Bioactive Diketopiperazines and Nucleoside Derivatives from a Sponge-Derived Streptomyces Species

Abstract

Fractionation and purification of the ethyl acetate extract of the culture of a sponge-derived actinomycete, Streptomyces species Call-36, resulted in the isolation and identification of a new diketopiperazine, actinozine A (1), cyclo(2-OH-d-Pro-l-Leu) (2), two new nucleosides, thymidine-3-mercaptocarbamic acid (3) and thymidine-3-thioamine (4), together with cyclo(d-Pro-l-Phe) (5) and cyclo(l-Pro-l-Phe) (6). The structure assignments of the compounds were carried out by interpretation of 1D and 2D NMR data and mass spectral determinations. The absolute configurations of 1 and 2 were determined by Marfey’s method and by comparison of the experimental and TDDFT-calculated ECD spectra. Actinozine A possesses an unprecedented hydroperoxy moiety at C-2 of the proline moiety, while 3 and 4 possess unusual mercaptocarbamic acid and thiohydroxylamine functionalities at N-3 of the thymine moiety. The isolated compounds displayed variable cytotoxic and antimicrobial activities.

1. Introduction

Secondary metabolites played a vital role in discovery and development of drugs and saved the lives of millions of people over the past decades [1,2]. Microbial secondary metabolites contributed significantly to drug discovery and pharmaceutical industry. There is a continuous need for novel chemical entities with new mechanisms of action to overcome drug resistance and to fulfill the medical needs for new drugs [3]. This need forced the chemists to search for understudied microbial resources that inhabit challenging environments [4]. The marine environment, with its great biological diversity, has been considered as under-investigated source of microbes with unlimited biosynthetic capabilities for bioactive compounds [4]. Marine actinobateria, including actinomycetes, have proven to be producers of novel chemical entities with diverse bioactivities such as antitumor, antimicrobial, anti-inflammatory, antiparasitic, antimalarial, antiviral, antioxidant, anti-angiogenesis and many others [5,6]. The 2,5-diketopiperazines (DKPs) belong to the smallest cyclic dipeptides that consist of a six-membered ring containing two amide linkages where the two nitrogen atoms and the two carbonyls are at opposite positions. DKPs are represented in structurally diverse groups of secondary metabolites, with different examples showing significant bioactivities including antimicrobial, antitumor, analgesic, and many other biomedical indications [7]. Prominent anticancer DKPs include the anti-microtubule phenylahistins [8], the cell cycle inhibitors tryprostatins [9], the chaetocins, which inhibit the lysine-specific histone methyltransferase [10] and the ardeemins with their reversal effects on multiple drug resistant (MDR) phenotype [11].

As a part of ongoing search on marine microbial secondary metabolites [12,13,14,15,16,17], the ethyl acetate extract of the sponge-derived Streptomyces species Call-36 was studied. A new diketopiperazine alkaloid with a hydroperoxy group at C-2 of the proline moiety, actinozine A (1), cyclo(2-OH-Pro-l-Leu) (2) [18], two new nucleoside derivatives, thymidine-3-mercaptocarbamic acid (3) and thymidine-3-thioamine (4), in addition to the previously reported compounds cyclo(d-Pro-l-Phe) (5) [19] and cyclo(l-Pro-l-Phe) (6) [20] (Figure 1) were isolated. The structure assignments of the compounds were supported by comprehensive investigations of the NMR and MS data. The relative configuration of 1 and 2 was determined using the DP4 probability approach. The absolute configuration of the amino acid moieties in 1 and 2 were established by Marfey’s method and by comparison of both experimental and predicted ECD spectra. In this paper, the structure determinations as well as the cytotoxic and antimicrobial activities of the compounds are discussed.

Figure 1.

Structures of compounds 1–6.

2. Results and Discussion

2.1. Purification of the Compounds

The ethyl acetate extract of the fermentation broth of the actinomycete, Streptomyces sp. Call-36, was subjected to extensive chromatographic separations over silica gel, Sephadex LH-20 and final HPLC purification to yield compounds 1–6.

