Effects of D-Glucosamine on the Growth of Human Gut-Dominant Microbiota in vitro and Bowel Movements in Healthy Individuals

Abstract

D-Glucosamine hydrochloride (GlcN), a monomer produced by the hydrolysis of chitosan, is a dietary supplement used worldwide to mitigate cartilage degeneration. Previous reports have shown that some dietary glucosamine migrates to the colon. However, the effect of glucosamine alone on colonic microbiota and bowel movements remains poorly understood. In this study, we evaluated the effect of glucosamine on the growth of 46 dominant human colonic bacterial species and 24 other important bacteria in vitro. Among the 70 gut bacterial species tested, the growth of 57 (81 %) was significantly enhanced by 0.5 g/L GlcN, with the most prominent growth activity (> 5-fold) observed in Anaerotruncus colihominis, Pseudoflavonifractor capillosus, and Roseburia hominis. These results indicate that a wide range of the tested gut bacteria can utilize GlcN, similar to the effect of conventional dietary fiber in improving bowel function. Next, we conducted an open-label, single-arm trial involving 29 healthy individuals to determine the effects of 1,500 mg GlcN/day, a commonly used dose. Stool color significantly changed during the 2 weeks of GlcN intake from brown to ocher (p < 0.01), suggesting enhanced colonic fermentation. The stool odor and the sensation of incomplete evacuation improved significantly (p < 0.05). Numerical measurements of bowel movements revealed significant increases in stool volume, defecation frequency, and the number of days of defecation during GlcN intake (p < 0.001). Thus, dietary glucosamine may stimulate gut microbiota growth in the colon and promote bowel movements. This study was registered with the University Hospital Medical Information Network (regd. no. UMIN000056757).

Abbreviations

GAM, Gifu anaerobic medium; GB medium, GAM and blood medium; GAM-wos, GAM without sugar medium; GlcN, glucosamine hydrochloride; SCFA, short-chain fatty acid; ATCC, American Type Culture Collection; DSMZ, German Collection of Microorganisms and Cell Cultures GmbH; JCM, Japan Collection of Microorganisms; OD₆₀₀, optical density of 600 nm

INTRODUCTION

D-Glucosamine (2-amino-2-deoxy-D-glucose) is a fundamental component of chitosan and chitin, which are naturally produced by arthropods [1]. Industrially, D-glucosamine hydrochloride (GlcN) is produced as a dietary supplement by hydrolyzing crustacean exoskeletons, which are primarily composed of chitin [2]. GlcN is commonly consumed worldwide as a supplement to alleviate symptoms of osteoarthritis and joint degeneration. However, a meta-analysis has found that GlcN may not be effective in treating joint disorders [3]. Therefore, the anti-osteoarthritis benefits of GlcN remain controversial. Despite this, many people continue to use GlcN as a dietary supplement. A large-scale epidemiological study of supplement users found that GlcN consumption was associated with reduced overall mortality [4, 5]. The results of this study are consistent with findings from animal experiments in mice and nematodes [6, 7]. Nevertheless, the molecular mechanisms behind GlcN's potential benefits on longevity remain unknown.

After ingestion, GlcN undergoes partial absorption in the small intestine, with a significant portion remaining unmetabolized and reaching the large intestine [8]. Once in the colon, GlcN may interact with gut microbiota, potentially serving as a substrate for microbial fermentation [9]. This process has been suggested to influence the composition and metabolic activity of intestinal microbiota, leading to the production of short-chain fatty acids (SCFAs). SCFAs play crucial roles in maintaining the intestinal barrier integrity, modulating immune responses, supporting overall metabolic health, and promoting bowel movements [10]. For this reason, it is possible that GlcN contributes to reduced mortality through the intestinal microbiota [11]; however, the details remain unknown.

Several studies have reported that GlcN supplementation may improve bowel movements, further linking its effects to gut microbiota modulation [12]. However, a common limitation of these reports is the relatively small number of study participants, despite significant differences in the intestinal microbiome of each participant, which could restrict the generalizability of the findings. In this study, we cultured and used the most dominant gut commensal microbial species from the large populations of Japan [13] and Europe [14], whose strains could be cultured simultaneously in vitro. We first evaluated the effects of GlcN on the growth of 70 human-dominant colonic bacterial species and other important human bacteria in vitro. Furthermore, we conducted an open-label single-arm trial on the effects of GlcN on bowel improvement in a sufficient number of healthy individuals, more than double the number of participants previously reported.

MATERIALS AND METHODS

In vitro assay for growth of human gut bacterium:

1. Chemicals.

All chemicals including GlcN (not mentioned in the following sections) were purchased from Fujifilm-Wako Co., Ltd. (Osaka, Japan).

