Antioxidant and Neuroprotective Xenicane Diterpenes from the Brown Alga Dictyota coriacea

Abstract

Five new xenicane diterpenes, including three rare nitrogen-containing derivatives, dictyolactams A (1) and B (2) and 9-demethoxy-9-ethoxyjoalin (3), a rare diterpene with a cyclobutanone moiety, named 4-hydroxyisoacetylcoriacenone (4), and 19-O-acetyldictyodiol (5), were isolated from an East China Sea collection of the brown alga Dictyota coriacea, along with 15 known analogues (6–20). The structures of the new diterpenes were elucidated by spectroscopic analyses and theoretical ECD calculations. All compounds had cytoprotective effects against oxidative stress in neuron-like PC12 cells. The antioxidant mechanism of 18-acetoxy-6,7-epoxy-4-hydroxydictyo-19-al (6) was related to the activation of Nrf2/ARE signaling pathway; it also showed significant neuroprotective effects against cerebral ischemia-reperfusion injury (CIRI) in vivo. This study provided xenicane diterpene as a promising lead scaffold for the development of potent neuroprotective agents against CIRI.

Oxidative stress is considered to be one of the key pathogenic factors contributing to the occurrence and development of neurological diseases, especially cerebral ischemia-reperfusion injury (CIRI) brought about by thrombolytic therapy for ischemic stroke.¹ Nrf2/ARE signaling pathway is the most important endogenous defense system protecting nerve cells against oxidative damage.² Some natural products derived from terrestrial plants, such as britanin,³ swertiamain,⁴ and dl-3-n-butylphthalide,⁵ have proven to be able to prevent and mitigate CIRI by activating Nrf2 and enhancing antioxidant gene expression in the preclinical studies. However, the potential of marine natural products as novel neuroprotective antioxidants against CIRI has been rarely studied yet.⁶

Marine brown algae of the genus Dictyota (family Dictyotacea) are a rich source of diterpenes with diverse scaffolds,⁷ many of which are biologically active (cytotoxic, antibacterial, antiviral, and antioxidant properties^6,8−14). Previously, we reported the structure elucidation of dictyospiromide, a unique spirosuccinimide alkaloid with antioxidant properties that are associated with the activation of the Nrf2/ARE signaling pathway from the EtOAc extract of D. coriacea.¹⁵ Further chemical investigation of this species led to the isolation of 20 xenicane diterpene derivatives (1–20) including five new compounds (1–5), three of which contain a γ-lactam moiety (1–3) (Chart 1). These diterpenes were tested for their protective activities against oxidative stress in neuron-like PC12 cells; one of the major components 6 was selected for detailed studies of the antioxidant mechanism of action in vitro and the neuroprotective effect against CIRI in vivo. The details of isolation, structure elucidation, and biological evaluations of these xenicane diterpenes are described below.

Chart 1. Chemical Structures of the Xenicane Diterpenes 1–20.

Results and Discussion

Sequential column chromatography of the EtOAc extract of the brown alga D. coriacea yielded 20 xenicane diterpenes (1–20) including five new compounds (1–5) (Chart 1). The structures were established by 1D and 2D NMR (HSQC, HMBC, COSY, NOESY), HRESIMS, IR, and ECD spectroscopic and spectrometric analyses. Fifteen known compounds were identified as 18-acetoxy-6,7-epoxy-4-hydroxydictyo-19-al (6),¹⁶ 18-acetoxy-4-hydroxydictyo-19-al (7),¹⁷ 6,7-epoxy-4-hydroxydictyolactone (8),¹⁸ 18-hydroxydictyolactone (9),¹⁹ 4-hydroxydictyolactone (10),²⁰ (2S*,3S*,4R*,10R*,19R*)-19-deoxo-4-hydroxy-19-methoxydictyolactone (11),⁸ (2S*,3S*,4R*,10R*,19S*)-19-deoxo-4-hydroxy-19-methoxydictyolactone (12),⁸ isodictyohemiacetal (13),¹⁷ 4-hydroxycrenulide (14),²¹ 4,18-dihydroxycrenulide (15),²² 4-acetoxycrenulide (16),²² 18-hydroxy-4-acetoxycrenulide (17),²² crenulacetal B (18),²³ 4-hydroxypachylactone (19),¹³ and 4-acetoxysanadaol (20)²⁴ by comparing their ¹H and ¹³C NMR and MS data, as well as specific rotations with those reported in the literature.

