Asperopiperazines A and B: Antimicrobial and Cytotoxic Dipeptides from a Tunicate-Derived Fungus Aspergillus sp. DY001

Investigation of the cytotoxic fractions of the ethyl acetate extract of the fermentation broth of the tunicate-derived Aspergillus sp. DY001 afforded two new dipeptides, asperopiperazines A and B (1 and 2), along with the previously reported compounds (+)-citreoisocoumarin (3) and (−)-6,8-di-O-methylcitreoisocoumarin (4). Analyses of the 1D and 2D NMR spectroscopic data of the compounds supported their structural assignments. Asperopiperazine A (1) is a cyclic dipeptide of leucine and phenylalanine moieties, which are substituted with an N-methyl and an N-acetyl group, respectively. On the other hand, asperopiperazine B (2) is a cyclic dipeptide of proline and phenylalanine moieties with a hydroxyl group at C-2 of the proline part. The absolute configuration of the amino acid moieties in 1 and 2 were determined by Marfey’s analyses and DFT NMR chemical shift calculations, leading to their assignment as cyclo(l-NMe-Leu-l-NAc-Phe) and cyclo(d-6-OH-Pro-l-Phe), respectively. Asperopiperazines A and B displayed higher antimicrobial effects against Escherichia coli and Staphylococcus aureus than Candida albicans. Furthermore, compounds 1–4 displayed variable growth inhibitory effects towards HCT 116 and MDA-MB-231 cells, with asperopiperazine A as the most active one towards HCT 116.


Introduction
Secondary metabolites played an important and inspiring role in the process of drug discovery and development [1][2][3][4][5]. Tunicates (ascidians) are well-known marine invertebrates and represent an extraordinary and major source of secondary metabolites with pharmaceutical and biomedical importance, including the marketed drugs Trabectedin and Plitidepsin [5].
The absolute configuration of the amino acid components of 1 was determined by Marfey's analysis [32]. After the treatment of 1 in 6 M HCl at 110 • C for 16 h, the hydrolysate was derivatized with 1-fluoro-2,4-dinitrophenyl-5-L-leucinamide (L-FDLA), and the retention times were compared with the FDLA derivatives of the standard amino acids phenylalanine and N-methylleucine (both D and L). The amino acids in 1 were found to possess the L-configuration. Figure 2 shows the reaction of L-FDLA with the hydrolytic products of 1. Accordingly, 1 was assigned as cyclo(L-NMe-Leu-L-NAc-Phe) and named asperopiperazine A.  Figures S16-S20, Supplementary Materials) spectra of 2 supported its planar structure. The 13 C NMR spectrum along with the HSQC showed signals for 14 carbons, which were classified into five aromatic methines, one aliphatic methine, four methylenes, and four quaternary carbons (Table 1). Again, the characteristic resonances for a cyclic dipeptide backbone [31] were noticed in 2. These included two carbonyls at δ C 166.1 (C, C-2) and 167.9 (C, C-5), a methine attached to heteroatom at δ H /δ C 4.45 (dd, J = 11.0, 3.3 Hz, H-3)/55.8 (CH, C-3), and an interchangeable signal at δ H 5.59 (1H, brs, NH-4). Moreover, the remaining 1 H and 13 C NMR in 2 supported the presence of a proline nucleus, which is substituted at C-2 (C-6 in 2) [31], and a phenylalanine part (Table 1)
Similarly, the existence of a phenylalanine part in 2 was supported from the amidic carbonyl 13 [31]. The 1 H-1 H COSY experiment supported the assignment of the substituted proline part from consecutive 3 J vicinal couplings from H 2 -7 to H 2 -8 and from H 2 -8 and H 2 -9. The absence of any further coupling within the proline part suggested the quaternary nature of C-6 [31]. The downfield shift of C-6 at δ C 87.6 [31] together with the one-proton interchangeable signal at δ H 2.99 (OH) supported the existence of a hydroxyl moiety at C-6 [31]. Finally, HMBC correlations of H 2 -8/C-6 (δ C 87.6), H 2 -7/C-6, and H 2 -9/C-6 supported the assignment of 6-OH-proline moiety (Table 1 and Figure 1).
Similarly, the existence of a phenylalanine part in 2 was supported from the amidic carbonyl 13  The vicinal coupling between the NH signal (δ H 5.58) and H-3 (δ H 4.45), which further couples with H-10a (δ H 3.63) and H-10b (δ H 2.74), as well as the consecutive couplings within the aromatic moiety from H-12 to H-16 secured the assignment of the phenylalanine moiety in 2. Further, the connection of the proline part with the phenylalanine parts was also supported by HMBC from NH (δ H/ 5.58) to C-6 (δ C 87.6) ( Table 1 and Figure 1). Similar to 1, the absolute configuration of the phenylalanine moiety in 2 was determined by Marfey's analysis [31]. After treatment of 2 in 6 M HCl at 110 • C for 16 h, the hydrolysate was derivatized with 1-fluoro-2,4-dinitrophenyl-5-L-leucinamide (L-FDLA), and the retention times were compared with the L-FDLA derivatives of the standard amino acids D-and L-Phe. The derivatized amino acid in 2 was found to possess the D-configuration. Thus, the existence of D-Phe unit was confirmed in 2. Figure 3 displays the reaction of L-FDLA with the hydrolytic product of 2.
Similar to 1, the absolute configuration of the phenylalanine moiety in 2 was determined by Marfey's analysis [31]. After treatment of 2 in 6 M HCl at 110 °C for 16 h, the hydrolysate was derivatized with 1-fluoro-2,4-dinitrophenyl-5-L-leucinamide (L-FDLA), and the retention times were compared with the L-FDLA derivatives of the standard amino acids D-and L-Phe. The derivatized amino acid in 2 was found to possess the Dconfiguration. Thus, the existence of D-Phe unit was confirmed in 2. Figure 3 displays the reaction of L-FDLA with the hydrolytic product of 2. Attempts to confirm the configuration of C-6 of the proline part by Marfey's method were impossible, due to the decomposition of 2 under acidic conditions.
To assign the absolute configuration of the OH moiety at C-6 of compound 2, the 13 C NMR chemicals shifts of the 6-OH proline moiety in 2 with cyclo(D-6-OH-Pro-L-Phe) and cyclo(L-6-OH-Pro-L-Phe) were compared. As shown in Table 2, the 13 C shifts of 2 are perfectly matching those of cyclo(D-6-OH-Pro-L-Phe) [33]. Thus, the D-configuration was assigned for the proline moiety in 2. Table 2. Comparison of the 13 C NMR data of the 6-OH-Pro moiety in 2 and cyclo(D-6-OH-Pro-L-Phe).

