        QIIME 2: Reproducible, interactive, scalable, and extensible microbiome data science.
        Peer J Prepr. 2018; 6
        • Kruskal W.H.
        • Wallis W.
        Use of ranks in one-criterion variance analysis.
        J Am Stat Assoc. 1952; 47: 583-621
      3. McKinney W. Data structures for statistical computing in Python. In: van der Walt S, Millman J, eds. Proceedings of the 9th Python in Science Conference, Austin, TX, June 28-July 3, 2010:51-56.

        • McDonald D.
        • Clemente J.C.
        • Kuczynski J.
        • et al.
        The Biological Observation Matrix (BIOM) format or: How I learned to stop worrying and love the ome-ome.
        GigaScience. 2012; 1: 7
        • Weiss S.
        • Xu Z.Z.
        • Peddada S.
        • et al.
        Normalization and microbial differential abundance strategies depend upon data characteristics.
        Microbiome. 2017; 5: 27
        • Amir A.
        • McDonald D.
        • Navas-Molina J.A.
        • et al.
        Deblur rapidly resolves single-nucleotide community sequence patterns.
        Am Soc Microbiol. 2017; 2 (e00191-16)
        • Bokulich N.A.
        • Subramanian S.
        • Faith J.J.
        • et al.
        Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing.
        Nat Methods. 2013; 10: 57-59
        • Price M.N.
        • Dehal P.S.
        • Arkin A.P.
        FastTree 2—Approximately maximum-likelihood trees for large alignments.
        PLoS One. 2010; 5: e9490
        • Lane D.J.
        16S/23S rRNA sequencing.
        in: Stackebrandt E. Goodfellow M. Nucleic Acid Techniques in Bacterial Systematics. John Wiley and Sons, New York1991: 115-175
        • Katoh K.
        • Standley D.M.
        MAFFT multiple sequence alignment software version 7: Improvements in performance and usability.
        Mol Biol Evol. 2013; 30: 772-780
        • Lozupone C.
        • Knight R.
        UniFrac: A new phylogenetic method for comparing microbial communities.
        Appl Environ Microbiol. 2005; 71: 8228-8235
        • Lozupone C.A.
        • Hamady M.
        • Kelley S.T.
        • Knight R.
        Quantitative and qualitative β diversity measures lead to different insights into factors that structure microbial communities.
        Appl Environ Microbiol. 2007; 73: 1576-1585
        • Chang Q.
        • Luan Y.
        • Sun F.
        Variance adjusted weighted UniFrac: A powerful beta diversity measure for comparing communities based on phylogeny.
        BMC Bioinformatics. 2011; 12: 118
        • Chen J.
        • Bittinger K.
        • Charlson E.S.
        • et al.
        Associating microbiome composition with environmental covariates using generalized UniFrac distances.
        Bioinformatics. 2012; 28: 2106-2113
        • McDonald D.
        • Vázquez-Baeza Y.
        • Koslicki D.
        • et al.
        Striped UniFrac: Enabling microbiome analysis at unprecedented scale.
        Nat Methods. 2018; 15: 847-848
        • Vázquez-Baeza Y.
        • Pirrung M.
        • Gonzalez A.
        • Knight R.
        EMPeror: A tool for visualizing high-throughput microbial community data.
        Gigascience. 2013; 2: 2013
        • Vázquez-Baeza Y.
        • Gonzalez A.
        • Smarr L.
        • et al.
        Bringing the dynamic microbiome to life with animations.
        Cell Host Microbe. 2017; 21: 7-10
        • Segata N.
        • Izard J.
        • Waldron L.
        • et al.
        Metagenomic biomarker discovery and explanation.
        Genome Biol. 2011; 12: R60
        • Lewis S.J.
        • Heaton K.W.
        Stool form scale as a useful guide to intestinal transit time.
        Scand J Gastroenterol. 1997; 32: 920-924
        • Hanifi A.
        • Culpepper T.
        • Mai V.
        • et al.
        Evaluation of Bacillus subtilis R0179 on gastrointestinal viability and general wellness: A randomised, double-blind, placebo-controlled trial in healthy adults.
        Benef Microbes. 2015; 6: 19-27
      4. QualityMetric Health Outcomes Scoring [computer program]. Version 4.5. QualityMetric Inc, Lincoln, RI2004
        • Flood A.
        • Chung A.
        • Parker H.
        • Kearns V.
        • O'Sullivan T.A.
        The use of hand grip strength as a predictor of nutrition status in hospital patients.
        Clin Nutr. 2014; 33: 106-114
      5. xPONENT [computer program]. Version 3.1. Luminex, Austin, TX2009
        • Hébuterne X.
        Gut changes attributed to ageing: Effects on intestinal microflora.
        Curr Opin Clin Nutr Metab Care. 2003; 6: 49-54
        • Odamaki T.
        • Kato K.
        • Sugahara H.
        • et al.
        Age-related changes in gut microbiota composition from newborn to centenarian: A cross-sectional study.