2.2. Structure Elucidation of the Compounds

Compound 1 (Figure 1) was isolated as a colorless solid with molecular formula C₁₁H₁₈N₂O₄. The NMR spectra of 1 (Figures S1–S6 and Table 1) displayed 11 signals including two methyls, four methylenes, two methines, and three quaternary carbons (two carbonyls and one oxygenated carbon) as supported by the ¹³C NMR and HSQC experiments (Table 1). The ¹H and ¹³C NMR spectra of 1 (Figures S1 and S2 and Table 1) displayed characteristic signals for a diketopiperazine ring, which included two amidic carbonyls at δ_C 168.0 (qC, C-2) and 167.2 (qC, C-5), a methine signal at δ_H/δ_C 4.04 (dt, 7.2, 3.0 Hz, H-3)/56.0 (CH, C-3) and an amidic singlet at δ_H 6.25 (1H, brs, NH). In addition, the remaining ¹H and ¹³C signals of 1 supported the presence of leucine and 2-substituted proline units. The presence of ¹H-¹H spin-spin coupling system in the COSY experiment of 1 (Figure S4) from H₂-7 to H₂-9 supported this assignment (Figure 2). The interruption of the coupling system at C-6 suggested its quaternary nature. The NMR signals at δ_H/δ_C 167.2 (qC, C-5), 95.5 (qC, C-6), 2.44, 2.15 (H₂-7)/33.4 (CH₂, C-7), 2.16, 2.08 (H₂-8)/19.2 (CH₂, C-8), and 3.90, 3.62 (H₂-9)/45.3 (CH₂, C-9) are characteristic for a 2-substituted proline residue. In addition, the presence of a leucine unit was supported from the NMR signals at δ_H/δ_C 168.0 (qC, C-2), 4.04 (H-3)/56.0 (CH, C-3), 6.25 (4-NH), 1.96, 1.84 (H₂-10)/44.1 (CH₂, C-10), 1.80 (H-11)/24.5 (C-11), 0.99 (H₃-12)/23.1 (C-12) and 0.95 (H₃-13)/21.1 (C-13). Again, the presence of a second ¹H-¹H spin-spin coupling system in the COSY experiment between H-3 and the isobutyl moiety at C-3 (H-10 → H₃-13) supported the presence leucine unit (Figure 2). Despite the similarity between the NMR data of 1 with those of 2 (Table 2), there is a significant downfield shift of C-6 in 1 by 8.8 ppm (resonating at δ_C 95.5 instead of δ_C 86.7 in 2) (Table 2), suggesting the presence of a hydroperoxy (OOH) (δ_H 9.27, brs) group at C-6 in 1 instead of OH in 2 (Table 2). Finally, all protonated carbons were assigned from HSQC (Figure S6) and COSY experiments (Figure 2 and Figure S4), while the placement of the OOH moiety at C-6 was supported by HMBC correlations of H₂-7/C-6 and H₂-8/C-6 (Table 1, Figure 2 and Figure S6). Similarly, the assignment of the quaternary amidic carbons (C-2 and C-5) was secured from HMBC correlations (Table 1, Figure 2 and Figure S6). The relative configuration at C-6 was determined by comparison between experimental and theoretical NMR chemical shifts as proposed by Smith and Goodman [21] as several recent applications of this approach demonstrated to be efficient [22,23].

Table 1.

NMR data of compound 1 (600 and 150 MHz, CDCl₃).

No.	δC (mult.)	δH [mult., J (Hz)]	HMBC
2	168.0, qC
3	56.0, CH	4.04 (dt, 7.2, 3.0)	C-2, C-5, C-10, C-11
4 (NH)	-	6.25 (brs)
5	167.2, qC
6	95.5, qC
7	33.4, CH2	2.44 (ddd, 13.2, 7.2, 3.0) 2.15 (m)	C-6, C-8, C-9
8	19.2, CH2	2.16 (m), 2.08 (m)	C-6, C-7, C-9
9	45.3, CH2	3.90 (ddd, 12.0, 8.6, 7.8) 3.62 (ddd, 12.0, 9.6, 4.2)	C-2, C-7, C-8
10	44.1, CH2	1.96 (m), 1.84 (t, 7.2)	C-2, C-3, C-12, C-13
11	24.5, CH	1.80 (nonet, 6.6)	C-3, C-12, C-13
12	23.1, CH3	0.99 (d, 6.6)	C-10, C-11
13	21.1, CH3	0.95 (d, 6.6)	C-10, C-11
OOH		9.27 (brs)

Figure 2.

Key COSY and HMBC correlations of 1, 2, and 3.

Table 2.

NMR data of compound 2 (600 and 150 MHz, CDCl₃).