2. Microbe strains.

Bacteria were obtained from the American Type Culture Collection (ATCC) (Manassas,VA, USA), German Collection of Microorganisms and Cell Cultures GmbH (DSMZ) (Braunschweig, Germany), and Japan Collection of Microorganisms (JCM; Table 1) (Tsukuba, Japan) [15, 16]. Bacteria were cultured at 37 °C in an anaerobic chamber (10 % CO₂, 10 % H₂, and 80 % N₂; InvivO₂ 400; Ruskinn Technology, Ltd., Bridgend, UK).

Table 1. Details of dominant and important human gut bacteria used in this study.

Used in this study	Occupancy rank of dominant bacterium		Bacterial species	Used strain	Note
	European	Japanese
✓	1	10	Bacteroides uniformis	JCM 5828T*
✓	3	25	Parabacteroides merdae	JCM 9497T
✓	4	12	Dorea longicatena	DSM 13814T
✓	5	-	Ruminococcus bromii	ATCC 27255T
✓	6	-	Bacteroides caccae	JCM 9498T
✓	8	49	Bacteroides thetaiotaomicron	JCM 5827T
✓	10	20, 26	Ruminococcus torques	ATCC 27756T
✓	13	23	Faecalibacterium duncaniae	JCM 31915	Faecalibacterium prausnitzii
✓	14	50	Ruminococcus lactaris	ATCC 29176T
✓	15	9	Collinsella aerofaciens	JCM 7790
✓	16	18	Dorea formicigenerans	ATCC 27755T
✓	17	13	Bacteroides vulgatus	JCM 5826T
✓	18	37	Roseburia intestinalis	DSM 14610T
✓	21	16	Parabacteroides distasonis	JCM 5825T
✓	23	42	Bacteroides ovatus	JCM 5824T
✓	26	5	Eubacterium rectale	JCM 17463
✓	27	-	Bacteroides xylanisolvens	JCM 15633T
✓	28	39	Coprococcus comes	ATCC 27758T
✓	31	-	Eubacterium ventriosum	ATCC 27560T
✓	32	22	Phocaeicola dorei	JCM 13471T	Bacteroides dorei
✓	33	19, 34	Blautia obeum	DSM 25238T	Ruminococcus obeum
✓	34	-	Subdoligranulum variabile	DSM 15176T
✓	35		Pseudoflavonifractor capillosus	ATCC 29799T
✓	38		Holdemania filiformis	DSM 12042T
✓	39		Bacteroides stercoris	JCM 9496T
✓	42		Bacteroides eggerthii	JCM 12986T
✓	43		Butyrivibrio crossotus	DSM 2876T
✓	44		Bacteroides finegoldii	JCM 13345T
✓	45		Parabacteroides johnsonii	JCM 13406T
✓	47	28	Clostridium nexile	ATCC 27757T
✓	49		Anaerotruncus colihominis	JCM 15631T
✓	50	14	Ruminococcus gnavus	ATCC 29149T
✓	51		Bacteroides intestinalis	JCM 13265
✓	52	33	Bacteroides fragilis	JCM 11019T
✓	53		Clostridium asparagiforme	DSM 15981T
✓	54		Enterococcus faecalis	ATCC 700802
✓	55		Clostridium scindens	JCM 6567T
✓	56		Blautia hansenii	JCM 14655T
✓		11	Anaerostipes hadrus	DSM 3319T
		7	Bifidobacterium adolescentis	JCM 1275T	Important bacterium (see below)
		3	Bifidobacterium longum	JCM 1217T	Important bacterium (see below)
		4	Bifidobacterium pseudocatenulatum	JCM 1200T	Important bacterium (see below)
		35	Clostridium bolteae	JCM 12243T	Important bacterium (see below)
✓		41	Clostridium innocuum	JCM 1292T
✓		43	Coprococcus catus	ATCC 27761T
✓		45	Enterocloster clostridioformis	JCM 1291T	Clostridium clostridioforme
✓		24	Flavonifractor plautii	ATCC 29863T
✓		46	Roseburia hominis	JCM 17582T
✓		27	Roseburia inulinivorans	DSM 16841T
✓		30	Streptococcus salivarius	JCM 5707T
Human important bacterium
✓		-	Lactobacillus casei subsp. casei	JCM1134T	Lactic acid bacterium
✓		-	Lactobacillus casei subsp. rhamnosus	ATCC7469T	Lactic acid bacterium
✓		-	Lactobacillus curvatus	KP3-4	Lactic acid bacterium#
✓		-	Lactobacillus paragasseri	JCM1130	Lactic acid bacterium
✓		-	Lactobacillus johnsonii	JCM8794	Lactic acid bacterium
✓		-	Lactobacillus plantarum	JCM1149	Lactic acid bacterium
✓		-	Lactobacillus reuteri	JCM1112T	Lactic acid bacterium
✓		-	Lactococcus lactis	JCM1158	Lactic acid bacterium
✓		7	Bifidobacterium adolescentis	JCM1275T	Bifidobacterium
✓		-	Bifidobacterium animalis subsp. lactis	JCM10602T	Bifidobacterium
✓		-	Bifidobacterium bifidum	JCM1254	Bifidobacterium
✓		-	Bifidobacterium breve	JCM1192T	Bifidobacterium
✓		-	Bifidobacterium catenulatum	JCM1194T	Bifidobacterium
✓		-	Bifidobacterium infantis	ATCC15697T	Bifidobacterium
✓		3	Bifidobacterium longum	JCM1217T	Bifidobacterium
✓		4	Bifidobacterium pseudocatenulatum	JCM1200T	Bifidobacterium
✓		-	Bifidobacterium pseudolongum	JCM1205T	Bifidobacterium
✓		35	Clostridium bolteae	JCM12243T	Butyrate producing bacterium
✓		-	Clostridium indolis	JCM1380T	Butyrate producing bacterium
✓		-	Clostridium ramosum	JCM1298T	Butyrate producing bacterium
✓		-	Clostridium difficile	JCM 1296T	Harmful bacterium
✓		-	Clostridium perfringens	JCM 1290	Harmful bacterium
✓		-	Fusobacterium nucleatum subsp. nucleatum	JCM 8532T	Harmful bacterium
✓		-	Akkermansia muciniphila	JCM30893	Beneficial bacterium

