Chemical Diversity from a Chinese Marine Red Alga, Symphyocladia latiuscula

Abstract

This study describes an investigation into secondary metabolites that are produced by a marine red alga, Symphyocladia latiuscula, which was collected from coastal waters off Qingdao, China. A combination of normal, reversed phase, and gel chromatography was used to isolate six citric acid derived natural products, aconitates A–F (1–6), together with two known and ten new polybrominated phenols, symphyocladins C/D (7a/b), and symphyocladins H–Q (8a/b, 9a/b and 10–15), respectively. Structure elucidation was achieved by detailed spectroscopic (including X-ray crystallographic) analysis. We propose a plausible and convergent biosynthetic pathway involving a key quinone methide intermediate, linking aconitates and symphyocladins.

1. Introduction

Historically, natural products have inspired the development of many pharmaceuticals and agrochemicals, which, have in turn, played an important role in improving human and animal health and agricultural productivity, enhancing the quality of life for communities across the globe [1]. One of the defining characteristics of natural products is their structure diversity, which can encompass complex carbocyclic and heterocyclic scaffolds, annotated with a wide array of functional groups and stereochemical features. As such, even a limited set of biosynthetic precursors can deliver remarkable chemical diversity. Illustrative of this phenomenon are bromophenols from marine red algae (Rhodophyta) [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]. For example, the red alga Symphyocladia latiuscula (Harvey) Yamada has been reported to produce a diverse array of bromophenols elaborated by sulfoxides, sulphones, sulfates, glutamines, pyrrolidin-2-ones, ureas, diketopiperazines, and aconitic acids, with biological properties spanning antibacterial [11,12], antifungal [10,11,12,13,14], antiviral [15], anticancer [16], free radical scavenging [9,17,18], aldose reductase inhibitory [19], and Taq DNA polymerase inhibitory activities [20]. S. latiuscula bromophenols typically contain at least one 2,3,6-tribromo-4,5-dihydroxybenzyl moiety, as consistent with a highly conserved biosynthetic pathway. This report described our efforts to further elaborate the chemical diversity of S. latiuscula.

2. Results and Discussion

The EtOAc extract of a Chinese collection of S. latiuscula was concentrated in vacuo and subjected to a sequence of normal, reversed phase, and gel chromatography, with HPLC-MS analysis being used to prioritize fractions of interest. Following this strategy, we isolated and characterized six citric acid derived natural products, aconitates A–F (1–6), together with two known and ten new polybrominated phenol adducts, symphyocladins C/D (7a/b), and symphyocladins H–Q (8a/b, 9a/b and 10–15), respectively (Figure 1). A spectroscopic analysis approach (see

Table 1. ¹H NMR Data for Compounds 7a/b–9a/b (600 MHz).

Position	7a a	7b a	8a b	8b b	9a c	9b c
3	3.74, m	3.74, m			3.50, m	3.50, m
4a	3.17, m	3.17, m	3.65, s	3.19, s	2.98, dd e	2.976, dd e
4b	2.490, dd d	2.489, dd d			2.41, dd f	2.40, dd f
5-OCH3	3.55, s	3.55, s	3.71, s	3.51, s	3.52, s	3.52, s
6-OCH3					3.541, s	3.535, s
7′	7.544, s	7.538, s	4.22, s	4.34, s	7.391, s	7.388, s

^a acetone-d₆, ^b methanol-d₄, ^c DMSO-d₆, ^d J = 16.8, 7.8 Hz, ^e J = 16.8, 10.8 Hz, ^f J = 16.8, 3.0 Hz.

,

Table 2. ¹³C NMR Data for Compounds 7a/b–9a/b (150 MHz).

