Enigmazole C: A Cytotoxic Macrocyclic Lactone and Its Ring-Opened Derivatives from a New Species of Homophymia Sponge

Abstract

A new macrolide, enigmazole C (1), and two additional analogues, enigmazoles E (2) and D (3), were obtained from a new species of the Homophymia genus as part of an ongoing discovery program at PharmaMar to study cytotoxic substances from marine sources. The structures were fully characterized by cumulative analyses of NMR, IR, and MS spectra, along with density functional theory computational calculations. All three of the new compounds feature an unusual 2,3-dihydro-4H-pyran-4-one moiety, but only enigmazoles C (1) and D (3) showed cytotoxic activity in the micromolar range against A-549 (lung), HT-29 (colon), MDA-MB-231 (breast), and PSN-1 (pancreas) tumor cells.

Marine sponges represent a prolific source of structurally unique macrolides possessing promising biological activities, including cytotoxic, anticancer, and neuroprotective properties, thus suggesting their potential value for the development of leads in drug discovery.¹ Particularly, sponges from the Neopeltidae family have not been extensively chemically investigated, with relatively few structures being described in the literature from each of the three genera Homophymia, Callipelta, and Daedalopelta. By way of illustration, the high molecular weight peptides homophymines A–E/A1–E1,^2,3 homophymamide A,⁴ pipecolidepsins A–C,⁵⁻⁷ and callipeltins A and B,⁸⁻¹⁰ have all been reported to have significant cytotoxic and anti-HIV activities. Others metabolites isolated from Neopeltidae sponges that were assigned in 2013 to the suborder Astrophorina^11,12 are the bioactive tetramic acid glycoside aurantoside C¹³ and the macrocycles callipeltosides A–C.^14,15 Furthermore, only two structures belonging to the Daedalopelta genus have been described, the cytotoxic cyclodepsipeptide daedophamide¹⁶ and the 14-membered macrolide neopeltolide,¹⁷ which has been extensively studied and synthesized because it is a potent cytochrome bc1 complex inhibitor, as well as a cytotoxic compound with activity against A-549 human lung adenocarcinoma, NCI/ADR-Res ovarian sarcoma, and P388 murine leukemia cell lines.¹⁸⁻²¹

During continuing efforts at PharmaMar to discover new cytotoxic compounds from marine natural sources, we have evaluated a new sponge species of the Homophymia genus (Vacelet & Vasseur, 1971) collected off the coast of Gorontalo, Indonesia. In this paper, we describe the isolation of a new macrocyclic compound as well as two open-chain analogues, all isolated from a specimen of this sponge collected in Indonesia. Despite the fact that the new macrocycle shows structural similarities to neopeltolide, a clear resemblance to the macrolide enigmazole A isolated by Oku et al. in 2010 from the marine sponge Cynachyrella enigmatica,²² has led us to designate the three new compounds as enigmazoles C (1), E (2), and D (3). The configurations of these enigmazoles have been solved using a combination of microscale chemical conversions and the use of an elegant J-based configurational analysis based on capillary NMR measurements.²³

Chart 1. Structures of Enigmazoles C (1), E (2), and D (3) and Known Enigmazole A (4).

A methanolic extract of the Homophymia sponge specimen showed cytotoxic activity against A-549 (lung), HT-29 (colon), MDA-MB-231 (breast), and PSN-1 (pancreas) tumor cells and was hence selected for a more detailed bioassay-guided chemical investigation.

Careful fractionation of the MeOH extract led to isolation of the new macrolide enigmazole C (1, 2.9 mg), and two open-chain derivatives, enigmazole D (3, 0.4 mg) and the methanolic adduct enigmazole E (2, 1.4 mg), which were purified by semipreparative HPLC.

Enigmazole C (1) was isolated as a colorless amorphous solid and showed a [M + H]⁺ ion at m/z 516.2964 (calcd for C₂₉H₄₂NO₇m/z 516.2956) in its (+)-HR-ESITOF-MS spectrum. The presence of the sodium adduct at m/z 538.2785 (calcd for C₂₉H₄₁NO₇Na m/z 538.2775) confirmed the molecular formula and 10 degrees of unsaturation required for this compound.

