Effect of L-citrulline, L-leucine, and multicomponent exercises on body compositions, physical 1 activity, and amino acid concentrations in older Japanese women with low body mass index: A 2 randomized double-blind placebo-controlled study 3

Background: The intake of citrulline (CIT) and leucine (LEU) can stimulate protein synthesis. However, the efficacy of the combined intervention of CIT and LEU intake with exercise on body 18 composition and physical activity (PA) remains unclear. This study aimed to investigate the combined effect of CIT and LEU intake and weight-bearing exercises (WBE) and square stepping exercise (SSE) 20 for 20 weeks on body composition, PA, and amino acid concentration in older women with low body 21 mass index (BMI) (16 to 21 kg/m 2 ). Methods: A total of 23 participants practiced WBE and SSE once a week for 75 minutes and were 23 administered supplement (Ex + CIT·LEU group: CIT 0.8 g and LEU 1.6 g; Ex + Placebo group: 3.5 g 24 carbohydrate) twice a day for 20 weeks. Body composition was measured using dual-energy X-ray 25 absorptiometry. PA, including leisure-time, household, and occupational PA, was assessed using the 26 Physical Activity Scale for the Elderly. Amino acid concentrations in the blood were analyzed by high-

Aging strongly affects the body's composition, particularly body cells, and bone and muscle mass 44 [1]. Muscle mass begins to decrease at age 25, further accelerating at age 50, and by 80 years when it 45 has decreased by an average of 40% compared with age 20 [2]. Asian women (-1.3 kg/m 2 ) and men (-46 1.4 kg/m 2 ) have a lower body mass index (BMI) than Europeans, which may lead to several 47 consequences [3]. A previous study reported that low BMI may lead to poor physical performance in 48 Japanese older adults [4]. In this population, abnormal BMI, either high or low, is a major cause of 49 mortality, and a previous study found a higher mortality rate in older adults with low BMI than in 50 those with normal BMI [5]. The low BMI of older individuals is strongly associated with sarcopenia, 51 and sarcopenia is paired with phenotypes such as decrease in muscle mass [6]. Sarcopenia has been 52 associated with disabilities in the performance of instrumental activities of daily living, mobility 53 limitation [7], poor physical functioning [8], high risk of falls or fractures [9], and mortality [10]. To 54 prevent the loss of body composition, the importance of exercise and a well-balanced diet have been 55 proven for Japanese adults with low BMI. The combined therapy of exercise and amino acid intake 56 has been proven to increase muscle protein synthesis [11] and leg muscle mass [12] in older adults. 57 Therefore, exercise programs such as weight-bearing exercises (WBE), which can be easily performed 58 without equipment at any time, are necessary for older adults. 59 Previous studies have shown that different exercise programs offer different benefits for older adults. 60 The WBE program has been proven to improve bone mineral density [13] and promote muscle 61 activation [14]. By contrast, the square stepping exercise (SSE), an aerobic exercise, has been reported 62 to improve both lower-extremity [15] and cognitive functioning [16] and prevent falls [17] in older 63 adults. Furthermore, the combined intervention of exercise and nutrition has shown more significance 64 in the improvement of the body composition of underweight older adults compared with the single use 65 of either one. 66 Leucine (LEU) is a well-known essential amino acid for protein metabolism and is composed of 67 branched-chain amino acids (BCAA), including isoleucine and valine. Additionally, a unique 68 characteristic of LEU is the stimulation of protein synthesis among BCAA. Through the activation of 69 the mammalian target of rapamycin (mTOR), LEU regulates the signaling pathway of insulin PI 3-70 kinase and stimulates the translational control of protein synthesis [18]. In addition, citrulline (CIT), 71 a non-protein amino acid, is a potent endogenous precursor of arginine. Direct intake of arginine can 72 cause gastrointestinal distress; however, intake of CIT can reduce gastrointestinal distress and promote 73 arginine production [19]. CIT is converted to arginine in the kidneys after ingestion, which is converted 74 into nitric oxide in the ornithine circuit [20]. Arginine increases nitric oxide causing vasodilatation, 75 which helps the circulation of both central and peripheral blood vessels, and increases growth hormone 76 secretion, which promotes protein anabolism and wound healing [21-23]. A previous animal study 77 showed that CIT intake increased protein synthesis and content in muscle [24]. Another study reported 78 that the intake of both CIT and LEU could stimulate muscle protein synthesis, which resulted from the 79 activation of the mTOR complex 1 (mTORC1) signaling pathway [25]. Therefore, the intake of both 80 CIT and LEU, combined with WBE and SSE, can be expected to prevent the decrease in body 81 composition of older adults with low BMI. 82 To our knowledge, no study has examined the effect of the combined intake of CIT and LEU with 83 exercise practice on older women with low BMI. We hypothesized that the simultaneous 84 administration of CIT and LEU with exercise practice may increase amino acid concentration in the 85 blood and improve the body composition and physical activity (PA) of older women. This study aimed 86 to examine the effect of multicomponent exercises (WBE and SSE) and amino acid intake (CIT and 87 LEU) on body compositions, PA, and amino acid concentrations of older Japanese women with low 88 BMI  were recruited through regional information magazines (Joyo Living Co., Ltd., Tsukuba, Japan). A 98 screening survey was conducted through telephone interview using self-reported, general health 99 questionnaires. The inclusion criteria were as follows: 1) age ranging from 65 to 80 years, 2) BMI 100 ranging from 16 to 21 kg/m 2 [3,26], 3) no exercise prohibition from doctors, and 4) independent 101 mobility and active participation in the exercise classes of the study. As regards the exclusion criteria, 102 participants were ineligible if they 1) take neurological medicine; 2) have a medical history of 103 comorbid diseases such as diabetes, brain, liver, kidney, heart, and peripheral vascular diseases; 3) 104 qualify to at least one question of the physical strength section (questions 6 to 10) of the "Kihon 105 Checklist" [27], which is a measure used to identify frailty; 4) have excessive intake of alcohol (>60 106 g/day) [28], 5) smoking, 6) have allergies to the supplements administered in this study; 7) have 107 undergone blood collection of more than 400 mL within 10 weeks, or 200 mL within 4 weeks, or have 108 donated blood in the past 2 weeks; and 8) have participated in other clinical studies in the past 10 109 weeks. A total of 43 older women applied for this study, but 15 applicants were excluded according to 110 the criteria; further, two applicants dropped out because of conflicting schedules. Eligible participants 111 were fully informed face-to-face about the study objectives, design, criteria of inclusion and exclusion, 112 intervention of exercise program and supplements, assessments, insurance compensation for injury, 113 withdrawal of consent and privacy protection. Finally, written consent to participation and data 114 publishing was obtained from 26 participants. 115

