Premna microphylla Turcz. reshapes nutritional metabolism via the Simulator of the Human Intestinal Microbial Ecosystem model

Abstract

BACKGROUND

Gastric cancer (GC) remains a leading cause of cancer mortality worldwide. Despite improved survival with surgery combined with chemotherapy and immunotherapy, these treatments frequently induce gastrointestinal mucosal injury and disrupt intestinal microecological homeostasis, resulting in severe dysbiosis. Premna microphylla Turcz. (PMT), a traditional medicinal and edible herb, exhibits anti-inflammatory and immunomodulatory properties. However, the metabolism of PMT bioactive components in a compromised intestinal microenvironment and their roles in microecology-mediated intestinal repair remain poorly understood.

AIM

To establish a Simulator of the Human Intestinal Microbial Ecosystem (SHIME) model for explicitly elucidating the segment-specific metabolism of PMT.

METHODS

Using the SHIME model, this study simulated the absorption of PMT in GC patients undergoing chemotherapy and immunotherapy, aiming to clarify its active components' intestinal recovery mechanism and nutritional intervention value.

RESULTS

Components of PMT were converted into small-molecule phenolic acids by microbiota. Their absorption and metabolism showed heterogeneity and dynamics across intestinal segments, e.g., flavonoid glycosides are activated rapidly in the small intestine, while hydrophobic terpenes are retained in the colon. The shifted primary absorption site in the intervention group (e.g., epicatechin from the small intestine in the control group to the colon in the treatment group) suggests that dysbiosis may reshape drug absorption pathways, providing a theoretical basis for natural nutrient-based nutritional strategies for cancer patients.

CONCLUSION

The SHIME model revealed PMT’s metabolism-repair dual pathway in chemotherapy and immunotherapy-damaged intestines, laying a theoretical foundation for its targeted nutritional rehabilitation strategies for cancer patients’ intestinal microecology.

Key Words: Gastric cancer; Chemotherapy combined with immunotherapy; Nutritional bioactivity; Simulator of the Human Intestinal Microbial Ecosystem; Premna microphylla Turcz

Core Tip: This study established an intestinal microbial ecosystem (Simulator of the Human Intestinal Microbial Ecosystem) model to uncover the differential absorption and metabolism profiles of Premna microphylla Turcz. (PMT) across various intestinal segments. We demonstrated that chemotherapy-induced gut microbiota dysbiosis leads to a shift in the primary absorption site of active ingredients toward the colon. These findings not only provide a scientific basis for developing novel strategies to improve cancer patient prognosis but also lay a solid theoretical foundation for utilizing PMT as a low-toxicity therapeutic agent.

INTRODUCTION

Gastric cancer (GC) is the fifth most prevalent cancer globally and third leading cause of cancer deaths, after lung and liver cancer. Its risk factors include Helicobacter pylori (H. pylori) infection, advanced age, high salt intake, and low intake of fruit and vegetables[1]. Diagnosis relies on endoscopy and biopsy. Preoperative chemotherapy reduces tumor size and burden, controls micrometastases, and improves resectability in locally advanced cases[2]. However, targeted therapy faces screening and drug resistance challenges, making immunotherapy a new hope[3]. However, long-term prognosis remains poor, demanding postoperative interventions[4]. Natural compounds, with low toxicity, show promise.

Premna microphylla Turcz. (PMT) is a perennial deciduous shrub belonging to the genus Premna (Verbenaceae). Its roots and stems contain medicinal components with heat-clearing, detoxifying, swelling-reducing, and hemostatic effects; its leaves are nutrient-rich, with amino acids and pectin[5].

Chemotherapy and immunotherapy often cause gastrointestinal mucositis, nausea, vomiting, and anorexia, significantly compromising nutrient intake. Its extracts, rich in polyphenols, flavonoids, and polysaccharides, have anti-inflammatory, antioxidant, and mucosal repair effects, alleviating therapy-induced gastrointestinal mucositis. Its dietary fiber and polyphenols also repair disrupted intestinal microecology.