NMR experiments of 1 were carried out in CD₃CN because a chemical transformation was observed using CD₃OD. ¹H, ¹³C, and edited-HSQC NMR analysis of 1 revealed the presence of 29 carbons assigned to seven sp² nonprotonated carbons (δ_C 193.6, 175.6, 175.5, 161.2, 152.7, 143.9, and 141.7), three sp² methine carbons (δ_H/δ_C 7.63/135.9, 6.20/113.6, and 5.30/106.8), a disubstituted exomethylene moiety (δ_H/δ_C 4.91; 4.86/113.9), six sp³ methine carbons (δ_H/δ_C 5.92/66.7, 5.20/75.5, 4.48/77.8, 3.74/69.5, 2.49/39.9, and 2.20/25.8), seven sp³ methylene carbons (δ_H/δ_C 2.72; 2.62/42.2, 2.49; 2.27/44.6, 2.49; 1.59/40.2, 2.41; 2.23/42.2, 1.81; 1.46/42.9, 1.69; 1.63/30.8 and 1.43; 1.31/34.5), and five methyl groups (δ_H/δ_C 3.15/56.6 (OMe), 1.86/17.7, 1.23/19.5, 1.10/17.6, and 0.95/21.2).

Two-dimensional COSY, 2D-TOCSY, and selective 1D-TOCSY experiments of 1 allowed us to identify five different spin systems, fragments A (C23–C24), B (C21–C25), C (C16–C19), D (C2–C6, C26), and E (C10–14, C27) (Figure 1). Special attention was paid to fragments D and E due to the presence of four stereogenic centers. Thus, the spin system D was identified by selective 1D-TOCSY from selective irradiation of proton H-5 at δ_H 3.74, which gave responses to the methylene protons at δ_H 1.69/1.63 (H₂-3), δ_H 1.43/1.31 (H₂-4), and δ_H 2.49/2.27 (H₂-6), as well as the methine at δ_H 2.49 (H-2), the methyl group at δ_H 1.10 (H₃-26), and a broad signal at δ_H 3.00 assigned to a hydroxy proton (Figure 2b).

Figure 1.

Spin systems deduced by COSY and 1D-TOCSY experiments and HMBC assembling in 1.

Figure 2.

(a) NMR spectrum of enigmazole C (1) in CD₃CN at 500 MHz. Selective 1D-TOCSY irradiation experiments at (b) δ_H 3.74 (H-5) and (c) δ_H 4.48 (H-11) to identify fragments D and E.

In the same way, fragment E was deduced by selective 1D-TOCSY irradiation of H-11 at δ_H 4.48, which showed coupling responses to three methylenes at δ_H 2.41/2.23 (H₂-10), δ_H 1.81/1.46 (H₂-12), and δ_H 2.49/1.59 (H₂-14), the methine H-13 (δ_H 2.20), and the methyl group δ_H 0.95 (H₃-27) (Figure 2c).

The interconnection between the determined fragments A–E was achieved by using an HMBC experiment as follows:

• (a)The methine at δ_H 5.20 (H-23) showed long-range correlation with three methyl groups at δ_C 19.5 (C-24), δ_C 17.7 (C-25), and δ_C 56.6 (C-29), along with the sp² methine carbon at δ_C 113.6 (C-21), implying the position of the methoxy group on C-23 and the connection between fragments A and B.
• (b)The HMBC cross-peak observed from H-21 (δ_H 6.20) to C-22 (δ_C 152.7) placed a nonprotonated carbon in the double bond. On the other hand, the singlet at δ_H 7.63 (δ_C 135.9) presented a typical chemical shift and ¹J_CH of 210 Hz of a 2,4-disubstituted oxazole ring, which was completed by inspection of the HMBC correlations observed between the oxazolic proton H-19 to carbons C-18 (δ_C 141.7) and C-20 (δ_C 161.2). The connection between fragments B and C was undoubtedly determined by the HMBC cross-peak from H-25 (δ_H 1.86) to C-20 (δ_C 161.2).
• (c)Fragments D and E were connected through a six-membered ring by the cross-peaks observed from δ_H 5.30 (H-8) to δ_C 44.6 (C-6), δ_C 42.2 (C-10), and to a nonprotonated oxygenated sp² carbon at δ_C 175.5 (C-7). A dihydropyranone ring was deduced by the observation of the HMBC correlations from both diastereotopic protons at H-10_a (δ_H 2.41) and H10_b (δ_H 2.23) to an α,β-unsaturated carbonyl signal at δ_C 193.6.
• (d)Fragments C and E were linked by a disubstituted exomethylene group as the HMBC correlations observed of H-28 (δ_H 4.86) with C-14 (δ_C 40.2) and C-16 (δ_C 42.2).
• (e)A cross-peak from H-17 (δ_H 5.92) to C-1 (δ_C 175.6) allowed us to connect fragments C and D, therefore assembling the cyclic macrolactone planar structure for 1 as is drawn in Figure 1.