We used the most dominant gut commensal microbial species of the Japanese [13] and European [14] populations. *Type strains are indicated by a superscript “T.”

^#This strain of lactic acid bacterium was previously isolated as a high producer of polyamine, a beneficial bioactive compound from turnip sushi [16].

3. Microbe medium.

The procedure for the preparation of GB medium (Gifu anaerobic medium (GAM) and blood medium) has been reported in detail previously [15]. Briefly, AccuDia^TM GAM Broth (Shimadzu Corporation, Kyoto, Japan) was completely dissolved in water at 95 % of the volume specified in the manufacturer's instructions and autoclaved, placed in a closed container together with Anaeropack Kenki (Mitsubishi Gas Chemical Co., Ltd., Tokyo, Japan), and allowed to stand overnight to remove dissolved oxygen. Horse blood (horse whole-blood, defibrinated and sterile; Nippon Bio-Supp. Center, Tokyo, Japan) stored anaerobically with Anaeropack Kenki was added to the GAM at 5 % (v/v) in an anaerobic chamber.

AccuDia^TM GAM Semisolid without Dextrose (Shimadzu Corporation) was dissolved following the manufacturer's instructions and filtered to remove agar [17]. Hereafter, this medium is referred to as GAM-wos (GAM without sugar medium). After autoclave sterilization, the medium was immediately placed in a closed container with Anaeropack Kenki and allowed to stand overnight to remove oxygen.

4. Microbe culture.

Bacteria were cultured at 37 °C in an anaerobic chamber (10 % CO₂, 10 % H₂, and 80 % N₂; InvivO₂ 400; Ruskinn Technology, Ltd.).

First, 5 µL of thawed frozen cells of the bacterial strains in 18 % glycerol at −80 °C was added to 500 µL of GB medium in a deep 96-well plate (Thermo Fisher Scientific Inc., Waltham, MA, USA) to inoculate each species, and pre-culture was performed at 37 °C under anaerobic conditions. For pre-culturing in vials, 500 µL of the pre-culture solution was transferred to a deep 96-well plate before using a copy stand. Approximately 2 µL of the respective culture collection was inoculated in 500 µL of GAM-wos or GAM-wos supplemented with 0.5 g/L GlcN in another deep 96-well plate using a copy plate 96 (Tokken, Inc., Chiba, Japan) and a copy plate stand (Tokken, Inc.). After 48 h of anaerobic incubation, growth was measured at an optical density of 600 nm (OD₆₀₀). For OD₆₀₀ determination, cultures on 96 deep-well plates were completely resuspended, then diluted 4-fold with PBS on a new 96 plate. We added 150 µL of PBS to 50 µL of culture medium to create a 4-fold diluted solution (200 µL), which was then measured in a 96-well plate using a plate reader. For the 96-well plate measurement, the optical path length was 6.5 mm. The diluted samples were measured using a Thermo Scientific^TM Multiskan^TM GO instrument (Thermo Fisher Scientific Inc.). A 96-well electric pipettor (VIAFLO 96; INTEGRA Biosciences AG, Zizers, Switzerland; Product No. 6001) was used for handling cultures on both the 96 deep-well plate and the 96 plate. The value measured at 600 nm for the uninoculated medium, similarly diluted 4-fold with PBS, was used as the blank. This blank value was subtracted from the value measured at 600 nm for cultures, and the result was converted to a value for a 1 cm optical path length to obtain the OD₆₀₀. We derived the result values from comparing the 96-well plate measurement values with those obtained using a cuvette with a 1 cm optical path length in advance.