Carbon No.
Compound 2 Cyclo(6-OH-D-Pro-L-Phe) [33] ΔδC As shown in Figure 4, the 3R,6S configurations for 2 were supported by all of the metrics used. The correlation between the experimental and theoretical chemical shifts was higher for 3R,6S (R2 = 0.9997) and the mean average error (MAE) value was lower for 3R,6S isomer and, finally, a 99.9% DP4 score for this isomer ( Figure 4). All of the calculations for both isomers (3R,6R and 3R,6S) of 2 can be found in Table S1 of the Supplementary Materials. Accordingly, 2 was assigned as cyclo(D-6-OH-Pro-D-Phe) and named asperopiperazine B. Attempts to confirm the configuration of C-6 of the proline part by Marfey's method were impossible, due to the decomposition of 2 under acidic conditions.
To assign the absolute configuration of the OH moiety at C-6 of compound 2, the 13 C NMR chemicals shifts of the 6-OH proline moiety in 2 with cyclo(D-6-OH-Pro-L-Phe) and cyclo(L-6-OH-Pro-L-Phe) were compared. As shown in Table 2, the 13 C shifts of 2 are perfectly matching those of cyclo(D-6-OH-Pro-L-Phe) [33]. Thus, the D-configuration was assigned for the proline moiety in 2.  As shown in Figure 4, the 3R,6S configurations for 2 were supported by all of the metrics used. The correlation between the experimental and theoretical chemical shifts was higher for 3R,6S (R2 = 0.9997) and the mean average error (MAE) value was lower for 3R,6S isomer and, finally, a 99.9% DP4 score for this isomer (Figure 4). All of the calculations for both isomers (3R,6R and 3R,6S) of 2 can be found in Table S1 of the Supplementary Materials. Accordingly, 2 was assigned as cyclo(D-6-OH-Pro-D-Phe) and named asperopiperazine B.  Compound 3 possesses molecular formula C14H14O6 as supported from (+)-HRESIMS ( Figure S15, Supplementary Materials). Its structure was assigned by interpretation of its 1D and 2D NMR spectra (Figures S16-S20, Supplementary Materials). The 1 H and 13 C NMR data of 3 are similar to those reported for (+)-citreoisocoumarin [28,29]. In addition,   Figure S15, Supplementary Materials). Its structure was assigned by interpretation of its 1D and 2D NMR spectra ( Figures S16-S20, Supplementary Materials). The 1 H and 13 C NMR data of 3 are similar to those reported for (+)-citreoisocoumarin [28,29]. In addition, the optical rotation value of compound 3 was [α] D = +43.5 • , which was in good agreement with the reported value ([α] D = +46.5 • ) for (+)-citreoisocoumarin [28]. Thus, 3 was assigned as (+)-citreoisocoumarin.