At this point, the resemblance of 1 to the known compound enigmazole A (4) was clear, as some structural features are common in both compounds: a disubstituted oxazole ring, a pyran ring, the presence of an exomethylene double bond, and a similar macrolactone size. Once the planar structure of 1 was established, the relative and absolute configuration were determined by ROESY experiments, derivatization reactions, and computational calculations.

First, the Z configuration of the C-21/C-22 double bond was deduced from the ROESY correlation between H₃-25 (δ_H 1.86) and H-21 (δ_H 6.20).

To deduce the relative configuration of the entire spin system D, we were nicely able to relate the two stereogenic centers at C-2 and C-5 located three carbon–carbon bonds away through a J-based configurational analysis (JBCA).²⁴ To make this possible, we had to change the NMR solvent to acetone-d₆, because it gave us a ¹H NMR spectrum where the two diastereotopic pairs at C-3 (H-3a and H-3b) and C-4 (H-4a and H-4b) are well separated and fully resolved.

Not many approaches of this kind have been applied to a natural compound with two sp³ methylenes surrounded by two stereogenic centers, even though this approximation is a very good tool when the diastereotopic protons can be unequivocally assigned with their corresponding chemical shifts and their sets of proton–proton and carbon–proton coupling constants. Therefore, ¹³C–¹H-HSQC-TOCSY-HECADE and J-HMBC experiments were needed to measure key small coupling constants for ³J_C26–H3a and ³J_C26–H3b to place H₃-26 in a gauche disposition to both H-3 diastereotopic protons at the C-2/C-3 single bond (Figure 3a). These two experiments along with selective ¹H NMR irradiation spectra were satisfactory to deduce relationships from diastereotopic protons H-3a/H-3b and H-4a/H-4b to C-2 and C-5 as is drawn Figure 3b) for the C-3/C-4 bond. The hydroxylated carbon at C-5 helped us to deduce the relative dispositions for H-4a and H-4b to the OH group with regard to the C-4/C-5 bond (Figure 3c), as well as H-6a and H-6b to the mentioned OH relative to the C-5/C-6 bond (Figure 3d).

Figure 3.

C-2 to C-6 fragment deduced by J-based configurational analysis.

The absolute configuration at position C-5 was elucidated by the application of the modified Mosher’s method (MMM).^25,26 The derivatization of the secondary alcohol with R- and S-MTPA-Cl was completed directly in pyridine-d₅ in an NMR tube to give compounds 1-S and 1-R, respectively. Comparison of the Δδ values (δ_S – δ_R) obtained from the MTPA esters (Figure 4) indicated that the absolute configuration of C-5 was R, which implies a C-2S configuration.

Figure 4.

Mosher’s derivatives of enigmazole C (1).

During the process of purification of enigmazole C, the use of MeOH in the HPLC separation procedure induced a macrolactone ring-opening that produces the methyl ester adduct 2, which we have named as enigmazole E.

The (+)-HR-ESIMS-TOF data of 2 at m/z 548.3249 [M + H]⁺ (calcd for C₃₀H₄₆NO₈, m/z 548.3218) and the ¹H NMR of 2 revealed the opening of the macrolactone ring.

Thus, the ¹H NMR spectrum of 2 displays a singlet signal at δ_H 3.61 (s, 3H) that was assigned to a methoxy group, which shows an HMBC correlation to an carbonyl carbon at δ_C 176.6 assigned to the new carbonyl ester at C-1. Moreover, the proton chemical shift at δ_H 5.92, assigned to H-17 in 1, was clearly shifted to δ_H 4.72 in 2 as expected due to the lack of the lactone moiety. The presence of a hydroxy group at C-17 in 2 due to the macrolactone ring opening of 1 could be used to determine the absolute configuration at this position by the application of the MMM using the α-methoxy-α-trifluoromethylphenylacetic (MTPA) esters.

In this case, the comparison between the proton chemical shifts of both MTPA esters at C-17 (δ_2-S – δ_2-R) established the absolute configuration of C-17 as S (see Figure 5). In addition, the esterification of 2 with MTPA-Cl generated also the MTPA esters at C-5; the comparison of proton chemical shift values confirmed again the configuration of this stereogenic center as R.

Figure 5.

Modified Mosher methodology (MMM) applied to enigmazole E (2) to deduce the absolute configuration at C-17 and confirming the absolute configuration at C-5.

