Aerobic Exercise Training with Brisk Walking Increases Intestinal Bacteroides in Healthy Elderly Women

This study examined the e ect of an exercise intervention on the composition of the intestinal microbiota in healthy elderly women. Thirty-two sedentary women that were aged 65 years and older participated in a 12-week, non-randomized comparative trial. The subjects were allocated to two groups receiving di erent exercise interventions, trunk muscle training (TM), or aerobic exercise training (AE). AE included brisk walking, i.e., at an intensity of 3 metabolic equivalents (METs). The composition of the intestinal microbiota in fecal samples was determined before and after the training period. We also assessed the daily physical activity using an accelerometer, trunk muscle strength by the modi ed Kraus–Weber (K-W) test, and cardiorespiratory tness by a 6-min. walk test (6MWT). K-W test scores and distance achieved during the 6MWT (6MWD) improved in both groups. The relative abundance of intestinalBacteroidesonly signi cantly increased in the AE group, particularly in subjects showing increases in the time spent in brisk walking. Overall, the increases in intestinalBacteroidesfollowing

strength by the modi ed Kraus–Weber (K-W) test, and cardiorespiratory tness by a 6-min. walk test (6MWT). K-W test scores and distance achieved during the 6MWT (6MWD) improved in both groups. The relative abundance of intestinalBacteroidesonly signi cantly increased in the AE group, particularly in subjects showing increases in the time spent in brisk walking. Overall, the increases in intestinalBacteroidesfollowing the exercise intervention were associated with increases in 6MWD. In conclusion, aerobic exercise training that targets an increase of the time spent in brisk walking may increase intestinalBacteroidesin association with improved cardiorespiratory tness in healthy elderly women. Keywords: intestinal microbiota; intestinalBacteroides; cardiorespiratory tness; trunk muscle training; aerobic exercise training; brisk walking 1. Introduction “All disease begins in the gut”., a quotation from the ancient Greek physician Hippocrates, highlights the potential roles of intestinal microbiota in various disease risks, which have recently Nutrients2019,11, 868; doi:10.3390 /nu11040868 /journal/nutrients

Nutrients2019,11, 868 2 of 17 attracted considerable attention from researchers. The presence of an imbalanced, low-diversity, intestinal microbiota is known as dysbiosis and it is associated with a variety of pathologies, including constipation [1], obesity [2], diabetes [3], colon cancer [4], coronary artery disease [5], in ammatory bowel disease [6], and depression [7]. Aging also strongly a ects the composition of the intestinal microbiota. In general, the intestinal microbiota of the elderly show reduced species diversity [8]. In addition, intestinalBi dobacteriumandBacteroides, which are known to be related to obesity, are also reduced [8], which potentially contributes to the high prevalence of obesity in the elderly population. Overall, the intestinal microbiota could be regarded as an indicator of host health. Multiple factors, including host genetics [9], method of childbirth (i.e., by vaginal delivery or caesarian section) [10], age [8], nutrition [11], and antibiotic intake [8], have been suggested to a ect the composition of the intestinal microbiota. Recent studies demonstrated the association between exercise training, i.e., a low-cost health strategy, and lower risks of colon cancer [12,13], a disease that is known to at least partly arise from imbalanced intestinal microbiota [4]. Therefore, exercise may also have potential for modifying the composition of the intestinal microbiota, although these studies did not directly examine the e ect of exercise on intestinal microbiota. In fact, animal studies have demonstrated the changes in the composition of the intestinal microbiota by exercise training [14–16]. A number of cross-sectional human studies have con rmed the associations between physical activity or cardiorespiratory tness and the composition of the intestinal microbiota [17–19]. For example, rugby players were found to have a greater diversity of intestinal microbiota and an enlarged abundance ofAkkermansia—which is known to prevent diabetes—when compared to sedentary adults [17]. Other studies showed that cardiorespiratory tness or physical activity level is associated with greater microbial diversity in healthy humans [18–20]. Furthermore, trained elite race walkers show increased relative abundance ofBacteroides,in combination with high fat diet [21]. However, these studies did not examine the e ect of exercise alone on intestinal microbiota independent of the dietary habits that