This study used the Simulator of the Human Intestinal Microbial Ecosystem (SHIME) model to investigate the absorption and metabolism of PMT’s active ingredients in GC patients, analyzing chemotherapy-induced intestinal microenvironment damage, the metabolic adaptability and repair effects of its components, and developing microbiota-oriented nutritional support strategies. SHIME, a multi-chamber dynamic simulator, is a key in vitro model for studying gut microbiota interactions and disturbances. With real-time monitoring and flexible programming, it stably simulates the gut ecosystem to evaluate microbiota-targeted treatments in controlled environments[6].

This study clarifies the absorption and metabolism of PMT’s active ingredients across intestinal segments in GC patients, providing a basis for its development and phytochemicals-assisted recovery (Figure 1).

Figure 1 Design of the current study. FMT: Fecal microbiota transplation; PMT: Premna microphylla Turcz.; PMTE: Premna microphylla Turcz. extract.

MATERIALS AND METHODS

Clinical samples

Fecal samples were collected from a 76-year-old male patient with GC. The patient was diagnosed with GC on October 7, 2022, with a clinical stage of cT3N2M0, stage III. From October 15, 2022, the patient received a total of 5 cycles of treatment, which consisted of SOX chemotherapy regimen (oxaliplatin plus tegafur, gimeracil, and oteracil potassium) in combination with sintilimab immunotherapy. The samples were collected during a grade 3 immune checkpoint inhibitor-related diarrheal flare in the patients before initiating corticosteroid treatment and without recent antibiotic use. The diagnostic criteria for GC followed the Guidelines for the Diagnosis and Treatment of Gastric Cancer (2022).

PMT extract

PMT leaf powder (200 g; 60-mesh sieved) was added to a 5 L flask, mixed with a 10-fold volume of methanol-water (70:30, v/v), and sonicated for 1.5 hours (500 W, 40 kHz). The residue was re-extracted under the same conditions. Combined filtrates were evaporated, concentrated, and lyophilized at -40 °C under 9-10 Pa to obtain the extract.

Main instruments

JP-080 ultrasound device (Shenzhen Jiemeng Cleaning Equipment Corp., China); Shimadzu LC-20A high performance liquid chromatograph (Shimadzu, Japan); SHIME (Prodigest, Belgium); Orbitrap Exploris 120 high-resolution mass spectrometer (Thermo Fisher Scientific, United States); Vanquish ultra-high performance liquid chromatography system (Thermo Fisher Scientific, United States); ACQUITY UPLC BEH C18 column (2.1 mm × 100 mm, 1.7 μm, Waters, United States); ZORBAX SB-C18 column (250 mm × 4.6 mm, 5 μm, Agilent, United States).

SHIME gastrointestinal simulated digestion

The treatment group and control group were introduced into the SHIME system via an oral reactor, respectively. The drug concentration was standardized to 400 mg/mL of PMTE. Based on the extraction rate, 16.67 g of PMTE was precisely weighed and dissolved in 50.00 mL of pure water to form the treatment group. An equal volume of deionized water served as the control and was introduced into the SHIME system. Specific SHIME parameters are provided in Table 1. The system was kept in a 37 °C water bath and rendered anaerobic by purging with nitrogen for 10 minutes twice daily. Continuous magnetic stirring within all reaction chambers ensured the chyme and microbiota were well mixed, thereby simulating intestinal peristalsis.

Table 1 Simulator of the Human Intestinal Microbial Ecosystem system parameters.

Compartment	Range of pH	Residence time (hour)	Volume (mL)	Supplementation
Stomach	4.6-3.0 (uniform speed)	2	140	Gastric juice, lecithin, pepsin
Duodenum	6.3-6.7	0.5	124.6	Pancreatic juice, terypsin, chymotrypsin, amylase, CaCl2
Small intestine	6.8-7.2	2	180	NaHCO3
Colon	5.8-6.5	24	419.25

The SHIME model has four sampling points: Gastric reactor, small intestine reactor, small intestinal extraintestinal reactor, and colon reactor. Samples were taken from the gastric reactor at 0 hour and 2 hours of digestion; from the small bowel reactor and small intestinal extraintestinal reactor at 0 hour and 2 hours; and from the colon reactor (for small intestinal digestive juice) at 0 hour, 6 hours, 8 hours, 18 hours, and 24 hours. All samples were frozen at -80 °C for analysis.