To establish the configuration between C-11 and C-13 at 1, and therefore of 2, homonuclear coupling constants of H-11/H-12b, H-11/H-12a, H-12b/H-13, and H-12a/H-13 were extracted from selective 1D-TOCSY experiments (Figure 2c). The multiplicity observed for H-11 as a dddd with two large and two small couplings (³J_H11–H10a 14.1 Hz, ³J_H11–H10b 2.7 Hz, ³J_H11–H12b 11.0 Hz, and ³J_H11–H12a 2.6 Hz), the heteronuclear coupling constants (³J_C10–H12a 1.7 Hz, ³J_C10–H12b 1.7, ²J_C11–H12a −5.6 Hz ²J_C11–H12b 0 Hz), and the ROESY correlation observed between H-12b and H-10 allowed us to deduce the presence of the rotamer represented for C-11/C-12 bonds in Figure 6a.

Figure 6.

Rotamers along C-11/C-12 (a) and C-12/C-13 (b) bonds in enigmazole C (1).

In the same way, a large proton–proton coupling constant of 11.0 Hz between H-13 and H-12a fixed these protons in an antiperiplanar disposition, making necessary the use of ROESY correlations between the pairs H-12a/H-27 and H-11/H-14 and the long-range ¹³C–¹H coupling constants ³J_C11–H13 0.5 Hz, ³J_C27–H12a 3.0 Hz, and ²J_C27–H12a 3.0 Hz to determine the presence of the rotamer drawn in Figure 6b. Once H-12a and H-12b are interrelated to conformers 6a and 6b, we were able to determine the relative configuration of the two stereogenic centers C-11 and C-13 as R* and R*.

While the absolute configurations of C-2, C-5, and C-17 are already known as S, R, and S, respectively, the relative configurations of C-11 and C-13 as R* and R* along with the possible configuration at C-23 of either R or S give us four possible enigmazole C diastereomers (1a–d) that can be discriminated by an NMR–density functional theory (DFT) approximation. Initially we submitted all diastereoisomers to a conformational search with the Macromodel program using the protocol of Daranas, Sarotti, et al.²⁷ Thus, 58 conformers for 1a-(2S,5R,11R,13R,17S,23R), 64 for 1b-(2S,5R,11S,13S,17S,23S), 60 for 1c-(2S,5R,11S,13S,17S,23R), and 62 for 1d-(2S,5R,11R,13R,17S,23S) were found within a 5.0 kcal/mol window, which were further classified by energy and frequencies using the B3LYP/6-31G(d) functional. Once the duplicates and conformers with imaginary frequencies were removed, a combination of MPW1PW91/6-31G(d,p) and the polarizable continuum model was used for proton and carbon chemical shift calculations using MeCN as solvent. The sets of ¹H and ¹³C chemical shifts were compared by the statistical DP4+ parameter developed by Sarotti and co-workers.^27,28 A 100% probability DP4+ value for all chemical shifts was in agreement with diastereoisomer 1a-(2S,5R,11R,13R,17S,23R) (Figure 7).

Figure 7.

DP4+ and iJ-DP4 statistical correlations of diastereoisomers 1a-(2S,5R,11R,13R,17S,23R), 1b-(2S,5R,11S,13S,17S,23S), 1c-(2S,5R,11S,13S,17S,23R), and 1d-(2S,5R,11R,13R,17S,23S) of enigmazole C.

However, when the iJ-DP4 was computed using the suggested combination of the MMFF-5 kcal/mol conformational search and B3LYP/6-31G(d) for chemical shifts and coupling constants (just the contact Fermi contribution), no clear discrimination between isomers 1a-(2S,5R,11R,13R,17S,23R) and 1d-(2S,5R,11R,13R,17S,23S) was achieved when six ³J_HH constraints were introduced.²⁷ Clearly both DP4+ and iJ-DP4 were able to discriminate 11R,13R diastereoisomers 1a and 1d from both 11S,13S-1b and 11S,13S-1c. At this point just C-23 remains to be assigned, and since the fragment C-18/C-25 showed similar NMR values (see chemical shifts in CD₃OD in the Supporting Information) to that for enigmazole A (4),²² we propose the final structure for enigmazole C (1) as 2S,5R,11R,13R,17S,23R (diastereoisomer 1a). In comparison with enigmazole A, 1 presents the same configurations of the stereogenic centers at the C-2, C-5, C-11, C-13, C-17, and C-23 positions.