5. Evaluation of GlcN on the growth of gut microbiome in vitro.

The growth of the most dominant and important intestinal bacteria in the intestinal microbiota of Westerners and Japanese people cultured in GAM-wos supplemented with 0.5 g/L GlcN was substituted to calculate the growth promotion and inhibition effects of GlcN on intestinal bacterial species [18]. Briefly, after 48 h of anaerobic incubation, growth was measured at an OD₆₀₀, mentioned above in detail. To compare the ability of the bacterial species to utilize GlcN, the OD₆₀₀ ratio was obtained by dividing the bacterial growth (OD₆₀₀) in GAM-wos supplemented with 0.5 g/L GlcN by that of GAM-wos.

All outcomes were presented as mean ± SD. Statistical analyses were performed using a two-sided Student's t-test, and the statistical significance was set at 5 %. The software used was Microsoft Excel 2010 (Microsoft Japan Co., Ltd., Tokyo, Japan).

Single arm trial on bowel movement in healthy individuals:

1. Study design, ethics, and participants.

This was an open-label, single-arm trial. The Shiba Palace Clinic Institutional Review Board approved the study protocol on January 16, 2025 (approval no. 155708_te-37811). The study was conducted with full consideration of medical ethics and in accordance with the Declaration of Helsinki (2013) [19] and the Ethical Guidelines for Medical and Health Research Involving Human Subjects. Testing was conducted by SOUKEN Co., Ltd. (Tokyo, Japan). This study was registered with the University Hospital Medical Information Network (regd. no. UMIN000056757). We performed a power analysis to estimate the required sample size for clinical trials. Using R software (version 4.3.1) and the pwr package (version 1.3-0), we conducted the analysis for a two-tailed paired-samples t-test, assuming a medium effect size (Cohen's d = 0.5) and a significance level of α = 0.05. The analysis indicated that a sample size of 34 participants would be needed to achieve a statistical power of 80 %.

2. Screening of participants.

A follow-up flowchart of the open-label, single-arm trial is shown in Fig. 1. The study participants were publicly recruited. A total of forty-nine individuals who agreed to participate were selected from a public database. A preliminary questionnaire was administered to those who provided written informed consent confirming their wish to participate in the study. Constipation-prone individuals aged 20-59 years with a mean frequency of bowel movements of about 3-5 times per week were enrolled in the study. Among them, those who did not meet the following 9 exclusion criteria were selected for participation in the study: 1) Those taking medicines that may affect the test results (medicines with intestinal regulating effects, antibiotics, etc.); 2) Those who regularly consumed health foods that may affect the test results (lactic acid bacteria, oligosaccharides, supplements that claim to regulate the intestines, etc.) or health foods that contain the same main ingredients as the test product; 3) Pregnant or potentially pregnant and breastfeeding women; 4) Those with alcoholism; 5) Those who may have allergic reactions to the test product ingredients; 6) Those participating in other clinical trials; 7) Those with a history of severe liver damage, kidney damage, or heart disease; 8) Those with a history of hepatitis or current illness; and 9) Those with severe anemia. After screening 40 participants, 29 underwent an intake test. The mean age of the 29 participants (22 women, 7 men) was 42.9 ± 9.8 years (mean age: women = 42.9 ± 9.9 years, men = 43.3 ± 10.4 years).

Fig. 1. The follow-up flow chart of the open label single arm trial.

　The study participants were publicly recruited. Forty-nine individuals who agreed to participate were selected from a public database. A preliminary questionnaire was administered to those who provided written informed consent confirming their wish to participate in the study. Constipation-prone persons aged 20-59 years or younger with a mean frequency of bowel movements of about 3-5 times per week were enrolled in the study. Those who did not meet the exclusion criteria were included in the study.

3. Subject management information.

Subjects were instructed to maintain their regular lifestyle throughout the study. All 29 subjects were instructed daily to maintain the same living environment (sleep, diet, and general lifestyle) as they had before the study began during the four-week period. Compliance with these instructions, along with test product intake status, was monitored through a daily intake diary. In the daily intake diary, subjects were asked to answer “yes” or “no” whether they had been able to maintain the same living environment (sleep, diet, general lifestyle) during the study period as they had before the study began. If they answered “no,” they were asked to write down the details of the living environment management in a free-form section provided in the diary. All subjects answered these questions once a day along with their intake of the test foods. As a result, no subjects reported failing to comply with the instructions to maintain the same living environment as they had before the study began during the four-week period.