Chromatographic separation was performed using a Waters HSS T3 C18 column. The mobile phase comprised 0.1% formic acid in water (phase A) and acetonitrile (phase B). The column temperature was 50 °C, with a flow rate of 0.3 mL/minute and an injection volume of 5 μL. The gradient program is shown in Table 2.

Table 2 Gradient program of mobile phases for chromatographic separation.

Time (minute)	Phase B (%)	Phase A (%)
0	5	95
10	10	90
22	18	82
36	50	50
37	95	5
40	95	5
40.01	5	95
45	5	95

RESULTS

Identification of digestive juice components

UHPLC-Q-Orbitrap-MS/MS qualitatively analyzed digestive juices from different regions, generating total ion current (TIC) chromatograms (Figure 2). TIC chromatograms showed compound variations across digestive sites. Thermo Scientific Compound Discoverer software was used to identify 103 plant-derived compounds and nutrients, including 17 flavonoids, 21 terpenes, 14 phenylpropanoids, 11 fatty acids, and 40 other derivatives (Table 3). The content variations of the main natural compounds across different digestive stages are displayed in Figure 3.

Figure 2 Total ion current chromatograms of digestive juice samples (stomach, small intestine, extraintestinal tract, and colon). A: Control group; B: Treatment group.

Figure 3 Waterfall chart of the content of the main natural compounds in different intestinal segments. ST: Stomach; EN: Small intestine (inner); PN: Small intestine (outer); CL: Colon.

Table 3 Total ion current chromatograms of digestive juice samples (stomach, small intestine, extraintestinal tract, and colon) in the control and treatment groups.