Compound 1 had a molecular formula of C₂₄H₃₉NO₃ as determined by HRESIMS data, requiring six degrees of unsaturation. The IR absorptions at 3273 and 1697 cm^–1 indicated the presence of amino and carbonyl groups. The ¹H NMR spectrum of 1 displayed three methyl singlets at δ_H 1.62 (s, H₃-16), 1.57 (s, H₃-15), 1.24 (s, H₃-20), a methyl doublet at δ_H 0.98 (d, J = 6.0 Hz, H₃-17), and two methyl triplets at δ_H 1.22 (t, J = 7.2 Hz) and 1.15 (t, J = 7.2 Hz) that attributed to two ethoxy groups (Table 1). The ¹³C NMR spectrum showed 24 carbon signals including a carbonyl (δ_C 172.7, C-19) and six olefinic carbons at δ_C 160.5 (C, C-2), 140.8 (C, C-6), 133.4 (C, C-1), 130.7 (C, C-14), 126.1 (CH, C-13), and 121.6 (CH, C-7) (Table 2), which accounted for four of the six degrees of unsaturation. Thus, 1 must be bicyclic. A γ-ethoxy-α,β-unsaturated-γ-lactam moiety was easily recognized by COSY correlations between an NH (δ_H 7.63, br s) and a heteroacetal proton H-18 (δ_H 5.41, d, J = 1.2 Hz) and between an oxygenated methylene (δ_H 3.76, m; 3.61, m) and a methyl triplet (δ_H 1.22, t, J = 7.2 Hz), as well as HMBC correlations from NH and H-18 to the two olefinic nonprotonated carbons C-1 and C-2, from H-18 to the carbonyl carbon C-19, and from the oxygenated methylene protons to a methine carbon C-18 (δ_C 85.6, CH) (Figure 1). COSY correlations of H-3 (δ_H 2.65, t, J = 9.0 Hz)/H₂-4 (δ_H 2.10, m; 1.95, m), H₂-4/H₂-5 (δ_H 2.19, m; 1.96, m), H-7 (δ_H 5.36, dd, J = 12.6, 4.2 Hz)/H₂-8 (δ_H 2.45, dt, J = 13.2, 4.2 Hz; 1.86, td, J = 13.2, 2.4 Hz), and H₂-8/H-9 (δ_H 4.64, dd, J = 4.2, 2.4 Hz) defined two proton spin systems. HMBC correlations from the olefinic methyl H₃-20 to an aliphatic methylene carbon C-5 (δ_C 41.4, CH₂) and two olefinic carbons C-6 and C-7 and from H-3 and H-9 to two olefinic carbons C-1 and C-2 allowed the construction of a nine-membered ring, in which C-1 and C-2 were also constituents of the α,β-unsaturated-γ-lactam, C-6/C-7 composed a double bond, C-6 was substituted by a methyl group, and C-9 (δ_C 77.4, CH) was oxygenated. In addition, the fusion of the ring system was revealed by HMBC correlations from H-9 to the carbonyl carbon C-19 and from H-3 to the methine carbon C-18. Moreover, a 6-methylhept-5-en-2-yl side chain was established by COSY correlations of H-13 (δ_H 5.08, t, J = 7.2 Hz)/H₂-12 (δ_H 2.04, m; 1.83, m), H₂-12/H₂-11 (δ_H 1.65, m; 0.86, m), H₂-11/H-10 (δ_H 1.84, m), and H-10/H₃-17, as well as HMBC correlations from two olefinic methyls H₃-15 and H₃-16 to two olefinic carbons C-13 and C-14 and from H₃-17 to an aliphatic methine carbon C-10 (δ_C 35.5, CH) and an aliphatic methylene carbon C-11 (δ_C 36.8, CH₂). All these NMR data are characteristic of a xenicane-type diterpene represented by the co-occurring analogues 18-hydroxydictyolactone (9)¹⁹ and 4-hydroxydictyolactone (10).²⁰ The side chain was connected to the nine-membered ring via C-3 by COSY correlation between H-3 and H-10 and HMBC correlation from H₃-17 to C-3, while the second ethoxy group was attached to C-9 according to COSY correlations between the oxygenated methylene protons (δ_H 3.49, m; 3.43, m) and the methyl triplet at δ_H 1.15 (t, J = 7.2 Hz) and HMBC correlations from the oxygenated methylene protons to the oxygenated methine carbon C-9. Thus, the planar structure of 1 was elucidated as 9,18-diethoxyxenica-1,6,13-trien-19,18-lactam.

Table 1. ¹H NMR Spectroscopic Data for 1–5 (Acetone-d₆)^a.

no.	1b	2c	3c	4c	5c
2					2.72, dd (9.5, 6.0)
3	2.65, t (9.0)	2.36, t (9.5)	2.78, d (10.5)	2.14, d (10.0)	1.72, d (9.5)
4	2.10, m	2.33, m	4.67, dd (3.0, 1.5)	4.55, br s	1.69, m
	1.95, m	1.90, m			1.56, m
5	2.19, m	2.13, m	2.32, dd (13.0, 3.5)	2.35, m	2.23, br d (12.0)
	1.96, m	1.96, m	2.15, br d (13.0)	2.17, m	1.95, m
7	5.36, dd (12.6, 4.2)	5.40, dd (12.5, 3.5)	5.36, br d (10.5)	5.27, dd (10.0, 4.5)	5.44, br d (12.0)
8	2.45, dt (13.2, 4.2)	2.49, dt (13.5, 4.0)	2.50, dt (13.0, 3.0)	2.69, m	3.04, t (13.5)
	1.86, td (13.2, 2.4)	2.12, m	2.20, td (13.0, 3.0)	2.32, m	2.57, dd (13.5, 8.5)
9	4.64, dd (4.2, 2.4)	4.37, dd (4.0, 2.5)	4.54, br t (2.5)	2.30, m	5.77, br d (9.0)
				2.27, m
10	1.84, m	2.02, m	1.78, m	2.16, m	1.83, m
11	1.65, m	1.35, m	1.52, m	1.38, m	1.26, dd (15.0, 9.0)
	0.86, m	0.91, m	1.17, m	1.10, m
12	2.04, m	2.04, m	2.11, m	1.93, m	2.00, m
	1.83, m	1.81, m	1.96, m	1.87, m
13	5.08, t (7.2)	5.03, t (7.0)	5.08, t (6.5)	5.06, t (7.0)	5.13, t (7.0)
15	1.57, s	1.55, s	1.58, s	1.56, s	1.62, s
16	1.62, s	1.61, s	1.65, s	1.64, s	1.68, s
17	0.98, d (6.0)	0.96, d (6.5)	1.05, d (6.5)	0.99, d (6.5)	0.86, d (6.5)
18	5.41, d (1.2)		5.63, br s		3.80, dd (10.0, 6.0)
					3.61, t (10.0)
19		5.07, d (2.0)		5.93, s	4.58, dt (13.0, 1.5)
					4.53, dt (13.0, 1.5)
20	1.24, s	1.26, s	1.56, s	1.55, s	1.75, s
9-OEt	3.49, m	3.53, m	3.51, m
	3.43, m	3.45, m	1.17, t (7.0)
	1.15, t (7.2)	1.15, t (7.0)
18/19-OEt	3.76, m	3.52, m
	3.61, m	3.50, m
	1.22, t (7.2)	1.16, t (7.0)
OAc				2.04, s	2.04, s
NH/OH	7.63, br s	7.43, br s	7.39, br s	3.72, br s

^aThe coupling constants (J) are in parentheses and reported in Hz; chemical shifts are given in ppm.