        BMC Microbiol. 2016; 16: 90
        • Zhu Q.
        • Huang L.
        • Zhu J.
        • Lu J.
        • Qin H.
        Analysis of gut microbiota in long-lived older adults and their relatives: A gradual change with ageing.
        Mech Ageing Dev. 2018; 178: 1-8
        • Tomova A.
        • Bukovsky I.
        • Rembert E.
        • et al.
        The effects of vegetarian and vegan diets on gut microbiota.
        Front Nutr. 2019; 6: 47
        • Brahe L.K.
        • Astrup A.
        • Larsen L.H.
        Is butyrate the link between diet, intestinal microbiota and obesity-related metabolic diseases?.
        Obes Rev. 2013; 14: 950-959
        • Martinez I.
        • Lattimer J.M.
        • Hubach K.L.
        • et al.
        Gut microbiome composition is linked to whole grain-induced immunological improvements.
        ISME J. 2013; 7: 269-280
        • Liu Y.
        • Ajami N.J.
        • El-Serag H.B.
        • et al.
        Dietary quality and the colonic mucosa-associated gut microbiome in humans.
        Am J Clin Nutr. 2019; 110: 701-712
        • Jiang S.
        • Xie S.
        • Lv D.
        • et al.
        A reduction in the butyrate producing species Roseburia spp. and Faecalibacterium prausnitzii is associated with chronic kidney disease progression.
        Antonie Van Leeuwenhoek. 2016; 109: 1389-1396
        • Machiels K.
        • Joossens M.
        • Sabino J.
        • et al.
        A decrease of the butyrate-producing species Roseburia hominis and Faecalibacterium prausnitzii defines dysbiosis in patients with ulcerative colitis.
        Gut. 2014; 63: 1275-1283
        • Kristensen N.B.
        • Bryrup T.
        • Allin K.H.
        • Nielsen T.
        • Hansen T.H.
        • Pedersen O.
        Alterations in fecal microbiota composition by probiotic supplementation in healthy adults: A systematic review of randomized controlled trials.
        Genome Med. 2016; 8: 52
        • Sommer F.
        • Anderson J.M.
        • Bharti R.
        • Raes J.
        • Rosenstiel P.
        The resilience of the intestinal microbiota influences health and disease.
        Nat Rev Microbiol. 2017; 15: 630-638
        • Cheung S.G.
        • Goldenthal A.R.
        • Uhlemann A.C.
        • Mann J.J.
        • Miller J.M.
        • Sublette M.E.
        Systematic review of gut microbiota and major depression.
        Front Psychiatry. 2019; 10: 34
        • Jie Z.
        • Xia H.
        • Zhong S.L.
        • et al.
        The gut microbiome in atherosclerotic cardiovascular disease.
        Nat Commun. 2017; 8: 845
        • Mu C.
        • Yang Y.
        • Luo Z.
        • Guan L.
        • Zhu W.
        The colonic microbiome and epithelial transcriptome are altered in rats fed a high-protein diet compared with a normal-protein diet.
        J Nutr. 2016; 146: 474-483
        • Ho J.
        • Nicolucci A.C.
        • Virtanen H.
        • et al.
        Effect of prebiotic on microbiota, intestinal permeability and glycemic control in children with type 1 diabetes.
        J Clin Endocrinol Metab. 2019; 104: 4427-4440
        • McFarland L.V.
        Use of probiotics to correct dysbiosis of normal microbiota following disease or disruptive events: A systematic review.
        BMJ Open. 2014; 4e005047
        • So D.
        • Whelan K.
        • Rossi M.
        • et al.
        Dietary fiber intervention on gut microbiota composition in healthy adults: A systematic review and meta-analysis.
        Am J Clin Nutr. 2018; 107: 965-983
        • Moens F.
        • De Vuyst L.
        Inulin-type fructan degradation capacity of Clostridium cluster IV and XIVa butyrate-producing colon bacteria and their associated metabolic outcomes.
        Benef Microbes. 2017; 8: 473-490
        • Skov A.R.
        • Toubro S.
        • Ronn B.
        • Holm L.
        • Astrup A.
        Randomized trial on protein vs carbohydrate in ad libitum fat reduced diet for the treatment of obesity.
        Int J Obes Relat Metab Disord. 1999; 23: 528-536
        • Sairanen U.
        • Piirainen L.
        • Grasten S.
        • et al.
        The effect of probiotic fermented milk and inulin on the functions and microecology of the intestine.
        J Dairy Res. 2007; 74: 367-373
        • Dahl W.J.
        • Wright A.R.
        • Specht G.J.
        • et al.
        Consuming foods with added oligofructose improves stool frequency: A randomised trial in healthy young adults.
        J Nutr Sci. 2014; 3: e7
        • Pedersen A.
        • Sandstrom B.
        • Van Amelsvoort J.M.
        The effect of ingestion of inulin on blood lipids and gastrointestinal symptoms in healthy females.