No.	RT (minute)	Compound	Formula	Theoretical (m/z)	Observed (m/z)	Adducts
1	27.339	(-)-Caryophyllene oxide	C15H24O	220.18269	221.18997	[M+H]+1
2	11.022	(+)-(S)-carvone	C10H14O	150.1045	151.11178	[M+H]+1
3	16.92	(+)-Epicatechin	C15H14O6	290.07906	291.08633	[M+H]+1
4	5.807	(E)-p-coumaric acid	C9H8O3	164.04731	163.04003	[M-H]-1
5	34.059	(E,E)-alpha-farnesene	C15H24	204.18777	205.19504	[M+H]+1
6	3.33	1-Caffeoyl-beta-D-glucose	C15H18O9	342.09477	341.08749	[M-H]-1
7	5.364	1-O-(4-coumaroyl)-beta-D-glucose	C15H18O8	326.10004	325.09276	[M-H]-1
8	31.839	2-(2,6-dimethoxyphenyl)-5,6-dimethoxy-4H-chromen-4-one	C19H18O6	342.11033	343.1176	[M+H]+1
9	34.241	3,5-Dihydroxy-4',7-dimethoxyflavone	C17H14O6	314.07904	315.08635	[M+H]+1
10	3.344	8-Hydroxyquinoline	C9H7NO	145.05277	146.06005	[M+H]+1
11	37.656	9(S)-HPODE	C18H32O4	312.23001	311.22274	[M-H]-1
12	38.307	9-Oxo-ODE	C18H30O3	294.21947	293.21218	[M-H]-1
13	0.876	Beta-D-mannopyranose	C6H12O6	180.06328	179.05602	[M-H]-1
14	0.884	Beta-D-xylopyranose	C5H10O5	150.05274	149.04552	[M-H]-1
15	27.339	Bisabolol oxide B	C15H26O2	238.1933	239.20058	[M+H]+1
16	18.062	Blumenol A	C13H20O3	224.14126	225.14854	[M+H]+1
17	33.341	Casticin	C19H18O8	374.10009	375.10738	[M+H]+1
18	1.093	Citric acid	C6H8O7	192.02693	191.01966	[M-H]-1
19	41.25	Erucamide	C22H43NO	337.33447	338.34175	[M+H]+1
20	37.156	Euscaphic acid	C30H48O5	488.35013	487.34268	[M-H]-1
21	39.001	Glycol palmitate	C18H36O3	300.26634	299.25906	[M-H]-1
22	39.57	Hexadecanamide	C16H33NO	255.25623	256.26351	[M+H]+1
23	4.636	Isopropylmalic acid	C7H12O5	176.06835	175.06108	[M-H]-1
24	20.822	Isorhamnetin	C16H12O7	316.05819	317.06546	[M+H]+1
25	0.978	L(-)-pipecolinic acid	C6H11NO2	129.07897	130.08623	[M+H]+1
26	1.312	Leucine	C6H13NO2	131.09458	132.10186	[M+H]+1
27	39.099	Linoleoyl ethanolamide	C20H37NO2	323.2824	324.28968	[M+H]+1
28	0.945	Mesaconic acid	C5H6O4	130.02662	129.01935	[M-H]-1
29	8.526	Methyl cinnamate	C10H10O2	162.06808	163.07536	[M+H]+1
30	23.214	Nootkatone	C15H22O	218.16701	219.17429	[M+H]+1
31	38.573	N-phenyl-1-naphthylamine	C16H13N	219.10477	220.11205	[M+H]+1
32	39.893	Oleamide	C18H35NO	281.27184	282.27911	[M+H]+1
33	39.383	Palmitylethanolamide	C18H37NO2	299.28238	300.28965	[M+H]+1
34	8.474	Propylparaben	C10H12O3	180.07844	179.07116	[M-H]-1
35	3.661	Protocatechuic aldehyde	C7H6O3	138.03169	137.02441	[M-H]-1
36	3.775	Salidroside	C14H20O7	300.12069	345.11878	[M+FA-H]-1
37	31.698	Scrophulein	C17H14O6	314.07905	315.08637	[M+H]+1
38	38.815	Senkyunolide B	C12H12O3	204.0786	205.08588	[M+H]+1
39	32.739	Tangeritin	C20H20O7	372.12083	373.12811	[M+H]+1
40	30.952	Traumatic acid	C12H20O4	228.13608	227.12879	[M-H]-1
41	0.883	Trigonelline	C7H7NO2	137.0477	138.05496	[M+H]+1
42	2.545	Vanillyl alcohol	C8H10O3	154.06296	153.05568	[M-H]-1
43	20.119	Verbascoside	C29H36O15	624.205	623.19772	[M-H]-1
44	38.571	α-linolenic acid	C18H30O2	278.22453	279.23181	[M+H]+1
45	20.436	α-sinensal	C15H22O	218.16703	219.17431	[M+H]+1
46	28.311	(-)-Andrographolide	C20H30O5	350.20928	351.21655	[M+H]+1
47	28.906	(-)-Beta-caryophyllene epoxide	C15H24O	220.18267	221.18995	[M+H]+1
48	26.235	(+)-Leucopelargonidin	C15H14O6	290.07912	291.0864	[M+H]+1
49	10.01	(+)-Riboflavin	C17H20N4O6	376.13829	377.1456	[M+H]+1
