Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP, https://old.tcmsp-e.com/tcmsp.php) was employed to search all herbal components of AOH [21]. Next, the drug absorption, distribution, metabolism, and excretion (ADME) parameters were applied to screen the potential active ingredients with an Oral Bioavailability (OB) $\ge$30% and Drug-Likeness (DL) $\ge$0.18 set as the main indicators.

Then, TCMSP and SwissTargetPrediction databases (http://www.swisstargetprediction.ch/) were employed to search for targets corresponding to the AOH active ingredients [22, 23]. The Uniprot database (https://www.uniprot.org/) [24] was used to set species as “Homo sapiens” and convert the target proteins to matching target genes. Finally, a database of AOH compounds and their targets was constructed.

The proteomic expression data of DGP patients and non-diabetic non-gastroparetic controls were retrieved from the GEO database of the National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/geo/, GSE130672; GPL23119, 14 samples) with p 0.05 and $|$log2 (fold change)$|$ $>$1, which were considered as target genes with a significant expression difference in DGP patients and with research value [25].

“Diabetic gastroparesis” was used as the keyword to screen the related genes from the following databases: GeneCards database (https://www.genecards.org) [26], Pharmacogenetics and Pharmacogenomics Knowledge Base (PharmGKB) database (https://www.pharmgkb.org/) [27], Therapeutic Target Database (TTD) (http://db.idrblab.net/ttd/) [28], DrugBank database (https://www.drugbank.ca/) [29] and the Online Mendelian Inheritance in Man (OMIM) database (https://www.omim.org/) [30]. Furthermore, the obtained data were combined and the duplicate entries were removed. Finally, the Venn diagram (http://bioinformatics.psb.ugent.be/webtools/Venn/) was applied to visualize the ultimate results.

The predicted targets of active ingredients of AOH and the targets related to DGP were depicted in a Venn diagram, with the overlapping portion denoting the common targets of AOH and DGP.

The common targets of disease and drug obtained above were imported into the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) 11.0 (https://string-db.org) [31], and the species column was selected as “homo sapiens”. Medium confidence = 0.4 was set as the lowest interaction score, and the rest of the settings were maintained at their default values. The PPI data were collected, collated, and imported into the Cytoscape 3.7.2 program [32] to construct the PPI network diagram. Subsequently, the degree, betweenness, and closeness in the CytoNCA analysis [33] were used to screen the network for primary targets.

To determine the specific targets and pathways of AOH that improve DGP, GO functional analysis and KEGG pathway enrichment analysis were conducted with reference to the Metascape database (http://metascape.org/gp/index.html) [34]. The results revealed that Biological Process (BP), Cellular Component (CC), Molecular Function (MF), and signaling pathways were closely associated with improvement of DGP by AOH.

The previously screened AOH active components, common targets, and main signaling pathways were imported into the Cytoscape 3.7.2 software (https://cytoscape.org/release_notes_3_7_2.html), and the “component–target–pathway” (C–T–P) network of AOH to ameliorate DGP was constructed. The “C–T–P” network comprised nodes and edges, wherein the nodes represented the major pharmacodynamic compounds of AOH, potential targets, and key pathways that acted concertedly, and the edges represented intermolecular interactions and mainly referred to the association between the active ingredients and key targets and signaling pathways.

Male C57BL/KsJ db/db and db/m mice (aged 8 weeks) were purchased from Jiangsu Jicui Yaokang Biotechnology Co., Ltd., China. The mice were housed at a temperature of 25 °C and a humidity of 50%–65%, with 12-h light/dark cycles, and had free access to standard water and laboratory diets. All animal experiments were approved by the Experimental Animal Ethics Committee of Hainan Medical University (Permit No. HYLL-2021-377).

After 1 week of adaptive feeding, the DGP mouse model was constructed by modifying a previously reported method [37]. Eight db/m mice constituted the control group and were regularly fed with chow diet, and sixteen db/db mice were fed a high-fat diet (HFD, 60% fat, 20% protein, and 20% carbohydrate) in an irregular manner (fasted on odd days, fed on even days) to promote the development of DGP [38]. During DGP modeling, AOH (75 mg/kg) was administered to prevent the development of DGP. Briefly, the mice were randomly categorized into three groups (n = 8), as follows: (1) control (CON) group; (2) DGP group; and (3) AOH group. The AOH was dissolved in distilled water and administered via oral gavage at a dose of 0.2 mL/10 g once a day for 9 weeks. The mice in the CON and DGP groups were given the same physiological volume of distilled water. During the experiments, fasting blood glucose (FBG, fasting for 8 h) and body weight were monitored and recorded weekly.