^b600 MHz.

^c500 MHz.

Table 2. ¹³C NMR Data for 1–5 (Acetone-d₆)^a.

no.	1b	2c	3c	4c	5c
1	133.4, C	150.9, C	132.8, C	173.2, C	144.1, C
2	160.5, C	142.6, C	162.6, C	160.5, C	47.8, CH
3	47.1, CH	44.9, CH	50.5, CH	49.0, CH	48.4, CH
4	35.8, CH2	34.3, CH2	94.5, CH	78.2, CH	28.9, CH2
5	41.4, CH2	41.7, CH2	45.6, CH2	49.1, CH2	42.1, CH2
6	140.8, C	141.6, C	140.5, C	140.1, C	135.6, C
7	121.6, CH	120.5, CH	121.7, CH	125.8, CH	126.6, CH
8	30.7, CH2	30.7, CH2	32.5, CH2	27.9, CH2	28.4, CH2
9	77.4, CH	78.0, CH	78.5, CH	24.4, CH2	129.0, CH
10	35.5, CH	35.2, CH	35.5, CH	32.3, CH	33.8, CH
11	36.8, CH2	36.5, CH2	36.4, CH2	35.9, CH2	39.2, CH2
12	27.2, CH2	26.7, CH2	26.5, CH2	25.5, CH2	26.9, CH2
13	126.1, CH	125.9, CH	125.3, CH	125.5, CH	125.8, CH
14	130.7, C	131.1, C	131.9, C	131.6, C	131.6, C
15	17.7, CH3	17.7, CH3	17.7, CH3	17.7, CH3	17.8, CH3
16	25.8, CH3	25.8, CH3	25.8, CH3	25.9, CH3	25.8, CH3
17	18.7, CH3	18.7, CH3	18.5, CH3	18.1, CH3	16.9, CH3
18	85.6, CH	173.0, C	86.8, CH	186.6, C	60.7, CH2
19	172.7, C	84.3, CH	174.8, C	83.9, CH	66.7, CH2
20	15.8, CH3	15.9, CH3	20.1, CH3	19.4, CH3	17.5, CH3
9-OEt	65.2, CH2	65.1, CH2	65.4, CH2
	15.8, CH3	15.8, CH3	15.7, CH3
18/19-OEt	65.8, CH2	61.1, CH2
	15.8, CH3	15.8, CH3
OAc				170.6, C	170.7, C
				20.7, CH3	21.0, CH3

^aThe assignments were based on HSQC, HMBC, and COSY spectra.

^b150 MHz.

^c125 MHz.

Figure 1.

Key COSY and HMBC correlations for 1–5.

The relative configuration of 1 was determined by coupling constant and NOE analysis. The coupling pattern of H-3 (δ_H 2.65, t, J = 9.0 Hz) indicated the anti orientations of H-3/H-4b (δ_H 1.95, m) and H-3/H-10 (δ_H 1.84, m). The E geometry of the C-6/C-7 double bond was suggested by the diagnostic chemical shift of C-20 (<20 ppm)^25,26 and confirmed by NOE correlations of H-7/H-5b (δ_H 1.96, m) and H₃-20/H-8b (δ_H 1.86, td, J = 13.2, 2.4 Hz) (Figure 2). NOE correlations of H-7/H-3, H-3/H₃-17, and H-4a (δ_H 2.10, m)/H₃-17 suggested the 3S*,10R* assignment, while the small coupling constant values of J_H-9/H-8a (4.2 Hz) and J_H-9/H-8b (2.4 Hz) indicated the gauche orientations of H-9/H-8a (δ_H 2.45, dt, J = 13.2, 4.2 Hz) and H-9/H-8b, requiring the 9R* configuration. The relative configuration of C-18 cannot be determined by NOE correlations. The absolute configuration at C-18 was established by the application of the electronic circular dichroism (ECD) helicity rule originally used for α,β-unsaturated-γ-lactones.²⁷ The negative Cotton effect at 217 nm (π–π* transition) in the ECD spectrum was consistent with the anticlockwise screw. Therefore, the absolute configuration at C-18 was assigned as R. In addition, the absolute configurations of 3S,10R were determined to be in agreement with those of 4-hydroxydictyolactone (10)²⁰ by biogenetic considerations. This assignment was further supported by the calculated ECD data of 1 at the B3LYP/Def2-TZVP level (Figure S36), which showed a similar ECD curve as that of the experimental ECD spectrum. Compound 1 was given the name dictyolactam A. The ethyl ether is possibly an artifact because EtOH was used during the extraction process; however no evidence of the nonetherified form of 1 was observed during the isolation procedure. In either case, the free hydroxy or ethyl ether form of 1 is the second natural nitrogen-containing xenicane diterpene.²⁸

Figure 2.

Key NOE correlations for 1–5.