50	22.112	(+)-Syringaresinol	C22H26O8	418.16254	417.15509	[M-H]-1
51	18.516	(E)-Ferulic acid	C10H10O4	194.05782	195.06516	[M+H]+1
52	1.842	Beta-D-glucopyranose	C6H12O6	180.0633	179.05603	[M-H]-1
53	38.474	12-oxo phytodienoic acid	C18H28O3	292.20389	293.21115	[M+H]+1
54	10.48	1-O-feruloyl-beta-D-glucose	C16H20O9	356.11047	355.1032	[M-H]-1
55	9.327	4-Methylumbelliferone	C10H8O3	176.04732	177.05459	[M+H]+1
56	9.17	5-Caffeoylshikimic acid	C16H16O8	336.08436	335.07708	[M-H]-1
57	18.336	6-O-feruloylcatalpol	C25H30O13	538.16874	539.17606	[M+H]+1
58	8.473	Abietin	C16H22O8	342.13113	341.12385	[M-H]-1
59	0.966	Adenine	C5H5N5	135.05449	136.06177	[M+H]+1
60	20.699	Beta-myrcene	C10H16	136.12523	137.1325	[M+H]+1
61	3.885	Caffeic acid	C9H8O4	180.04217	181.04947	[M+H]+1
62	9.123	Coniferyl alcohol	C10H12O3	180.07852	179.07124	[M-H]-1
63	31.31	Curcumene	C15H22	202.17195	203.17936	[M+H]+1
64	1.107	D-(+)-pyroglutamic acid	C5H7NO3	129.04262	128.03535	[M-H]-1
65	1.045	DL-arginine	C6H14N4O2	174.11163	175.1189	[M+H]+1
66	0.883	DL-glutamic acid	C5H9NO4	147.05316	148.06044	[M+H]+1
67	0.94	DL-glutamine	C5H10N2O3	146.06917	147.07652	[M+H]+1
68	2.103	D-pantothenic acid	C9H17NO5	219.11059	218.10326	[M-H]-1
69	34.858	Embelin	C17H26O4	294.18311	293.17584	[M-H]-1
70	3.611	Esculetin	C9H6O4	178.02651	179.03387	[M+H]+1
71	24.718	Eucommin A	C27H34O12	550.20478	549.19756	[M-H]-1
72	26.974	Eugenol	C10H12O2	164.0837	163.07642	[M-H]-1
73	17.608	Fisetin	C15H10O6	286.04721	287.05449	[M+H]+1
74	18.612	Geniposide pentaacetate	C27H34O15	598.18994	597.18258	[M-H]-1
75	8.037	Gentiopicrin	C16H20O9	356.1105	355.10316	[M-H]-1
76	24.803	Geranylacetone	C13H22O	194.16704	195.17432	[M+H]+1
77	0.96	Guanine	C5H5N5O	151.0494	152.05671	[M+H]+1
78	0.863	Hex-2-ulofuranosyl hexopyranosyl-(1->6)hexopyranoside	C18H32O16	504.16871	503.16147	[M-H]-1
79	41.901	Hex-2-ulose	C6H12O6	180.06339	179.05611	[M-H]-1
80	17.342	Kaempferol-7-O-glucoside	C21H20O11	448.10037	447.09309	[M-H]-1
81	9.024	Kanokoside A	C21H32O12	476.18911	475.18178	[M-H]-1
82	1.388	L-5-hydroxytryptophan	C11H12N2O3	220.08447	221.09174	[M+H]+1
83	0.879	L-alpha-glycerylphosphorylcholine	C8H20NO6P	257.10265	258.11002	[M+H]+1
84	38.348	Lauramide	C12H25NO	199.19362	200.20089	[M+H]+1
85	0.777	L-histidine	C6H9N3O2	155.06948	156.07677	[M+H]+1
86	0.946	L-phenylalanine	C9H11NO2	165.07898	166.08633	[M+H]+1
87	33.177	Lycofawcine	C18H29NO4	323.20957	324.21686	[M+H]+1
88	1.256	N-acetylornithine	C7H14N2O3	174.10042	175.10774	[M+H]+1
89	1.81	Nicotianamine	C12H21N3O6	303.14289	302.13559	[M-H]-1
90	1.08	Nicotinic acid	C6H5NO2	123.03201	124.03929	[M+H]+1
91	7.433	Paeonolide	C20H28O12	460.15782	459.15048	[M-H]-1
92	24.335	Phrymarolin I	C24H24O11	488.13166	487.12437	[M-H]-1
93	15.589	Plastoquinol-1	C13H18O2	206.13024	207.13792	[M+H]+1
94	1.719	Pyrogallol	C6H6O3	126.03169	127.03897	[M+H]+1
95	16.488	Quercetin	C15H10O7	302.04262	303.04989	[M+H]+1
96	20.471	Quercitrin	C21H20O11	448.1005	447.09305	[M-H]-1
97	5.225	Safrole	C10H10O2	162.06809	163.07536	[M+H]+1
98	13.685	Schaftoside	C26H28O14	564.14772	563.1404	[M-H]-1
99	22.646	Sorbifolin	C16H12O6	300.06295	301.0703	[M+H]+1
100	1.841	Trans-p-coumaraldehyde	C9H8O2	148.05237	149.05973	[M+H]+1
101	1.964	Umbelliferone	C9H6O3	162.03173	163.03901	[M+H]+1

