The structure file of VK2 was downloaded in the PubChem database (https://pubchem.ncbi.nlm.nih.gov, searched on June 22, 2020) by searching for the keyword as “vitamin K₂.” The targets of VK2 were obtained by SwissTargetPrediction (http://www.swisstargetprediction.ch, searched on June 22, 2020) and DRAR-CPI (https://cpi.bio-x.cn/drar/, searched on June 22, 2020) using the structure file of VK2. Then the target library of VK2 was constructed using WPS Office software 2019. Salt-sensitive hypertension related therapeutic targets were obtained using the GeneCards database (https://www.genecards.org, searched on June 22, 2020) and Omim database (https://omim.org, searched on June 22, 2020). The “salt-sensitive hypertension” was used to act as searching keyword. Then, the names of target proteins were transformed into the corresponding gene symbols by the UniProt database (https://www.uniprot.org, searched on June 23, 2020).

The common targets library between putative targets of VK2 and the known therapeutic targets on salt-sensitive hypertension were amalgamated. Then the PPI network was integrated and conducted using the String database (https://string-db.org) according to the common targets.

The KEGG pathway enrichment analysis of common targets was carried out using clusterProfiler package which was used as a software package for pathway enrichment analysis and visualization in the R language (version 3.6.1). The screening condition was set as p < 0.001.

The network construction of VK2-targets-pathways-disease network was conducted using Cytoscape_v3.7.1. The detailed methods were described in previous reports (13, 14).

Eighteen 8-week-old male C57BL/6J mice were purchased from the experimental animal center of Guangzhou University of traditional Chinese medicine and raised in the animal center of Jinan University. All mice were naturally reared in a barrier environment with 12/12 h light cycle, 20–24°C and 40–60% humidity. This animal experiment was approved by the experimental animal ethics committee of Jinan University, which conforms to the principles of animal protection, animal welfare, and ethics, and the relevant provisions of national experimental animal welfare ethics. All mice were randomly divided into three groups: normal group (ND), high salt model group (HS), high salt diet plus VK2 supplementation group (HS_VK2), six mice per group. ND group was fed with a natural diet (containing 0.5% NaCl); HS group was fed with a high salt diet (containing 8% NaCl); HS_VK2 group was fed with high salt diet (containing 8% NaCl and additional 0.025% VK2) for 4 weeks. Finally, the mice were injected with 3% pentobarbital sodium to avoid suffering pain. After the necks were removed and sacrificed, the colon contents of the mice were quickly collected in sterile 1.5-ml EP tubes and stored at −80°C for testing.

The systolic blood pressure of all mice before and after the experiment was monitored by tail artery manometry and took the average value using the blood pressure analysis program (BP2000, USA) according to the operation instructions. The temperature of the mouse platform was set to 37°C in advance and test 15 times in each round.

The mouse aortic tissue was fixed in 2.5% glutaraldehyde at 4°C for 2–4 h under the condition of minimizing the mechanical injury such as traction, contusion, and extrusion. Then, the mouse aortic tissue was fixed and rinsed three times, 15 min each time, using 0.1 M phosphate buffer Pb (pH7.4). After dehydration and infiltration, the aortic tissue of mice was cut into 60–80 nm ultrathin sections. These sections were stained with uranium and lead (2% uranium acetate saturated alcohol solution, lead citrate, each staining for 15 min), and further observed under the TEM (HT7700; Hitachi; Tokyo, Japan), while three images were collected and analyzed from each group.

Three mice were randomly selected from each group and the proteins in their aortic tissues were detected through Western blotting. Ren antibody (abcam, ab212197), ACE (abcam, ab254222), AT1R (abcam, ab124734), and AT2R (abcam, ab227851) antibodies were used in the current study. According to the kit instructions, the total proteins were obtained by conventional tissue. Thus, PAGE separation was conducted after equivalent sampling. PAGE separins used 10% separating and 5% stacking gel. All proteins separated by PAGE were transferred to PVDF membranes with primary and secondary antibodies added. Moreover, the proteins were exposed, developed, and fixed. GAPDH was used as an internal control. Quantity One 4.0 software was used to analyze the imaging map and the ratios of Ren, ACE, AT1R, and AT2R proteins in each group were calculated.