Compound 2 had the same molecular formula as that of 1 according to its HRESIMS data. It was revealed to be an isomer of 1 with respect to the α,β-unsaturated-γ-lactam ring by extensive 2D NMR spectroscopic analysis and comparison of the NMR data with those of 1 (Tables 1 and 2). HMBC correlations from H-3 (δ_H 2.36, t, J = 9.5 Hz) to a carbonyl carbon C-18 (δ_C 173.0, C) and from the heteroacetal proton H-19 (δ_H 5.07, d, J = 2.0 Hz) to an oxygenated methine carbon C-9 (δ_C 78.0, CH), the carbonyl carbon C-18, and two olefinic nonprotonated carbons C-1 (δ_C 150.9, C) and C-2 (δ_C 142.6, C) disclosed that the fusion of the α,β-unsaturated-γ-lactam ring was reversed in 2 (Figure 1). The relative configurations at C-3, C-9, and C-10 and the geometry of the C-6/C-7 double bond were suggested to be the same as those of 1 based on the similar coupling constants and NOE relationships (Figure 2). The NOE correlation between H-19 and H-8b (δ_H 2.12, m) indicated the 19R* configuration. The experimental ECD spectrum of 2 was similar to that of 1 (Figure S36), implying the same absolute configuration at the heteroacetal carbons. This was supported by the overall match of the experimental and calculated ECD curves for 2. It was given the name dictyolactam B.

The NMR data of 3 were comparable to those of joalin,²⁸ the first nitrogen-containing xenicane from the genus Dictyota, with the exception of additional signals for an ethoxy [δ_H 3.51 (m), 1.17 (t, J = 7.0 Hz); δ_C 65.4 (CH₂), 15.7 (CH₃)] and loss of signals for a methoxy group (Tables 1 and 2), indicating that 3 was an ethoxylated 9-demethoxyjoalin. The ethoxy group was attached to C-9 (δ_C 78.5, CH) by HMBC correlations from the oxygenated methylene protons to an oxygenated methine carbon C-9 (Figure 1). The relative configuration of 3 was suggested to be in agreement with that of joalin²⁸ by NOE correlations of H-3 (δ_H 2.78, d, J = 10.5 Hz)/H-7 (δ_H 5.36, br d, J = 10.5 Hz), H-7/H-5b (δ_H 2.15, br d, J = 13.0 Hz), H-4 (δ_H 4.67, dd, J = 3.0, 1.5 Hz)/H-18 (δ_H 5.63, br s), H-18/H-10 (δ_H 1.78, m), and H-4/H₃-17 (δ_H 1.05, d, J = 6.5 Hz) (Figure 2) and similar coupling constant values J_H-3/H-10 (10.5 Hz, 3; 10.3 Hz, joalin), J_H-4/H-5 (3.5, 1.5 Hz, 3; 3.6, 2.2 Hz, joalin), and J_H-8/H-9 (2.5 Hz, 3; 3.1 Hz, joalin). The absolute configuration 3S,4R,9R,10R,18R was assigned based on the high similarity between the measured and calculated ECD spectra for 3 (Figure S36). Thus, compound 3 was established as 9-demethoxy-9-ethoxyjoalin.

The molecular formula of compound 4 was determined to be C₂₂H₃₂O₄ by its HRESIMS data. The ¹H and ¹³C NMR data of 4 closely paralleled those of previously reported xenicane diterpenes with a cyclobutanone moiety, acetylcoriacenone and isoacetylcoriacenone,²⁹ except for the loss of signals for an aliphatic methylene and the appearance of additional signals for a secondary alcohol [δ_H 4.55 (br s, H-4), 3.72 (br s, OH); δ_C 78.2 (CH, C-4)] (Tables 1 and 2), indicating that 4 was a hydroxylated derivative of acetylcoriacenone or isoacetylcoriacenone. This assumption was supported by the additional oxygen relative to the molecular formula of these two xenicanes. The location of the secondary alcohol was assigned at C-4 based on the HMBC correlation from the 4-OH proton to an aliphatic methine carbon C-3 (δ_C 49.0, CH), along with COSY correlations between the oxygenated methine proton H-4/H-3 (δ_H 2.14, d, J = 10.0 Hz) and H-4/H₂-5 (δ_H 2.35, m; 2.17, m) (Figure 1). The configurations at the C-3, C-10, and C-6/C-7 double bonds (3S*,10R*,6E) were assigned as in agreement with those of acetylcoriacenone and isoacetylcoriacenone²⁹ based on biogenetic considerations and the similar NMR data, including the large coupling constant value J_H-3/H-10 (10.0 Hz, indicating the anti orientations of H-3/H-10) and NOE correlations of H-3/H-7 (δ_H 5.27, dd, J = 10.0, 4.5 Hz) and H-7/H-5b (δ_H 2.17, m) (Figure 2). The broad singlet observed for H-4 (weak vicinal coupling between H-3 and H-4) suggested an H3–C3–C4–H4 dihedral angle value of approximately 90°, requiring the 4R* configuration. This was supported by the NOE correlation between H-4 and H₃-17 (δ_H 0.99, d, J = 6.5 Hz). In addition, the NOE correlations between H-19 (δ_H 5.93, s) and H₃-20 (δ_H 1.55, s) revealed the 19R* configuration. The absolute configuration at C-19 was determined to be the same as that of isoaceylcoriacenone²⁹ by the similar measured ECD curves and confirmed by the calculated ECD curve for 19R-enantiomer of 4 (Figure S36). Therefore, 4 was defined as 4-hydroxyisoacetylcoriacenone.