No.	δC (mult.)	δH [mult., J (Hz)]	HMBC
2	168.0, qC
3	56.4, CH	3.97 (td, 10.4, 4.2)	C-2, C-5, C-10, C-11
4 (NH)	-	6.02 (brs)
5	167.2, qC
6	86.7, qC
7	37.6, CH2	2.30 (m), 2.15 (m)	C-6, C-8, C-9
8	19.2, CH2	2.18 (m), 2.08 (m)	C-6, C-7, C-9
9	45.6, CH2	3.73 (m), 3.66 (m)	C-2, C-7, C-8
10	44.4, CH2	1.88 (dd, 10.2, 7.2) 1.80 (m)	C-2, C-3, C-12, C-13
11	24.5, CH	1.80 (m)	C-3, C-12, C-13
12	23.0, CH3	1.00 (d, 6.0)	C-10, C-11
13	21.2, CH3	0.94 (d, 6.0)	C-10, C-11
OH		2.97 (brs)

As shown on Figure 3, the 3S*,6S* relative configuration for 1b was supported by all the metrics used. The correlation between experimental and theoretical chemical shifts was higher for 1b (R² = 0.9994) and the mean average error (MAE) value was lower for 1b and finally a 100% DP4 score for 1b (Figure 3). All calculations for both isomers (3S,6S and 3R,6R) of 1 can be found at (https://figshare.com/s/d626b72364548b03e11b). The absolute configuration of the leucine residue was determined by Marfey’s method [24]. HPLC analysis of the derivative of 1-fluoro-2,4-dinitrophenyl-5-l-alanine amide (FDAA, Marfey’s reagent) with the hydrolytic products of 1 gave the same retention time as the derivative prepared from an authentic l-leucine, and therefore the l-configuration was assigned to the leucine residue. Due to the hydrolysis of 1 under acidic condition, the absolute configuration at C-6 was not determined by Marfey’s method. The absolute configuration at C-3 and C-6 asymmetric centers was determined from analysis of the ECD spectrum of 1. Comparison between experimental and theoretical ECD spectra gave a very good agreement as displayed on Figure 4. Two Cotton effects with alternative signs were observed due to the n→π* transition (Figure 4). Thus, 1 was assigned as (3S,8aS)-8a-hydroperoxy-3-isobutylhexahydropyrrolo[1,2-a]pyrazine-1,4-dione {cyclo(2-OOH-d-Pro-l-Leu } and is reported here as a new natural product and named actinozine A.

Figure 3.

Calculated and experimental NMR chemical shifts of 1.

Figure 4.

Experimental and predicted ECD spectra of 1.

Compound 2 (Figure 1) was purified as colorless oil with molecular formula of C₁₁H₁₈N₂O₃. The combined one- and two-dimensional NMR data of 2 (Table 2, Figure 2 and Figures S8–S13) are in good agreement with those reported for cyclo(2-OH-Pro-l-Leu) [18]. However, the absolute configuration at C-6 was missing. For the determination of the relative configuration, we followed the same approach as described above. As expected, the same 3S* and 6S* was obtained with a high probability (R² = 0.9995) (Figure 5). All calculations for both isomers (3S,6S and 3R,6R) of 2 can be found at (https://figshare.com/s/d626b72364548b03e11b). Again, the absolute configuration of the leucine residue was determined by Marfey’s method [24] as discussed above and was found to be l-Leucine. Attempts to determine the stereochemistry at C-6 by Marfey’s method was not successful because of the decomposition of 2 under acidic conditions. Therefore, a CD spectrum for compound 2 (Figure 6) was recorded in order to determine the absolute configuration of the hydroxylated C-6. The absolute configuration at C-3 and C-6 in 2 was determined to be 3S and 6S by comparison of the experimental and predicted ECD spectra (Figure 6). Thus, 2 was assigned as cyclo(2-OH-d-Pro-l-Leu) {(3S,8Sa)-8a-hydroxy-3-isobutylhexahydropyrrolo[1,2-a]pyrazine-1,4-dione}.

Figure 5.

Calculated and experimental NMR chemical shifts of 2.

Figure 6.

Experimental and predicted ECD spectra of 2.