4. Test food and intake.

The test food consisted of a tablet containing GlcN hydrochloride (Houkouen Seiyaku Co., Ltd., Kagawa, Japan). The tablets contained 1,500 mg GlcN hydrochloride, 312 mg starch, 300 mg cellulose, 240 mg sucrose fatty acid ester, and 48 mg silicon dioxide. The participants ingested the GlcN (1,500 mg) tablets each day with normal or warm water. The study schedules included a no-intake period (2 weeks) and an intake period (2 weeks). All 29 participants completed the study. Twenty-four participants had 100 % compliance; the remaining five had 92 % compliance. Data from all 29 participants were included in the final analysis.

5. Bowel movement questionnaire.

Participants maintained a bowel diary throughout the 4-week study to record bowel movements and assess defecation and constipation symptoms. The bowel diary recorded the presence and status of bowel movements, including both defecation and constipation-related items. The procedure is described in detail elsewhere [20] (Fig. 1).

6. Defecation items assessed on a scale.

The bowel movement diary of the four defecation items (stool characteristics, color, odor, and residual stool sensation) was assessed on a scale for each item. The stool characteristics were scored using the Bristol stool scale [21]. It classified the stool form into seven scales: 1) “Separate hard lumps, like nuts,” 2) “Sausage-shaped, but lumpy,” 3) “Like a sausage but with cracks on its surface,” 4) “Like a sausage or snake, smooth and soft,” 5) “Soft blobs with clear cut edges,” 6) “Fluffy pieces with ragged edges, a mushy stool,” and 7) “Watery, no solid pieces, entirely liquid.” Each participant selected a score corresponding to the stool, and the scores for each answer were tallied. The closer it was to 4, the more likely it was for the patient to have a better bowel movement status.

Stool color was assessed on a 6-point scale (1 = yellow, 2 = light ocher, 3 = ocher, 4 = brown, 5 = dark brown, and 6 = dark brown almost black), the scores for each answer were tallied. Stool odor intensity was assessed on a 5-point scale (1 = very weak, 2 = weak, 3 = normal, 4 = strong, and 5 = very strong), the scores for each answer were tallied. Residual Stool Sensation was assessed on a 4-point scale (1 = no sensation, 2 = almost no sensation, 3 = a little sensation, and 4 = full sensation). The higher the scores for color, odor, and residual stool sensation, the more likely it was for the patient to have a worse bowel movement status. The score of all 4 items (stool characteristics, color, odor, and residual stool sensation) was assessed as the score for 2 weeks (average of the 2-week period).

7. Constipation items assessed on a value.

The bowel movement diary of the four items (amount of stool, frequency of defecation and flatulence, and days of defecation) was subjectively assessed for each item, and the higher the score, the less likely it was to be constipation. The amount of stool per week was assessed as the estimated number (average of the 2-week period) of chicken eggs (large) converted from the amount of stool. The frequency of defecation (average of the 2-week period) was assessed as the number of defecation events per week. The frequency of flatulence per week was assessed as the number (average of the 2-week period) of flatulence events in the tested 2 weeks. The number of defecation days was assessed as the number (average/week) of days for defecation events per week.

8. Statistical analysis.

All outcomes were presented as median values and mean ± SD. All items were examined using participant testing in the pre-ingestion-2 week and post-ingestion-2 week periods. All statistical analyses were performed using two-sided testing, and the statistical significance was set at p < 0.05.

For the scales of four defecation items and the values of four constipation items, a paired Student's t-test was performed. The Microsoft Excel 2010 (Microsoft Japan Co., Ltd.) was used for all the analyses.

RESULTS

Effects on gut microbe growth by GlcN.

The growth values of the most dominant and important intestinal bacteria in the intestinal microbiota of Westerners and Japanese people, which were cultured in GAM-wos and GAM-wos supplemented with 0.5 g/L GlcN, were calculated. To investigate the effect of existing prebiotics on the growth of human gut microbiota, we cultured beneficial, pathogenic, and prominent bacteria (Table 1) in GAM-wos supplemented with GlcN. The prominent bacteria used in this test included 46 dominant species that could be cultured in GAM-wos.