Position	7a a	7b a	8a b	8b b	9a c	9b c
1	167.18, C	167.18, C	170.3, C	166.5, C	166.8, C	166.8, C
2	135.46, C	135.34, C	145.0, C	144.4, C	133.5, C	133.5, C
3	41.39, CH	41.37, CH	127.4, C	137.8, C	40.2, CH	40.2, CH
4	35.40, CH2	35.35, CH2	35.0, CH2	29.5, CH2	33.9, CH2	33.9, CH2
5	172.80, C	172.76, C	168.7, C	166.8, C	171.5, C	171.5, C
6	172.54, C	172.49, C	172.4, C	169.0, C	171.5, C	171.5, C
5-OCH3	51.87, CH3	51.83, CH3	52.8, CH3	53.1, CH3	52.0, CH3	52.0, CH3
6-OCH3					51.6, CH3	51.6, CH3
1′	129.79, C	129.76, C	128.8, C	127.9, C	128.07, C	128.07, C
2′	115.38, C	115.23, C	118.9, C	118.4, C	113.9, C	113.9, C
3′	113.98, C	113.63, C	114.8, C	114.2, C	113.7, C	113.6, C
4′	144.17, C	144.10, C	146.2, C	145.5, C	143.9, C	143.8, C
5′	145.36, C	145.32, C	145.6, C	144.8, C	145.2, C	145.0, C
6′	110.71, C	110.67, C	114.7, C	114.3, C	110.7, C	110.6, C
7′	141.52, C	141.48, C	41.1, CH	35.9, CH	140.7, C	140.7, C

^a acetone-d₆, ^b methanol-d₄, ^c DMSO-d₆.

,

Table 3. ¹H NMR Data for Compounds 10–11 (600 MHz).

Position	10 a δH, m (J in Hz)	11 b δH, m (J in Hz)
2	4.98, dd (11.4, 3.0)	4.98, dd (11.4, 3.0)
4	6.76, s	6.73, s
1-OCH3	3.66, s	3.70, s
5-OCH3	3.44, s	3.45, s
6-OCH2CH3		4.26, br q (7.2)
6-OCH2CH3		1.31, t (7.2)
7′a	3.87, dd (14.4, 3.0)	3.81, dd (14.4, 3.0)
7′b	3.61, dd (14.4, 11.4)	3.56, dd (14.4, 11.4)

^a acetone-d₆, ^b methanol-d₄.

,

Table 4. ¹H NMR Data for Compounds 12–15 (600 MHz).

Position	12 a δH, m (J in Hz)	13 b δH, m (J in Hz)	14 b δH, m (J in Hz)	15 a δH, m (J in Hz)
2a	6.79, br t (6.6)	3.35, m	3.75, dd (9.0, 6.0)	3.79, dd (9.6, 5.4)
2b		3.27, dd (20.4, 7.8)
3		3.34, m
4a	3.60, br s	2.86, dd (17.4, 7.2)	6.22. d (1.2)	6.17, d (1.2)
4b		2.73, dd (17.4, 6.0)	5.44, br s	5.52, s
1-OCH3			3.64, s	3.62, s
5-OCH3	3.66, s	3.68, s
6-OCH3		3.71, s		3.71, s
7′a	4.05, s		3.65, dd (13.8, 6.0)	3.69, dd (14.4, 5.4)
7′b			3.54, dd (13.8, 9.0)	3.56, dd (14.4, 9.6)

^a acetone-d₆, ^b methanol-d₄.

and

Table 5. ¹³C NMR Data for Compounds 10–15 (150 MHz).

Position	10 a δC, Type	11 b δC, Type	12 a δC, Type	13 b δC, Type	14 b δC, Type	15 a δC, Type
1	171.9, C	173.5, C			174.5, C	172.4, C
2	43.0, C	43.7, C	141.3, C	44.9, CH2	48.8, CH	48.2, CH
3	142.4, C	142.4, C	128.1, C	37.4, CH	139.3, C	138.3, C
4	130.5, CH	131.3, CH	33.2, CH2	35.6, CH2	129.1, CH2	128.7, CH2
5	165.8, C	166.7, C	171.3, C	173.9, C
6	167.0, C	167.0, C	167.9, C	175.5, C	169.2, C	166.6, C
1-OCH3	52.3, CH3	52.9, CH3			52.8, CH3	52.3, CH3
5-OCH3	52.1, CH3	52.5, CH3	52.0, CH3	52.5, CH3
6-OCH3				52.8, CH3		52.3, CH3
6-OCH2CH3		63.2, C
6-OCH2CH3		14.5, CH3
1′	130.7, C	130.6, C	131.0, C	136.1, C	131.4, C	131.3, C
2′	118.5, C	118.8, C	117.3, C	114.5, C	118.3, C	118.0, C
3′	113.7, C	114.4, C	114.0, C	110.5, C	114.5, C	113.9, C
4′	144.1, C	145.1, C	144.4, C	147.3, C	145.0, C	144.0, C
5′	143.8, C	144.8, C	144.3, C	145.3, C	144.8, C	143.9, C
6′	114.3, C	114.8, C	113.0, C	106.3, C	114.3, C	113.8, C
7′	39.0, CH2	39.4, CH2	39.0, CH2	202.2, C	39.4, CH2	39.1, CH2

^a acetone-d₆, ^b methanol-d₄.