Study design 116
We commissioned THF Co., Ltd. to divide 26 participants into two groups by the simple 117 randomization method: 13 female participants were assigned to the Ex + CIT·LEU group (exercises 118 and L-CIT and L-LEU) and 13 to the Ex + Placebo group (exercises and placebo). First, random 119 numbers were generated by a computer and distributed to the participants. Random numbers less than 120 0.5 were determined as A (Ex + CIT·LEU group), and numbers greater than 0.5 were determined as B 121 (Ex + Placebo group). No difference in age and BMI was found between the two groups. However, 122 three participants in the Ex + CIT·LEU group withdrew their consent to participate because of personal 123 reasons. Finally, 23 participants were analyzed for this study (Fig. 1). To ensure the efficacy of the 124 double-blinding method, all data (class attendance records, distribution and collection of food, diary, 125 and measurement results) were managed by THF Co., Ltd. Members of each group remained 126 completely anonymous to both participants and researchers until the key codes were revealed after 127 completion of the trials at 20 weeks. Measurements were obtained at baseline (pre-intervention 128 measurement: Pre), after 10 weeks (mid-intervention measurement: Mid), and after 20 weeks (post-129 intervention measurement: Post). 130 131 Within the 20-week intervention period, the two groups equally performed WBE and SSE with the 132 guidance of a professional instructor once a week. Exercise sessions lasted for 75 min (10-min warm-133 up, 25-min SSE, 30-min WBE, and 10-min cool down). WBE consisted of two training patterns 134 (pattern A, five exercises on the chair including knee extension, knee raise, squat, heel raise, and 135 abdominal roll-up; pattern B, five exercises on the mat including pressing towel between palms, supine 136 bridge, side-lying leg raise, pressing towel between knees, and sit-up), and only one set of WBE was 137 performed with each exercise repeated 10 times. In addition, to reduce possible burden of this exercise 138 intervention, which lasts for 6 months, participants were required to practice pattern A or pattern B 139 alternately on a daily basis at home. Then, their performance was evaluated at four levels, i.e., 1, 140 certainly; 2, moderately; 3, slightly; 4, could not. Following this, their daily records were checked and 141 feedback provided once every 2 weeks. SSE, detailed description of which has been introduced in 142 previous studies [15], was practiced on a thin mat measuring 250 cm × 100 cm with 40 squares (25 × 143 25 cm). SSE contained step patterns of forward, backward, lateral, and oblique movements; in addition, 144 step patterns increasingly became more complex and were categorized into six levels: junior, basic, 145 semiregular, regular, senior, and master. Participants were required to memorize the patterns and then 146 step forward continuously without treading on the front and side lines of squares. 147 All participants were required to take 4.1 g of supplements (KYOWA HAKKO BIO Co., Ltd., Tokyo, 148 Japan) with 100 mL of water twice a day (8.2 g) for the duration of the trial. The 16.0 kcal total calorie 149 supplementation in the Ex + CIT·LEU group consisted of 0.8 g CIT, 1.6 g LEU, 0.3 g valine, 0.3 g 150 isoleucine, and 1.1 g carbohydrate. The 16.1 kcal total calorie supplementation in the Ex + Placebo 151 group consisted only of 0.3 g valine, 0.3 g isoleucine, and 3.5 g carbohydrate. These supplements were 152 distributed to all participants once every 2 weeks; their supplement intake was confirmed using self-153 reported diaries. 154 We calculated the participation rate of exercise classes, practice rate of WBE at home, and intake 155 rate of amino acid supplement. 156