Effect of the addition of 0.5 g/L GlcN to GAM-wos on the growth of the most dominant species of human intestinal microbiota was observed (Fig. S1; see J. Appl. Glycosci. Web site). The addition of 0.5 g/L GlcN to GAM-wos promoted the growth of 42 of the 50 most dominant species of human intestinal microbiota (Anaerostipes hadrus, Anaerotruncus colihominis, Bacteroides caccae, Bacteroides dorei, Bacteroides eggerthii, Bacteroides finegoldii, Bacteroides fragilis, Bacteroides intestinalis, Bacteroides ovatus, Bacteroides stercoris, Bacteroides thetaiotaomicron, Bacteroides uniformis, Bacteroides vulgatus, Bacteroides xylanisolvens, Blautia hansenii, Butyrivibrio crossotus, Clostridium asparagiforme, Clostridium clostridioforme, Clostridium innocuum, Clostridium nexile, Collinsella aerofaciens, Coprococcus comes, Dorea longicatena, Enterococcus faecalis, Eubacterium rectale, Eubacterium ventriosum, Faecalibacterium prausnitzii, Holdemania filformis, Parabacteroides johnsonii, Parabacteroides merdae, Pseudoflavonifractor capillosus, Roseburia hominis, Roseburia intestinalis, Roseburia inulinivorans, Ruminococcus gnavus, Ruminococcus lactaris, Ruminococcus obeum, Ruminococcus torques, Streptococcus salivarius, Subdoligranulum variabile, Bifidobacteorium adolescentis, Bifidobacterium longum, Clostridium bolteae) (Fig. S1; see J. Appl. Glycosci. Web site). The addition of 0.5 g/L GlcN to GAM-wos promoted the growth of 17 of the 24 important species (seven out of eight lactic acid bacteria species (Lactobacillus casei subsp. casei, Lactobacillus casei subsp. rhamnosus, Lactobacillus curvatus, Lactobacillus paragasseri, Lactobacillus johnsonii, Lactobacillus plantarum, Lactococcus lactis), 5 out of 9 bifidobacterial species (Bifidobacterium bifidum, Bifidobacterium breve, Bifidobacterium infantis, Bifidobacterium longum, Bifidobacterium pseudocatenulatum), 3 out of 3 butyric acid-producing bacteria species (Clostridium bolteae, Clostridium indolis, Clostridium ramosum), and 2 out of 3 harmful bacteria species (Clostridium perfringens and Fusobacterium nucleatum subsp. nucleatum) (Fig. S2; see J. Appl. Glycosci. Web site).

The growth data were substituted into the above formula to calculate the growth promotion and inhibition effects of GlcN on the bacterial species (Fig. 2). Of the total 70 bacterial species tested, the growth of 57 (81 %) was statistically significantly promoted by 0.5 g/L GlcN. Especially, the growth of A. colihominis, P. capillosus, and R. hominis was promoted more than five-fold by 0.5 g/L GlcN (Fig. 2). In contrast, the growth of six species (9 %; Clostridium scindens, Coprococcus catus, Dorea formicigenerans, Lactobacillus reuteri, Bifidobacterium adolescentis and Clostridium difficile) was statistically significantly suppressed by 0.5 g/L GlcN.

Fig. 2. Evaluation of the effect of GlcN on growth of the 70 human gut important bacteria species.

　The growth of the most dominant and important intestinal bacteria in the intestinal microbiota, cultured in GAM-wos supplemented with 0.5 g/L GlcN, was used to calculate the growth promotion and inhibition effects of GlcN on intestinal bacterial species. Briefly, after 48 h of anaerobic incubation, growth was measured at an optical density of 600 nm (OD₆₀₀). To compare the ability of the bacterial species to utilize GlcN, the OD₆₀₀ ratio was obtained by dividing the bacterial growth (OD₆₀₀) in the GAM-wos supplemented with 0.5 g/L GlcN by that without GAM-wos. All outcomes were presented as mean ± SD.

Effects on human bowel movements by GlcN.

Next, we conducted an open-label, single-arm trial on bowel movements in 29 healthy individuals to determine the effects of the most common dose [22, 23]. The bowel movement diary of the four defecation items (stool characteristics, color, odor intensity, and residual stool sensation) was assessed using the respective scales (Table 2).

Table 2. Defecation items assessed on scales.

Parameter	Pre	Post	Difference	p value
Stool characteristics	3.60 ± 0.87	3.75 ± 0.76	+0.15 ± 0.59	p = 0.200
Stool colors	4.14 ± 0.54	3.93 ± 0.43	−0.20 ± 0.34	p = 0.004＊＊
Stool odor intensity	3.25 ± 0.34	3.09 ± 0.31	−0.16 ± 0.34	p = 0.020＊
Residual stool sensation	2.46 ± 0.56	2.33 ± 0.54	−0.13 ± 0.29	p = 0.025＊

Stool characteristics, color, odor, and residual stool sensation were assessed in 29 healthy participants during the 2-week pre-intake (Pre) and during the 2-week GlcN-intake (Post). Stool characteristic (Bristol stool scale) classified the stool form into 7 scales, from “Separate hard lumps, like nuts,” to “Watery, no solid pieces, entirely liquid.” Each answer was tallied. The closer it is to 4, the more likely the patient is to have a better bowel movement status. Stool color was assessed on a 6-point scale (from 1 = yellow to 6 = dark brown to almost black), and the scores were tallied. Stool odor intensity was assessed on a 5-point scale (1 = very weak, 5 = very strong), and the scores for each answer were tallied. Residual stool sensation was assessed on a 4-point scale (1 = no sensation to 4 = full sensation). The difference was calculated by subtracting the Pre score from the Post score.