Compound 3 (Figure 1) was isolated as a white solid with molecular formula of C₁₁H₁₅N₃O₇S as established from the HRESIMS pseudomolecular ion peak at m/z 334.0708 [M + H]⁺, suggesting six degrees of unsaturation. The ¹H and ¹³C NMR spectrum of 3 (Table 3 and Figures S15 and S16) showed signals for 2′-deoxyribose unit at δ_H/δ_C 6.15 (1H, t, J = 7.0 Hz, H-1′)/83.7 (CH, C-1′), 2.07 (1H, m, H-2′a) and 2.02 (1H, m, 2′b)/39.5 (CH₂, C-2′), 4.22 (1H, quin, J = 3.0 Hz, H-3′)/70.3 (CH, C-3′), 5.24 (1H, brs, OH-3′), 3.74 (1H, q, J = 3.5 Hz, H-4′)/87.2 (CH, C-4′) and 3.57 (1H, dd, J = 12.0, 3.5 Hz, H-5′a), 3.53 (1H, dd, J = 12.0, 4.0 Hz, H-5′b)/61.3 (CH₂, C-5′) and 5.05 (1H, brs, OH-5′). This was supported by the presence of a continuous ¹H-¹H spin-spin coupling system in the COSY experiment (Figure S17) from H-2′ to H₂-5′ (Figure 2), ¹J_CH correlations in the HSQC experiment (Figure S18) and HMBC correlations (Table 3, Figure 2 and Figure S19). The NMR data of the 2′-deoxyribose moiety (Table 3) are in good agreement with the literature [25,26,27]. In addition, the signals at δ_H/C 150.4 (qC, C-2), 163.7 (qC, C-4), 109.3 (qC, C-5), 7.68 (1H, s, H-6)/136.1 (CH, C-6), 1.75 (3H, s, H₃-7)/12.2 (CH₃, C-7) supported the presence of a substituted thymine moiety in 3 [25,26]. Again, the COSY and HMBC correlations in 3 (Table 3 and Figure 2) supported the presence of a thymine moiety. The remaining portion of 3 includes the elemental composition of CH₂NO₂S (S-NH-COOH), which corresponds to a mercaptocarbamic acid (Figure 1). The existence of the mercaptocarbamic acid moiety was supported by the presence of NMR signals at δ_H/C 7.26 (1H, brs, NH) and 10.90 (1H, hump, H-3′′)/158.1 (qC, C-3′′) (Table 3) and by MS/MS fragmentation ion peaks at m/z 316 [M − OH]⁺, 288 [M − COOH]⁺, 273 [M − NHCOOH]⁺ and 242 [M − SNHCOOH + H]⁺ (Figure 7 and Figure 8), completing the degrees of unsaturation and the molecular formula of 3. The ¹³C chemical shift of the carboxylic acid moiety of the carbamic acid is in good agreement with the literature [28]. The placement of the 2′-desoxyribose moiety at N-1 was supported by HMBC long-range correlations from H-1′ to C-2 and C-6 (Table 3 and Figure 2). Since N-3 is the only available and free place in 3, thus, the mercaptocarbamic acid moiety was linked to N-3 to complete the degrees of unsaturation and the planner structure of 3. The strong ROESY correlation (Figure S20) displayed between H-1′ and H-4′ protons and the absence of any ROESY correlations between H-3′ and H-1′ as well as H-3′ and H-4′ strongly supported the β configuration of the nucleoside (Table 3 and Figure 9). Detailed additional and expected ROESY correlations in 3 are presented in Table 3 and Figure 9. The similarity of the value and sign of the optical rotation of 3 (+11°) with those of d-thymidine (+18.5°) [29] supported the d-sugar moiety in 3. Accordingly, 3 was assigned as thymidine-3-mercaptocarbamic acid and is reported here as a new natural product.

Table 3.

NMR data of compound 3 (600 and 150 MHz, DMSO-d₆).

No.	δC (mult.)	δH [m., J (Hz)]	HMBC	ROESY
2	150.4, qC	-
4	163.7, qC	-
5	109.3, qC	-
6	136.1, CH	7.68 (s)	C-2, C-4, C-7	H3-7, H-1′, H-2′a, H-4′, H-5′a
7	12.2, CH3	1.75 (s)	C-4, C-5, C-6	H-6
1′	83.7, CH	6.15 (t, 7.0)	C-2, C-6, C-2′, C-3′	H-2′a, H-4′, H-6
2′a 2′b	39.5, CH2	2.07 (m) 2.02 (m)	C-4′	H-1′, H-6 H-3′, H-4′, H-5′b
3′	70.3, CH	4.22 (quin, 3.0)	C-1′, C-5′	H-2′b
4′	87.2, CH	3.74 (q, 3.5)		H-2′b, H-5′a, H-6
5′a 5′b	61.3, CH2	3.57 (dd, 12.0, 3.5) 3.53 (dd, 12.0,4.0)	C-3′	H-4′, H-6 H-2′b
OH-3′		5.24 (brs)
OH-5′		5.05 (brs)
NH		7.26 (brs)
3′′ (COOH)	158.1, qC	10.90 (hump)

Figure 7.

Key MS/MS fragmentation ion peaks of 3 and 4.

Figure 8.

MS/MS fragmentation spectra of 3 (A) and 4 (B).

Figure 9.

Observed ¹H-¹H ROESY correlations for 3.