Gastric emptying was visualized by slightly modifying a previous study protocol [39]. In the 6th week of treatment, the mice were depilated before the initiation of fluorescence imaging. The mice were fasted for 12 h, and 100 µL of orally gavaged indocyanine green was administered at a concentration of 1 mg/mL. Subsequently, the animals were anesthetized with isoflurane, and live images were obtained using IVIS Lumina XR (Perkin Elmer, Excitation: 780 nm, Emission: 831 nm, Waltham, MA, USA). Finally, fluorescence intensities in the mice at different time points (5 min, 30 min, and 50 min) were monitored.

One week before the end of the animal experiment, an oral glucose tolerance test (OGTT) was conducted. Briefly, after fasting for 12 h in each group, the blood glucose levels were measured at 0, 15, 30, 60, and 120 min after administering 2 g/kg of glucose. Finally, the glucose level was plotted as a curve, and the area under the curve (AUC) of OGTT was calculated.

All mice were euthanized in the 9th week of treatment, and the blood and stomach were collected after dissecting the mice. The blood samples were allowed to stand at room temperature for 2 h and then centrifuged at 3500 rpm for 15 min to obtain the serum. The stomach tissues were collected and stored at –80 °C for Western blotting (WB) and quantitative polymerase chain reaction (qPCR) analysis.

An automatic biochemistry analyzer was used to analyze the serum biochemical parameters, including total cholesterol (TC), triglyceride (TG), and alanine aminotransferase (ALT) levels. Serum motilin (MTL), malonic dialdehyde (MDA), and superoxide dismutase (SOD) levels were measured according to the instructions in the enzyme-linked immunosorbent assay (ELISA) kit.

Gastrointestinal propulsion was detected using the phenol red method as previously described [40]. Before euthanasia, the mice were treated orally with 300 µL of phenol red solution (0.05% w/v phenol red suspension in 1.5% w/v suspension of carboxymethylcellulose). After another 20 min, the mice were sacrificed and the entire stomach was carefully isolated and ligated just above the cardia and below the pylorus and separated. Its contents were poured into a beaker and washed with 20 mL of 0.9% normal saline. Subsequently, the contents were added to 20 mL of 0.5 mol/L NaOH, and 4 mL of trichloroacetic acid (20% w/v) was added for deproteinization. The absorbance of the supernatant was read at 560 nm using a Molecular Devices Spectra Max Plus automatic plate reader (Molecular Device, Sunnyvale, CA, USA) after centrifugation (3500 rpm for 10 min). Next, 18 mL of distilled water, 20 mL of 0.5 mol/L NaOH, and 4 mL of trichloroacetic acid (20% w/v) were added to 2 mL of phenol red solution and mixed well to measure the absorbance as a blank control. The gastric emptying rate was calculated according to the following equation: gastric emptying rate (%) = (1 – absorbance of phenol red collected from the stomach 20 min after gavage with phenol red/absorbance of blank control) $\times$ 100.

According to the manufacturer’s instructions, total RNA was isolated from gastric tissues using the Eastep®Super Total RNA Extraction Kit. Later, a single-stranded cDNA was synthesized from 2 µg of total RNA using reverse transcriptase. Afterward, quantitative polymerase chain reaction (qPCR) was performed using the following amplification conditions: a precycling stage at 95 °C for 5 min, followed by 40 cycles of the second step (i.e., denaturing at 95 °C for 10 sec, annealing at 60 °C for 20 sec, and elongation at 72 °C for 20 sec). Ultimately, nonspecific amplifications were monitored with the melting curves, and the 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{Ct}}$ method was used to analyze the relative expression of mRNA. The relative expression abundance of all genes was normalized with GAPDH. The primer sequences used for qPCR are presented in Table 1.