Compounds with significant trend changes were selected based on hierarchical clustering analysis (Figure 4), and 12 with significantly increased gastrointestinal levels in GC patient post-administration were identified: Epicatechin, casticin, quercitrin, quercetin, 1-caffeic acid, salidroside, verbascoside, eugenol, citric acid, embelin, and esculetin.

Figure 4 Heatmap of clustering analysis for all products. ST: Stomach; EN: Small intestine (inner); PN: Small intestine (outer); CL: Colon.

The proportional distribution of these main natural compounds in different intestinal parts is shown in Figure 5.

Figure 5 Proportion of the main natural compounds in different intestinal segments. ST: Stomach; EN: Small intestine (inner); PN: Small intestine (outer); CL: Colon.

The content changes of phenylpropane, phenols, flavonoids and terpenes were determined by chemical stability, solubility and conversion efficiency (Figure 6).

Figure 6 Diagram of dynamic changes in vivo of different compounds. ST: Stomach; PN: Small intestine (outer); CL: Colon.

Phenols and terpenes decreased from the stomach to the colon. Phenols’ free phenolic hydroxyl groups protonated in the stomach’s strong acidity, reducing stability.

Terpenes’ alicyclic and chain structures were disrupted by acidity. In the weakly alkaline small intestine, phenols ionized to enhance water solubility; hydrophobic terpenes formed micelles with bile acids, accelerating absorption.

Flavonoids increased from the stomach to the colon. Their glycosidic bonds were stable in the stomach. In the colon’s neutral environment, flora-secreted β-glucosidase and rhamnosidase hydrolyzed glycosidic bonds to lipid-soluble free flavonoids, raising content.

Phenylpropanes increased from the stomach to the small intestine and then decreased. Their bonds were stable in the stomach. In the weakly alkaline small intestine, enzymes hydrolyzed bonds to release free forms. Free phenylpropanes, moderately polar, were rapidly absorbed in the small intestine; colon content dropped as most were consumed by small intestinal absorption, despite further metabolism.

DISCUSSION

Aerobic glycolysis (Warburg effect) is a hallmark of cancers including GC. Under hypoxic conditions, tumor cells shift glucose metabolism from oxidative phosphorylation to glycolysis, and the produced lactic acid further promotes tumor progression. Phytochemicals can inhibit aerobic glycolysis in GC cells via PI3K/Akt, c-Myc, and p53 signaling pathways, thereby suppressing cell proliferation/migration and promoting apoptosis, indicating promising prospects of natural glycolysis-targeting compounds for GC treatment[7-9].

SHIME model data showed that glycosides and terpenoids in PMT only undergo preliminary hydrolysis in the stomach and small intestine, with approximately 90% of active components transported to the colon as parent prodrugs, whose bioactivation highly depends on colonic microbial enzymes. Thus, developing a colon-targeted drug delivery system is practically valuable. Delaying drug release to the colon can precisely match the distally shifted absorption window of active components, expected to enhance local drug concentration, systemic exposure, and therapeutic efficacy.

An alternative complementary strategy is co-administering PMT with probiotics to construct a synbiotic formulation. Its core hypothesis is that modulating colonic microbial community and microenvironmental homeostasis can improve the physiological status of the absorption site, thereby enhancing the solubility, stability, and intestinal mucosal permeability of PMT, and overcoming absorption site alteration challenges from a microbial regulation perspective.

