**2. Results and Discussion**

The EtOAc extract of the rice solid culture of *A. versicolor* DY180635 was fractionated by column chromatography (CC) on macroporous adsorbent resin, silica gel, and octadecyl silane (ODS), as well as by preparative HPLC, to afford 15 DKPS and two intermediates. Six new DKPs, named as aspamides A**–**F (**1–6**), were determined by comprehensive spectroscopic analysis including 1H nuclear magnetic resonance (NMR), 13C NMR, HSQC, heteronuclear multiple bond correlation (HMBC), rotating frame Overhauser effect spectroscopy (ROESY), and high resolution electrospray mass spectrometry (HRESIMS) spectra. By comparing the NMR and ESIMS data to the reported literatures in detail, 11 known compounds were determined as brevianamides K, N, and M (**7**, **16**, **17**) [28], brevianamide Q (**8**) [29], brevianamides V, U (**9–10**) [20], brevianamide F (**11**) [30], deoxybrevianamide E (**12**) [31], *N*-Prenyl-*cyclo*-l-tryptophyl-l-proline (**13**) [32], 2-(2-methyl-3-en-2-yl)-1*H*-indole-3-carbaldehyde (**14**) [33], and 2-(1,1-Dimethyl-allyl)-1*H*-indol-3-ylmercuric acetate (**15**) [34]. Herein, the details of the isolation, structural elucidation of these new compounds, and their bioactivities are described.

Aspamide A (**1**) was isolated as a yellowish powder. The UV spectrum with λmax (logε) in methanol at 200 (6.13), 224 (6.13), 284 (5.46), and 341 (5.68) nm was indicative of indole functionality with an extended conjugation [26]. Its molecular formula was determined as C23H27N3O3 on the basis of high-resolution ESIMS (*m*/*z* 394.2117 [M + H]<sup>+</sup>, calcd. for C23H28N3O3, 394.2125) and 13C NMR data, requiring 12 degrees of unsaturation. The 1H NMR, 13C NMR, and heteronuclear multiple quantum correlation (HMQC) spectra (Table 1 and Figure S3 in Supporting Information) showed three methyl groups (δ<sup>C</sup> 27.4, δ<sup>H</sup> 1.45; δ<sup>C</sup> 27.8, δ<sup>H</sup> 1.49; δ<sup>C</sup> 15.2, δ<sup>H</sup> 1.13), three sp<sup>3</sup> methylenes (including one oxygenated methylene) (δ<sup>C</sup> 29.6, δ<sup>H</sup> 1.75, δ<sup>H</sup> 1.97; δ<sup>C</sup> 25.9, δ<sup>H</sup> 2.13, δ<sup>H</sup> 2.27; δ<sup>C</sup> 63.6, δ<sup>H</sup> 3.65), two sp3 methine carbon signals (including one oxygen-bearing carbon) (δ<sup>C</sup> 56.5, δ<sup>H</sup> 4.57; δ<sup>C</sup> 86.7, δ<sup>H</sup> 5.59), six sp2 methines, one sp2 methylene, seven sp2, and one sp3 non-protonated carbon. The NMR data and UV absorptions were close to those of brevianamide V [20], with the exception that there was an additional oxygenated aza-acetal structure located at the proline motif (δ<sup>C</sup> 86.7, δ<sup>H</sup> 5.59; δ<sup>C</sup> 63.6, δ<sup>H</sup> 3.65; δ<sup>C</sup> 15.2, and δ<sup>H</sup> 1.13).

Further information about the structure was derived from heteronuclear multiple bond correlation (HMBC) spectra analyses (Figure S4). The key HMBC correlations (Figure 2A) from OCH2CH3-6 (δ<sup>H</sup> 3.65) to C-6 (δ<sup>C</sup> 86.7), H-6 (δ<sup>H</sup> 5.59) to C-9 (δ<sup>C</sup> 56.5), H-9 (δ<sup>H</sup> 4.57) to C-8 (δ<sup>C</sup> 25.9) and C-1 (δ<sup>C</sup> 165.9), and H-8α (δ<sup>H</sup> 2.27) to C-9 were observed. These data suggested that an oxethyl group was located at C-6 and confirmed the oxygenated aza-acetal structure at the proline motif, which was previous unpresented in the brevianamide analogues. Thus, the planar structure of **1** was determined as shown in Figure 1.

In order to determine the relative configuration of **1**, the rotating frame Overhauser effect spectroscopy (ROESY, Figure S5) experiment was performed. The ROESY correlation (Figure 2B) between NH-2 (δ<sup>H</sup> 9.01) and H-13 (δ<sup>H</sup> 7.29) revealed the *Z* configuration about Δ3,10, and the ROESY signals between H-8β (δ<sup>H</sup> 2.13) and H-6, and between H-8α (δ<sup>H</sup> 2.27) and H-9 suggested that H-6 and H-9 were *trans* form. The absolute configuration of **1** was assigned as (6*R*,9*S*) by comparing the experimental and calculated electronic circular dichroism (ECD) values obtained using Time-dependent Density functional theory (TD-DFT) at the B3LYP/6–31+g (d, p) level (Figure 2C).