All outcomes were presented as mean ± SD.

Statistical analysis was performed using the paired Student's t-test before and during the 2 weeks of intake.

^＊: p < 0.05, ^＊＊: p < 0.01

The individual effects of dietary GlcN on the four defecation items are shown in Table S1 (see J. Appl. Glycosci. Web site). Stool characteristics were assessed using the Bristol stool scale. There was no difference between the stool characteristics during the 2-week pre-intake and during the GlcN-intake (3.60 ± 0.87 and 3.75 ± 0.76, respectively; p = 0.200) (Table 2). The color of the stool was significantly different from that during the 2-week pre-intake and during the GlcN-intake (4.14 ± 0.54 and 3.93 ± 0.43, respectively; p = 0.004) (Table 2). There was a significant difference in odor between the stools during the two periods (3.25 ± 0.34 and 3.09 ± 0.31, respectively; p = 0.020) (Table 2). Residual stool sensation was significantly different between the two experimental periods (2.46 ± 0.56 and 2.33 ± 0.54, respectively; p = 0.025) (Table 2).

The individual effects of dietary GlcN on the four constipation parameters are shown in Table S2 (see J. Appl. Glycosci. Web site). Numerical measurements of the four constipation items (amount of stool, frequency of defecation and flatulence, and days of defecation) were assessed in the 29 participants (Table S2; see J. Appl. Glycosci. Web site). The stool amount was significantly different between that during the pre-intake 2-week period and during the intake of GlcN (9.3 ± 5.0 and 13.6 ± 7.3, respectively; p < 0.001) (Table 3). The frequency of defecation was significantly different between the two periods (4.05 ± 1.26 and 5.59 ± 1.55, respectively; p < 0.001) (Table 3). The frequency of farts tended to increase with GlcN intake, but the difference was not statistically significant (p = 0.059). The days of defecation were significantly different between the two experimental periods (3.69 ± 1.06 and 4.97 ± 1.03; p < 0.001) (Table 3).

Table 3. Constipation items assessed by values.

Parameter	Pre	Post	Difference	p value
Amount of stool	9.3 ± 5.0	13.6 ± 7.3	+4.3 ± 4.3	p < 0.001＊＊＊
Frequency of defecation	4.05 ± 1.26	5.59 ± 1.55	+1.53 ± 1.44	p < 0.001＊＊＊
Frequency of flatulence	36.4 ± 25.0	42.3 ± 25.8	+5.9 ± 15.9	p = 0.059
Number of days of defecation	3.69 ± 1.06	4.97 ± 1.03	+1.28 ± 1.08	p < 0.001＊＊＊

The amount of stool, frequency of defecation and flatulence, and number of days of defecation were assessed in 29 healthy participants during the 2-week pre-intake (Pre) and during the 2-week GlcN-intake (Post). The amount of stool per week was assessed as the estimated number (average of the 2 weeks) of large chicken eggs (large size) converted from the stool samples. The frequency of defecation (average of the 2 weeks) was assessed as the number of defecation events per week. The frequency of flatulence was assessed as the number (average of the 2 weeks) of flatulence events in the 2 weeks tested. The number of days of defecation per week was assessed as the number (average of the 2 weeks) of days of defecation events in 2 weeks. The difference was calculated by subtracting the Pre value from the Post value.

All outcomes were presented as mean ± SD.

Statistical analysis was performed using the paired Student's t-test before and during the 2 weeks of intake.

^＊＊＊: p < 0.001

DISCUSSION

We found that a wide range of tested gut bacteria utilized GlcN in vitro. Conventionally available prebiotics, such as raffinose, 1-kestose, lactulose, galacto-oligosaccharides, and fructo-oligosaccharides, have been reported to promote the growth of a wide range of gut bacteria using similar assay systems [17]. On the other hand, a new prebiotic disaccharide, D-galactosyl-β1→4-l-rhamnose, selectively enhanced the growth of Bifidobacterium and specifically suppressed the growth of the harmful bacterium Clostridioides difficile [17]. Thus, D-galactosyl-β1→4-l-rhamnose was thought to become a next generation prebiotic, in contrast to conventional prebiotics that promote the growth of a broad spectrum of gut bacteria. Considering the above, in this study GlcN appears to function similarly to that of conventional prebiotics and not like next generation prebiotics.

