*2.1. Structural Determination*

Compound **1** was isolated as a white needle crystal and had the molecular formula C12H11ClO4 as determined by the HRESIMS spectrum, which showed a cluster of protonated ion peaks at *m*/*z* 255.0422/257.0394 [M + H]+ with the ratio of 3:1, indicative of a monochlorinated compound. The 1D NMR (Table 1) and HSQC spectrum of **1** showed signals of a carbonyl carbon (*δ*C 181.1), seven non-protonated *sp*<sup>2</sup> carbons (*δ*C 163.9, 160.0, 159.9, 151.6, 114.6, 103.7 and 97.4), one aromatic methine (*δ*H/C 6.62/95.9), one oxygenated methyl (*δ*H/C 3.94/57.0), and two methyls (*δ*H/C 2.42/18.3 and 1.91/8.8). The above-mentioned data combined with seven degrees of unsaturation suggested that **1** presented a chromone skeleton. The above NMR characteristics showed grea<sup>t</sup> similarity to those of the co-isolated **2**, which was reported as a chromone compound. The main difference was the presence of the chlorine atom instead of a hydrogen atom at C-8 in **1** and this deduction was supported by the above 2D NMR data. The HMBC correlations (Figure 2) from H3-9 to C-2 and C-3, from H3-10 to C-2, C-3, and C-4, revealed that the methyl groups CH3-9 and CH3-10 were located at C-2 and C-3, respectively. The HMBC correlations from 5-OH to C-4a, C-5, and C-6 indicated the location of the phenolic hydroxyl group (C-5). The chemical shift of C-8 (*δ*C 97.4) revealed the substitution of chlorine instead of the oxygenated methyl, which was attached at C-8. The methoxy group was deduced to link with C-7 by the HMBC signal of H3-11/C-7 and the chemical shift of C-7 (*δ*C 159.9). The X-ray crystal structure of

**1** (CDCC 2221470), obtained by slow evaporation in CH3OH, further confirmed the above elucidation of the planar structure. Compound **1** was unambiguously characterized as shown in Figure 1 and defined as 8-chlorine-5-hydroxy-2,3-dimethyl-7-methoxychromone.

Compound **3** was isolated as a red needle crystal, and its molecular formula of C4HCl2NO2 was determined by HRESIMS data at *m*/*z* 163.9318 [M–H]– (calcd for C4Cl2NO2, 163.9312). The analysis of its structure accurately from the NMR spectrum was difficult, because of its symmetrical structure and simplicity of carbons (*δ*C 164.1, 132.8) and hydrogens (*δ*H 11.71 (s, 1H)). However, we obtained X-ray crystal data (Cu K*α* radiation) of **3,** and the structure was accurately determined as 3,4-dichloro-1*H*-pyrrole-2,5-dione (Figure 3). Compound **3** has been previously identified as a synthetic product, which was synthesized to illustrate the insecticidal structure-activity relationship of the *N*-amino-maleimide derivatives [20]. Here we are reporting the first time this compound explored from nature, as a new metabolite or new natural product.

Compound **4**, obtained as yellow needles, was found to have the molecular formula C30H42O7 on the basis of HRESIMS data at *m*/*z* 515.3003 [M+H]+ (calcd C30H43O7, 515.3003). The 13C NMR and DEPT data displayed 30 carbon signals including eight methyls, five methylenes, two *sp*<sup>3</sup> methine, two *sp*<sup>2</sup> methine, three oxygenated *sp*3, one *sp*<sup>3</sup> quaternary, two oxygenated *sp*<sup>3</sup> quaternary, four *sp*<sup>2</sup> quaternary and three carbonyls. Detailed comparison of the above NMR data with the literature [21], **4** was identified as stemphone C. However, the absolute configuration of **4** had not been reported. We determined the absolute configuration of **4** as 4*S*, 5*S*, 13*R*, 14*R*, 17*R*, 18*R*, 21*R* (Figure 3) by X-ray crystallographic analysis using Cu K*α* radiation.

Meanwhile, the other seven known compounds were identified as 5-hydroxy-2,3- dimethyl-7-methoxychromone (**2**) [22], *cis*-cyclo (Tyr-Ile) (**5**) [23],4,8-dihydroxy-1-tetralone (**6**) [24], cyclo (Phe-Tyr) (**7**) [25], tenuissimasatin (**8**) [26], 4-methyl-5,6-dihydro-2H-pyran-2- one (**9**) [27], respectively, by comparison of their NMR data (supplementary information) with previous reports.