Body compositions 157
We measured systolic and diastolic blood pressure plus heart rates (OMRON HEM-7111, Kyoto, 158 Japan). Participants were required to roll up their sleeves, and the sensor of the blood pressure monitor 159 was placed on the area of their left arm where the brachial artery was located; thereafter, systolic blood 160 pressure, diastolic blood pressure, and heart rate were measured. The unit of blood pressure was 161 millimeters of mercury (mmHg) and that of the heart rate was beats per minute (bpm). To measure 162 height, participants were required to stand on the height scale with bare feet and look straight. 163 Centimeter was used as the unit of height, and the value was specified to only one decimal place. As 164 for the body weight, a weighing scale (MC-980A, TANITA, Tokyo, Japan) was used and weight was 165 measured in kilograms. Participants were required to wear light clothes and take off their shoes before 166 measurement. Considering the weight of the clothes, 0.5 kg was subtracted from the obtained body 167 weight and specified to only one decimal place. BMI (kg/m 2 ) is calculated as body weight divided by 168 height squared. Using a dual-energy X-ray absorptiometry (DXA, QDR Discovery Wi, Hologic, Tokyo, 169 Japan) performed by a clinical technician, we obtained data regarding body fat, bone area, bone 170 mineral density, lean mass, fat mass, bone mass, and body mass (the sum of the lean mass, fat mass, 171 and bone mass) at baseline and after intervention. Participants were required to wear light clothes, take 172 off their shoes, and lie in a supine position on the DXA machine for 7 min. 173 Physical activity 174 PA was assessed using the Japanese version of the Physical Activity Scale for the Elderly (PASE), 175 the validity and reliability of which have been analyzed by previous studies [29]. PASE is a 12-item 176 questionnaire that measures the average time spent on daily physical activities during the past 7 days. 177 The 12 items comprised sections on leisure time PA (such as walking, recreational activities, and 178 strength training), household PA (such as home repair and garden maintenance), and occupational PA 179 (such as job and volunteer work). These items are weighted based on the intensity of each activity, and 180 the PASE (total PA) score is the sum of the 12 weighted items [29]. 181

Amino acid concentrations 182
We entrusted the analysis of blood samples to a microbiological institute (Kotobiken Medical  183 Laboratories Inc., Tsukuba, Japan). Participants were instructed to fast for 10 h in preparation for 184 venipuncture of the brachial veins. Blood was collected using the EDTA-2NA containing tubes by 185 nurses. To separate the plasma, 2 mL of blood was collected and centrifuged at 3000 rpm for 10 min 186 at 4°C, and 0.5 ml of plasma was stored at -80°C until analysis. Amino acid concentrations were 187 analyzed using high-performance liquid chromatography (HPLC) with an Acquity Ultra Performance 188 Liquid Chromatography with both bone mineral density (r = 0.66) and CIT change amount (r = 0.72). In the Ex + Placebo 255 group, significantly negative correlations were shown in the intake rate of supplementation with valine 256 (r = -0.59), and practice rate of WBE at home with both LEU (r = -0.70) and isoleucine (r = -0.64). 257 In addition, the participation rate of exercise classes showed significantly positive correlations with 258 arginine (r = 0.59), asparagine (r = 0.73), glutamine (r = 0.81), and histidine (r = 0.57), but 259 significantly negative correlations with height (r = -0.60), leisure time PA (r = -0.72), and LEU (r = 260 -0.57