Statistical analysis on the gut microbiota was performed using R package (MathSoft, Inc., United States). All other data are presented as mean ± standard error of mean (SEM). Statistical analysis was performed using Student t-test or one-way analysis of variance (ANOVA) through GraphPad Prism 5 (San Diego, CA, USA). p < 0.05 was considered as statistically significant.

The molecular formula of VK2 was C₃₁H₄₀O₂, and the molecule structure was shown as Figure 1A. A total of 195 reported pharmacological targets related to VK2 were obtained from the databases (Supplementary Table 1). A total of 493 salt-sensitive hypertension related targets were obtained from the databases (Supplementary Table 2). Combining the common targets of VK2 and salt-sensitive hypertension, 42 targets were screened to be the targets of VK2-treated salt-sensitive hypertension (Figure 1B; Supplementary Table 3). The function-related PPI network was conducted using the STRING database and shown in Figure 1C. The network containing 42 nodes and 185 edges, had significantly more interactions (p < 1.0 × e⁻¹⁶). Additionally, the top 10 core genes were screened as ALB, REN, IL10, SERPINE1, SRC, EGFR, F2, MAPK8, ESR1, and PLG (Figure 1D; Supplementary Table 4).

The KEGG pathway enrichment involved in predicted targets is systematically elucidated based on the KEGG pathway database. In the current study, KEGG pathway enrichment was performed according to the common 42 genes, and the top 20 enrichment data were plotted as a bar diagram (Figure 2A; Supplementary Table 5). The KEGG pathway enrichment mainly involved in neuroactive ligand-receptor interaction, complement and coagulation cascades, calcium signaling pathway, endocrine, and other factor-regulated calcium reabsorption, and renin–angiotensin system, etc. The integrative network of VK2 for salt-sensitive hypertension through the network pharmacology-based findings was conducted and shown in Figure 2B.

Vascular endothelial cells maintain the normal flow of blood and act as a barrier between blood and tissue fluid, whose injury is related to the occurrence of hypertension (16, 17). Therefore, the blood pressure and the ultrastructural changes of aortic vessels were observed in animal experiment. The results of animal experiment showed that significantly increased blood pressure was found in the HS group compared with the control group, whereas a significantly decreased blood pressure was found after VK2 supplementation compared with the HS group (Figure 3A). Moreover, the results of ultrastructural changes of aortic vessels are shown in Figure 3B. Compared with the control group, the vascular endothelial cells showed obvious edema, the intracellular matrix became lighter, the electron density of large area decreased, the internal elastic membrane appeared obvious local fracture; the nucleus showed irregular shape, local depression, heterochromatin edge set; the number of mitochondria decreased, swelling, mitochondrial cristae became shorter, and disappeared; and the rough endoplasmic reticulum showed obvious expansion and degranulation (Figure 3B). After VK2 supplementation, the results showed that the edema of vascular endothelial cells was alleviated; the electron density of intracellular matrix was low and the transparency was reduced; the internal elastic membrane was not obviously broken; the damage of cell structure such as nucleus, mitochondria, and rough endoplasmic reticulum was alleviated (Figure 3B).

In order to further verify the potential mechanism of VK2 resistance to salt-sensitive hypertension, we supplemented VK2 to intervene in high-salt induced mice, and detected the characteristics of gut bacteria by 16S rRNA analysis (Figure 4A) The results showed that there were 690 total OTUs in the control group, 771 total OTUs in the HS group, 663 total OTUs in VK2_HS group, and 530 common OTUs in the three groups (Figure 4B). Among them, there were 585 common OTUs in the control group and HS group, 59 personalized OTUs in the control group and 138 personalized OTUs in the HS group, which indicated that OTUs of high salt diet were higher than those of normal diet. There were 667 common OTUs in the HS group and VK2_ HS group, 39 personalized OTUs in the VK2_HS group, 139 personalized OTUs in the HS group, which indicated that VK2 supplementation reduced OTUs in the gut bacteria of mice fed with high salt diet.