**Figure 2.** The key heteronuclear multiple bond correlation (HMBC) (**A**) and rotating frame Overhauser effect spectroscopy (ROESY) (**B**) correlations, and experimental and calculated electronic circular dichroism (ECD) spectra (**C**) of compound **1**.

Aspamide B (**2**) was obtained as a yellowish powder. Its molecular formula was determined as C23H27N3O3 on the basis of HRESIMS (*m*/*z* 394.2120 [M + H]<sup>+</sup>, calcd. for C23H28N3O3, 394.2125) and 13C NMR data, corresponding to 12 degrees of unsaturation. By comparing the 1H, 13C, and HMQC data (Table 1, Figure S11) of **2** with those of **1**, it was discovered that **2** possessed the identical planar structure as that of **1**. Further analyses of the 2D NMR data of **2**, the key HMBC correlations from H-6 (δ<sup>H</sup> 5.35) to OCH2CH3-6 (δ<sup>C</sup> 64.3) and C-9 (δ<sup>C</sup> 58.7), as well as H-9 (δ<sup>H</sup> 4.44) to C-1 (δ<sup>C</sup> 166.8) revealed that the oxethyl group was located at C-6. Furthermore, the difference between H-6 (δ<sup>H</sup> 5.35, dd, *J* = 2.8, 1.7 Hz) in **2** and H-6 (δ<sup>H</sup> 5.59, dd, *J* = 5.7, 1.7 Hz) in **1** indicated that **2** was the C-6 epimer of **1**. Furthermore, the same ECD cotton effects (Figure 3C) of **2** compared to **1** indicated that the absolute configuration of C-9 in **2** was consistent with that in **1**. This result suggested that within the used spectral window, the ECD cotton effects were mainly caused by the chiral center of C-9 in both

compounds **1** and **2**, and it could also be confirmed by the experimental ECD data of (+)-brevianamide V and (−)-brevianamide V [35]. Thus, the absolute configuration of compound **2** was ascertained as (6*S*,9*S*).

Aspamide C (**3**) was obtained as a yellowish powder. The molecular formula was established as C21H23N3O3 by HRESIMS (*m*/*z* 366.1807 [M + H]<sup>+</sup>, calcd. for C21H24N3O3, 366.1812), indicating 12 degrees of unsaturation. The UV and NMR spectra were very similar to those of compound **1**. A comparison of the NMR data for **3** with **1**, together with characteristic HMBC signals (Figure 3A), suggested that the oxethyl group was replaced by a second OH in **3**. However, the OH group was not located at C-6, which was the same as the **1**, to form the aza-acetal structure for which the chemical shift of oxygenated methylene (δ<sup>C</sup> 66.6) was far below the shift of C-6 (δ<sup>C</sup> 86.7) in **1**. Thus, the second OH was distributed to C-7, and the planar structure of **3** was determined as shown in Figure 1. The ROESY correlation (Figure S20) between NH-2 (δ<sup>H</sup> 8.93) and H-13 (δ<sup>H</sup> 7.29) confirmed the *cis* form of the double bond between C-3 and C-10. Furthermore, the ROESY signals (Figure 3B) between H-8β (δ<sup>H</sup> 1.98) and H-7 (δ<sup>H</sup> 4.35), and between H-8α (δ<sup>H</sup> 2.12) and H-9 (δ<sup>H</sup> 4.65) revealed that H-7 and H-9 were *trans* form. Finally, the absolute configuration of **3** was determined as (7*R*,9*S*) by comparison of the experimental ECD curve of **3** with that of **1** (Figure 3C).

**Figure 3.** The key HMBC correlations (**A**) and partial enlarged view of ROESY spectra (**B**) of **3**, and experimental ECD spectra (**C**) of compounds **1–3**.