Ex + CIT·LEU ( n = 10 ) Ex + Placebo ( n = 13 )
Note: Spearman rank correlation, † P < 0.05, ‡ P < 0.001 Change rate of all parameters were calculated using the formula (post intervention value -baseline value) / baseline value) *100 ) # Body fat and citrulline change amount were calculated using the formula (post intervention value -baseline value) the combination of exercise and nutrition intervention is more essential for older women with low 298 BMI to prevent sarcopenia. The efficacy of the combined intervention of exercise and nutrition is 299 controversial. Combined intervention of exercise and LEU-rich essential amino acid for 3 months was 300 reported to increase leg muscle mass in Japanese women with sarcopenia [12]. They defined 301 sarcopenia as BMI ≤22 kg/m 2 and found that exercise improved the muscle mass of older women with 302 low BMI, which is consistent with our study. In another study, supplement intake (6 g CIT maleate, 5 303 g creatine, 3 g LEU, 3 g isoleucine, 1.5 g valine, etc.) and resistance exercise practice for 4 weeks 304 significantly increased body mass and lean body mass compared with the placebo group (digestion-305 resistant maltodextrin) [40]. The participants of the previous study were recreationally active men, 306 whereas our study focused on older women with low BMI, but the effects of CIT and LEU intake and 307 exercise practice were consistent. However, other studies have reported conflicting results. A previous 308 study reported that a high dose of whey protein (45 g) intake, rather than combined intake of low dose 309 whey protein (15 g) and CIT (10 g), promoted postprandial muscle protein synthesis after resistance 310 exercise in older men [41]. Another study reported that CIT intake (0.18 g/kg/day) increased plasma 311 CIT and promoted arginine availability, but exerted no influence on LEU oxidation and whole-body 312 protein synthesis [42]. This finding contradicted a previous study reporting that CIT intake increased 313 arginine availability in the urea cycle, which promotes protein anabolism, indicating that CIT may 314 affect protein synthesis [24,25]. 315 These conflicting findings may result from various doses and durations of CIT and LEU intake. 316 According to a review study, an effective dose for daily CIT supplementation was recommended at a 317 minimum of 3 g to a maximum of 10 g [43]. The dose of CIT administered in our study was less than 318 that of the review study. In addition, the WHO recommended doses for daily intake of LEU (0.039 319 g/kg), isoleucine (0.020 g/kg), and valine ( that CIT and LEU supplements were well tolerated by the participants, and a long-term intake of CIT 331 and LEU effectively increased their plasma levels. In addition, the concentration of tyrosine 332 significantly increased in both groups, but the effect size was larger in the Ex + CIT·LEU group (d = 333 1.12) than in the Ex + Placebo group (d = 0.51). The concentration of phenylalanine increased 334 significantly only in the Ex + CIT·LEU group. Phenylalanine is an essential amino acid converted to 335 tyrosine during protein metabolism, which plays a vital role in transmitting signals between the brain 336 and nerve cells. After a mixed diet (49% carbohydrate, 22% protein, 29% lipid), leucine and 337 phenylalanine-tyrosine tracers are produced in the plasma or intracellularly increased whole-body 338 protein synthesis [45]. With reference to previous studies, LEU intake may have increased 339 phenylalanine and tyrosine concentrations in our study; however, this requires further studies. 340 Our study has several limitations. First, there was sampling bias and a relatively small number of 341 participants. Because our participants were Japanese older women with low BMI, the results may not 342 be generalizable to the population of older women with low BMI. Furthermore, 23 of 35 participants 343 were included in this study based on the exclusion criteria and personal reasons. Therefore, a study 344 with a larger population that adheres to the same intervention design is warranted. Second, a nutritional 345 survey was not conducted among the participants. Results of the main effect of time showed significant 346 differences in most variables of amino acid concentrations, indicating that in both groups, amino acid 347 levels in the plasma increased after the intervention (Pre < Post). However, this may have resulted 348 from the leniency toward the participant's lifestyle during the intervention. Diet, sleep, and other types 349 of exercises were not limited to verify the effect of the combined intervention without interfering with 350 the patient's usual lifestyle. Therefore, rather than diet limitation, it is necessary to conduct a 351 nutritional survey to investigate the increase in amino acid concentrations of the placebo group. Third, 352 the amount of PA was measured using the PASE questionnaire, which is a reliable tool to evaluate 353 different types of PA in the elderly [29]. However, reporting bias may exist, and other methods to 354 obtain objective data, such as using an accelerometer, should be considered in future research. 355

Conclusion 357
This study demonstrated that body weight, BMI, lean mass, bone mineral density, bone mass, 358 household and total PA, tyrosine, and phenylalanine were increased significantly after 20 weeks in the 359 Ex + CIT·LEU group. Our findings suggest that the combination practice of multicomponent exercises 360 (WBE + SSE) and intake of L-citrulline and L-leucine can improve body weight, muscle mass, and 361 PA in older women with low BMI, contributing to the prevention of sarcopenia and frailty. 362 363 List of Abbreviations 364