) to the structure elucidation of all of these metabolites is summarized below.

HRESI(+)MS measurements confirmed that 1 (C₇H₈O₆, Δmmu +0.2), 2 (C₇H₈O₆, Δmmu +0.1) and 3 (C₇H₈O₆, Δmmu +0.1) were isomeric, while analysis of the one-dimensional (1D) and two-dimensional (2D) NMR (methanol-d₄) data (Figures S8–S13, Tables S2 and S3) suggested they were mono methyl esters of E-aconitic acid, for which we attribute the trivial names aconitates A–C. Assignment of E ∆^3,4 configurations were inferred from diagnostic chemical shifts for H-4 and C-2 in 1 (δ_H 6.92; δ_C 33.8), 2 (δ_H 6.89; δ_C 33.9), and 3 (δ_H 6.91; δ_C 33.9), when compared to the authentic standards for E (δ_H 6.90; δ_C 33.8) and Z (δ_H 6.26; δ_C 40.2) aconitic acid (Figures S1–S7, Table S1). HMBC correlations permitted assignment of the methyl ester regiochemistry across 1–3 with correlations from (i) the OMe (δ_H 3.67) and H₂-2 (δ_H 3.89) to C-1 (δ_C 172.7) confirming a C-1 CO₂Me in aconitate A (1), (ii) the OMe (δ_H 3.81) and H₂-2 (δ_H 3.89) to C-6 (δ_C 168.4) confirming a C-6 CO₂Me in aconitate B (2), and (iii) the OMe (δ_H 3.77) and H-4 (δ_H 6.91) to C-5 (δ_C 167.5) confirming a C-5 CO₂Me in aconitate C (3) (Figure 2).

HRESI(+)MS measurements suggested that 4 (C₈H₁₀O₆, Δmmu +0.2) and 5 (C₈H₁₀O₆, Δmmu +0.2) were isomeric dimethyl esters, and 6 (C₉H₁₂O₆, Δmmu +0.1) was a trimethyl ester, of aconitic acid. Analysis of the NMR (methanol-d₄) data for 4–6 (Figures S14–S19, Tables S3 and S4) confirmed these assignments, with E ∆^3,4 configurations being inferred from diagnostic chemical shifts for H-4 and C-2 in aconitate D (4) (δ_H 6.91; δ_C 33.8), aconitate E (5) (δ_H 6.92; δ_C 33.9) and aconitate F (6) (δ_H 6.92; δ_C 33.8), and HMBC correlations permitting the assignment of the dimethyl ester regiochemistry across 4 and 5. For example, correlations from an OMe (δ_H 3.67) and H₂-2 (δ_H 3.91) to C-1 (δ_C 172.5), and from an OMe (δ_H 3.80) and H₂-2 to C-6 (δ_C 168.2), confirmed the presence of C-1 CO₂Me and C-6 CO₂Me moieties in 4, whereas correlations from an OMe (δ_H 3.67) and H₂-2 (δ_H 3.91) to C-1 (δ_C 172.5), and from an OMe (δ_H 3.76) and H-4 (δ_H 6.92) to C-5 (δ_C 167.5), confirmed C-1 CO₂Me and C-5 CO₂Me moieties in 5 (Figure 2).

