Dissection of the Functional Mechanism of Human Gut Bacterial Strain AD16 by Secondary Metabolites’ Identification, Network Pharmacology, and Experimental Validation

Gut microbiota plays important roles in several metabolic processes, such as appetite and food intake and absorption of nutrients from the gut. It is also of great importance in the maintenance of the health of the host. However, much remains unknown about the functional mechanisms of human gut microbiota itself. Here, we report the identification of one anticancer gut bacterial strain AD16, which exhibited potent suppressive effects on a broad range of solid and blood malignancies. The secondary metabolites of the strain were isolated and characterized by a bioactivity-guided isolation strategy. Five new compounds, streptonaphthalenes A and B (1-2), pestaloficins F and G (3-4), and eudesmanetetraiol A (5), together with nine previously known compounds, were isolated from the effective fractions of AD16. Structures of the new compounds were established by 1D and 2D NMR and MS analysis, and the absolute configurations were determined by the CD method. The analysis of network pharmacology suggested that 3, 2, and 13 could be the key components for the anti-NSCLC activity of AD16. In addition to the PI3K–Akt signaling pathway, the proteoglycans in cancer pathway could be involved in the anti-NSCLC action of AD16.


INTRODUCTION
The human gut microbiota is composed of an enormous diversity of microorganisms, including bacteria, fungi, and other microbes, which together play important roles in maintaining the dynamic homeostatic and healthy micro-environment of the host (Johnson et al., 2016;Thomas et al., 2017;Liang et al., 2018). In recent years, numerous discoveries have been reported on the human gut bacteria affecting human health and diseases, such as cardiovascular diseases, inflammatory diseases, obesity, and especially cancer (Hasan et al., 2020;Moritz et al., 2020). There has been mounting evidence supporting the roles of the gut bacteria in response to cancer (Johnson et al., 2016;Li et al., 2019), such as producing anticancer metabolites (Zhou et al., 2017). Although several bioactive metabolites from animal gut bacteria have been reported, such as sannastatin (Yang et al., 2011), few therapeutic metabolites have been identified from human gut bacteria (Rahim et al., 2019).
In this paper, we describe the anticancer activities of gut bacterial strain AD16, the isolation and structural elucidation of five new compounds, along with nine known ones via bioactivity-guided isolation, and the network analysis of the compounds from AD16. The chemical structures of the isolated compounds were deduced by means of their physicochemical properties, as well as the analysis of their spectroscopic data. This work demonstrated that gut microbiota is a rich source of potential cancer therapeutics for further studies and future clinical applications.

General Experimental Procedures
Optical rotations were measured on a Nicolet iS5 (Thermo, United States) spectrometer, and UV spectra were recorded on an Evolution 220 (Thermo, United States) UV/Vis spectrometer. IR spectra were obtained using a JASCO FT/IR-480 plus spectrometer. 1 H-NMR and 2D NMR spectra were measured on a Bruker AV-600 spectrometer, while 13 C-NMR spectra were measured on a Bruker AV-400 spectrometer. CD spectra were recorded on MOS 450 (Bio-Logic, France). HRESIMS data were determined by an Agilent Q-TOF 6520 mass spectrometer. Open column chromatography (CC) was performed using silica gel Qingdao Haiyang Chemical Group Corp.,Qingdao,China), ODS (50 μm, YMC, Japan), and HW-40 (Tosoh, Japan). Thin-layer chromatography (TLC) was performed using precoated silica gel plates (silica gel GF254, 1 mm, Yantai).

Isolation and Identification of AD16
The detailed collection and isolation procedures of the bacteria from human fecal specimens were done as previously reported (Zhuang et al., 2019). A colony of bacteria that showed potent anticancer activities was identified as closely related to Streptomyces and was given the strain name AD16 (gene bank No. KU883604.1). This strain was isolated from the fecal specimen of a healthy girl (5 years old) and stocked in the Laboratory of Genomics Research Center of Harbin Medical University (Harbin, China). All the experiments of the study were consistent with standard biosecurity and institutional safety procedures. All microbes were handled in the BSL-2 laboratory.