Biological Activities of the Compounds
The antimicrobial effects of 1-4 were determined in a disc diffusion assay at a concentration of 50 µg/disc against three organisms. Compounds 1 and 2 showed moderate antibacterial effects towards E. coli (inhibition zones = 17-22 mm) and S. aureus (inhibition zones = 16-18 mm) and a lower effect against C. albicans (inhibition zones = 11-12 mm), suggesting their higher effects towards E. coli and S. aureus (Table 3).  On the other hand, compound 3 displayed a higher activity against S. aureus (inhibition zone of 19 mm) than C. albicans (17 mm) and E. coli (11 mm). On the contrary, compound 4 was less active than 3 (inhibition zones 6-9 mm) against these microbes. This difference in the activity between 3 and 4 could be attributed to the lack of the phenolic moieties in 4.
Furthermore, 1 and 2 displayed minimum inhibitory concentration (MIC) values of 8 and 4 µM against E. coli, and 8 and 8 µM against S. aureus, respectively. On the contrary, compound 3 was active against C. albicans with MIC value of 8 µM (Table 3).
The results represent the mean of three independent experiments; ( b ) 5-Flourouracil, a positive cytotoxic control.

General Experimental Procedures
Optical rotations were acquired on a digital DIP-370 polarimeter (JASCO, Oklahoma City, OK, USA). The IR spectra were recorded on a Shimadzu Infrared-400 spectrophotometer (Shimadzu, Kyoto, Japan). One-and two-dimensional NMR spectra were acquired on Bruker Avance DRX 600 MHz (600 MHz for 1 H and 150 MHz for 13

Purification of the Fungal Isolate
The fungal isolate DY001 ( Figure 5) was purified from the internal tissue Didemnum species, as described earlier [34]. In brief, surface sterilization of the tis the Didemnum species was performed using 70% EtOH. A small piece of the in tissue of the tunicate was mixed with sterile seawater (10 mL). The homogenized tis the tunicate was diluted serially to 1:10, 1:100, and 1:1000. About 100 μL of each di was used for culturing the isolate on Sabouraud dextrose agar medium [34]. The me

Purification of the Fungal Isolate
The fungal isolate DY001 ( Figure 5) was purified from the internal tissue of the Didemnum species, as described earlier [34]. In brief, surface sterilization of the tissue of the Didemnum species was performed using 70% EtOH. A small piece of the internal tissue of the tunicate was mixed with sterile seawater (10 mL). The homogenized tissue of the tunicate was diluted serially to 1:10, 1:100, and 1:1000. About 100 µL of each dilution was used for culturing the isolate on Sabouraud dextrose agar medium [34]. The medium was prepared in sterile seawater containing 0.25% chloramphenicol (w/v). The cultured plates were stored for 1-2 weeks at 30 • C until complete growth of the fungal isolate. Subsequent purification steps were carried out until a pure strain was obtained.

Purification of gDNA from Fungal Isolates
The fungal isolate was cultured in Sabouraud dextrose broth at 28 • C for 3-4 days. The mycelia were separated from the broth and freeze-dried separately. The fungal gDNA was extracted using the DNA Mini Kit QIAamp, according to the instructions of the manufacturer and as reported earlier [34].

Fungal ITS-rDNA Fragments' Amplification
The gDNA of the fungal strain DY001 represented the template for the amplification of the Internal Transcribed Spacer-rDNA (ITS-rDNA) fragments using the primers ITS1 and ITS4 [34,35]. The mixture for amplification and the conditions of the PCR were used as previously described [34,35].

Sequence of ITS-rDNA Fungal Region
The sequence of the ITS-rDNA fungal region of the isolate was used for the initial characterization through comparing it with related sequences in the NCBI database [36]. The Clustal X (Version 1.83) program [37] was used to edit and align the fungal ITS-rDNA sequence with the best n-BLAST hits from GenBank [38]. Further manual adjustments were carried out, using BioEdit software [38]. The base composition of the fungal sequences was calculated, using the MEGA v.6 program [39].