The ¹H and ¹³C NMR data recorded for 5 were comparable to those of dictyodiol,³⁰ except for the appearance of additional signals attributed to an acetyl group [δ_H 2.04 (s); δ_C 170.7 (C), 21.0 (CH₃)] (Tables 1 and 2), indicating that 5 was an acetylated derivative of dictyodiol. The acetoxy group was located at C-19 (δ_C 66.7, CH₂), as supported by HMBC correlations from the oxygenated methylene H₂-19 (δ_H 4.58, dt, J = 13.0, 1.5 Hz; 4.53, dt, J = 13.0, 1.5 Hz) to the carbonyl carbon, two olefinic carbons C-1 (δ_C 144.1, C) and C-9 (δ_C 129.0, CH), and an aliphatic methine carbon C-2 (δ_C 47.8, CH) (Figure 1). The relative configuration of 5 was suggested to be the same as that of dictyodiol³⁰ based on the similar weak vicinal couplings between H-2 (δ_H 2.72, dd, J = 9.5, 6.0 Hz)/H-3 (δ_H 1.72, d, J = 9.5 Hz) and H-3/H-10 (δ_H 1.83, m) (indicating the H2–C2–C3–H3 and H3–C3–C10–H10 dihedral angle values of approximately 90°), in combination with significant NOE correlations of H-2/H₃-20 (δ_H 1.75, s), H-3/H-7 (δ_H 5.44, br d, J = 12.0 Hz), H₂-18 (δ_H 3.80, dd, J = 10.0, 6.0 Hz; 3.61, t, J = 10.0 Hz)/H₃-17 (δ_H 0.86, d, J = 6.5 Hz), and H₂-19/H-10 (Figure 2). Thus, compound 5 was determined as 19-O-acetyldictyodiol.

All of the xenicane diterpenes 1–20 were evaluated for their cytoprotective effects against hydrogen peroxide (H₂O₂)-induced oxidative damage in neuron-like pheochromocytoma PC12 cells by the MTT method.^31,32 As shown in Figure 3, the cell survival rate of PC12 cells treated with H₂O₂ was around 57% compared with the control group, and it was increased to 77%–84% by treatments with 2 μM 1–11, 13,15, 16, and 18–20, while 12, 14, and 17 showed weaker cytoprotection than the other analogues in this assay.

Figure 3.

Cytoprotection of 1–20 against H₂O₂-induced oxidative damage in PC12 cells. PC12 cells were pretreated with 2 μM concentrations of compounds for 6 h and then cocultured with 500 μM of H₂O₂ for 24 h. The cell viability was determined by the MTT assay. Data are expressed as the mean ± SD, n = 3. ^####P < 0.0001 vs control; ***P < 0.001, **P < 0.01 vs H₂O₂.

Compound 6, a major active component of this alga, probably had higher water solubility comparing with the other analogues due to its highly oxygenated nature. It was thus selected for further studies to investigate the antioxidant properties of xenicane diterpenes. It showed no cytotoxic activity toward PC12 cells at the concentration of 5 μM in the MTT assay (Figure S68). Thus, the maximum concentration of 6 was set as 5 μM in the following antioxidant evaluations in PC12 cells. As shown in Figure 4A, the viability of PC12 cells injured by H₂O₂ was significantly increased by 6 (2.5 and 5 μM), and the cytoprotection of 6 (2.5 μM) was comparable to that of the positive control tert-butylhydroquinone (TBHQ, 10 μM). The cytoprotection of 6 was confirmed by a colony formation assay.^32,33 Compound 6 (5 μM) significantly promoted colony formation of PC12 cells treated by H₂O₂ with the effect comparable to that of the positive control TBHQ (10 μM) (Figure 4B). In addition, 6 (2.5 and 5 μM) could significantly reduce the lactate dehydrogenase (LDH) level (which correlates with the number of dead cells³⁴) released by PC12 cells treated with H₂O₂ (Figure 4C), indicating a potent cytoprotective effect against oxidative damage in PC12 cells.

Figure 4.

Protective effect of 6 against H₂O₂-induced oxidative damage in PC12 cells. (A) 6 increased cell survival rate in a H₂O₂ damage model. (B) Pretreatment with 5 μM 6 promoted colony formation. (C, D) 6 decreased the levels of LDH and MDA, respectively. (E–G) 6 increased the levels of GSH, GSH-Px, and SOD, respectively. (H, I) 6 inhibited the excessive production of intracellular ROS stimulated by H₂O₂ in PC12 cells. ROS levels were detected using flow cytometry. PC12 cells were preincubated with 6 for 6 h prior to another 24 h exposure to 500 μM H₂O₂. TBHQ (10 μM) was used as a positive control. Data are expressed as the mean ± SD, n = 3. ^###P < 0.001 vs control; ***P < 0.001, **P < 0.01, *P < 0.05 vs H₂O₂.

The antioxidant property of 6 was further evaluated using the malondialdehyde (MDA), glutathione (GSH), glutathione peroxide (GSH-Px), and superoxide dismutase (SOD) assay kits in the H₂O₂-induced PC12 cell damage model. As shown in Figure 4D–G, 6 h of pretreatment with 6 significantly inhibited the MDA production and upregulated the levels of GSH, GSH-Px, and SOD in a dose-dependent manner. In addition, pretreatment with 6 significantly reduced the intracellular ROS production induced by H₂O₂ based on the dichlorodihydrofluorescein diacetate (DCFH-DA) assay³² (Figure 4H, I). Overall, 6 has potent antioxidant and cytoprotective activities in PC12 cells.

