240-day-old, male KM mice obtained from the Guangzhou Regal Biotechnology Co., Ltd., SCXK [Yue] 2018–0182, SYXK [Yue] 2021–0059) were pair-housed in plastic cages in a temperature-controlled (25 ± 2°C) colony room under a 12/12-h light/dark cycle, with free access to food and water. The aged KM male mice were administrated with distilled water throughout the whole course and 6 mice were grouped as the normal controls without any other intervention(N). The rest mice were treated with 5 mg/kg loperamide (imodium, SOURCE: Xian Janssen Pharmaceutical Ltd. IDENTIFIER: LGJ0549.) for 1 week and randomly divided into model group (M), positive drug group (Y), hawthorn group (S), probiotic group (F) and postbiotic of hawthorn-probiotic group (FS). Mice in Y group were intragastrically treated with 10% lactulose (0.2 ml/day/per mouse), S group with 1 g/ml pure hawthorn solution (0.2 ml/day/per mouse), F group with Lactobacillus paracasei supernatant (0.2 ml/day/per mouse) and FS group with postbiotic of hawthorn-probiotic (0.2 ml/day/per mouse) for another week.

The inclusion criteria were normal development, appropriate age, and weight. The exclusion criteria were abnormal development, inappropriate age, or weight. Any single criterion not met resulted in exclusion. Grouping was conducted randomly according to cage order, and blind method was employed to select one individual for gavage. The experimenter assigned a number to the gavage material, and the individual performing the gavage administered it based on the assigned number. During the observation period, the condition of all KM mice was closely monitored. If any adverse reactions were detected, the gavage was immediately discontinued. However, throughout the entire experiment, the mice remained in good health and no abnormalities were observed.

These KM mice were sacrificed after anesthesia with pentobarbital sodium. The structure of colon, and fecal metabolite of mice were determined. All experimental protocols were approved by the Animal Center, Guangzhou University of Chinese Medicine.

RNA was extracted from colon tissue using RNAiso Plus according to the instructions. Then, cDNA was obtained using the ImProm-II™ Reverse Transcription System (Promega) and RT-qPCR was carried out with custom designed oligonucleotides using the SweScript RT I First Strand cDNA Synthesis Kit and 2X Universal Blue SYBR Green qPCR Master Mix in a total volume of 20 μl: 95°C for 1 min and 40 cycles of denaturation (95°C for 15 s) and extension (60°C for 1 min). Experiments were performed in triplicates. Following amplification, dissociation curve analyses were performed to confirm the amplicon specificity for each PCR run. The relative level of gene expression in mouse colon tissue was normalized against mouse β-actin, respectively. Analysis of relative expression was performed using the 2(−ΔΔCT) method.

KM mouse (M group; with 5 mg/kg loperamide for 1 week; Constipation was diagnosed by insufficient fecal water content, long defecation time, small fecal volume and light weight) sacrificed after anesthesia with pentobarbital sodium. The colon were collected. The feces were removed with sterile syringes in a continuously ventilated (5% CO₂ + 95% O₂) Krebs-Ringer Buffer, and then cut colon into small segments 1 cm long. These colonal segments were then cut into sheets and placed in 24-well plates, respectively. Each well contained Krebs-Ringer Buffer continuously ventilated at 37°C and corresponding drug with a final concentration of 10 M and a volume of 1 ml. The culture was kept at 37°C, 5%CO₂ + 95%O₂, PH 7.4 for 5 h, and the tissue status was examined under the microscope every 15 min. Further experiments were carried out 5 h later.

The collected samples were thawed on ice, and metabolite were extracted with 80% methanol Buffer. Briefly, 50 mg of sample was extracted with 0.5 ml of precooled 80% methanol. The extraction mixture was then stored in 30 min at −20°C. After centrifugation at 20,000 g for 15 min, the supernatants were transferred into new tube to and vacuum dried. The samples were redissolved with 100 μl 80% methanol and stored at −80°C prior to the LC–MS analysis. In addition, pooled QC samples were also prepared by combining 10 μl of each extraction mixture.

Chromatographic mass spectrometry condition parameters are as follows: Chromatographic column: Waters ACQUITY UPLC BEH C18 (3.0 × 100 mm, 1.7 μm). Column temperature: 40°C. Sample size: 2 μl. Mobile phase: A: primary aqueous solution; B: Methanol: acetonitrile (volume ratio 1:1). The mass spectrum condition parameters are as follows: Sampling mode: MRM mode. See Materials section for standards. The standard curve is shown in Supplementary Table S2.

The TIC (Total Ion Chromotogram) of the metabolites of Lactobacillus (F), postbiotic of hawthorn-probiotic(FS), hawthorn (S) is shown in Figure 1A. We found that after fermentation, the FS group was significantly different from the F and S groups in both anion and cationic tests. According to the Mz-rt figrure (Figure 1B), it can be found that the main detection period of metabolic ions in this test is 0–5 min, and the main detection m/z is 100–500. In this test, there were a total of 12,009 anion peaks and 20,611 cation peaks. The types of metabolic ion peaks are mainly lipids and lipid-like molecules, Organoheterocyclic compounds, Organic acids and derivatives, Benzenoids, Phenylpropanoids and polyketides, and Organic oxygen compounds.