H. pylori contributes to gastric carcinogenesis by regulating non-coding RNA expression to induce gastric epithelial cell proliferation/apoptosis[10], and promoting GC cell proliferation, migration, and invasion via upregulating circMAN1A2 in AGS and BGC823 cells[11], suggesting H. pylori inhibition as a potential therapeutic approach. Systematic functional analysis was performed to clarify the biological activities of 12 previously identified compounds: (1) Flavonoids: Epicatechin exerts anti-GC effects indirectly via anti-inflammation and H. pylori inhibition (H. pylori induces gastritis/ulcers that may progress to carcinogenesis), and by reversing DNA methylation of tumor suppressor genes (e.g., p16INK4a and Cip1/p21) to upregulate their transcription[12,13]. Casticin upregulates RECK mRNA/protein expression in MGC803 cells by inhibiting DNMT1 expression and impairing Sp1 DNA-binding activity, thereby reducing RECK promoter methylation and suppressing cell proliferation[14]. Quercitrin (a quercetin rhamnoside) is hydrolyzed by colonic microbes to increase quercetin levels, which inhibits cyclin D1, P21, and Twist expression to affect the G1 phase, and suppresses P38MAPK phosphorylation via the P38MAPK pathway to exert anti-proliferative effects[15,16]; (2) Phenylpropanoids: 1-Caffeoyl-β-D-glucose is metabolized by gut flora into caffeic acid, which induces anti-proliferation and apoptosis in tumor cells by increasing cytosolic free calcium ion concentration [(Ca²⁺)i] via PLC-dependent endoplasmic reticulum Ca²⁺ release and store-operated calcium channel-mediated Ca²⁺ influx[17]. Salidroside inhibits GC cell proliferation and induces apoptosis/autophagy via the PI3K/Akt/mTOR pathway[18], while verbascoside suppresses tumor metastasis/invasion by inhibiting Rac-1, HIF-1α, and Zeb-1 signaling pathways[19]; and (3) Other compounds: Eugenol targets the TGF-β signaling pathway to inhibit GC metastasis. TGF-β promotes metastasis via c-Jun amino-terminal kinase and extracellular signal-regulated kinase-mediated Fascin-1 expression[20]. Embelin induces S and G2/M phase arrest by downregulating CDC25, CDC2, and CDK2, and reduces cell viability/induces apoptosis by downregulating X-linked inhibitor of apoptosis protein[21]. Esculetin exerts dose-dependent anti-proliferative and pro-apoptotic effects on GC cells (low toxicity to normal gastric epithelial cells) via inhibiting the IGF-1/PI3K/Akt pathway[22]. Citric acid cycle and pyruvate metabolism, as core energy production and biosynthesis units, are briefly mentioned here as they are not directly linked to PMT’s mechanism[23].

This study used the SHIME model to comprehensively explore PMT’s metabolic fate and its role in restoring intestinal homeostasis in GC patients, demonstrating its potential as an adjuvant for patients receiving chemoimmunotherapy. Methodologically, it aligns with and extends the traditional Chinese medicine (TCM) framework in integrative oncology, providing mechanistic insights into natural products’ regulation of the gut-microbiota-immune-metabolism axis for systemic anti-cancer effects.

The SHIME model faithfully simulates the human gastrointestinal tract’s dynamic physicochemical conditions and microbial ecosystems via its multi-compartment, continuous-flow design (surpassing single-chamber models with physiological pH gradients, transit times, and region-specific microbiota), enabling real-time profiling of metabolites, microbial shifts, and metabolic outputs. However, due to the inherent limitations (lack of mucosal immunity, and small bioreactor scale), further validation in in vivo models (e.g., GC xenografts) and large clinical cohorts is required.