**Table 1.** 1H (600 MHz) and 13C (150 MHz) NMR data of **1**–**3** in DMSO-*d*6.


**Table 1.** *Cont*.

The racemic (±)-aspamide D (**4**) was isolated as colorless gum with the molecular formula of C23H27N3O3 from an HRESIMS peak at *m*/*z* 394.2120 [M + H]<sup>+</sup> (calcd. for C23H28N3O3, 394.2125). Its NMR data and UV absorption were similar to compound **1**. Comparing to **1**, the major change was that two sp<sup>3</sup> methines were replaced by one sp3 methylene and one sp<sup>3</sup> non-protonated carbon, suggesting that the oxethyl group was connected to C-9 rather to the C-6. Additionally, the key HMBC signals (Figure 4) from OCH2CH3-9 (δ<sup>H</sup> 3.53) to C-9 (δ<sup>C</sup> 91.0), H-6 (δ<sup>H</sup> 3.62) to C-4 (δ<sup>C</sup> 159.4) and C-9, and H-8a (δ<sup>H</sup> 2.02) to C-1 (δ<sup>C</sup> 163.1) confirmed the aforementioned planar structure. There was no Cotton effect observed on its ECD spectra (Figure S29), which in accordance with the racemic (±)-brevianamide X [35], indicating that **4** might be a pair of enantiomers. Furthermore, the chiral HPLC resolution of **4** contributed to the separation of a pair of enantiomers (+)-**4** and (−)-**4**, which exhibited nearly mirror-image ECD spectra (Figure 5). The absolute configurations of (+)-**4** and (−)-**4** were discriminably determined as 9*R* and 9*S* by comparing the experimental and calculated ECD data obtained using TD-DFT at the B3LYP/6-31+g (d, p) level (Figure 5). Correspondingly, we named (+)-**4** and (−)-**4** as (+)-aspamide D and (−)-aspamide D, respectively.

**Figure 4.** The key HMBC correlations of compounds **4**–**6**.

Aspamide F (**5**) was obtained as a brown powder with the molecular formula of C19H17N3O3 from an HRESIMS peak at *m*/*z* 336.1336 [M + H]<sup>+</sup> (calcd. for C19H18N3O3, 336.1343), requiring 13 indices of hydrogen deficiency. The 1H NMR, 13C NMR, and HMQC spectra (Table 2 and Figure S35) suggested the presence of one oxygenated methyl group, one sp<sup>3</sup> methylene, two sp3 methines carbon signals (including one oxygen-bearing carbon), nine sp2 methines, and six sp<sup>2</sup> non-protonated carbons. These NMR data of **5** were similar to those of brevianamide M [28], except for the different chemical shifts for C-2 and C-3 due to the presence of a methoxy at C-2 in **5**, implying that **5** was an analogue of brevianamide M with a methoxy at C-2. Additionally, the key HMBC correlations (Figure 4) from OCH3-2 (δ<sup>H</sup> 3.53) to C-2 (δ<sup>C</sup> 83.9), from H-2 (δ<sup>H</sup> 5.27) to C-3 (δ<sup>C</sup> 146.9)/C-14 (δ<sup>C</sup> 170.0), and from H-13

(δ<sup>H</sup> 5.53) to C-15 (δ<sup>C</sup> 40.0)/C-16 (δ<sup>C</sup> 135.9) confirmed the planner structure of **5** as shown in Figure 4. The absolute configuration of **5** was determined as (2*S*,13*S*) via comparing the ECD curve (Figure 6) of **5** with the brevianamide M (**17**).

**Figure 5.** Experimental and calculated ECD spectra of **4**.



Aspamide G (**6**) was isolated as a brown powder. The molecular formula was determined as C20H19N3O3 by HRESIMS (*m*/*z* 350.1498 [M + H]<sup>+</sup>, calcd. for C20H20N3O3, 350.1499), which was 14 Dalton more than **5**. The 13C NMR data of **6** showed a close resemblance to those of **5**, except for an additional oxygenated sp<sup>3</sup> methylene, suggesting that there was an ethoxy group located at C-2 in **6**. The key HMBC signal (Figure 4) from H-2 (δ<sup>H</sup> 6.40) to OCH2-2 (δ<sup>C</sup> 66.2) verified that **6** was an analogue of brevianamide M (**17**) with an ethoxy motif at C-2. In addition, similar Cotton effects at 212, 220, and 237 nm in the ECD spectra (Figure 6) of **6** suggested that **6** and **17** had the same counterpart absolute configurations. Thus, the absolute configuration of **6** was assigned as (2*S*,13*S*), and it was elucidated as 6-ethoxy-aspamide F.

**Figure 6.** Experimental ECD spectra of **5**, **6**, and **17**.

All isolated compounds were tested for their anti-inflammatory activities in *P. acnes*-induced THP-1 cells; unfortunately, none of the compounds showed moderate anti-inflammatory properties. Aiming to give our contribution to the COVID-19 research, all compounds were selected for the virtual screening on the 3CL hydrolase (Mpro) of SARS-CoV-2, which had been exploited as a potential drug target to fight COVID-19 [27]. The docking scores of compounds **1–2**, **5**, **6**, **8** and **17** were top among all screened molecules (docking scores: −5.389, −4.772, −5.146, −4.962, −5.158), and the score of ritonavir [36] (a potent inhibitor in vitro of human immunodeficiency virus type 1 protease) was −7.039, which suggested that these compounds may be helpful in fighting COVID-19 after further studies.
