Novel indole diketopiperazine stereoisomers from a marine-derived fungus Aspergillus sp

ABSTRACT Four dimeric diketopiperazine stereoisomers (1–4) including two new ones (1–2) had been isolated from the culture broth of one marine-derived fungus Aspergillus sp. Z3, which was found in the gut of a marine isopod Ligia exotica. The planner structures and absolute configurations of the new compounds were determined by combination of NMR, HRESIMS, electronic circular dichroism calculation, Marfey’s method as well as single-crystal X-ray diffraction. Their cytotoxicity against the prostate cancer PC3 cell line was assayed by the MTT method.


Introduction
Indole diketopiperazines, which were formed by the condensation of tryptophan with other different amino acids, are an important family of secondary metabolites commonly produced by fungi. Most of these alkaloids are characterised by diverse structural scaffolds and ring systems derived from dimerisation by two indole units and conjugation with another building block as cyclic dipeptides (Li et al. 2009;Ma et al. 2016;Klas et al. 2018). This class of alkaloids possess diverse biological activities such as antimicrobial (Zheng et al. 2007;Jia et al. 2019;Yang et al. 2021), anticancer (Wang et al. 2015;Zhang et al. 2019), immunomodulatory (Fujimoto et al. 2004), anti-inflammatory (Barrow et al. 1993;Wen et al. 2018;Yang et al. 2021), antioxidant (Sun et al. 2018) and antiviral (Wibowo et al. 2021;Nishiuchi et al. 2022), which attracted much attention. Our ongoing research on marine fungal metabolites revealed one diketopiperazine dimer WIN 64821 (4), a competitive antagonist to substance P (Barrow et al. 1993) and an antagonist of cholecystokinin type-B receptor (Sedlock et al. 1994), with a series of its stereoisomers (1-3) from the marine-derived fungus Aspergillus sp. Z3 (Figure 1). Compounds 1 and 2 are new ones.
Herein, the purification, structure elucidation, and biological activities of the four compounds are described.

Fungal material and fermentation
The fungal colony (Z3) was isolated from marine isopod Ligia exotica which was collected in seaside of Dinghai in Zhoushan, Zhejiang Province of China, in December 2011. The fungal Z3 was determined as Aspergillus sp. by ITS analysis.

Marfey's method
Compounds 1-3 (0.5 mg) were hydrolysed in 1 mL HCl (6 M) at 110°C for 24 h, respectively. The cooled reaction mixture was evaporated to dryness and HCl was removed. The dried hydrolysate was dissolved in 120 μL ddH 2 O and then 20 μL NaHCO 3 (1 M) and 400 μL 1% (w/v) FDAA in acetone were added to the solution of the hydrolysate. The mixture was stirred at 42°C for 2 h and the reaction was terminated by an addition of 10 μL HCl (1 M). The reaction product was evaporated to dryness and dissolved with 1 mL MeOH to give amino acid-FDAA derivatives for the HPLC analysis. Each standard amino acid (0.5 mg) of L-phenylalanine and D-phenylalanine was also converted to its FDAA derivatives. For Marfey′s analysis, each of the amino acid-FDAA derivatives (15 μL) was analysed by HPLC (SHIMADZU column: C 18 , 5 μm, 4.6 × 250 mm; flow rate: 1.0 mL/min; detection wavelength: 340 nm) with the linear gradient programme from 10% to 100% acetonitrile in ddH 2 O over 60 min. Each chromatographic peak was identified by comparing its retention time with the FDAA derivatives of the standard amino acids.  127.2 a in CDCl 3 ; b in CH 3 OH-d 4 ; b data of C-1-C-23 can be changed with that of C-1'~ C-23'.
The retention time of the standard derivatives was as follows: 23.2 min for FDAA, 28.8 min for L-phenylalanine and 31.1 min for D-phenylalanine.

ECD calculation
The CIF structure of compound 1was initially optimised at B3LYP/6-31 + g (d, p) level in MeOH using the CPCM polarisable conductor calculation model. The theoretical calculation of ECD was conducted in MeOH using Time-dependent Density functional theory (TD-DFT) at the B3LYP/6-31 + g (d, p) level for one conformer of compound 1. Rotatory strengths for a total of 30 excited states were calculated. ECD spectrum were generated using the program SpecDis 1.6 (University of Würzburg, Würzburg, Germany) and GraphPad Prism 5 (University of California San Diego, USA) from dipole-length rotational strengths by applying Gaussian band shapes with sigma = 0.3 eV.