HRESI(+)MS measurements confirmed that 7a/b (C₁₄H₁₁Br₃O₈, Δmmu +0.5) and 8a/b (C₁₄H₁₁Br₃O₈, ∆mmu +0.4) were isomeric, and suggested that 9a/b (C₁₅H₁₃Br₃O₈, Δmmu +0.4) and 10 (C₁₅H₁₃Br₃O₈, Δmmu +0.5) were CH₂ homologues, and 11 (C₁₇H₁₇Br₃O₈, Δmmu +0.6) was a CH₂CH₂ homologue of 7a/b and 8a/b. Analysis of the NMR (acetone-d₆) data for 7a/b (Figures S20 and S21, Table 1, Table 2 and Table S5) confirmed them as symphyocladins C/D, first reported in 2012 from S. latiuscula as an inseparable mixture of Z/E ∆^2,7′ isomers [13]. Analysis of the NMR (methanol-d₄) data for symphyocladins H/I (8a/b) (Figures S22–S27, Table 1, Table 2 and Table S6) revealed ∆^2,3 and C-5 CO₂Me moieties, as evidenced by diagnostic HMBC correlations (Figure 3). Significantly, these data also revealed an interconverting mixture of E/Z ∆^2,3 isomers, in which the minor Z isomer, symphyocladin H (8a), as evidenced by a ROESY correlation between H₂-4 and H₂-7′ (Figure 3), was in equilibrium with the major E isomer, symphyocladin I (8b). Further analysis of this NMR data revealed chemical shift differences diagnostic for ∆^2,3 geometric isomers; H₂-4 (E δ_H 3.65, δ_C 35.0; Z δ_H 3.19, δ_C 29.5), C-1 (E δ_C 170.3; Z δ_C 166.5), C-3 (E δ_C 127.4; Z δ_C 137.8), and C-5 CO₂Me (E δ_H 3.71; Z δ_H 3.51). Remarkably, the NMR (acetonitrile-d₃) data for 8a/b revealed a single Z isomer 8a, as evidenced by simplified spectra, a ROESY correlation between H₂-4 and H₂-7′, and diagnostic chemical shifts (Figures S28 and S29, Table S7). We speculate that in aprotic solvents (i.e., acetonitrile-d₃), hydrogen bonding between adjacent CO₂H moieties exclusively favors the lower energy Z ∆^2,3 isomer. By contrast, in protic solvents (i.e., methanol-d₄), the disruption of this hydrogen bonding favor equilibration to an E/Z ∆^2,3 mixture dominated by the less sterically constrained E isomer. This observation highlights the critical importance that NMR solvents can play in the analysis and structure elucidation of natural products.

Analysis of the NMR (DMSO-d₆) data for symphyocladins J/K (9a/b) (Figures S30 and S31, Table 1, Table 2 and Table S8) identified an inseparable mixture of C-6 CO₂Me homologues of 7a/b and 8a/b, as evidenced by spectroscopic comparisons and diagnostic HMBC correlations from the additional CO₂Me resonances to C-6 (Figure 3). In this instance, as hydrogen bonding does not stabilize double bond isomers, the Z/E mixture prevails even in an aprotic solvent (i.e., DMSO-d₆). Analysis of the NMR (acetone-d₆) data for symphyocladin L (10) (Figures S32 and S33, Table 3, Table 5 and Table S9) revealed a ∆^3,4 isomer and 1-CO₂Me homologue of 7a/b and 8a/b, as evidenced by COSY correlations between H-2 and H₂-7′, and diagnostic HMBC correlations positioning both C-1 CO₂Me and C-5 CO₂Me (Figure 3). The structure of 10 inclusive of an E ∆^3,4 configuration and its racemic nature were confirmed by single crystal X-ray analysis with the compound crystallizing in a centrosymmetric space group (Figures S34 and S35). Analysis of the NMR (methanol-d₄) data for symphyocladin M (11) (Figures S36 and S37, Table 3, Table 5 and Table S10) revealed it as a C-6 CO₂Et homologue of 10, as evidenced by spectroscopic comparisons and an HMBC correlation from the CO₂Et moiety to C-6.