Gastric tissues were homogenized on ice and centrifuged at 15,000 rpm for 10 min at 4 °C after weighing them and cutting them with scissors. The total protein was extracted from the mice’s gastric tissue using radioimmunoprecipitation assay (RIPA) lysis buffer in each group, and protein concentrations were measured using the bicinchoninic acid protein assay kit. Protein samples were mixed with 5$\times$ loading buffer and denatured at 100 °C for 5 min. Then, gel preparation and sample loading were performed. Subsequently, the proteins in the gel were transferred to a polyvinylidene difluoride membrane by wet transfer after electrophoretic separation. The membranes were blocked with 5% skim milk powder in Tris-buffered saline with 0.1% Tween 20 blocking solution for 1 h. After the membrane was washed, it was incubated with PI3K (1:1000), AKT (1:2000), p-PI3K (1:1000), p-AKT (1:2000), and GAPDH (1:10,000) overnight at 4 °C. After washing, the membrane was incubated with secondary antibodies for 1 h at room temperature. Eventually, images were analyzed, and relative expression levels of proteins were determined using the Image Lab analysis software (Bio-Rad Laboratories, Inc., Hercules, CA, USA). GAPDH served as the internal reference.

Searching the TCMSP database revealed that 130 compounds were currently known to be found in AOH. Subsequently, 13 active compounds were screened across all compounds based on two key ADME parameters (OB $\ge$30% and DL $\ge$0.18) (Table 2). In addition, the TCMSP and SwissTargetPrediction databases were used to obtain the targets corresponding to the active components of AOH, and finally 498 targets were obtained (Supplementary Table 1).

Table 2.Detailed information on the 13 active components from AOH.

MOL ID	Molecule name	Structure	OB (%)	DL
MOL001771	Poriferast-5-en-3beta-ol		36.91	0.75
MOL002543	(2S,3R)-2-(3,4-dimethoxyphenyl)-5,7-dimethoxychroman-3-ol		51.89	0.37
MOL002544	1,7-diphenyl-5-hydroxy-3-heptanone		61.9	0.18
MOL002554	5-methoxy-1,7-diphenyl-3-heptanone		68.29	0.20
MOL002556	7-Methoxy-8-(2${}^{\prime }$-ethoxy-3${}^{\prime }$-hydroxy-3${}^{\prime }$-methybutyl)coumarin		40.36	0.21
MOL002563	Galangin		45.55	0.21
MOL002565	Medicarpin		49.22	0.34
MOL002575	Butyl-2-ethylhexyl phthalate		44.52	0.22
MOL000354	Isorhamnetin		49.6	0.31
MOL000358	Beta-sitosterol		36.91	0.75
MOL000359	Sitosterol		36.91	0.75
MOL000422	Kaempferol		41.88	0.24
MOL000098	Quercetin		46.43	0.28

A total of 136 differentially expressed targets related to DGP were obtained from the GEO database search, of which 76 were downregulated and 60 were upregulated (Fig. 1A,B).

Fig. 1.

Network pharmacology analyses of AOH on DGP. (A) A volcano map of differentially expressed genes in DGP patients (The abscissa represents the fold changes in gene expression, and the ordinate represents the statistical significance of the gene expression change). (B) The number of differential genes in DGP patients (DGP stands for DGP patients; C stand for non-diabetic non-gastroparetic controls). (C) Venn diagram of differentially expressed genes in DGP. (D) Venn diagram and candidate targets of AOH and DGP. (E) PPI network (D, Degree; B, Betweenness; C, Closeness). (F) Degree values of the top 13 target proteins. (G) GO enrichment analysis of targets of AOH. (H) Analysis of KEGG enrichment in 20 pathways as targets of AOH. (I) “component-target-pathway” of AOH ameliorating DGP (Blue nodes: overlapping genes between AOH and DGP; pink nodes: the KEGG pathway; orange nodes: the active ingredients of AOH). AOH, alpinia officinarum hance; DGP, diabetic gastroparesis; PPI, protein–protein interaction; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.

In addition, using “diabetic gastroparesis” as the keyword, a total of 674 possible candidate targets were screened from different databases, including 581 in GeneCards, 38 in PharmGKB, 5 in TTD, 14 in DrugBank, and 61 in OMIM (Fig. 1C).

Venn analysis of 498 targets of AOH active components and 674 DGP-related targets resulted in the identification of 89 drug–disease intersection targets, which were important targets that played a key role in the alleviation of DGP by AOH (Fig. 1D).