Compound 4 (Figure 1) was purified as a white powder with molecular formula of C₁₀H₁₅N₃O₅S as established by the HRESIMS pseudomolecular ion peak at m/z 290.0812 [M + H]⁺, being 44 mass unit less than 3. Comparison of the ¹H and ¹³C NMR data of 4 with those of 3 (Table 4 and Figures S23 and S24) displayed identical signals, suggesting similar structures of both compounds. The protonated and quaternary carbons were assigned from 2D NMR experiments including COSY, HSQC and HMBC experiments (Figures S25–S27). The loss of 44 mass unit in 4 suggested the absence of the COOH moiety (Table 4). The presence of a thiohydroxylamine moiety in 4 was also supported by MS/MS fragmentation ion peaks at m/z 273 [M − NH₂]⁺ and 242 [M − SNH₂ + H]⁺ (Figure 7 and Figure 8), completing the planner structure of 4. The configuration of the sugar moiety of 4 was established by comparison of the NMR data of 3 and 4 (Table 3 and Table 4). In addition, both compounds displayed positive optical rotation values (+11° and +13°, respectively), supported the same β configuration of the sugar part and the d-nucleoside in both compounds. Therefore, 4 was assigned as thymidine-3-thioamine and is reported here as a new natural product. It is worth to mention that, the occurrence of the mercaptocarbamic acid and thiohydroxylamine moieties in 3 and 4 are presented here for the first time from a natural source.

Table 4.

NMR data of compound 4 (850 and 213 MHz, DMSO-d₆).

No.	δC (mult.)	δH [m., J (Hz)]	HMBC
2	150.4, qC	-
4	163.7, qC	-
5	109.3, qC	-
6	136.1, CH	7.70 (s)	C-2, C-4, C-7
7	12.2, CH3	1.77 (s)	C-4, C-5, C-6
1′	83.7, CH	6.16 (t, 6.8)	C-2, C-6, C-2′, C-3′
2′a 2′b	39.5, CH2	2.09 (ddd, 13.6, 7.7, 6.0) 2.04 (ddd, 13.6, 6.0, 3.4)	C-4′
3′	70.4, CH	4.24 (m)	C-1′, C-5′
4′	87.2, CH	3.76 (q, 4.2)
5′a 5′b	61.3, CH2	3.58 (brd, 12.0) 3.54 (brd, 12.0)	C-3′
OH-3′		5.26 (brs)
OH-5′		5.06 (brs)
NH2		a

^a Not observed.

Compounds 5 and 6 were identified by interpretation of their NMR (Figures S29–S38) and MS data as well as their optical rotations and by comparison with available data in the literature. The absolute configuration of the amino acid moieties in 5 and 6 was determined by HPLC analysis of the derivatives of 1-fluoro-2,4-dinitrophenyl-5-l-alanine amide (FDAA, Marfey’s reagent) with the hydrolytic products of 5–7 as previously reported [24]. Accordingly, the compounds were assigned as cyclo(d-Pro-l-Phe) (5) [19] and cyclo(l-Pro-l-Phe) (6) [20].

The compounds were evaluated for their cytotoxic activities in the sulforhodamine B (SRB) assay [30] against HCT-116 (colorectal carcinoma, ATCC CCL-247) and MCF-7 (breast cancer, ATCC HTB-22). Compound 5 showed moderate activity with IC₅₀ of 32.7 µM against HCT-116, while other compounds were weakly active against this cell line. On the other hand, all compounds showed weak activities against breast cancer cell line (Table 5).

Table 5.

Cytotoxic and antimicrobial activities of compounds 1–6.

	IC50 (μM)		Inhibition Zones (mm) at 100 μg/disc
Compound	HCT 116	MCF 7	S. aureus	C. albicans
1	146	88.8	23.0	19.0
2	150	117	20.0	16.0
3	≥50	90	NT	NT
4	≥50	112	NT	NT
5	32.7	81.9	14.0	11.0
6	131	123	10.0	9.0
Doxorubicin a	1.62	0.77
Ciprofloxacin b			22.0
Ketoconazole c				30.0

NT = Not tested; ^a Positive cytotoxicity control; ^b Positive antibacterial control (5.0 µg/disc); ^c Positive antifungal control (50 µg/disc).

Furthermore, the antimicrobial activities of the compounds were evaluated against S. aureus and C. albicans using disc diffusion assay [31] at 100 μg/disc. Compounds 1 and 2 were potently active against S. aureus with inhibition zones of 23 and 20 mm, respectively, while these compounds displayed moderate activity against C. albicans with inhibition zones of 19 and 16 mm, respectively (Table 5). On the other hand, compounds 5 and 6 displayed moderate activity against S. aureus and C. albicans with inhibition zones of 9.0–14.0 mm (Table 5).

As a result, the moderate cytotoxicity of 5 and the weak cytotoxicity of other diketopiperazines (1, 2, and 6) may suggest the need to screen these compounds against a variety of other cell lines or completely different biological target. Moreover, the strong antimicrobial activities of 1 and 2 support the use of such compounds as a template or scaffold to develop structurally related antimicrobial drugs.