Compound 6 was tested for the free radical scavenging capacity using a DPPH (2,2-diphenyl-1-picrylhydrazyl) assay.³ It showed no scavenging effect even at a high concentration of 20 μM (Figure S69), indicating an indirect antioxidant that cannot remove excessive ROS via chemical ways. Compound 6 was also detected for the activation of the Nrf2/ARE, a key signaling pathway in the antioxidant defense response.² As shown in Figure 5A–C, 6 upregulated the levels of Nrf2 and heme oxygenase-1 (HO-1), an antioxidant protein regulated by Nrf2,³⁵ in a dose-dependent manner based on a Western-blot assay. In addition, 6 significantly promoted the nuclear translocation of Nrf2 as indicated by the strong green fluorescence observed in the nucleus in an immunofluorescence assay,^32,33 and the promotion effect of 5 μM 6 was comparable to that of the positive control TBHQ (Figure 5D). These findings confirmed that 6 is a potent activator of the Nrf2/ARE signaling in PC12 cells. Moreover, Nrf2 siRNA was applied to investigate if the antioxidant activity of 6 is dependent on Nrf2. Cell viability in the control siRNA group (si-NC) was increased by 6, while the cytoprotection was reversed by the knockdown of Nrf2 (Figure 5E), which suggested that Nrf2 contributes to the antioxidant effect of 6. This conclusion was further supported by the fact that cytoprotection of 6 was suppressed by the maximum noncytotoxic concentration of the HO-1 inhibitor zinc protoporphyrin (ZnPP) (Figure 5F). Thus, compound 6 exerted an antioxidant effect in PC12 cells involved activation of the Nrf2/ARE signaling and upregulated expression of HO-1.

Figure 5.

Compound 6 exerts antioxidant effect in PC12 cells by activating the Nrf2/ARE pathway. (A–C) 6 increased the expressions of Nrf2 and HO-1 proteins. The Nrf2 and HO-1 protein levels were determined by Western blotting. (D) 6 induced Nrf2 nuclear translocation in PC12 cells. Nrf2 was marked using specific antibody (green) immunofluorescence. Nuclei were labeled using DAPI (blue). Typical photographs of merge and enlarge are shown. Scale bars represent 50 μm, the magnification of merge is 400×. (E) Down-regulation of Nrf2 expression with siRNA decreased the cytoprotective effect of 6. Cells were transfected with 100 nM control siRNA (si-NC) or Nrf2 siRNA (si-Nrf2). The survival rates were determined by MTT assay. (F) Pretreatment with HO-1 inhibitor ZnPP (0.12 μM) suppressed the cytoprotective effect of 6. PC12 cells were preincubated with 6 for 6 h prior to another 24 h of exposure to 500 μM H₂O₂. TBHQ (10 μM) was used as a positive control. Data are expressed as the mean ± SD, n = 3. ^###P < 0.001, ^##P < 0.01 vs control; ***P < 0.001, **P < 0.01 vs H₂O₂; ^&&&P < 0.001 vs 6+H₂O₂.

The oxygen-glucose deprivation and reperfusion (OGD/R)-induced oxidative stress is the main pathological process of cerebral ischemia-reperfusion injury (CIRI).³ Compound 6 was further evaluated for its neuroprotective effect against CIRI in vitro using the OGD/R-induced PC12 cell damage model. As shown in Figure S70, pretreatment with 5 μM 6 significantly reduced cell damage and the productions of LDH, MDA, and ROS. Its inhibitory activity against OGD/R-induced oxidative stress in PC12 cells is also related to the Nrf2 pathway based on the fact that the cytoprotection was reversed by the knockdown of Nrf2 (Figure S70).

Finally, the neuroprotective effect of 6 against CIRI was evaluated using a rat model of transient middle artery occlusion (MCAO).^36,37 Compound 6 was preinjected into lateral ventricles 3 h after the ischemia-reperfusion injury. As shown in Figure 6, treatment with 6 (80 μg/kg) significantly reduced the size of brain infarction and neurological deficit score, implying potent neuroprotection against CIRI.

Figure 6.

Protective effect of 6 against CIRI in a rat model of MCAO. (A) Representative images of triphenyl tetrazolium chloride (TTC)-stained brain slices of sham-operated (sham) rats and MCAO rats treated with normal saline (NS), vehicle (DMSO/NS, 1:100), and 80 μg/kg 6. (B) Quantitative analysis of infarct size. (C) Neurological deficit score. Data are expressed as the mean ± SD, n = 6. ^####P < 0.0001 vs sham-operated group; ***P < 0.001, **P < 0.01 vs vehicle-treated group.

Conclusions

In summary, a further phytochemical investigation on the brown alga D. coriacea afforded five new xenicane diterpenes including three rare nitrogen-containing derivatives as the EtOH-adduct artifacts (1–3) and a rare xenicane with cyclobutanone moiety (4). All of the xenicanes (1–20) showed cytoprotective properties against H₂O₂-induced oxidative stress in PC12 cells. Compound 6 also had inhibitory activity against OGD/R-induced oxidative damage in PC12 cells. The antioxidant mechanism of 6 was suggested to be the activation of Nrf2/ARE signaling pathway. It is interesting to note that 6 showed significant neuroprotective effect against CIRI in a rat model of MCAO. This is the first report of neuroprotective properties of xenicane diterpenes against oxidative stress, providing evidence that xenicane might be a valuable lead scaffold for neuroprotective agents against CIRI.

Experimental Section

General Experimental Procedures

Optical rotations were measured with a PoLAAR 3005 digital polarimeter. UV and IR spectra were recorded with a TU 1901 spectrometer and an FTIR-850 spectrometer, respectively. ECD spectra were obtained with a Chirascan circular dichroism spectrometer. NMR spectra were acquired with Bruker Avance III NMR spectrometers running at 500 or 600 MHz for ¹H and 125 or 150 MHz for ¹³C. Tetramethylsilane was used as an internal standard. (+)-HRESIMS data were recorded on a Thermo Scientific Q Exactive hybrid quadrupole-Orbitrap mass spectrometer. Column chromatography was performed with silica gel (200–300 mesh, Qingdao Marine Chemistry Co. Ltd.), ODS (50 μm, YMC), and Sephadex LH-20 (GE Healthcare Biosciences AB). High-performance liquid chromatography (HPLC) was performed using an Agilent 1100 series instrument equipped with a VWD G1314A detector and a YMC-Pack C₁₈ column (10 μm, 250 × 10 mm).