The 89 drug–disease intersection gene targets obtained above were imported into the STRING database, where reliable interaction information could be predicted and imported into Cytoscape for analysis and construction of PPI networks. After hiding the unconnected nodes, a PPI network comprising 88 nodes and 2517 edges was obtained. The larger the node, the greater its biological functions in the PPI network (Fig. 1E). Therefore, according to the results of degree, betweenness, and closeness, 13 key nodes were further screened out and defined as the “key targets” for subsequent research. The degree values of the 13 most closely related target proteins are presented in Fig. 1F.

The results indicated that the enriched BPs were mainly responses to xenobiotic stimulus, positive regulation of protein phosphorylation, transferase activity and MAPK cascades, regulation of kinase and protein kinase activity, and cellular responses to various types of compounds; the enriched CCs were mainly related to membrane raft, micromembrane domain, and synaptic composition; and the enriched MFs were predominantly related to kinase activity, protein phosphatase binding, cytokine receptor binding, and transcription factor binding (Fig. 1G).

Furthermore, the KEGG pathway enrichment analysis was used to decipher the potential role of signaling pathways. Of the 91 signaling pathways, the top 20 differentially expressed pathways were identified under the screening condition of ascending p-value sorting, and the enrichment situation was visually displayed using the enrichment bubble chart. The analysis revealed that the related signaling pathways of AOH for improving DGP chiefly included the PI3K-AKT, AGE-RAGE in diabetic complications, HIF-1, FoxO, p53, and TNF signaling pathways (Fig. 1H).

To further investigate the potential molecular mechanism of AOH in improving DGP, a “component–target–pathway” network was constructed based on the top 20 significant signaling pathways and their corresponding targets (Fig. 1I). The network contained 124 nodes (1 plant, 13 active ingredients, 89 targets, 20 signaling pathways, and 1 disease) and 575 edges. The size of a node was proportional to its topological score in the network; the higher the degree value, the larger the node, and the lower the degree value, the smaller the node. Therefore, the top five active ingredients (quercetin, kaempferol, medicarpin, galangin, and 1,7-diphenyl-5-hydroxy-3-heptanone) were screened according to the size of the nodes in the network, which were considered to be the most likely compounds effective in the treatment of DGP.

Over time, symptoms of lethargy, polydipsia, polyuria, loose stools, and abdominal distension were observed in DGP mice. Additionally, throughout the experiment, FBG levels of mice in each group were evaluated at a fixed time point every week. The alterations in FBG levels in each group are shown in Fig. 3A. In the CON group, the FBG level remained at a relatively stable and normal level throughout the experimental period. However, compared with the CON group, the FBG levels of the DGP and AOH groups increased continuously, with the uptrend of the AOH group being more gradual than that of the DGP group. Moreover, compared with the DGP group, the FBG level was significantly decreased in AOH-treated mice after 9 weeks of drug administration (p 0.01).

Fig. 3.

The effects of AOH on Blood Glucose and Gastric Emptying in DGP Mice. (A) FBG levels. (B) OGTT. (C) The AUC of the OGTT. (D) Fluorescence imaging. (E) Gastric emptying. All values are expressed as the means $\pm$ SD (n $\ge$ 3). ##p 0.01, and ###p 0.001, when compared with the CON group; and **p 0.01, and ***p 0.001, when compared with the DGP group. FBG, fasting blood glucose; OGTT, oral glucose tolerance test.

OGTT was performed in different groups, and the corresponding AUCs were also analyzed. As shown in Fig. 3B,C, the results of OGTT were ameliorated to some extent in the AOH group. The AUC of the DGP group was significantly higher than that of the CON group (p 0.001) according to the calculation, and the AUC of the AOH group was significantly lower than that of the DGP group (p 0.001) but could not reach the glucose tolerance level of the CON group.

Real-time ICG fluorescence imaging results are depicted in Fig. 3D. The symptoms of gastroparesis in the mice were judged by the intensity and location of the fluorescence. The mice in the DGP group showed delayed gastric emptying compared with those in the CON group. Administration of AOH, however, significantly accelerated gastric emptying. As expected, the liquid gastric emptying test validated this result (Fig. 3E). Compared with the CON group, the gastric emptying rate decreased significantly in the DGP group (p 0.01), thereby suggesting that the DGP model was established successfully. Furthermore, compared with the DGP group, the gastric emptying rate was increased significantly in the AOH group (p 0.01). The above results demonstrate that AOH significantly promoted gastric emptying in DGP mice, exhibiting a definite alleviating effect.