sedentary adults [17]. Other studies showed that cardiorespiratory tness or physical activity level is associated with greater microbial diversity in healthy humans [18–20]. Furthermore, trained elite race walkers show increased relative abundance ofBacteroides,in combination with high fat diet [21]. However, these studies did not examine the e ect of exercise alone on intestinal microbiota independent of the dietary habits that may have the greater impact on intestinal microbiota than exercise. Therefore, the potential impact of exercise interventions on human intestinal microbiota has not been fully clari ed. In the present study, we examined the e ects of exercise interventions on intestinal microbiotic composition in healthy elderly women. We hypothesized that an improvement of cardiorespiratory tness would be crucial to exercise-induced changes in the intestinal microbiota. We compared the e ects of two exercise modalities on the intestinal microbiota: aerobic exercise, which speci cally enhances cardiorespiratory tness, and trunk muscle training as a control condition to verify this hypothesis. 2. Materials and Methods 2.1. Subjects Thirty-two healthy sedentary women that were aged 65 years and over were recruited from the residents of Osaka City, Japan, by an advertisement in a local magazine. The selected 32 subjects voluntarily opted for enrollment in either of the two exercise programs, aerobic exercise training (AE) or trunk muscle training (control condition; TM). Prior to the study, none of the subjects engaged in a regular exercise for more than 1 h per week. Health status and the use of medication were assessed by structured interview. Applicants presenting a history of ischemic heart disease, chronic heart failure, stroke, severe hypertension, diabetes, or neuropsychiatric disorder were excluded from the study. Applicants who were judged by a physician to be unable or ill-equipped to participate in the exercise program were also excluded. Consequently, none of the 32 subjects was excluded. The Institutional Review Board of Osaka City University Graduate School of Medicine approved the study protocol (approval no. 3501, approved on August 30, 2016). The authors also con rm that all of the ongoing and related trials for this intervention are registered in the University Hospital Medical Information Network

were also excluded. Consequently, none of the 32 subjects was excluded. The Institutional Review Board of Osaka City University Graduate School of Medicine approved the study protocol (approval no. 3501, approved on August 30, 2016). The authors also con rm that all of the ongoing and related trials for this intervention are registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN 000023930). Written informed consent was obtained from all

Nutrients2019,11, 868 3 of 17 of the participants after explanation of the study purpose. The study protocol also conformed to the ethical guidelines of the 1975 Declaration of Helsinki. 2.2. Study Design The study design involved a 12-week non-randomized, comparative trial, in which the allocation of the participants to either of the two exercise groups, AE and TM, was based on their own preference. This study was conducted between the rst recruitment of the participants on 12 September 2016 and the nal follow-up of the participants on 24 January 2018. Before study enrollment, all of the applicants visited our research center at Osaka City University for baseline measurements, e.g., body composition, motor ability, and clinical laboratory analyses, as well as an assessment of daily physical activity levels, nutrient intake, and bowel habits. In addition, fecal samples were collected. All of the baseline assessments were conducted at least 1 week before the rst training session. Finally, 18 and 14 applicants who met the inclusion criteria were enrolled in the AE group and the TM group, respectively, after which they were started on the selected 12-week exercise programs. The measurements during the baseline session were repeated at least one week after the nal session of the exercise program. 2.3. Exercise Intervention The subjects in the TM group received a 1-h group training weekly for 12 weeks, which aimed at strengthening the trunk muscles. All of the sessions were held at Sumiyoshi Sports Center, a gymnasium located in Osaka City, and supervised by a trained instructor. A training session comprised 5–10 min. of warm-up, followed by 45 min. of targeted resistance training of the trunk muscles and 5–10 min. of cool down exercises. Figure was composed of several kinds of exercises, including arching–swaying, plank, pelvic rotation in the supine position, and diagonal lifting while standing on all fours. The contraction duration was set at 3 to 5 s, and each exercise was performed in two sets of 10 repetitions. The subjects were also instructed to work out at home daily. Adherence to group sessions, as well as to the home exercises, was recorded

plank, pelvic rotation in the supine position, and diagonal lifting while standing on all fours. The contraction duration was set at 3 to 5 s, and each exercise was performed in two sets of 10 repetitions. The subjects were also instructed to work out at home daily. Adherence to group sessions, as well as to the home exercises, was recorded weekly by the instructor throughout the 12-week intervention period.Nutrients 2019, 11, x FOR PEER REVIEW 3 of 18 Information Network Clinical Trials Registry (UMIN 000023930). Written informed consent was obtained from all of the participants after explanation of the study purpose. The study protocol also conformed to the ethical guidelines of the 1975 Declaration of Helsinki. 2.2. Study Design The study design involved a 12-week non-randomized, comparative trial, in which the allocation of the participants to either of the two exercise groups, AE and TM, was based on their own preference. This study was conducted between the first recruitment of the participants on September 12, 2016 and the final follow-up of the participants on January 24, 2018. Before study enrollment, all of the applicants visited our research center at Osaka City University for baseline measurements, e.g., body composition, motor ability, and clinical laboratory analyses, as well as an assessment of daily physical activity levels, nutrient intake, and bowel habits. In addition, fecal samples were collected. All of the baseline assessments were conducted at least 1 week before the first training session. Finally, 18 and 14 applicants who met the inclusion criteria were enrolled in the AE group and the TM group, respectively, after which they were started on the selected 12-week exercise programs. The measurements during the baseline session were repeated at least one week after the final session of the exercise program. 2.3. Exercise Intervention The subjects in the TM group received a 1-h group training weekly for 12 weeks, which aimed at strengthening the trunk muscles. All of the sessions were held at Sumiyoshi Sports Center, a gymnasium located in Osaka City, and supervised by a trained instructor. A training session comprised 5–10 min. of warm-up, followed by