Along with 1, an analogue of 3, enigmazole D, was also found during the process of isolation and purification. Its HR-ESITOF-MS data showed the [M + H]⁺ adduct at m/z 516.2932, which was in perfect agreement for a molecular formula of C₂₉H₄₂NO₇. Although the formula of 3 matches that found for enigmazole C, its HPLC retention time and ¹H NMR spectrum were different. Three new ¹H NMR signals corresponding to two vinylic protons (δ_H 6.58 and 6.00) and one oxygenated methine at δ_H 4.74, in addition to the absence of the signals for H-17 (δ_H 5.92), H-5 (δ_H 3.74), and the OH group at δ_H 3.00 in 1, lead us to suspect the opening of the macrolactone ring accompanied by an elimination of H₂O. COSY correlations of the sp² methine group at δ_H 6.58 with δ_H 6.00 and 2.21 and two extra signals observed from sp² methines, with the loss of the hydroxy group, clearly suggested the formation of a new double bond in 3. The movement of the signal at δ_H 5.92 (H-17) found in 1 to a lower value of δ_H 4.74 in 3 along with the expected chemical shift of the new methine proton is in agreement with the opening of the lactone ring. With all these data on hand, we were able to assign the structure of enigmazole E to 3, a metabolite resulted from the opening of enigmazole C and a further dehydration at position C-5. It is important to notice that this new double bond has an E configuration, based on the 15.6 Hz coupling constant found between H-5 and H-6.

Compounds 2 and 3 can be considered derivatives of enigmazole C, and therefore we assumed the absolute configurations at positions C-2, C-5 (just for compound 2), C-11, C-13, C-17, and C-23 were the same as for 1; thus (2S,5R,11R,13R,17S,23R) for enigmazole E (2) and (2S,11R,13R,17S,23R) for enigmazole D (3) were determined.

All of the new compounds were tested in a panel of four human cancer cell lines: A-549 (lung), HT-29 (colon), MDA-MB-231 (breast), and PSN-1 (pancreas). Enigmazole C (1) showed activity on the order of micromolar range with a GI₅₀ of 9.9 μM against the A-549 cell line (Table 2). Although enigmazole D (3) is the open form with an additional double bond of compound 1, cytotoxicity was observed in all cell lines (Table 3), while the methyl ester 2 of the open form of 1 renders the compound totally devoid of cytotoxicity.

Table 2. Cytotoxic Activities for Enigmazole C (1).

cell line	GI50	TGI	LC50
A-549	9.9 μM	>19 μM	>19 μM
HT-29	18 μM	>19 μM	>19 μM
MDA-MB-231	>19 μM	>19 μM	>19 μM
PSN-1	87 μM	>19 μM	>19 μM

Table 3. Cytotoxic Activities for Enigmazole D (3).

cell line	GI50	TGI	LC50
A-549	1.4 μM	>19 μM	>19 μM
HT-29	1.0 μM	3.7 μM	>19 μM
MDA-MB-231	4.1 μM	>19 μM	>19 μM
PSN-1	1.1 μM	>19 μM	>19 μM

In summary, the investigation of the Homophymia sp. sponge led to the isolation of one new macrolide lactone, enigmazole C (1), along with two open form congeners, enigmazoles E (enigmazole C seco acid methyl ester) (2) and enigmazole D (3). Spectroscopic data, chemical derivatization, and NMR-DFT methods were necessary to deduce the configurations of the six stereogenic centers present in their carbon skeletons. Notably, compounds 1 and 3 exhibited cytotoxic activities against A-549 (lung), HT-29 (colon), MDA-MB-231 (breast), and PSN-1 (pancreas) cell lines.

Experimental Section

General Experimental Procedures

Optical rotations were measured on a JASCO DIP-1000 polarimeter, with a Na (589 nm) lamp and filter. UV spectra were measured on a JASCO V-650 spectrophotometer. IR spectra were measured on a FTIR Bruker Vector 22 spectrometer. ¹H, ¹³C, and 2D NMR spectra were recorded on a Varian “Unity 500” (500 MHz for ¹H and 125 MHz for ¹³C), Varian “Unity 400” (400 MHz for ¹H and 100 MHz for ¹³C), and Bruker Avance 500 (500 MHz for ¹H and 125 MHz for ¹³C) with a dual cryoprobe or a BBI probe. CD₃OD, CD₃CN, C₅D₅N, and acetone-d₆ were used as deuterated solvents. Chemical shifts are reported in δ scale relative to methanol-d₄ (δ 3.31 ppm for ¹H NMR, δ 49.0 ppm for ¹³C NMR), acetonitrile-d₃ (δ 1.94 ppm for ¹H NMR, δ 1.32 ppm for ¹³C NMR), acetone-d₆ (δ 2.05 ppm for ¹H NMR, δ 29.84 ppm for ¹³C NMR), and pyridine-d₅ (δ 8.74 ppm for ¹H NMR, δ 150.35 ppm for ¹³C NMR). HSQC-TOCSY-HECADE experiments were acquired with 32 scans and 256 increments with a mixing time of 60 ms and 4K points. J-HMBC was run with 16 scans, 200 increments, and 2 K points in F2.