Cell Culture and CCK-8 Assay
Human solid cancer cell lines, including cervical cancer HeLa, ovarian cancer A2780, lung cancer A549, and colorectal cancer HCT116, were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum. Ovarian cancer cell lines ES-2 and OV-90 were cultured in McCoy's 5A medium with 10% fetal bovine serum. All the cultures were maintained in an incubator at 37°C with 5% CO 2 in a humidified atmosphere.
Cell viability was measured by the Cell Counting Kit (CCK)-8 (Dojindo, Tokyo, Japan) assay. A549 cells (5.0 × 10 3 cells per well) were seeded into 96-well plates (Corning, NY) and cultured for 24 h. The cells were then incubated with fresh media containing the compounds under study at various concentrations for 24, 48, or 72 h. After incubation, the media were removed and the wells were washed twice with PBS to remove non-adherent cells. Then, 100 μL fresh medium and 10 μL CCK-8 were added to each well at the indicated time points. The cells were further incubated at 37°C for 60 min. The absorbance of the samples was measured at 492 nm using a Bio-Rad model 3550 microplate reader (Richmond, CA).

Morphological Assessment
Morphological changes of cells treated with AD16 supernatant or metabolites were inspected by phase-contrast inverted microscopy (Zeiss Axiocam ERc 5s, Germany). The performance of the experiments and the determination of experimental results were completed blindly and separately by at least two different persons.

Cell Apoptosis Analysis
The cells were incubated in the medium containing culture supernatant of AD16 for 6 h. The cells were harvested, washed twice with cold 1 × PBS, and re-suspended in 100 μL 1 × binding buffer at a density of 1 × 10 5 cells/mL. The cells were then stained with 5 μL Annexin V and 5 μL PI (BD Biosciences) for 15 min in dark condition at room temperature. After staining, we added 400 µL of 1 × binding buffer to each tube. The samples were subjected to analysis by flow cytometry (BD FACSCanto TM II). The early apoptosis was evaluated based on the percentage of Annexin V-positive and PI-negative cells, while the late apoptosis was evaluated based on the percentage of Annexin V-positive and PI-positive cells.

Statistical Analysis
Statistical analysis was presented as the mean ± standard deviation (SD) of at least three independent experiments. Student's t-test, chi-square test, and Spearman's rank correlation analysis were used to assess the means of the different samples with SPSS statistical software version 17.0 and GraphPad Prism software. The statistical significance was accepted at p < 0.05. Our study closely followed the line of randomness and preciseness to ensure reproducibility.

Fermentation, Extraction, and Isolation of AD16
Strain AD16 was inoculated in 500 ml conical flasks (497 bottles in total) containing 300 ml GRC1 medium (20 g of soluble starch, 1 g of KNO 3 , 0.5 g of KH 2 PO 4 , 0.5 g of MgSO 4 7H 2 O, and 0.5 g of NaCl in 1 L of distilled water) for 15 days at 150 rpm/min at room temperature. D101 macroporous resin was soaked with the whole culture for 24 h and then eluted with water and EtOH-H 2 O (95:5, V/V), respectively. The EtOH-H 2 O eluate was concentrated by a rotary evaporator in vacuum to afford 33.8 g of dry material. An aliquot (31.7 g) was applied to an ODS column (3.5*46 cm; 50 μm) and eluted with MeOH-H 2 O in gradient to give 13 fractions (K1-K13 for 1 H-NMR (CD 3 OD, 600 MHz) and 13 C-NMR (CD 3 OD, 100 MHz) data, see Table 2.  H-NMR (CD 3 OD, 600 MHz) and 13 C-NMR (CD 3 OD, 100 MHz) data, see Table 2.