the exercise program. 2.3. Exercise Intervention The subjects in the TM group received a 1-h group training weekly for 12 weeks, which aimed at strengthening the trunk muscles. All of the sessions were held at Sumiyoshi Sports Center, a gymnasium located in Osaka City, and supervised by a trained instructor. A training session comprised 5–10 min. of warm-up, followed by 45 min. of targeted resistance training of the trunk muscles and 5–10 min. of cool down exercises. Figure 1 shows examples of the trunk muscle training. The training was composed of several kinds of exercises, including arching–swaying, plank, pelvic rotation in the supine position, and diagonal lifting while standing on all fours. The contraction duration was set at 3 to 5 s, and each exercise was performed in two sets of 10 repetitions. The subjects were also instructed to work out at home daily. Adherence to group sessions, as well as to the home exercises, was recorded weekly by the instructor throughout the 12-week intervention period. Figure 1. Exercises during trunk muscle training. (A) Arching–swaying while standing on all fours, (B) plank, (C) lying pelvic rotation, and (D) diagonal lifting while standing on all fours. Figure 1. Exercises during trunk muscle training. (A) Arching–swaying while standing on all fours, (B) plank, (C) lying pelvic rotation, and (D) diagonal lifting while standing on all fours. The subjects in the AE group were instructed to perform 60 min. of brisk walking at an intensity of 3 metabolic equivalents (METs) daily for 12 weeks. They wore a three-axis accelerometer (Mediwalk ®

Nutrients2019,11, 868 4 of 17 MT-KT02DZ, TERUMO, Tokyo, Japan [22,23]) throughout the intervention period, except while sleeping and bathing, to record their daily number of steps and time that is spent in brisk walking. The instructor shared the accelerometer data with the participants once a week and was encouraged them to increase the intensity and duration of their brisk walking regimen gradually as much as possible. The subjects were also instructed to keep good posture while walking. 2.4. Analysis of Intestinal Microbiota The fecal samples were collected in a container with guanidine thiocyanate as a preservative solution (TechnoSuruga Laboratory, Shizuoka, Japan) and refrigerated at 4 C until transfer to the laboratory within seven days. We conformed to the protocol [24] for the representative extraction of DNA from bacterial populations in feces. Terminal restriction fragment length polymorphism (T-RFLP) analyses to determine the relative abundance of intestinal microbiota phylogenetic groups from each fecal sample were performed at the TechnoSuruga Laboratory (Shizuoka, Japan) [25,26]. T-RFLP analysis is one of the most well-established and reliable 16S ribosomal RNA-based methods, especially when considering its high throughput and reproducibility. Brie y, the fecal samples (approximately 4 mg each) were suspended in a 1200 L solution containing 100 mM Tris-HCl (pH 9.0), 40 mM ethylenediaminetetraacetic acid, 4 M guanidine thiocyanate, and 0.001% bromothymol blue. A FastPrep 24 device homogenized the Fecal solids in the suspension (MP Biomedicals, Irvine, CA, USA) with zirconia beads being set at 5 m/s for 2 min. DNA was then extracted from a 200 L suspension using magLEAD 12gC (Precision System Science; Chiba, Japan). MagDEA ® Dx SV (Precision System Science) was used as the reagent in automatic nucleic acid extraction. PCR was performed with a Takara Thermal Cycler Dice TP650 (Takara Bio, Shiga, Japan) in 20 L of a reaction mixture containing 1 PCR bu er, with each deoxynucleotide triphosphate at a concentration of 200 M, 1.5 mM MgCl2, each primer at a concentration of 0.2 M, 10 ng of fecal DNA, and 0.2 U of HotStarTaq DNA polymerase (Qiagen, Hilden, Germany). 5 0 FAM-labeled 516f (5'-TGC-CAGCAGCCGCGGTA-3'; Escherichia coli positions 516 532)