LRESIMS and HRESIMS experiments were performed on the Applied Biosystems QSTAR Elite system or on an Agilent 6230 TOF LC/MS. HPLC separation was performed on an Agilent 1100 or 1200 using reversed-phase chromatographic columns.

Animal Material

The sponge belongs to the genus Homophymia Vacelet and Vasseur, 1971. It was collected by hand using a rebreather diving system in Gorontalo, Indonesia (01° 19.836′ S/122° 45.022′ E) at depths ranging between 40 and 80 m. The animal material was identified by Dr. María Jesús Uriz (Center for Avanced Studies of Blanes). A sample of the specimen was deposited in the Center for Advanced Studies of Blanes in Girona, Spain, with the reference code GORO-093. In addition, a voucher specimen (ORMA136284) was deposited at PharmaMar.

Description: Thickly globular sponge (3 cm high, 4 cm wide, 2 cm tick), smooth without conulous projections. Hard consistency, white color, and small single oscule in apical position. Megascleres: fine monoactines spicules broken in their preparations. Desmas pseudotetraclones with very tuberculated zygomes. Ectosomal megascleras are smooth pesudophyllotriaenes with different irregular pseudocladis tuberculated in the same specimen. Microscleres consist of a single type of microspinose amphiasters occurring in different size classes: the smaller 10–15 μm long, the larger 12–25 μm long.

Extraction and Isolation

A specimen of Homophymia (33 g) was triturated and exhaustively extracted with MeOH/CH₂Cl₂ (50:50, 3 × 500 mL). The combined extracts were concentrated to yield a mass of 2.5 g, which was subjected to VLC on Lichroprep RP-18 (Merck KGaA) with a stepped gradient from H₂O to MeOH and then CH₂Cl₂. The fraction eluting with MeOH (96 mg) was subjected to semipreparative HPLC (Ascentis C18, 5 μm, 410 × 150 mm, isocratic H₂O/MeCN (60:40) for 3 min, gradient H₂O/MeCN from 50% to 65% MeCN in 25 min, UV detection, flow 1 mL/min, t_r: 19.5 min to give compound 1 (2.9 mg), 12.0 min to give compound 2 (1.4 mg), and 13.0 min to give compound 3 (0.4 mg)).

Enigmazole C (1):

amorphous, yellow oil; [α]_D +56 (c 0.03, MeOH); UV (MeOH) λ_max 261 nm; IR (ATR) ν_max 3378, 2966, 2311, 1675, 1140 cm^–1; ¹H NMR (500 MHz) and ¹³C NMR (125 MHz), Table 1; (+)-HRESI-TOFMS m/z 516.2964 [M + H]⁺ (calcd for C₂₉H₄₂NO₇, 516.2956); m/z 538.2785, [M + Na]⁺ (calcd for C₂₉H₄₁NO₇Na).

Table 1. NMR Data of Enigmazole C (1) in CD₃CN and Acetone-d₆.