Plant Material

Specimens of the brown alga Dictyota coriacea were collected off the coast of Nanji Island, Wenzhou, Zhejiang Province, China, in May 2018. The identification was carried out by one of the authors (Z.S.). A voucher specimen of the alga (ZN201801) was deposited at the Laboratory of Marine Natural Products Chemistry, Wenzhou Medical University, China.

Extraction and Isolation

The air-dried alga material (1.64 kg) was extracted with 95% EtOH at room temperature. The concentrated extract was partitioned between H₂O and EtOAc. Evaporation of EtOAc in vacuo afforded a dark residue of 130.5 g. The EtOAc fraction (100.2 g) was separated by silica gel vacuum column chromatography, eluting with a gradient of EtOAc/petroleum ether (1:15, 1:10, 1:4, and 1:2), to obtain five fractions (A–E). A portion of fraction C (2.0 g) was separated on a Sephadex LH-20 column, eluting with CH₂Cl₂/MeOH (1:1), to obtain three fractions (C1–C3). Fraction C2 (948.8 mg) was subjected to an ODS column, eluting with a stepwise gradient of MeOH/H₂O (65:35, 70:30, 75:25, 80:20, 85:15, and 90:10), to afford six fractions (C2a–C 2f). Fractions C2b (19.1 mg) was purified by C₁₈ HPLC, using MeCN/H₂O (70:30) as eluent, to yield 4 (7.9 mg). Fraction C2c (86.7 mg) was purified by HPLC (MeCN/H₂O, 78:22) to obtain 1 (1.5 mg), 2 (4.5 mg), and 5 (1.6 mg). Fraction C 2d (154.2 mg) was separated by HPLC (MeCN/H₂O, 75:25) to yield 6 (80.8 mg) and 7 (28.4 mg). Fraction C2e (133.1 mg) was purified by HPLC (MeCN/H₂O, 80:20) to obtain 13 (6.7 mg). A portion of fraction D (3.1 g) was subjected to a Sephadex LH-20 column, eluting with CH₂Cl₂/MeOH (1:1), to afford three fractions (D1–D3). Fraction D2 (2.0 g) was further separated on an ODS column, eluting with a stepwise gradient of MeOH/H₂O (60:40, 65:35, 70:30, 75:25, 80:20, 85:15, and 90:10), to obtain 11 fractions (D2a–D2k). Fraction D2b (18.5 mg) was purified by HPLC (MeCN/H₂O, 60:40) to afford 20 (4.2 mg). Fraction D2c (349.5 mg) was separated by HPLC (MeCN/H₂O, 62:38) to yield 8 (16.0 mg), 10 (76.1 mg), and 15 (28.5 mg). Fraction D2d (432.2 mg) was purified by HPLC (MeCN/H₂O, 65:35) to obtain 14 (41.5 mg), 16 (20.0 mg), 17 (61.5 mg), and 19 (10.1 mg). Fractions D2e (29.0 mg) and D2f (134.3 mg) were purified by HPLC (MeCN/H₂O, 70:30) to yield 9 (4.7 mg) and 11 (20.4 mg), respectively. Compounds 3 (1.4 mg), 12 (2.5 mg), and 18 (2.4 mg) were separated from fraction D2i (56.1 mg) by HPLC using MeCN/H₂O (75:25) as eluent.

Dictyolactam A (1)

Colorless oil; [α]²⁵D, −124 (c 0.05, MeOH); UV (MeOH) λ_max (log ε) 221 (3.85) nm; ECD (MeOH) λ_max (Δε) 217 (−59.26) nm; IR (KBr) ν_max 3273, 2972, 2920, 2861, 1697, 1451, 1383, 1088 cm^–1; ¹H and ¹³C NMR data, Tables 1 and 2; HRESIMS m/z 412.2825 [M + Na]⁺ (calcd for C₂₄H₃₉NO₃Na, 412.2822).

Dictyolactam B (2)

Colorless oil; [α]²⁵D −54 (c 0.05, MeOH); UV (MeOH) λ_max (log ε) 203 (4.18) nm; ECD (MeOH) λ_max (Δε) 212 (−27.08) nm; IR (KBr) ν_max 3278, 2968, 2926, 2858, 1696, 1446, 1377, 1088 cm^–1; ¹H and ¹³C NMR data, Tables 1 and 2; HRESIMS m/z 412.2804 [M + Na]⁺ (calcd for C₂₄H₃₉NO₃Na, 412.2822).

9-Demethoxy-9-ethoxyjoalin (3)

Colorless oil; [α]²⁵D +12 (c 0.05, MeOH); UV (MeOH) λ_max (log ε) 203 (4.07) nm; ECD (MeOH) λ_max (Δε) 254 (+17.94), 208 (−62.29) nm; IR (KBr) 3346, 2970, 2929, 1708, 1446, 1379, 1267, 1090, 982, 735, 636 cm^–1; ¹H and ¹³C NMR data, Tables 1 and 2; HRESIMS m/z 382.2339 [M + Na]⁺ (calcd for C₂₂H₃₃NO₃Na, 382.2353).

4-Hydroxyisoacetylcoriacenone (4)

Colorless oil; [α]²⁵D −78 (c 0.05, MeOH); IR (KBr) ν_max 3514, 2926, 2860, 1758, 1664, 1606, 1446, 1375, 1227, 1020 cm^–1; UV (MeOH) λ_max (log ε) 230 (3.80) nm; ECD (MeOH) λ_max (Δε) 315 (−6.99), 238 (+52.50), 210 (−47.37) nm; ¹H and ¹³C NMR data, Tables 1 and 2; HRESIMS m/z 383.2175 [M + Na]⁺ (calcd for C₂₂H₃₂O₄Na, 383.2193).