HRESI(+)MS measurements suggested that 12 (C₁₃H₁₁Br₃O₆, Δmmu +0.5) was a decarboxy analogue of 7a/b and 8a/b; 13 (C₁₄H₁₃Br₃O₇, Δmmu +0.5) was a dihydro oxidized methyl ester of 12; 14 (C₁₃H₁₁Br₃O₆, Δmmu +0.5) was a decarboxymethyl analogue of 10; and, 15 (C₁₄H₁₃Br₃O₆, Δmmu +0.5) was a CH₂ homologue of 14. Comparison of the NMR (methanol-d₄) data for symphyocladin N (12) (Figures S38 and S39, Table 4, Table 5 and Table S11) with that for 8a/b revealed the key difference as replacement of the C-1 CO₂H moiety in 8a/b with an H-2 olefinic methine (δ_H 6.79) coupled to H₂-7′ (δ_H 4.05). The presence of a C-5 CO₂Me moiety in 12 was evident from an HMBC correlation from the OMe (δ_H 3.66) to C-5 (δ_C 171.3), while an E Δ^2,3 configuration was confirmed by a ROESY correlation between H₂-4 (δ_H 3.60) and H₂-7′ (Figure 4). Analysis of the NMR (methanol-d₄) data for symphyocladin O (13) (Figures S40 and S41, Table 4, Table 5, and Table S12) revealed it to be a saturated oxidized analogue of 12, as evidenced by COSY correlations between a diastereotopic H₂-2 (δ_H 3.35/3.27), through H-3 (δ_H 3.34) to a diastereotopic H₂-4 (δ_H 2.86/2.73). Likewise, replacement of the C-7′ sp³ methylene in 12 (δ_C 39.0) with a carbonyl resonance in 13 (δ_C 202.2) was evidence of a 7-oxo moiety. Diagnostic HMBC correlations also established the presence of incorporated C-5 CO₂Me (δ_H 3.68) and C-6 CO₂Me (δ_H 3.71) moieties (see Figure 3).

Comparison of the NMR (methanol-d₄) data for symphyocladin P (14) (Figures S42 and S43, Table 4, Table 5 and Table S13) with that for 10 revealed the key difference as replacement of the C-5 CO₂Me moiety in 10 with a diastereotopic H₂-4 olefinic methylene (δ_H 6.22/5.44). Comparison of the NMR (acetone-d₆) data for symphyocladin Q (15) (Figures S44 and S45, Table 4, Table 5 and Table S14) with that for 14 revealed the key difference as an additional resonance, attributed to a C-6 CO₂Me moiety (δ_H 3.71). Structure assignments for 14 and 15 were further supported by diagnostic 2D NMR correlations (Figure 4).

Structural similarities across 1–15 suggest a highly conserved biosynthesis. Building on this observation, we propose a biosynthetic relationship (Figure 5), in which the aconitates A–F (1–6) are viewed as mono, di, and tri methyl esters of the precursor E-aconitic acid, itself a dehydration product of citric acid. Likewise, metabolites 7–15 can be viewed as adducts between aconitates and an intermediate quinone methide that is generated from 2,3,6-tribromo-4,5-dihydroxybenzyl alcohol, further elaborated by a combination of 1,3-hydride shifts, decarboxylations and oxidations. Although 7a/b, 9a/b, 10–11, and 13 incorporate a single chiral sp³ center, as they do not exhibit measurable optical rotations they are presumed to be racemic, as confirmed for 10 by X-ray crystallography. The absence of double bond migrations (i.e., racemization) during isolation and handling suggests that this racemic character is a function of achiral adduct addition. The proposed biosynthetic relationship informs a possible biomimetic synthesis of 7–15, although, in our hands, synthetic 2,3,6-tribromo-4,5-dihydroxybenzyl alcohol proved stable to both acid and base conditions indicative of a requirement to activate the benzyl alcohol moiety to effect dehydration and the formation of a quinone methide.

3. Materials and Methods

General Experimental Procedures. Specific optical rotations ([α]_D) were measured on a polarimeter in a 100 × 2 mm cell at 22 °C. NMR spectra were obtained on a Bruker Avance DRX600 or DRX500 spectrometers, in the solvents indicated and referenced to residual ¹H and ¹³C signals in deuterated solvents. Electrospray ionization mass spectra (ESIMS) were acquired using an Agilent 1100 Series separations module equipped with an Agilent 1100 Series LC/MSD mass detector in both positive and negative ion modes. High-resolution ESIMS measurements were obtained on a Bruker micrOTOF mass spectrometer by direct infusion in MeCN at 3 mL/min using sodium formate clusters as an internal calibrant. HPLC was performed using an Agilent 1100 Series separations module equipped with Agilent 1100 Series diode array and/or multiple wavelength detectors and Agilent 1100 Series fraction collector, controlled using ChemStation Rev.B02.01 and Purify version A.1.2 software.