Shiga, Japan) in 20 L of a reaction mixture containing 1 PCR bu er, with each deoxynucleotide triphosphate at a concentration of 200 M, 1.5 mM MgCl2, each primer at a concentration of 0.2 M, 10 ng of fecal DNA, and 0.2 U of HotStarTaq DNA polymerase (Qiagen, Hilden, Germany). 5 0 FAM-labeled 516f (5'-TGC-CAGCAGCCGCGGTA-3'; Escherichia coli positions 516 532) and 1510r (5'-GGTTACCTTGTTACGA-CTT-3';E. colipositions 1510 1492) were the primers used. The ampli cation program used was as follows: preheating at 95 C for 15 min, 35 cycles of denaturation at 95 C for 30 s, annealing at 50 C for 30 s, extension at 72 C for 90 s, and nally, terminal extension at 72 C for 10 min. Electrophoresis and puri ed using a MultiScreen PCR 96 Filter Plate veri ed ampli ed DNA (Millipore, Billerica, MA, USA). The puri ed 16S rDNA amplicons were treated with 10 U of FastDigest BseLI (Thermo Fisher Scienti c, Waltham, MA, USA) for 10 min. An ABI PRISM 3130xl genetic analyzer (Thermo Fisher Scienti c) was used to analyze the resultant DNA fragments, i.e., uorescent-labeled terminal restriction fragments (T-RFs). GeneMapper software (Thermo Fisher Scienti c) was used to determine the T-RF length and the peak area for each sample. T-RFs were divided into 29 operational taxonomic units (OTUs). The individual OTUs were quanti ed as the percentage of all OTUs combined based on the area under the curve (% AUC). The reference database, Human Fecal Microbiota T-RFLP pro ling (http: //www.tecsrg-lab.jp/t_r p_hito_OTU.html), was used to putatively match the bacteria in each classi cation unit to the corresponding OTU. T-RFLP analyses enabled the classi cation of the sampled intestinal microbiota into the following 10 groups: Bi dobacterium,Lactobacillales,Bacteroides,Prevotella,Clostridiumcluster IV,Clostridiumsubcluster XIVa, Clostridiumcluster IX,Clostridiumcluster XI,Clostridiumcluster XVIII, and others. 2.5. Anthropometrical Measurements The body mass index (BMI) was calculated as body weight/(height) 2 , as expressed in kg/m 2 . Bioelectrical impedance analysis using a body composition analyzer estiated the percentages of fat and muscle mass of the trunk and lower extremities (Nippon Shooter Ltd., Physion MD, Tokyo, Japan). 2.6. Physiological Performance Quadriceps muscle strength was assessed

and others. 2.5. Anthropometrical Measurements The body mass index (BMI) was calculated as body weight/(height) 2 , as expressed in kg/m 2 . Bioelectrical impedance analysis using a body composition analyzer estiated the percentages of fat and muscle mass of the trunk and lower extremities (Nippon Shooter Ltd., Physion MD, Tokyo, Japan). 2.6. Physiological Performance Quadriceps muscle strength was assessed using a strain gage dynamometer (ST-200S, MUL-TECH, Tokyo, Japan). Each subject performed two attempts on each leg and the maximum value of these

Nutrients2019,11, 868 5 of 17 four trials was marked for later analysis. The modi ed Kraus–Weber (K-W) test was used to assess trunk muscle strength [27]. This simple exercise test was based on the K-W Minimum test that was developed by Drs. Hans Kraus and Sonja Weber in the 1950s [28] to assess the strength and endurance of the trunk muscles. The trunk muscle strength of each subject was rated based on the total scores (full marks=40) of the test (Supplementary Figure S1). Four physical performance tests were conducted to evaluate motor ability and tness: maximal step length (MSL), Timed Up and Go (TUG) test, single-leg standing, and the 6-min. walk test (6MWT). MSL was determined as the maximum possible stride per step of a subject. In the TUG test, we measured the time that is required for a subject to stand up from a chair, walk 3 m, turn, walk back to the chair, and sit down. In single-leg standing, we measured the maximum time that a subject could stand on one leg. In case a subject continued single-leg standing for over 120 s, the test was discontinued. All of the functional tests were conducted twice and the best scores were marked for analysis. Cardiorespiratory tness was evaluated by the 6MWT according to the guidelines of the American Thoracic Society [29]. In short, he subjects were instructed to walk back and forth on a 25-m course as fast as possible for 6 min under the supervision of a medical doctor. They were permitted to stop and rest in case of fatigue. The investigator encouraged the subjects with routine phrases (e.g., “you are doing well” and “keep up the good work”) once per minute during the test. The total distance (in meters) walked after 6 min. (6MWD) was recorded and used as an indicator of cardiorespiratory tness, since performance on the 6MWD strongly correlates with peak oxygen uptake [30,31]. 2.7. Daily Physical Activity Level The parameters re ecting the daily physical activity level of the participants included the number of steps and the time spent in brisk walking—i.e., at an

The study investigates how aerobic exercise affects intestinal microbiota in elderly women.