**Figure 2.** Key HMBC (arrows) correlations of **1**.

**Figure 3.** X-ray single-crystal structures of compounds **1**, **3**, and **4**.


**Table 1.** The NMR data of compound **1** (500 and 125 MHz, *δ* in ppm, DMSO-*d*6).

#### *2.2. Antimicrobial and Antiproliferative Activities*

All of the obtained compounds were evaluated for the activities of pathogenic fungi and bacteria commonly found in crop plants (Table 2). Compound **3** exhibited antifungal activities against *Botrytis cinerea*, *Verticillium dahlia* kieb., *Fusarium graminearum* schw., *Fusarium oxysporum* f.sp. niveum, *Rhizoctonia solani,* and *Septoria nodorum* Berk., with the MIC values of 25–50 μg/mL. Compound **4** exhibited antibiotic activity against bacteria *Erysipelothrix rhusiopathiae* WH13013 and *Streptococcus suis* SC19, with the MIC values of 1.56 and 6.25 μg/mL. In particular, the strength of **4** against *E. rhusiopathiae* was stronger than that of penicillin with the MIC value of 6.25 μg/mL.


**Table 2.** Antifungal, antibacterial, and cytotoxic activities of **3** and **4**.

a Cycloheximide; b Penicillin; c Streptomycin; d Docetaxel

Two human prostate cancer cell lines, PC-3 (androgen receptor negative) and 22Rv1 (androgen receptor positive), were used in the antiproliferative assay for all the obtained compounds. Compound **3** exhibited antiproliferative activity against 22Rv1 and PC-3 cells with IC50 values of 8.35 and 9.60 μM, respectively, while **4** showed activities against 22Rv1 and PC-3 cells with IC50 values of 5.81 and 2.77 μM, respectively. In order to evaluate whether **4** selectively inhibits prostate cancer cells, we screened other cancer cells for antiproliferative activity. The result showed that **4** was also active against other cells (HepG2, A549, Hela, WPMY-1, MC3T3-E1) with IC50 values of 3.63–11.68 μM. Thus, **4** had the most significant effect on PC-3 cells compared to other compounds with broader inhibitory activity.

To further evaluate the inhibitory effect of **4** on PC-3 cells, we performed a plate clone formation assay. The results showed that **4** significantly inhibited the formation of clonal colonies of PC-3 cells and its inhibitory effect was positively correlated with the dose (Figure 4). In addition, we examined the effect of **4** on PC-3 cell apoptosis by flow cytometry. It was revealed that **4** could significantly induce apoptosis in PC-3 cells. As shown in Figure 5, when PC-3 cells were treated with 10 μM of **4** for 48 h, 14.64% of the cells were induced to early apoptosis and 36.84% of the cells were induced to late apoptosis. This finding suggests that the induction of apoptosis in PC-3 cells is a mode of action for the production of antiproliferative activity by **4**.

**Figure 4.** Compound **4** reduced PC-3 cells colony formation in a dose-dependent manner.

**Figure 5.** Compound **4** triggered PC-3 cells apoptosis in a dose-dependent manner (**A**,**B**). All results were presented as mean ± standard deviation (SD). Statistical significance was determined with One-Way ANOVA. \*\* *p* < 0.01 was considered statistically significant.

To identify the inhibitory process of the proliferation of prostate cancer cells, we detected the cell cycle distribution of PC-3 cells. As shown in Figure 6, when treated with the M-phase blocker docetaxel, a large number of PC-3 cells were blocked in M-phase, with a dramatic increase in the ratio of G2/M, up to 71.23 percent. However, unlike in the case of docetaxel, the percentage of S-phase was significantly increased in cells treated with **4**. When the concentration of **4** reached 10 μM, the percentage of cells in the S-phase was as high as 59.14 percent. These results suggested that **4** blocked the cell cycle at Sphase, impairing cell proliferation. Consequently, it is revealed that **4** is a promising lead compound for pharmacotherapy of prostate cancer.

**Figure 6.** Compound **4** induced PC-3 cell cycle arresting at S phase (**A**,**B**). All results were presented as mean ± standard deviation (SD).