Characterization of the Fungal Isolate
The sequence analyses of the fungal isolate DY001 displayed 99.5% sequence similarity with Aspergillus flavipes (KF986416) in BLAST. This sequence was placed in the GenBank under Accession Numbers MN818770 and it was released on 17 December 2019.

Large-Scale Culture of Aspergillus sp. DY001
A culture of five-liter fermentation broth of the fungus was prepared in Sabouraud Dextrose broth amended by 3% NaCl (w/v) by shaking for 15 days at 30 • C and 180 rpm in a shaker incubator. Later, each flask of the culture broth was partitioned with 200 mL of EtOAc three times at room temperature. The combined EtOAc extracts were evaporated under reduced pressure and the resulting extract was used for further purification of the compounds.

Computational Details
All of the DFT calculations were performed using Gaussian 16 [40]. A conformation analysis was conducted using the GMMX plugin followed by a geometry optimization at the B3LYP/6-31g(d) level. A frequency check was performed at the same level of theory. The GIAO NMR properties were calculated at the mpw1pw91/6-311+g(2d,p) level. The DP4 probabilities were calculated using our own implementation of the algorithm published by Smith and Goodman [41].

Evaluation of the Antimicrobial Effects of 1-4 Disk Diffusion Assay
The antimicrobial effects of compounds 1-4 were evaluated using disc diffusion assay at 50 µg/disc against E. coli (ATCC 25922), C. albicans (ATCC 14053), and S. aureus (ATCC 25923), and as described before [42,43]. Ciprofloxacin and ketoconazole served as the positive controls.

Evaluation of the MIC Values
The determination of the MIC values of 1-4 was performed using a macro-dilution assay, as previously reported [44]. Briefly, The MIC values of 1-4 were evaluation of using a macro-dilution method [44], using MeOH to dissolve the compounds at a final concentration of 2000 µg/mL, while distilled H 2 O was used to dissolve ciprofloxacin and ketoconazole at final concentrations of 100 µg/mL. All of the solutions were sterilized using syringe filters (0.2 µm). Two-fold serial dilutions of the solutions were used in MHB to afford concentrations between 1.0 and 1000 µg/mL for the compounds, and between 0.125 and 64 µg/mL for the ciprofloxacin and ketoconazole. From the 10 6 CFU/mL microbial suspensions, 500 µL were added in sterile tubes giving inoculua of 5 × 10 5 CFU/mL. In addition, 100 µL of each stock solution of the compounds and antibiotics were added into the tubes. A control tube, which contained only the test microorganisms and methanol, was prepared [44]. The MeOH displayed no antimicrobial effect. Incubation of the tubes was accomplished at 37 • C for 48 h. The lowest concentrations of the compounds/antibiotics, which showed no microbial growth, were considered as MIC.

Evaluation of the Growth Inhibition Effects of the Compounds Culture of Cell Lines
The cell lines HeLa (human cervical carcinoma, ATCC CCL-2) and HCT 116 (colorectal carcinoma, ATCC CCL-247) were cultured in RPMI 1640 medium, including 1% penicillin-streptomycin and 10% FBS. The cell line MDA-MB-231 (triple-negative breast cancer, ATCC HTB-26) was cultured in DMEM medium, including 10% FBS and 1% penicillin-streptomycin.

MTT Assay
The evaluation of the antiproliferative and growth inhibitory effects of the compounds was carried out using MTT assay, as described before [45][46][47][48]. Briefly, the cells were incubated overnight in 5% CO 2 /air at 37 • C. The compounds were added at the top row of a 96-well microtiter plate, and descendant serial dilutions (1:4) of the concentration were carried out. After that, the cells with the compounds were incubated for 72 h. Using the Cell Titer 96 AQueous non-radioactive cell proliferation protocol, the cells' viability was estimated at 490 nm, using a Molecular Devices Emax microplate reader.
The IC 50 values of the compounds were evaluated using the program SoftMax ® PRO (https://www.moleculardevices.com/products/microplate-readers/acquisition-andanalysis-software/softmax-pro-software#gref; accessed on 1 July 2022). SoftMax ® Pro Software is a program that is designed to provide simple, flexible, and powerful methods for advanced data analysis. It provides analysis algorithms, ready-to-run protocols, with 21 different curve fit options. Every step is optimized for data acquired from a Molecular Devices microplate reader, or data imported from another source to simplify the analysis and reporting. Compliance tools are available for regulated laboratories providing end-toend chain of custody (https://www.moleculardevices.com/products/microplate-readers/ acquisition-and-analysis-software/softmax-pro-software#gref; accessed on 1 July 2022).