The results of serum biochemical assays are displayed in Fig. 4. In Fig. 4A,B, there was a significant increase in the levels of TC and TG in the DGP group compared with the CON group. In contrast, the TC and TG levels were decreased in the serum after AOH treatment (p 0.001). The ALT activity of the DGP group showed a nearly two-fold increase compared with that of the CON group (p 0.001, Fig. 4C), which implied that the liver function of the mice in the DGP group was also affected. Unfortunately, AOH could not effectively improve liver function.

Fig. 4.

Effects of AOH on Biochemical Indicator in DGP Mice. (A) TC. (B) TG. (C) ALT. (D) MTL. All values are expressed as the means $\pm$ SD, (n $\ge$ 3). ##p 0.01, and ###p 0.001, when compared with the CON group; and *p 0.05, and **p 0.01, when compared with the DGP group. TC, total cholesterol; TG, triglyceride; ALT, alanine amino-transferase levels; MTL. Serum motilin.

MTL accelerates gastric emptying and is released during the digestive interval [42]. The levels of MTL in different groups were detected using the MTL kit, and the results are presented in Fig. 4D. The MTL level in the DGP group was 3.2 times lower than that in the CON group (p 0.01), which indicated that severe delayed gastric emptying had occurred in the DGP group because of the irregular high-fat diet. However, after the AOH intervention, the MTL level was significantly higher than that of the DGP group (p 0.01), which again proved that AOH can protect the DGP mice from delayed gastric emptying.

These results indicate that AOH could effectively improve the pathological indicators related to type 2 diabetes and further reduce the delayed gastric emptying induced by the disease. However, the molecular mechanism of AOH in ameliorating DGP remains unclear. Hence, we attempted to explore the potential molecular mechanism.

Based on the results of network pharmacology and molecular docking analysis, we discovered that AOH may play a role in the treatment of DGP by regulating the PI3K-AKT signaling pathway, in which PI3K and AKT are the key proteins. Therefore, to further evaluate the molecular mechanisms by which AOH ameliorates DGP, WB and PCR were performed to examine the related targets of AOH. As demonstrated in Fig. 5A–C,E–G, the expressions of AKT and PI3K were significantly lower in the DGP group compared with those in the CON group (p 0.01). However, AOH significantly reversed this decrease in the expression (p 0.01). Furthermore, compared with the DGP group, the phosphorylation of PI3K and AKT was promoted in the AOH group. Moreover, the mRNA expressions of PI3K and AKT were confirmed by this result (Fig. 5D,H), which indicated that AOH might exert its therapeutic effects by regulating the PI3K-AKT signaling pathway.

Fig. 5.

The effects of AOH on the PI3K-AKT Signaling Pathway in DGP Mice. (A) The protein expression of PI3K and p-PI3K. (B) Quantification of PI3K. (C) Quantification of p-PI3K. (D) mRNA expression of PI3K. (E) Protein expression of AKT and p-AKT. (F) Quantification of AKT. (G) Quantification of p-AKT. (H) mRNA expression of AKT. All values are expressed as the means $\pm$ SD (n $\ge$ 3). ##p 0.01, and ###p 0.001, when compared with the CON group; and *p 0.05, and **p 0.01, when compared with the DGP group.

Oxidative stress is an important factor in the development of DGP, and oxidative damage that occurs in many diseases can be improved via the PI3K-AKT signaling pathway [43, 44, 45]. Therefore, the effect of AOH on oxidative stress in DGP mice was investigated. As shown in Fig. 6A,B, the amount of MDA was higher in the DGP mice and the SOD activity was lower (p 0.01). However, AOH significantly enhanced the SOD activity and reduced the MDA levels in stomach homogenates (p 0.05). These results signify that AOH may regulate the related antioxidant enzymes to combat oxidative stress in DGP mice via the PI3K-AKT signaling pathway.

Fig. 6.

The effects of AOH on Oxidative Stress in DGP Mice. (A) MDA. (B) SOD. All values are expressed as the means $\pm$ SD (n $\ge$ 3). ##p 0.01 when compared with the CON group; and *p 0.05, when compared with the DGP group. MDA, malonic dialdehyde.