Furthermore, phyla level results showed that the proportions of dominant phyla Firmicutes, Bacteroidetes, and Proteobacteria in the control group were 41.26, 40.06, and 5.823%, whereas 43.32, 36, and 4.899% were in the HS group, and 49.99, 34.89, and 4.665% were in the VK2-HS group, respectively, (Figure 4C; Supplementary Table 6). In addition, 20 dominant genera were found in bacteria (Figure 4D; Supplementary Table 7). For instance, the relative abundances of norank_f__Muribaculaceae, Lactobacillus, Lachnospiraceae_NK4A136_group, Helicobacter in the control group were 34.49, 11, 10.54%, and 6.87; 29.07, 15.60, 5.83, and 5.24% were in the HS group; and 27.29, 2.17, 12.16, and 2.33% were in the VK2_HS group. All these findings revealed that there were differences in bacterial composition and structure after VK2 supplementation.

Similarities of the bacterial composition and structure after VK2 supplementation among groups were compared using ANOSIM and PLS-DA based on Bray–Curtiss (18). The results distance box plot showed that there were significant differences in the bacterial composition and structure (Figure 5A; r = 0.2639; p = 0.025) among different groups. Moreover, there was an obvious tendency for separating the bacterial profiles on genus level after VK2 supplementation (Figure 5B), and a great difference in the individual samples was found in the HS group, but little difference in the individual samples after VK2 supplementation (Figure 5B), which promulgated a positive action on genera after VK2 supplementation. Additionally, a total of 29 different bacteria after VK2 supplementation were screened among the three groups (Figure 5C; Supplementary Table 8).

To identify signaling pathways found through the previous network of VK2-treated salt-sensitive hypertension, functional analysis was performed using PICRUSt2 after VK2 supplementation. A total of 10 common signaling pathways were enriched and identified. As shown in Figure 5A, compared with the control group, decreased relative abundance was found in the signaling pathways in epithelial cell signaling in Helicobacter pylori infection, estrogen signaling pathway, and prolactin signaling pathway in the HS group; increased relative abundance was found in the renin–angiotensin system, cGMP-PKG signaling pathway, dopaminergic synapse, endocrine, and other factor-regulated calcium reabsorption, inflammatory mediator regulation of TRP channels, vascular smooth muscle contraction, and calcium signaling pathway in the HS group. Increased relative abundance was found in the signaling pathways in epithelial cell signaling in Helicobacter pylori infection, estrogen signaling pathway, and prolactin signaling pathway after VK2 supplementation; decreased relative abundance was found in the renin–angiotensin system, cGMP-PKG signaling pathway, dopaminergic synapse, endocrine, and other factor-regulated calcium reabsorption, inflammatory mediator regulation of TRP channels, vascular smooth muscle contraction, and calcium signaling pathway after VK2 supplementation (Figure 6A). Moreover, there was a greatest change multiple and statistical significance of the relative abundance in renin-angiotensin system (p < 0.05). Also, the mapper of the renin–angiotensin system is shown in Figure 6B combined with the findings in the previous network of VK2-treated salt-sensitive hypertension. All these results further revealed the fact that renin-angiotensin system was the potential mechanisms of VK2-treated salt-sensitive hypertension.

To further verify the effect of VK2 on the renin–angiotensin system, the renin–angiotensin system-related proteins expression (including REN, ACE, AT1R, and AT2R) in salt-induced mice were conducted. Significantly increased REN, ACE, AT1R, and AT2R proteins expression were found in the HS group compared with the control group. Moreover, significantly decreased REN, ACE, AT1R, and AT2R proteins expression were found after VK2 supplementation (Figures 7A–E). Obviously, the changes in the renin–angiotensin system-related proteins expression exhibited VK2 inhibited renin-angiotensin system-related proteins expression in salt-induced hypertensive mice, which was statistically significantly consistent with the previous biological and functional prediction analysis.

To reveal the relationship between differential bacteria and functional proteins in renin-angiotensin system-related proteins, Spearman correlation analysis of the renin–angiotensin system-related proteins and gut microbiota were performed. The results of spearman correlation analysis indicated that the 20 different bacteria such as Dubosiella, Ileibacterium, etc., had a positive or negative correlation with REN, ACE, AT1R, and AT2R in salt-induced mice (Figure 8).