3. Materials and Methods

3.1. General Experimental Procedures

Optical rotations were measured on a JASCO DIP-370 digital polarimeter at 25 °C at the sodium D line (589 nm). UV spectra were recorded on a Hitachi 300 spectrometer (Hitachi High-Technologies Corporation, Kyoto, Japan). The ECD spectra were obtained on a JASCO J-810 spectropolarimeter with a 0.5 cm cell in MeOH. IR spectra were measured on a Shimadzu Infrared-400 spectrophotometer (Shimadzu, Kyoto, Japan). 1D and 2D NMR spectra (chemical shifts in ppm, coupling constants in Hz) were recorded on Bruker Avance DRX 600 MHz (600 MHz for ¹H and 150 MHz for ¹³C NMR) (Bruker, Rheinstetten, Germany) and Bruker Ascend™ 850 MHz (850 MHz for ¹H and 213 MHz for ¹³C NMR) (Bruker BioSpin, Billerica, MA, USA) spectrometers using CDCl₃ or DMSO-d₆ as solvent. NMR spectra were referenced to the residual protonated solvent signals (for CHCl₃: 7.26 ppm for ¹H and 77.0 ppm for ¹³C; for DMSO: 2.51 ppm for ¹H and 39.6 ppm for ¹³C). Positive ion HRESIMS data were obtained with a Micromass Q-ToF equipped with leucine enkaphalin lockspray, using m/z 556.2771 [M + H]⁺ as a reference mass. For column chromatography, silica gel (Merck, 70-230 mesh ASTM) and Sephadex LH-20 (0.25–0.1 mm, Pharmacia) were used. Precoated silica gel 60 F-254 plates (Merck) were used for TLC. HPLC purifications were performed on a semi-preparative HPLC column (RP18, 5 μm, ARII Cosmosil, 250 × 10 mm, Waters).

3.2. Biological Materials

The actinomycete Streptomyces species Call-36 was isolated from the Red Sea sponge Callyspongia species. To purify the actinomycete strain from the sponge materials, the previously reported protocol from our laboratory was used [16]. Identification of the actinomycete strain was based on 16S rRNA gene sequence analysis. Purification of the genomic DNA, the amplification of the 16S rRNA gene by PCR, and sequence alignment of the strain were performed as reported before [32]. The 16S rRNA gene sequence of the strain displayed 99% similarity with type strains of Streptomyces albus subsp. albus (AB184781) and Streptmyces almquistii (AB184258).

3.3. Fermentation and Extraction

The spores of Streptomyces sp. Call-36 were cultured in Erlenmeyer flasks (2 L) containing 500 mL of ISP-2 fermentation media. The media contained 4.0 g of yeast extract, 10 g of malt extract and 4.0 g of dextrose and 3.3% sea salt in 1 L distilled water (pH 7.2). The cultures were incubated on a rotatory shaker at 180 rpm at 28 °C for 15 days. The combined fermentation broths (10 L) were partitioned against EtOAc and the resulting EtOAc solutions were concentrated under vacuum to give a brown residue (2.4 g).

3.4. Purification of the Compounds

The EtOAc extract (2.4 g) was subjected to SiO₂ VLC using n-hexane/CH₂Cl₂/MeOH gradients to give five fractions (A–E). Fraction C (370 mg) was purified by gel filtration over Sephadex LH-20 with MeOH to give four subfractions (C1-C4). Fraction C2 (120 mg) was purified on C18 HPLC column with 40% CH₃CN to yield 1 (3.5 mg) and 2 (2.3 mg). Fraction C4 (60 mg) was purified on C18 HPLC column with 30% CH₃CN to yield 3 (1.8 mg) and 4 (1.5 mg). Fraction D (410 mg) was purified by gel filtration over Sephadex LH-20 with MeOH to give five subfractions (D1-D5). Fraction D3 (110 mg) was further purified on C18 HPLC column with 40% CH₃CN to yield 5 (2.5 mg) and 6 (3.6 mg).

3.5. Spectral Data of the Compounds

Actinozine A (1): colorless solid; [α]_D −52° (c 0.1, MeOH); UV (MeOH) λ_max (log ε): 230 (4.15), 303 (4.09) nm; ECD (MeOH) [Δε]_203nm −15.40, [Δε]_224nm +13.27; IR (film) ν_max 3450, 1656, 1627 cm⁻¹; NMR data: Table 1; HRESIMS m/z 243.1344 (calcd for C₁₁H₁₉N₂O₄, [M + H]⁺, 243.1345).