After further analysis of primary metabolites (Figure 2A), we found that group FS integrated some components of group F and group S, and the part with effective increase in content was in the lower right corner of the way. As can be seen from Figures 2B–D, the differences within the test group were small, while the differences between the groups were large, and the biological repeatability was relatively good.

We analyzed the secondary metabolites (Figure 3A) and found that the three groups were still very different. FS/S mainly increased, while FS/F mainly decreased (Figure 3B). However, in general, the metabolite content of FS was higher than that of F and S (Figure 3C). As can be seen from Figure 4A, the difference within the test group is small, while the difference between groups is large, which is close to the real data.

With VIP > 1 and Pvalue <0.05, differential secondary metabolites were screened, and Figure 4B was obtained. Then, K-mean cluster analysis was performed on these differential secondary metabolites (Figure 4C), and it was found that Organic acids and derivatives, Organoheterocyclic compounds, lipids and lipid-like molecules, Benzenoids, Phenylpropanoids and polyketides are the main rising components of FS, but many components are unknown.

Then we went on target hunting. Through network pharmacology, we found 20 common targets for metabolic mixtures and constipation (Figure 5A). The PPI network between them is plotted (Figure 5B). In the previous study (Figure 5C), we also found and verified several other targets, including TNF-α, Mucin2, Enac-γ, and AQP3. Among them, TNF-α has attracted our attention, and after reviewing the literature, we also included IL-6 and IL-17A in the experiment. By conducting RT-qPCR on the colons of mice in normal group (N), model group (M), and postbiotic of hawthorn-probiotic group (FS), the primers are listed in Table 1. We found that the postbiotic of hawthorn-probiotic was capable of decreasing the expression of NR1I2 and ST1A1 (SULT1A1), and increased the expression of GLP-2r, a receptor for GCG. However, the expression of ABCB1B, NR1I3, IL-6, and IL-17A showed a decreased trend, with no statistical significance (Figure 5D).

We found 131 significantly elevated secondary metabolites (see Supplementary material), and then searched for the postbiotic of hawthorn-probiotic that was significantly higher (VIP > 1.2, FC > 2or < 0.5, p < 0.01). We then used 112 metabolites for which we found a clear chemical structure to intermingle with previously screened targets (AQP3, Enac-γ, SULT1A1, TNF) and those we are still interested in (IL17A, IL6R). NR1I2 was found in the nucleus, and we did not focus on this target because we had no evidence that our metabolites could enter the nucleus. Subsequently, we found (Figure 6A) that these metabolites are generally combined with SULT1A1 and TNF. They bind well to parts of AQP3, GLP-2R, Enac-γ (SCNIG), and IL17A. But they do not combine well with IL6R. We used the index of binding energy to sort from large to small, and selected the top 10 compounds in each target for presentation (Table 2). Then, we combined the best groups and visualized them (Figures 6B–G). These are TNF-Baicalin, AQP3-Quinacridone, Enacγ-Quinacridone, IL17A-Baicalin, SULT1A1-N-Phenyl-2 naphthylamine, GLP2R-L-Isoleucine. We also performed molecular dynamics analysis on them (Figure 7). The results of RMSD, RMSF, and RG of the above small molecules and targets were all good.

We tried our best to purchase the 11 compounds that were virtually screened (Quinacridone, vitexin-2 “-o-rhamnoside, L-Isoleucine, Baicalin, Exemestane, Hyperoside, Quercetin, Apigenin, N-Phenyl-2-naphthylamine, Genistein, Daidzein. See the Materials section for specific information), and the colons of loperamide modeled mice were removed for experiments. Finally, we found (Figure 8) that in general, the compound regulated the intestinal targets of the model mice in a favorable direction, especially Enac-γ. However, some compounds resulted in increased expression of the target, and there is a worsening trend. Daidzein, for example, increased TNF expression.

So we found that the postbiotic of hawthorn-probiotic regulated the gut microbiome structure of loperamide in mice and their ability to produce biofilms (13). So we conducted targeted metabolomic analysis on feces, and we found (Figure 9A) that the short chain fatty acids profiling of normal group (N), model group (M), postbiotic of hawthorn-probiotic (FS) was completely different. The three groups had good differentiation and intra-group repeatability (Figure 9B). By analyzing short chain fatty acids separately (Figures 9C–I), Butyric acid and Propionic acid contents rise significantly in the intestine of constipated mice, while Isovaleric acid and Hexanoic acid contents decline. The postbiotic of hawthorn-probiotic can reduce Butyric acid content to normal level, and it can also increase Valeric acid and Isobutyric acid content. However, Acetic acid and Propionic acid did not change significantly.