Target Network Analysis
The ingredients isolated were imported into the PubChem database and ChemBio3D Ultra 14.0, and the 3D molecular structures were exported in the form of SDFs. The targets were retrieved from the online target prediction platform PharmMapper (http://www.lilab-ecust.cn/pharmmapper/). Human species was used for target prediction, and the targets with Norm Fit ≥ 0.75 were collected. Thereafter, the targets were converted to gene names using the UniProt Knowledgebase (UniProtKB, http://www.uniprot.org/), and species were restricted to "Homo sapiens." Meanwhile, the NSCLC-related targets were obtained from the DisGeNET database (http:// www.disgenet.org/) and TTD (http://database.idrb.cqu.edu.cn/ TTD/). The STRING database (version 11.0, https://string-db. org/) was used to explore the protein-protein interactions (PPIs), and protein interactions with a confidence score > 0.4 were selected in the designed setting after eliminating duplicates and independent ones. Cytoscape software (version 3.7.2) was applied to construct the chemical-target network and protein-protein interaction (PPI) network. All genes were subjected to pathway enrichment analysis (KEGG analysis) using DAVID Bioinformatics Resources 6.8, and those pathway terms with a p-value < 0.05 were regarded as significant and interesting .

RESULTS AND DISCUSSIONS Cytotoxic Effects of the Extract of AD16
The anticancer activity of the EtOAc extract of AD16 was investigated. Strain AD16 exhibited a broad killing spectrum of cancers including lung cancer (A549), ovarian cancer (A2780, ES-2/OV-90), colorectal cancer (HCT116), and cervical cancer (HeLa) at the concentration of 5 μL/ml ( Figure 1A). The CCK-8 result of A549 cells incubated with AD16 demonstrated that the effects of AD16 were dose-and time-dependent against A549 as judged by cell proliferation percentages in comparison with the control ( Figure 1B). The colony formation activity against A549 cells was also investigated, which indicated that AD16 could strongly inhibit colony formation of the A549 cell line (Figures 1C,D). To determine the possible mechanism of the anticancer effects of AD16, we detected the induction to apoptosis after treatment with AD16. Six hours after treatment with different concentrations, cells were double-stained with Annexin V and PI and subjected to flow cytometry to quantitatively analyze the apoptotic effects. As The paired-sample t-test was used to analyze whether there was a significant difference in the number of colony formation between each AD16-added group and control (*p < 0.05, ***p < 0.001). Ctrl, control.
Frontiers in Pharmacology | www.frontiersin.org November 2021 | Volume 12 | Article 706220 illustrated in Figure 2A, the percentages of total apoptotic cells, including the early apoptotic portion (Annexin V positive) and the late apoptotic portion (Annexin V and PI positive), were dose-dependently increased with increasing concentrations of AD16 in the A549 cell line ( Figure 2B). These results suggested that the AD16 culture could suppress cell proliferation by inducing cell apoptosis.

Cytotoxic Effect of the Subfractions
Based on bioactivity-guided isolation, a large quantity of the AD16 extract was partitioned with ODS by MeOH-H 2 O gradients. All the fractions were examined to determine their anticancer effects at 100 μl/ml ( Figures 3A,B). When compared to other fractions, fractions 9-11 showed the highest inhibitory activities ( Figure 3B). Eventually, we isolated and identified 14 compounds, including five new compounds and nine previously known ones.