Fecal metabolite analysis highlighted the link between dysbiosis and disease progression. PMT-derived bioactive compounds (anti-cancer, anti-inflammatory, antimicrobial, gut-barrier protective, and antioxidant) act via a multi-target network, consistent with TCM’s “holistic regulation” paradigm (e.g., Astragalus and PHY906 modulate immunity and mitigate therapy-induced toxicity)[24]. As a complex botanical matrix, PMT exerts synergistic effects via flavonoids, polysaccharides, and terpenoids to target the gut-immune-metabolic interface.

Developing safe and effective adjuvant therapies to alleviate chemoimmunotherapy side effects and enhance anti-tumor immunity is critical in oncology. TCM has proven efficacy in managing cancer-related complications[25], providing a reference for PMT’s clinical development. Notably, PMT’s “prodrug-microbiota activation” model and colonic absorption shift offer innovative solutions to natural products’ low oral bioavailability[26], which can be maximized via colon-targeted delivery systems or synbiotic formulations with probiotics.

While the present study has preliminarily delineated the metabolic profiles and potential targets of PMT using the SHIME system and bioinformatics analysis, direct experimental validation with GC cell lines or animal models remains unperformed. The specific mechanisms through which PMT metabolites regulate related signaling pathways require further verification in the subsequent experiments. Furthermore, the highly complex chemical composition of PMT leaves its pharmacodynamic material basis incompletely defined. Notably, the specific active monomeric compounds underlying its anticancer effects and their precise structure-activity relationships have not been fully characterized—a gap that has impeded the advancement of its standardized pharmaceutical development to a certain extent.

In the future studies, we plan to utilize the following experimental systems and analytical approaches to further explore the mechanisms of action and developmental potential of PMT. First, the colonic effluent fermented by the SHIME model will be co-cultured with GC organoids or cell lines. This will be coupled with phosphorylated proteomic profiling and pathway rescue experiments to directly elucidate the specific molecular pathways involved. Second, GC-bearing mouse models will be established to systematically assess its in vivo antitumor activity and safety profiles. Meanwhile, in-depth chemical isolation and identification studies will be conducted on PMT to isolate and characterize key active monomers, thereby clarifying its intrinsic pharmacodynamic material basis. Finally, through systematic preclinical toxicological assessments and pharmacodynamic evaluations, we aim to comprehensively explore the potential of PMT as an adjuvant therapeutic agent for alleviating chemotherapy-induced toxicity and reversing drug resistance in clinical practice. The overarching objective is to provide safer, more effective, and novel integrated treatment strategies for patients with GC.

CONCLUSION

Utilizing the advanced SHIME model, this study systematically mapped, for the first time, the metabolic landscape of PMT in the compromised intestinal microenvironment of GC patients undergoing chemotherapy. Our findings not only corroborated the stability and biotransformation kinetics of PMT-derived flavonoids, terpenoids, and polysaccharides across simulated digestive fluids but also highlighted a critical phenomenon: Chemotherapy-induced dysbiosis precipitates a distal shift in the primary absorption site of bioactive compounds toward the colon. We demonstrated that PMT efficacy is intrinsically governed by metabolic processing via the gut microbiota. Specifically, flavonoid glycosides require hydrolysis by colonic bacteria to release bioactive aglycones, such as quercetin and epicatechin, thereby validating a “prodrug-microbiota activation” pharmacological paradigm. PMT may function through a synergistic “metabolism-repair” dual mechanism. Initially, the gut microbiota can convert dietary plant constituents into potent small-molecule metabolites. These metabolites are then likely to mediate therapeutic effects, potentially by inhibiting tumor aerobic glycolysis, antagonizing H. pylori, restoring the intestinal barrier, and modulating inflammatory responses. Recognizing the observed distal shift in absorption sites, future clinical translation of PMT should prioritize colon-targeted delivery systems or “symbiotic” formulations incorporating glycosidase-producing probiotics. Such strategies are essential to maximize bioavailability and therapeutic efficacy within the colon. In summary, this study identifies PMT as a promising, low-toxicity, and medicinal-edible agent with the potential to restore intestinal microecological homeostasis and mitigate chemotherapy-induced cytotoxicity in GC patients, thereby providing a scientific basis for its future development as a targeted medical nutritional therapy in oncology.