Among the 70 human gut bacteria tested in this study, the greatest growth activity (over 5-fold) was shown by A. colihominis, P. capillosus, and R. hominis (Fig. 2). Anaerotruncus colihominis is a butyrate producer that ameliorated experimental autoimmune encephalomyelitis, which is associated with the induction of regulatory T cells in the lymph, in a mouse model [24]. Pseudoflavonifractor capillosus is a gram-negative, non-sporulating species belonging to the Clostridium cluster IV [25]. This cluster includes species that produce acetate and butyrate [26]. Roseburia hominis is also a butyrate producer [27]. Roseburia hominis, a host gut colonizer, shows upregulation of its bacterial genes involved in metabolism and motility; in addition, its colonization upregulates host genes related to antimicrobial peptides, gut barrier function, and Toll-like receptor signaling [28]. Roseburia hominis was also recently reported to be depleted in stool samples of obese individuals compared to lean controls, and its abundance was negatively correlated with body mass index and serum triglycerides [29]. All three species are related to the production of SCFAs, such as butyrate and acetic acid.

A previous clinical report of 10 individuals showed that dietary 1,500 mg GlcN and 1,200 mg chondroitin increased the abundance of four Lachnospiraceae genera, two Prevotellaceae genera, and Desulfovibrio, compared to the placebo [30]. This study included chondroitin as a possible confounding factor in the test substance, making it difficult to interpret the effects of GlcN on gut bacteria. Another clinical report with a sample size of 11 individuals showed that dietary 3,000 mg GlcN, which was double the dose used in this trial, showed a trend towards reducing constipation with no significant increase in distinctive genera [12]. Rather, in the study, an indicator of phylogenetic diversity and the proportions of Pseudomonadaceae, Peptococcaceae, and Bacillaceae were significantly reduced after GlcN intake and did not affect SCFA production. Thus, a few studies have explored the potential relationship between GlcN intake and changes in gut microbiota composition; however, the details of the effects of GlcN on the gut microbiota and bowel movements in humans are still unknown. To address the limitations of previous studies [12, 30], we employed a single-arm design, providing a larger sample size than that of previously reported studies [12, 30]. As a results of this study, although the ideal sample size was 34, a total of 49 candidates applied to participate, and 29 individuals were selected after screening (see MATERIALS AND METHODS). The achieved statistical power with this sample size (29 participants) was approximately 73 %, which is slightly below the conventional threshold of 80 %. Nonetheless, we considered this power sufficient to detect medium-sized effects with reasonable accuracy. The increased number of participants may enhance the statistical power of the analysis, allowing for a more comprehensive assessment of the effect of supplementation with GlcN alone, at a normal dose of 1,500 mg/day, on bowel movements.

In human trials, the Bristol stool scale score is associated with gut microbiota richness and composition [31]. In this study, the stool statistics evaluated using the Bristol stool scale did not change during GlcN intake (Table 2); therefore, the gut microbiota richness and composition might not change. However, it is possible that GlcN increases the number of total bacteria or specific bacteria without changing their richness. An increase in bacterial count in vitro may lead to an increase in fecal volume in clinical trials (Table 3). In addition, the significant increase in SCFA-producing bacteria observed in vitro may have led to accumulating SCFA [32]. Conventional prebiotics such as inulin as use as symbiotic was reported to changed color of stool simultaneously when increasing SCFA-producing bacteria and SCFA [33]. Taken together, the results of this study may suggest a potential increase in almost all human gut predominant bacteria including SCFA-producing bacteria and improved bowel movements following GlcN consumption.

This study, however, has certain limitations. It did not measure the metabolites contained in the culture medium or human feces; therefore, future research is needed to measure the production of SCFAs, which have a significant effect on human gut environment. In addition, as the in vitro experiments in this study involved culturing intestinal bacteria alone, the interactions between the intestinal bacteria themselves or with the host were not considered. Additionally, in this clinical trial, measurements of fecal moisture content and total bacterial count, which are thought to affect bowel movements, have not been carried out. Finally, and most importantly, a blinded randomized study with dietary records should be conducted to accurately evaluate the results and eliminate placebo effects [34].

In conclusion, we revealed that dietary GlcN could stimulate the growth of a wide range of gut microbiota components in vitro, including beneficial butyrate-producing bacteria. Additionally, to our knowledge, we provide the first clinical data showing that a small dose of GlcN alone favorably influences colon health. GlcN significantly improved defecation and constipation. This could be attributed to the effects of GlcN on improving bowel movements, potentially mediated by the metabolites of the gut microbiota, indicating a possible probiotic-like effect.

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.