	CD3CN δH, m (J in Hz)		acetone-d6
pos.	δC, type	δH, m (J in Hz)	δC	δH, m (J in Hz)
1	175.6, C		174.9
2	39.9, CH	2.49, m	39.7	2.53 ddq (10.6, 6.9, 3.4)
3	30.8, CH2	a: 1.69, m	30.8	a: 1.79, dddd (13.8, 9.9, 5.4, 3.4)
		b: 1.63, m		b: 1.68, dddd (13.8, 10.6, 5.3, 5.3)
4	34.5, CH2	a: 1.43, m	34.5	a: 1.51, dddd (14.4, 9.9, 5.3, 2.2)
		b: 1.31, m		b: 1.38, dddd (14.4, 10.4, 5.4, 5.3)
5	69.5, CH	3.74, ddddd (10.4, 7.6, 5.5, 5.5, 2.5)	69.6	3.83, ddddd (10.4, 8.1, 5.8, 4.5, 2.2) 2.2)
6	44.6, CH2	a: 2.49, dd (13.2, 5.5)	44.9	a: 2.59, dd (13.1, 4.5)
		b: 2.27, dd (13.2, 7.6)		b: 2.31, dd (13.1, 8.1)
7	175.5, C		174.8
8	106.8, CH	5.30, d (1.0)	106.8	5.28, bs (1.0)
9	193.6, C		192.4
10	42.2, CH2	a: 2.41, dd (16.7, 14.0)	42.3	a: 2.41, dd (16.7, 14.0)
		b: 2.23, dd (16.7, 2.6)		b: 2.26, dd (16.7, 2.7)
11	77.8, CH	4.48, ddt (14.0, 11.0, 2.6, 2.6)	77.4	4.53 ddt (14.0, 11.0, 2.7, 2.7)
12	42.9, CH2	a: 1.81, ddd (14.2, 10.9, 2.6)	43.1	a: 1.85, ddd (14.1, 11.0, 2.7)
		b: 1.46, dd (14.2, 11.0)		b: 1.54, ddd (14.2, 11.1, 2.3)
13	25.8, CH	2.20, m	25.6	2.29, m
14	40.2, CH2	a: 2.49, dd (15.8, 1.5)	40.0	a: 2.27, dd (16.0, 2.7)
		b: 1.59, dd (15.8, 10.7)		b: 1.61, dd (16.0, 11.0)
15	143.9, C		143.8
16	42.2, CH2	a: 2.72, dd (13.7, 9.6)	42.3	a: 2.75, dd (13.8, 9.7)
		b: 2.62, dd (13.7, 5.2)		b: 2.67, dd (13.8, 5.1)
17	66.7, CH	5.92, ddd (9.4, 5.2, 0.8)	66.3	6.00, dd (9.7, 4.9)
18	141.7, C		142.0
19	135.9, CH	7.63, d (0.7)	135.6	7.77, s
20	161.2, C		161.0
21	113.6, CH	6.20, dd (1.5, 0.8)	113.5	6.21, bs
22	152.7, C		152.4
23	75.5, CH	5.20, dq (6.5, 0.8)	75.3	5.29, bq (6.5)
24	19.5, CH3	1.23, d (6.5)	19.4	1.24, d (6.5)
25	17.7, CH3	1.86, d (1.5)	17.5	1.88, d (1.6)
26	17.6, CH3	1.10, d (6.9)	17.6	1.11, d (6.9)
27	21.2, CH3	0.95, d (6.7)	21.2	0.98, d (6.6)
28	113.9, CH2	a: 4.91, s	113.5	a: 4.94, s
		b: 4.86, s		b: 4.88, s
29	56.6, CH3	3.15, s	56.5	3.17, s
OH	-	3.00, d (6.0)	-	3.90 d (5.8)

Enigmazole E (enigmazole C seco acid methyl ester, 2):

yellow oil; UV (MeOH) λ_max 265 nm; ¹H and ¹³C NMR, Table S1; (+)-HRESI-TOFMS m/z 548.3249 [M + H]⁺ (calcd for C₃₀H₄₆NO₈, 548.3218).

Enigmazole D (3):

yellow oil; UV (MeOH) λ_max 254, 277, 286 nm; ¹H and ¹³C NMR, Table S2; (+)-HRESI-TOFMS m/z 516.2932 [M + H]⁺ (calcd for C₂₉H₄₂NO₇, 516.2956).

Computational Calculations

Conformational searches were performed by using the corresponding module implemented in the Maestro Quantum mechanical software. The MMFF force field with acetonitrile as solvent was used, and torsional enhanced sampling with 10 000 steps was fixed using an energy window of 5 kcal/mol.

DP4+

Molecular geometry optimizations were performed for 1a–d at the DFT theoretical level using the Gaussian 16W package first at the B3LYP/6-31G(d) level for energy and frequency calculations. After removing redundant conformers and those with imaginary frequencies, theoretical Boltzmann energy population-weighted ¹H and ¹³C NMR were calculated by using the combination MPW1PW91/6-31G(d,p). Results were input in Sarotti’s and Darana’s Excel spreadsheet²⁷ to calculate the best fit.

iJ-DP4

MMFF conformers at 5 kcal/mol were filtered by six proton–proton vicinal coupling restriction: ³J_H11H12a, ³J_H11H12b, ³J_H5H4b, ³J_H5H4a, ³J_H2H3b, and ³J_H2H3a. Those conformers were chosen to calculate the population-weighted Boltzmann MMFF energy. ¹H and ¹³C NMR and ¹H–¹H coupling constants (just the Fermi’s contact contribution) were then calculated by the combination B3LYP/6-31G(d). Results were input in Sarotti’s and Darana’s Excel spreadsheet²⁷ to calculate the best fit.