Algal material. Symphyocladia latiuscula was collected on the coast of Qingdao, Shandong Province, China, in May 2004. The specimen identification was verified by Dr. Kui-Shuang Shao (Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China). A voucher specimen (No. 2004X16) was deposited at the Herbarium of the Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China.

Extraction and isolation. The air-dried red alga Symphyocladia latiuscula (4.3 kg) was extracted with 95% EtOH at room temperature (3 × 72 h). After the solvent was removed under reduced pressure at <40 °C, a dark residue (610 g) was obtained. The residue was partitioned between EtOAc and H₂O, and the EtOAc-soluble partition (320 g) was chromatographed over silica gel, eluting with a gradient of 0–100% Me₂CO/petroleum ether to yield 85 fractions (F1–F85) (see Supporting Information Scheme S1 for the fractionation scheme). Fraction F54 was further fractionated over Sephadex LH-20 using CHCl₃–MeOH (2:1) to afford 21 fractions.

The sixth fraction from Sephadex LH-20 chromatography was fractionated on an ODS column, eluted by a stepwise gradient (0–100% MeOH/H₂O) to afford 11 fractions. The third fraction was subjected to HPLC separation (Zorbax Eclipse XDB-C₈, 5 μm, 250 × 9.4 mm column, 3.0 mL/min, gradient elution from 10 to 80% MeCN/H₂O over 15 min, with isocratic 0.01% TFA modifier) to yield 1–6; the sixth fraction was subjected to HPLC separation (Zorbax SB-C₁₈, 5 μm, 250 × 9.4 mm column, 3.0 mL/min, gradient elution from 30 to 75% MeCN/H₂O over 14 min, with isocratic 0.01% TFA modifier) to yield 8a/b and 9a/b; and the seventh fraction was subjected to HPLC separation (Zorbax Eclipse XDB-C₈, 5 μm, 250 × 9.4 mm column, 3.0 mL/min, gradient elution from 40 to 55% MeCN/H₂O over 20 min, with isocratic 0.01% TFA modifier) to yield 11 and 13.

The seventh fraction from Sephadex LH-20 chromatography was subjected to HPLC fractionation (Zorbax SB-C₁₈, 5 um, 250 × 9.4 mm column, 3.0 mL/min, gradient elution from 30 to 80% MeCN/H₂O over 14 min, with isocratic 0.01% TFA modifier) to yield 10.

The eighth fraction from Sephadex LH-20 chromatography was fractionated on an ODS column, eluted with a stepwise gradient (0–100% MeOH/H₂O) to afford 11 fractions; the sixth fraction was subjected to HPLC fractionation (Zorbax SB-C₁₈, 5 δm, 250 × 9.4 mm column, 3.0 mL/min, gradient elution from 35 to 50% MeCN/H₂O over 15 min, with isocratic 0.01% TFA modifier) to yield 14; the seventh fraction was subjected to HPLC fractionation (Zorbax SB-C₁₈, 5 μm, 250 × 9.4 mm column, 3.0 mL/min, gradient elution from 50 to 60% MeCN/H₂O over 12 min, with isocratic 0.01% TFA modifier) to yield 12 and 15.

The ninth fraction from Sephadex LH-20 chromatography was fractionated on an ODS column, eluted with a stepwise gradient (0–100% MeOH/H₂O) to afford 11 fractions; the third fraction was subjected to HPLC fractionation (Zorbax SB-C₁₈, 5 μm, 250 × 9.4 mm column, 3.0 mL/min, gradient elution from 30 to 80% MeCN/H₂O over 20 min, with isocratic 0.01% TFA modifier) to yield 7a/b.

Aconitate A (1): White solid; NMR (600 MHz, methanol-d₄) see Table S2; HRESIMS m/z 189.0396 [M + H]⁺ (calcd. for C₇H₈O₆ 189.0394).

Aconitate B (2): White solid; NMR (600 MHz, methanol-d₄) see Table S2; HRESIMS m/z 189.0395 [M + H]⁺ (calcd. for C₇H₈O₆ 189.0394).