Cyclo(2-OH-d-Pro-l-Leu) (2): colorless oil; [α]_D −50° (c 0.1, MeOH); UV (MeOH) λ_max (log ε): 231 (4.17), 305 (4.00) nm; ECD (MeOH) [Δε]_203nm −30.25, [Δε]_223nm +29.82; IR (film) ν_max 3445, 1659, 1628 cm⁻¹; NMR data: Table 2; HRESIMS m/z 227.1395 (calcd for C₁₁H₁₉N₂O₃, [M + H]⁺, 227.1396).

Thymidine-3-mercaptocarbamic acid (3): white solid; [α]_D +11° (c 0.1, MeOH); UV (MeOH) λ_max (log ε): 265 (3.45), 255 (3.50) nm; NMR data: Table 3; HRESIMS m/z 334.0708 (calcd for C₁₁H₁₆N₃O₇S, [M + H]⁺, 334.0709).

Thymidine-3-thioamine (4): white solid; [α]_D +13° (c 0.1, MeOH); UV (MeOH λ_max (log ε): 266 (3.50), 256 (3.55) nm; NMR data: Table 3; HRESIMS m/z 290.0812 (calcd for C₁₀H₁₆N₃O₅S, [M + H]⁺, 290.0811).

3.6. Determination of the Configuration of the Leucine Moiety in 1 and 2

Compounds 1 and 2 (each about 0.5 mg) were heated separately in 1 mL of 6 N HCl at 100 °C for 16 h, followed by removal of the excess HCl under vacuum. To each dry hydrolysate, 200 µL of 1% solution of 1-fluoro-2,4-dinitrophenyl-5-l-alanine amide (FDAA, Marfey’s reagent) [24] in acetone and 40 µL of 1.0 M NaHCO₃ were added. The reaction mixture was heated at 45 °C for 1.5 h, cooled, and acidified with 20 µL of 2.0 M HCl. Similarly, standard amino acids (d and l) of phenylalanine and leucine were derivatized separately. The derivatized standard amino acids and hydrolysates of 1 and 2 were subjected to HPLC on Nova-Pak C18 reverse-phase column (150 × 3.9 mm i.d., 4 mm particle size; Waters, Milford, MA, USA) using the following gradient program. Solvent A was a 50 mM triethylamine–phosphate buffer (pH 3.5) containing 25% (v/v) MeOH, and solvent B was the same buffer containing 70% MeOH. The mobile phase was a linear gradient from 0 to 100% B (100 to 0% A) in 40 min, at a flow rate of 0.65 mL/min at 25 °C. The eluted peaks were monitored at 340 nm. The retention times for FDAA derivatives of standards and compounds 1 and 2 were as follows: (l)-leucine (t_R 27.2 min), (d)-leucine (t_R 36.0 min), compound 1 (t_R 27.2 min) and compound 2 (t_R 27.2 min) (Figure S39).

3.7. Computational Details

All DFT calculations have been performed using Gaussian 16 [33]. 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. GIAO NMR properties were calculated at the mpw1pw91/6-311+g(2d,p) level. DP4 probabilities were calculated using our own implementation of the algorithm published by Smith and Goodman [21]. Rotational strengths were calculated on 20 excited states using the b3lyp/6-31g(d) level of theory. The ECD spectra were plotted using Gaussview 6. All calculations for both isomers of 1 and 2 (3S,6S and 3R,6R) can be found at (https://figshare.com/s/d626b72364548b03e11b).

3.8. Cytotoxicity Evaluation

The cytotoxicity of the compounds was evaluated against three tumorous human cell lines including colorectal carcinoma (HCT-116, ATCC CCL-247) and breast cancer (MCF-7, ATCC HTB-22) (Figure S40). The evaluation of the cytotoxicity of the compounds was carried out by sulforhodamine B (SRB) assay as reported before [30].

3.9. Antimicrobial Evaluation

A disc diffusion assay was used to determine the antimicrobial activity of the compounds [31] with replication (n = 3). Staphylococcus aureus and Candida albicans served as target models for bacteria and fungi. A total of 100 µg of each compound was loaded onto 6-mm sterile circular filter-paper discs. The paper discs were left to air-dry. The dried paper discs were placed onto nutrient agar plates that had already been inoculated with a lawn of target microorganisms. After 24 h of incubation, the antimicrobial activity of the compounds was calculated.