Preparation of the (S)-MTPA Ester of Enigmazole C (1-S)

R-(−)-MTPA chloride (10 μL) was added to a solution of 1 (1.0 mg) in 0.5 mL of pyridine-d₅ in an NMR tube. The resulting mixture was allowed to stand at room temperature (rt) for 8 h to yield the (R)-MTPA ester of 1, which was monitored by recording ¹H NMR spectra at 500 MHz. ¹H NMR (500 MHz, C₅D₅N) δ 5.66 (s, H-8), 5.56 (m, H-5), 2.83 (dd, J = 13.7, 5.1 Hz, H-6a), 2.71 (m, H2), 2.64 (dd, J = 13.7, 7.2 Hz, H-6b), 1.79 (m, H-4), 1.66 (m, H-3), 1.15 (d, J = 6.8 Hz, H-26).

Preparation of the (R)-MTPA Ester of Enigmazole C (1-R)

Treatment of 1 (1.0 mg) in the same manner as before with S-(+)-MTPA chloride (10 μL) gave the (R)-MTPA ester of 1. ¹H NMR (500 MHz, C₅D₅N) δ 5.71 (s, H-8), 5.57 (m, H-5), 2.90 (dd, J = 13.8, 5.0 Hz, H-6a), 2.59 (m, H-2), 2.74 (dd, J = 13.8, 7.0 Hz, H-6b), 1.71 (m, H-4), 1.42 (m, H-3), 1.07 (d, J = 6.8 Hz, H-26).

Preparation of the (S)-MTPA Ester of Enigmazole E (2-S)

R-(−)-MTPA chloride (10 μL) was added to a solution of 2 (0.4 mg) in 0.5 mL of pyridine-d₅ in an NMR tube. The resulting mixture was allowed to stand at rt for 8 h to yield the (S)-MTPA ester of 2, which was monitored by recording ¹H NMR spectra at 500 MHz. ¹H NMR (500 MHz, C₅D₅N) δ 6.60 (dd, J = 7.2, 7.2 Hz, H-17), 5.69 (m, H-5), 5.67 (s, H-21), 5.45 (d, J = 6.4 Hz, H-23), 4.92 (s, H-28), 2.93 (m, H-16), 2.68 (m, H-6), 2.51 (m, H-2), 2.18 (m, H-14), 1.87 (s, H-25), 1.86 (m, H-3), 1.86 (m, H-4), 1.12 (d, 7.0 Hz, H-26), 0.91 (d, 5.9 Hz, H-27).

Preparation of the (R)-MTPA Ester of Enigmazole E (2-R)

Treatment of 2 (0.4 mg) in the same manner as before with S-(+)-MTPA chloride (5 μL) gave the (R)-MTPA ester of 2. ¹H NMR (500 MHz, C₅D₅N) δ 6.62 (dd, J = 7.1, 7.1 Hz, H-17), 5.70 (m, H-5), 5.66 (s, H-21), 5.40 (d, J = 6.3 Hz, H-23), 5.04 (s, H-28), 2.99 (m, H-16), 2.73 (m, H-6), 2.41 (m, H-2), 2.24 (m, H-14), 1.85 (s, H-25), 1.73 (m, H-3), 1.73 (m, H-4), 1.04 (d, 7.0 Hz, H-26), 0.93 (d, 5.9 Hz, H-27).

Biological Assays

The cytotoxic activities of compounds 1, 2, and 3 were tested against A-549 human lung carcinoma cells, MDA-MB-231 human breast adenocarcinoma cells, HT-29 human colorectal carcinoma cells, and PSN-1 pancreatic adenocarcinoma cells. The concentration giving 50% inhibition of cell growth (GI₅₀) was calculated according to the procedure described in the literature. Cell survival was estimated using the National Cancer Institute (NCI) algorithm. Three dose–response parameters were calculated for 1, 2, and 3.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jnatprod.1c01179.

• ¹H, ¹³C, and 2D NMR data of the new compounds and J configurational analysis of 1 (PDF)

The authors declare no competing financial interest.