Network Pharmacology Analysis
Network pharmacology is a systems biology-based methodology focused on the complex interaction network composed of diseases, genes, protein targets, and drugs using holistic and systemic views in a biological system, offering an effective strategy to uncover the overall action mode of multiple compounds (Bu et al., 2021;Tu et al., 2021). Therefore, to predict the underlying mechanism of AD16, a network pharmacology approach was applied. All the isolated compounds were used for target prediction, and the targets with the probability more than 0.75 were used for analysis. As a result, a total of 89 targets were summarized. The target-compound network was constructed as well ( Figure 7A). The DisGeNET database and TTD search was performed to predict 592 targets associated with NSCLC. Then, eleven targets were screened out by looking for the overlapping targets from the compound-related targets and NSCLC-related targets ( Figure 7B). The connections of the targets are shown in Figure 7C. Ten targets were identified in the PPI network based on their topological parameters. The gene products AURKA, CHEK1, PGR, ESR1, MAPK1, CASP3, FGFR1, CDK2, KDR, and NOS3 with high node degree were considered the key targets of AD16 against NSCLC. Among them, three targets with a higher degree value among the anti-NSCLC activity of AD16 are caspase 3 (CASP3), estrogen receptor 1 (ESR1), and mitogen-activated protein kinase 1 (MAPK1).
It can be seen from the results of the interactions between components of AD16 and NSCLC targets ( Figure 7D) that 12 in 14 ingredients could correspond to multiple targets within multiple pathways, which were responsible for the anti-NSCLC effect of AD16. The compounds pestaloficin F (3), streptonaphthalene B (2), and 4-hydroxy-8-[6-hydroxy-1,3,7trimethyl-2-oxo-oct-3-enyl]-5-methyl-oxocan-2-one (13) having the highest degree value (6, 5, 5), which attribute nodes in the network graph, could be considered the core ingredients in the network with a major anti-NSCLC effect. Over the years, butenolides and tetralones have played an important role in drug discovery, design, and development of plentiful pharmacologically active moieties. A lot of natural butenolides have been isolated from endophytic fungus and other microbial sources, which covered a broad range of therapeutic activities,  including anticancer effects (Kornsakulkarn.et al., 2011;Kil et al., 2018;Husain et al., 2019;Wang et al., 2019;Yang et al., 2019). Compounds 3 (a new butenolide) and 2 (a new tetralone), proposed to be active constituents of AD16 herein, could act as leading compounds for further structural modification and drug design.
In addition, our result showed that cinnamic acid (9) played function on the target KDR. According to the references, 9 significantly increased the ratio of tumor growth inhibition, mean survival time, and percentage of the lifespan of the treated mice (Almeer et al., 2019). Furthermore, 9 induced angiogenesis in vivo and in vitro, which is related to VEGF and Flk-1/KDR expressions of endothelial cells (Choi et al., 2009). It was also reported that DBP (12) could inhibit the PI3K/Akt signaling pathway in INS-1 cells to induce cell apoptosis . These results partially supported these biological processes predicted by network pharmacology.
Furthermore, potential regulated biological processes and signaling pathways of AD16 treatment were predicted by KEGG analysis, and anti-NSCLC-related signal pathways were summarized ( Figure 8). In addition to the PI3K-Akt signaling pathway and proteoglycans in cancers, pathways in cancer, viral carcinogenesis, and oocyte meiosis were the other main patterns for AD16 to achieve its anti-NSCLC effects (Zhang et al., 2017;Chen et al., 2018;Kim et al., 2018).
To summarize, we isolated one bacterial strain AD16 from human gut microbiota that had significant cytotoxic effects on A549. Fourteen compounds were isolated and identified by various chromatographic methods. Among them, five compounds were new, and their structures were determined by NMR, HRESIMS, and CD methods. However, as the amount of components isolated was limited, we inferred the anti-NSCLC mechanism of the AD16 compounds mainly based on network pharmacology. Network pharmacology analysis revealed that the regulation of AD16 on NSCLC could be via acting on multiple targets, multiple pathways, and multiple biological processes. Compounds 3, 2, and 13 might possibly be the key components of AD16 for its anti-NSCLC effects. In addition, the PI3K-Akt signaling pathway and proteoglycans in cancer pathway were the main patterns for AD16 to achieve its anti-NSCLC effects. Our work demonstrated the function mechanism of the human gut bacterial strain AD16 by secondary metabolites' identification, network pharmacology, and experimental validation. It not only expanded the chemical and pharmacological diversities of metabolites from gut microbiota but also recommended that gut microbiota is of great potential for the discovery of new anticancer agents.

DATA AVAILABILITY STATEMENT
We obtained a written informed consent from participant or their guardian, consistent with the 1975 Declaration of Helsinki. All experimental protocols were reviewed and approved by The Ethics Committee, Harbin Medical University, and all experiments were performed in accordance with relevant guidelines and regulations.

ETHICS STATEMENT
The animal study was reviewed and approved by the Harbin Medical University Ethics Review Committee.