        Br J Nutr. 1997; 78: 215-222
        • Nickols-Richardson S.M.
        • Coleman M.D.
        • Volpe J.J.
        • Hosig K.W.
        Perceived hunger is lower and weight loss is greater in overweight premenopausal women consuming a low-carbohydrate/high-protein vs high-carbohydrate/low-fat diet.
        J Am Diet Assoc. 2005; 105: 1433-1437
        • Halton T.L.
        • Hu F.B.
        The effects of high protein diets on thermogenesis, satiety and weight loss: A critical review.
        J Am Coll Nutr. 2004; 23: 373-385
        • Litchford M.D.
        Counteracting the trajectory of frailty and sarcopenia in older adults.
        Nutr Clin Pract. 2014; 29: 428-434
        • Tieland M.
        • van de Rest O.
        • Dirks M.L.
        • et al.
        Protein supplementation improves physical performance in frail elderly people: A randomized, double-blind, placebo-controlled trial.
        J Am Med Dir Assoc. 2012; 13: 720-726
        • Houston D.K.
        • Nicklas B.J.
        • Ding J.
        • et al.
        Dietary protein intake is associated with lean mass change in older, community-dwelling adults: The Health, Aging, and Body Composition (Health ABC) Study.
        Am J Clin Nutr. 2008; 87: 150-155
        • Culpepper T.
        • Rowe C.C.
        • Rusch C.T.
        • et al.
        Three probiotic strains exert different effects on plasma bile acid profiles in healthy obese adults: Randomised, double-blind placebo-controlled crossover study.
        Benef Microbes. 2019; 10: 497-509
        • Giannetti E.
        • Maglione M.
        • Alessandrella A.
        • et al.
        A mixture of 3 bifidobacteria decreases abdominal pain and improves the quality of life in children with irritable bowel syndrome: A multicenter, randomized, double-blind, placebo-controlled, crossover trial.
        J Clin Gastroenterol. 2017; 51: e5-e10
        • Spaiser S.J.
        • Culpepper T.
        • Nieves Jr., C.
        • et al.
        Lactobacillus gasseri KS-13, Bifidobacterium bifidum G9-1, and Bifidobacterium longum MM-2 ingestion induces a less inflammatory cytokine profile and a potentially beneficial shift in gut microbiota in older adults: A randomized, double-blind, placebo-controlled, crossover study.
        J Am Coll Nutr. 2015; 34: 459-469
        • Macfarlane S.
        • Cleary S.
        • Bahrami B.
        • Reynolds N.
        • Macfarlane G.T.
        Synbiotic consumption changes the metabolism and composition of the gut microbiota in older people and modifies inflammatory processes: A randomised, double-blind, placebo-controlled crossover study.
        Aliment Pharmacol Ther. 2013; 38: 804-816
        • Asemi Z.
        • Khorrami-Rad A.
        • Alizadeh S.A.
        • Shakeri H.
        • Esmaillzadeh A.
        Effects of synbiotic food consumption on metabolic status of diabetic patients: A double-blind randomized cross-over controlled clinical trial.
        Clin Nutr. 2014; 33: 198-203
        • Murakami K.
        • Habukawa C.
        • Nobuta Y.
        • Moriguchi N.
        • Takemura T.
        The effect of Lactobacillus brevis KB290 against irritable bowel syndrome: A placebo-controlled double-blind crossover trial.
        Biopsychosoc Med. 2012; 6: 16
        • Guandalini S.
        • Magazzu G.
        • Chiaro A.
        • et al.
        VSL#3 improves symptoms in children with irritable bowel syndrome: A multicenter, randomized, placebo-controlled, double-blind, crossover study.
        J Pediatr Gastroenterol Nutr. 2010; 51: 24-30
        • Taverniti V.
        • Koirala R.
        • Dalla Via A.
        • et al.
        Effect of cell concentration on the persistence in the human intestine of four probiotic strains administered through a multispecies formulation.
        Nutrients. 2019; 11

      Biography

      A. L. Ford is a nutrition manager, Red Lobster Corp, Orlando, FL; at the time of the study, she was a doctoral student, Food Science and Human Nutrition Department, University of Florida, Gainesville.
      V. Nagulesapillai is a clinical research specialist, Rosell Institute for Microbiome and Probiotics, Montréal, Québec, Canada.
      A. Piano is a research scientist, Rosell Institute for Microbiome and Probiotics, Montréal, Québec, Canada.
      J. Auger is a clinical bioinformatician, Rosell Institute for Microbiome and Probiotics, Montréal, Québec, Canada.
      S.-A. Girard is a clinical program manager, Rosell Institute for Microbiome and Probiotics, Montréal, Québec, Canada.
      T. Tompkins is a research director, Rosell Institute for Microbiome and Probiotics, Montréal, Québec, Canada.
      M. Christman is owner, MCC Statistical Consulting LLC, Gainesville, FL.
      W. J. Dahl is an associate professor, Department of Food Science and Human Nutrition, University of Florida, Gainesville.