Aconitate C (3): White solid; NMR (600 MHz, methanol-d₄) see Table S3; HRESIMS m/z 189.0395 [M + H]⁺ (calcd. for C₇H₈O₆ 189.0394).

Aconitate D (4): White solid; NMR (600 MHz, methanol-d₄) see Table S3; HRESIMS m/z 203.0552 [M + H]⁺ (calcd. for C₈H₁₁O₆, 203.0550)

Aconitate E (5): White solid; NMR (600 MHz, methanol-d₄) see Table S4; HRESIMS m/z 203.0551 [M + H]⁺ (calcd. for C₈H₁₁O₆, 203.0550).

Aconitate F (6): White solid; NMR (600 MHz, methanol-d₄) see Table S4; HRESIMS m/z 217.0708 [M + H]⁺ (calcd. for C₉H₁₃O₆, 217.0707).

Symphyocladins C/D (7a/b): Light brown solid; NMR (600 MHz, acetone-d₆) see Table 1, Table 2 and Table S5; HRESIMS m/z 544.8082 [M + H]⁺ (calcd. for C₁₄H₁₂Br₃O₈, 544.8077).

Symphyocladins H/I (8a/b): Light brown solid; NMR (600 MHz, methanol-d₄, acetonitrile-d₃) see Table 1, Table 2, Tables S6 and S7; HRESIMS m/z 544.8081 [M + H]⁺ (calcd. for C₁₄H₁₂Br₃O₈, 544.8077).

Symphyocladins J/K (9a/b): Light brown solid; NMR (600 MHz, DMSO-d₆) see Table 1, Table 2 and Table S8; HRESIMS m/z 558.8237 [M + H]⁺ (calcd. for C₁₅H₁₄Br₃O₈, 558.8233).

Symphyocladin L (10): Light brown solid; NMR (600 MHz, acetone-d₆) see Table 3, Table 5 and Table S9; HRESIMS m/z 558.8238 [M + H]⁺ (calcd. for C₁₅H₁₄Br₃O₈, 558.8233).

Symphyocladin M (11): Light brown solid; NMR (600 MHz, methanol-d₄) see Table 3, Table 5 and Table S10; HRESIMS m/z 586.8552 [M + H]⁺ (calcd. for C₁₇H₁₇Br₃O₈, 586.8546).

Symphyocladin N (12): Light brown solid; NMR (600 MHz, acetone-d₆) see Table 4, Table 5 and Table S11; HRESIMS m/z 500.8184 [M + H]⁺ (calcd. for C₁₃H₁₂Br₃O₆, 500.8179).

Symphyocladin O (13): Light brown solid; NMR (600 MHz, methanol-d₄) see Table 4, Table 5 and Table S12; HRESIMS m/z 530.8289 [M + H]⁺ (calcd. for C₁₄H₁₄Br₃O₇, 530.8284).

Symphyocladin P (14): Light brown solid; NMR (600 MHz, methanol-d₄) see Table 4, Table 5 and Table S13; HRESIMS m/z 500.8184 [M + H]⁺ (calcd. for C₁₃H₁₂Br₃O₆, 500.8179).

Symphyocladin Q (15): Light brown solid; NMR (600 MHz, acetone-d₆) see Table 3, Table 4 and Table S14; HRESIMS m/z 514.8340 [M + H]⁺ (calcd. for C₁₄H₁₄Br₃O₆, 514.8335).

X-ray crystallography. X-ray crystallographic data were collected on an Oxford Diffraction Gemini CCD diffractometer with Mo-Kα radiation (0.71073 Å) operating within the range 2 < 2θ < 50°. The sample was cooled to 190 K with an Oxford Cryosystems Desktop Cooler. Data reduction and empirical absorption corrections were performed using CrysAlisPro (Rigaku Oxford Diffraction, Yarnton, Oxfordshire, UK). The structure was solved by Direct Methods and refined with SHELX [21] and all of the calculations and refinements were carried out by WinGX package [22]. All non-H atoms were refined aniostropically. The thermal ellipsoid plot was produced with ORTEP [23] and the unit cell diagram was drawn with PLATON [24]. Crystallographic data including structure factors in CIF format have been deposited with the Cambridge Crystallographic Data Centre (CCDC 1569026).