Cytotoxicity bioassays
The cytotoxicity was measured by the MTT assay against the prostate cancer PC3 cell line. Tumour cell lines were seeded in 96-well plates (4 × 10 3 per well in 100 μL). After 24 h of incubation cells were treated with gradient concentrations (100 μM, 50 μM, 25 μM, 12.5 μM, 6.25 μM, 3.125 μM) for another 72 h. Afterwards, MTT solution (5.0 mg/mL in RPMI-1640 media, Sigma, St. Louis, MO, USA) was added (20 μL/ well) and then plates were incubated for another 4 h at 37°C. The compounds were dissolved in DMSO and cell growth inhibition assay was performed as reported previously. The growth inhibitory ability of the compounds was calculated and expressed using the IC 50 value by the software of Dose-Effect Analysis. Doxorubicin (ADR) was used as a positive control (Wang et al. 2006).
Comparison of the 1 H-NMR data between compounds 1 and 4 revealed a significant difference that there was an unusual up-field proton signal for H-11 (δ H 1.89) in compound 1. It was possibly caused by the positive shielding effect of the adjacent aromatic ring as the α-H of amino acid residue found in asperazine (Varoglu et al. 1997). It was reported that the positive or negative cotton effect at 254 nm and 300 nm in the electronic circular dichroism (ECD) spectrum depended on the stereochemistry of C-2/ C-2' and C-3/C-3' (Barrow et al. 1993). Compound 1 showed positive cotton effect at 254 nm and 300 nm, same as WIN 64821, which indicated the chirality of C-2/C-2' and C-3/C-3' has not changed, both were R configuration (Figure 3a). The calculated ECD spectrum of this configuration at 254 nm was the same as experiment result, while the theoretical data at 300 nm did not yield (Figure 3b). It is speculated that the n-π transition signal of carbonyl is about 300 nm. The theoretical calculations suggested the carbonyl group had little effect on the configuration change may be the reason that no calculation results were obtained at 300 nm. The acid hydrolysates of compound 1 reacted with Marfey's reagent (FDAA) showed the presence of L-phenylalanine residue by HPLC analysis (Figure S10, Supplementary data). Therefore, the absolute configuration at C-15/C-15' was determined as S. The up-field shifting was speculated to be caused by the positive shielding effect of the aromatic ring of phenylalanine residue, so that H-11 and H-15 were in the opposite side of the piperazine ring. Because there was no change of chirality occurred at C-15/C-15', the configuration at C-11/ C-11' were inverse to R, which was different from the reported compound WIN 64821. Finally, the speculation described above was proved by the X-ray crystallographic data (Figure 4) and the absolute stereochemistry of compound 1, the symmetrical dimer, was determined as 2 R, 2ʹR, 3 R, 3ʹR, 11 R, 11ʹR, 15S, 15ʹS.
Detailed analysis of the 1 H NMR spectrum of compound 2 revealed two unusual up-field proton signals for H-11 (δ H 2.31) and H-11' (δ H 2.21), which were estimated to be affected by the positive shielding effect of the phenyl ring of phenylalanine residue as well according to the configuration of compound 1. Therefore, H-11 and H-15, H-11' and H-15' were on the opposite side of the respective piperazine ring. Hydrolysis in HCl, followed by Marfey's derivatisation, showed the presence of D-and L-phenylalanine residues in the structure of 2 ( Figure S19, Supplementary data), which indicated the absolute configurations at C-15 and C-15' were different. The ECD spectrum of compound 2 showed positive cotton effect at 254 nm and 300 nm (Figure 6), suggesting the chirality of C-2/C-2' and C-3/C-3' has not changed, still were R. Then the NOESY cross peak of H-2 (δ H 4.83) and H-11 suggested the cis-orientation of the two protons, which further defined the absolute configuration at C-11 as S. For H-11 and H-15 were transorientation, the chirality of C-15 was determined as R.
Compound 3, sharing the same planner structure of the afore-mentioned compounds, was an asymmetric dimer (Figure 1). 1 H NMR spectrum revealed one unusual up-field proton signals for H-11 (δ H 1.98) and one normal proton for H-11' (δ H 4.02) (Table S1, Supplementary data). The up-field shifting was estimated to be caused by the same reason as that for compounds 1 and 2, the positive shielding effect of the phenyl ring. The absolute configurations at C-15/ C-15' were both defined as S depending on Marfey's results ( Figure S29, Supplementary data). Thus, the chirality of C-11 and C-11' was determined as R and S, respectively. The ECD spectrum of compound 3 showed positive cotton effect at 254 nm and 300 nm ( Figure S30, Supplementary data), suggesting the chirality of C-2/C-2' and C-3/C-3' has not changed, still were R. Therefore, the absolute configuration of compound 3, which was consistent with the compound reported in the literature (Gu et al. 2019), was defined as 2 R, 2ʹR, 3 R, 3ʹR, 11 R, 11ʹS, 15S, 15ʹS.
All the four compounds were tested for their in vitro cytotoxicity against the prostate cancer PC3 cell line by MTT method. Unfortunately, all of them were found to be inactive, displaying IC 50 values >40 μM. Complex structural and stereoisomeric modification might influence the biological effects of the target compounds through the alteration of the shape or rigidity of the scaffold (Giessen et al. 2013;Giessen and Marahiel 2015). As some members of dimeric diketopiperazine encompass a vast spectrum of biological properties, it is suggested that the four compounds' probability possesses an alternative and undefined biology that can be studied in further research.

Conclusion
Diketopiperazine scaffold, the smallest of all cyclopeptides, frequently embedded in complex metabolic architectures. Dimeric diketopiperazines are a relevant variation of this class of natural products occupying an enormous chemical complexity. So far, the biosynthesis of mainframe structure of diketopiperazine was found to be catalysed by nonribosomal peptide synthetases (NRPSs) and cyclodipeptide synthases (CDPSs) (Gondry et al. 2009;Belin et al. 2012;Borgman et al. 2019). Tailoring enzymes, including oxidoreductases and prenyltransferases, modify the initially assembled diketopiperazine, and cytochrome P450 responsible for the dimerisation of diketopiperazine monomers (Mishra et al. 2017;Shende et al. 2020). Compounds 1-4 were dimeric diketopiperazine stereoisomers, whose configuration was mainly different at C-11/C-11' and C-15/C-15' positions, while C-2/C-2' and C-3/C-3' positions were identical, indicating the biosynthesis of the four stereoisomers might be carried out by using different chiral phenylalanine and tryptophan as substrates, then dimerised by a possible P450 enzyme at C-3/C-3' spot. Although the four dimeric diketopiperazine stereoisomers showed noncytotoxic against prostate cancer PC3 cell line, due to the diverse biological activities this class of compounds possess, they still provide attractive new targets for developing on the diketopiperazine scaffold to support future drug discovery.