4. Conclusions

In conclusion, investigation of a sponge-derived actinomycete, Streptomyces sp. Call-36, yielded a new diketopiperazine alkaloid, actinozine A (1), with a hydroperoxy functionality at C-2 of the proline moiety, two new thymidine derivatives, thymidine-3-mercaptocarbamic acid (3) and thymidine-3-thioamine (4), with unprecedented functionalities at N-3 of the thymine moiety. Additionally, the previously reported compounds cyclo(2-OH-d-Pro-l-Leu) (2), cyclo(d-Pro-l-Phe) (5) and cyclo(l-Pro-l-Phe) (6) were isolated. The one- and two-dimensional NMR data and MS spectral determinations supported the assignment of the compounds. The absolute stereochemistry of 1 and 2 were established by Marfey’s method and by comparison of the experimental and TDDFT-calculated ECD spectra with the experimental ones. Compounds 1 and 2 displayed potent activity against S. aureus and were moderately active against C. albicans. Finally, compound 5 displayed moderate and selective activity against HCT-116 with an IC₅₀ of 32.7 μM.

Supplementary Materials

The Supplementary Materials are available online at https://www.mdpi.com/1660-3397/17/10/584/s1, Figure S1: ¹H NMR spectrum of compound 1 (CDCl₃), Figure S2: expansion of the ¹H NMR spectrum of compound 1, Figure S3: ¹³C NMR spectrum of compound 1 (CDCl₃), Figure S4: ¹H-¹H COSY NMR spectrum of compound 1 (CDCl₃), Figure S5: HSQC spectrum of compound 1 (CDCl₃), Figure S6: HMBC spectrum of compound 1 (CDCl₃), Figure S7: HRESIMS spectrum of compound 1, Figure S8: ¹H NMR spectrum of compound 2 (CDCl₃), Figure S9: expansion of ¹H NMR spectrum of compound 2 (CDCl₃), Figure S10: ¹³C NMR spectrum of compound 2 (CDCl₃)^, Figure S11: ¹H-¹H COSY NMR spectrum of compound 2 (CDCl₃), Figure S12: HSQC spectrum of compound 2 (CDCl₃), Figure S13: HMBC Spectrum of compound 2 (CDCl₃), Figure S14: HRESIMS spectrum of compound 2, Figure S15: ¹H NMR spectrum and expansion of ¹H NMR spectrum of compound 3 (DMSO-d₆), Figure S16: ¹³C NMR spectrum of compound 3 (DMSO-d), Figure S17: ¹H-¹H COSY NMR Spectrum of compound 3 (DMSO-d), Figure S18: HSQC spectrum of compound 3 (DMSO-d₆), Figure S19: HMBC spectrum of compound 3 (DMSO-d₆), Figure S20: ROESY spectrum of compound 3 (DMSO-d₆), Figure S21: HRESIMS spectrum of compound 3, Figure S22: ¹H NMR Spectrum of compound 4 (DMSO-d₆), Figure S23: expansion of ¹H NMR Spectrum of compound 4 (DMSO-d₆), Figure S24: ¹³C NMR spectrum of compound 4 (DMSO-d₆), Figure S25: ¹H-¹H COSY NMR Spectrum of compound 4 (DMSO-d₆), Figure S26: HSQC Spectrum of compound 4 (DMSO-d₆), Figure S27: HMBC spectrum of compound 4 (DMSO-d₆), Figure S28: HRESIMS spectrum of compound 4, Figure S29: ¹H NMR spectrum of compound 5 (CDCl₃), Figure S30: ¹³C NMR spectrum of compound 5 (CDCl₃), Figure S31: ¹H-¹H COSY NMR spectrum of compound 5 (CDCl₃), Figure S32: HSQC spectrum of compound 5 (CDCl₃), Figure S33: HMBC spectrum of compound 6 (CDCl₃), Figure S34: ¹H NMR spectrum of compound 6 (CDCl₃), Figure S35: ¹³C NMR Spectrum of compound 6 (CDCl₃), Figure S36: ¹H-¹H COSY NMR spectrum of compound 6 (CDCl₃), Figure S37: HSQC spectrum of compound 6 (CDCl₃), Figure S38: HMBC spectrum of compound 6 (CDCl₃), Figure S39: HPLC trace for Marfey’s-derivatized hydrolysates of compounds 1,2,5,6 and derivatized standard amino acids, Figure S40: Concentration-response profiles for compounds 1–6.

Author Contributions

L.A.S. and D.T.A.Y. conceived and designed the experiments; L.A.S., J.M.B., and S.M.H. performed the experiments; L.A.S., J.M.B., and D.T.A.Y. interpreted and analyzed the NMR and MS data; G.G.-J. analyzed the calculated NMR and ECD data; L.A.S. and D.T.A.Y. wrote the paper; D.T.A.Y. edited and revised the paper.

Funding

This project has received funding from the National Plan for Science, Technology and Innovation (MAARIFAH)—King Abdulaziz City for Science and Technology—the Kingdom of Saudi Arabia under award number (12–BIO2251–03).

Conflicts of Interest

The authors declare no conflict of interest.