19-O-Acetyldictyodiol (5)

Colorless oil; [α]²⁵D +56 (c 0.05, MeOH); IR (KBr) ν_max 3458, 2925, 2854, 1734, 1664, 1587, 1446, 1377, 1234, 1026 cm^–1; ¹H and ¹³C NMR data, Tables 1 and 2; HRESIMS m/z 371.2541 [M + Na]⁺ (calcd for C₂₂H₃₆O₃Na, 371.2557).

ECD Calculation

All density functional theory (DFT) calculations were performed using Gaussian 16 (G16).³⁸ Conformers of molecules 1–4 were obtained using the GMMX package implemented in GaussView 6.0 with the MMFF94 force field. A total of 114, 168, 268, and 102 conformers were yielded for 1–4, respectively, with an energy window of 3.5 kcal/mol. The conformational energy plots for 1–4 were shown in Figures S71–S74. In addition, 11 lower-energy conformers of 1–4 were further optimized at different levels of theory. The semiempirical calculation PM6 in the G16 was used for structure optimization in MeOH with the polarizable continuum model (PCM). Conformers of 1–4 were fully optimized at the CAM-B3LYP/6-311G(2d,p) level. The vibrational frequencies analysis showed no imaginary frequency, and the locally stable structure was obtained. Then the ECD calculations were carried out at the B3LYP/Def2-TZVP level using the time-dependent DFT theory (TD-DFT), in which 30 singlet excited states were investigated to obtain the ECD data of each conformer. Moreover, the electronic spectra of mixture in GaussView 6.0 was applied to analyze the Gibbs free energy and Boltzmann distribution for the 11 lower-energy conformers of 1–4 (Table S1). The dominant conformations (No. 7 for 1; 11 and 7 for 2; 1, 4, 10, and 3 for 3; 8, 2, and 1 for 4) were computed to generate the weighted ECD curves for 1–4.

Cell Culture, MTT Assay, Colony Formation Assay, Western Blot Analysis, and Transfection Assay

The previously reported protocols^32,33 were followed except that the treatment time and concentration of H₂O₂ were modified as indicated in the figure captions.

Detections of LDH, MDA, GSH, GSH-Px, and SOD

PC12 cells in logarithmic growth phase were cultured in six-well plates at a density of 5 × 10⁵ cells/mL and underwent various treatments. The supernatant was collected for LDH assay and the cells were collected for MDA, GSH, GSH-Px, and SOD assays following the manufacturer’s instructions for assay kits (Nanjing Jiancheng Bioengineering Institute).

DPPH Radical Scavenging Assay

The ethanol solutions of 6 (2.5, 5, 10, and 20 μM) were mixed with 0.15 mM DPPH in ethanol in a 96-well plate. The absorbances of the reaction mixtures were recorded at 517 nm using a microplate reader after the plate was stored at 25 °C in the dark for 30 min. TBHQ (20 μM) was used as a positive control.

Intracellular ROS Analysis

PC12 cells were cultured in six-well plates at a density of 5 × 10⁵ cells/mL. Each well was incubated with 1 mL DCFH-DA (the final concentration was 10 μM) for 20 min. Cells were washed three times with medium and followed by drug treatments. Subsequently, the ROS level was quantified using a flow cytometry (Becton, Dickinson and Company) and analyzed with FlowJo software.

Immunostaining of Nrf2

A previously reported protocol^32,33 was followed except that the secondary antibody was replaced by goat antirabbit IgG (ab150077, Abcam).

OGD/R-Treatment

The PC12 cells were cultured in DMEM for 24 h and washed with PBS for three times. The medium was then replaced by a glucose-free DMEM and the cells were transferred to an incubator with 1% O₂, 5% CO₂, and 94% N₂ at 37 °C. After 10 h, the medium was changed back to high-glucose DMEM. The cells were cultured in a normal incubator humidified with 5% CO₂ balanced with air at 37 °C for 24 h of recovery time. A control group was also run in which PC12 cells were cultured in DMEM and left untreated.

Experimental Animals

Adult male Sprague–Dawley (SD) rats (250–280 g) were obtained from the Beijing Vital River Laboratory Animal Technology Co., Ltd. The rats were raised under a 12/12 h dark/light cycle and had free access to adequate food and sterile water. The indoor temperature was controlled at around 24 °C and the relative humidity was set as 40%–50%. Animals were adapted for 1 week before starting the experiments. All animal procedures were approved by the Animal Welfare Committee of Wenzhou Medical University (Protocol No.:WYDW2019–0419).

CIRI Based on MCAO Model, Intracerebroventricular Injection, Neurological Deficit Score, and TTC Staining

The previously reported protocols^31,33 were followed except that the drugs were injected 3 h after reperfusion and the neurological deficit score and brain infarct size by TTC staining were evaluated 24 h after drug treatment.

Statistical Analysis

Experimental data were expressed as mean ± standard deviation (SEMs). SPSS 20.0 software was used for statistical analysis. The statistical significance was analyzed by using a Student’s t test or one-way ANOVA for multiple comparisons. P values less than 0.05 were considered as statistically significant.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c07891.

• 1D and 2D NMR and HRESIMS spectra of the new compounds 1–5, NMR spectra of the known compounds 6–20, ECD and calculated ECD spectra for 1–4, details for the ECD calculation, the cytotoxicity against PC12 cells, DPPH radical scavenging effect, and inhibition against OGD/R-induced oxidative damage of 6 (PDF)

Author Contributions

^▽ Y. Qi, G. Liu, and C. Fang made equal contributions to this work.

The authors declare no competing financial interest.