2-((((R)-3-(((9E,17Z)-18-Bromooctadeca-9,17-dien-7,15-diynoyl)oxy)-2-hydroxypropoxy)(hydroxy)phosphoryl)oxy)-N,N,N-trimethylethan-1-aminium

Abstract

The Saudi Arabian Red Sea has been a focus of ongoing scientific exploration due to its extensive, largely untapped marine biodiversity, particularly marine sponges. Marine sponges have long been recognised as a valuable source of unique compounds. In this study, we isolated a new brominated compound (1) from the marine sponge Aiolochroia crassa, collected from the Saudi Arabian Sea, using chromatographic analyses. Molecular networking analysis revealed the presence of brominated molecules in the extract. The identified compound belongs to a class of phosphatidylcholine derivatives. The structure of compound 1 was elucidated using 1D and 2D NMR spectroscopy and high-resolution ESI-Q-TOF mass spectrometry. Further structure confirmation studies were performed using MS/MS fragmentation analysis and DFT calculations for the ¹H and ¹³C NMR chemical shifts. This is the first report of 1 from this species of marine sponges.

1. Introduction

Globally, the number of studies focusing on sponges (Porifera) has expanded, highlighting their significant relevance and scientific value [1]. Marine sponges are a rich source of unique compounds, producing hundreds of new molecules each year for natural product research [2]. They are an ecologically significant and varied component of marine natural products from benthic habitats, having an estimated 25,000 species [3]. To date, approximately 10,818 secondary metabolites have been identified from marine sponges according to the current MarinLit databases [4].

There are four major classes of marine sponge, including Demospongiae, Calcarea, Hexactinellida, and Homoscleromorpha [5]. Demospongiae is the biggest class, with almost 8000 recognised species divided into three subclasses: Verongimorpha, Heteroscleromorpha, and Keratosa [6,7].

The Red Sea is a semi-enclosed, extended warm water body of 2000 km in length, with a maximum width of 355 km, a surface area of around 458,620 km², and a volume of 250,000 km³ [8]. It is the youngest ocean zone on the planet, and, in general, it has been the subject of numerous studies, including chemical characterisation of its organisms. The Red Sea is one of the most diverse coral reefs, with marine biodiversity niches owing to its high species richness and endemism [9]. Surprisingly, the coastline of Saudi Arabia’s Red Sea seems to have remained mostly unexplored until now. It harbours more than 300 sponge species [10]. One of these species is Aiolochroia crassa, a marine sponge in the Verongida class, previously named Pseudoceratina crassa, [11]. The metazoan has the ability to biosynthesise interesting, brominated compounds possibly for defensive and signalling purposes [12].

To date, fifteen brominated metabolites, mainly belonging to the alkaloid class of organic compounds, have been isolated and characterised from the marine sponge Aiolochroia crassa [13,14]. The first brominated polyunsaturated fatty acid was isolated from the sea sponge Xestospongia testudinaria, collected from Pandora Reef near Townsville, Queensland, Australia [15], and later from the Red Sea, the Gulf of Eilat [16].

The sponge sample Aiolochroia crassa was collected from the Red Sea coral reefs near Alqunfudah City, Kingdom of Saudi Arabia. The collection process involves permissions from the National Centre for Wildlife (Supplementary Information S1). The sponge material was subjected to sequential extraction, fractionation, and purification, leading to the isolation of one new brominated acetylenic fatty acid derivative. Herein, we report for the first time on the marine-derived brominated acetylenic fatty acids.

2. Results and Discussion

The marine sponge Aiolochroia crassa was extracted with 500 mL methanol (100%) and dichloromethane (100%), yielding a crude extract of 1.4 g. Following the extraction process, a 0.1 mg/mL concentration of the extract was prepared in 50% methanol and subjected to LC-MS/MS analysis to determine the chemical profile of the extract. Interestingly, many brominated compounds were dereplicated, which displayed isotope patterns consistent with the presence of one Br. The crude extract was fractionated by solid phase extraction (SPE), yielding fractions F1–5 (449.0 mg, 179.0 mg, 83.0 mg, 98.0 mg, 235.0 mg, respectively). Fraction F5 (0.235 g) showed an interesting chemical profile (Table S1), indicating the presence of the targeted brominated compounds. LC-MS/MS data of the fraction was further analysed by molecular networking using the Global Natural Products Social Networking (GNPS) platform (Figure 1) [17]. The result shows many brominated compounds in a cluster of phosphatidylcholines consisting of 22 compounds including potentially brominated metabolites, as shown in Figure 1 and the Supplementary Information (Table S1).

Subsequently, the fraction was subjected to purification by C-18 reversed-phase HPLC, yielding a new brominated lysophosphatidylcholine, 1 (Figures S1–S7).

2.1. Structure Elucidation

Compound 1 is an oily brown colour. The molecular formula was confirmed as C₂₆H₄₂BrNO₇P⁺ at m/z (590.1885 and 592.1868) [M + H]⁺ (Δ −1.3 ppm), requiring 8 degrees of unsaturation, based on HRESI-MS (Figure S1). The full structure as it shows in Figure 2 was confirmed by detailed analysis of 1D and 2D NMR data which showed four olefinic protons, nine methylene protons, three oxygenated methylene protons, one N-methylene proton, three N-methyl protons, one hydroxy proton, four acetylenic quaternary carbon signals (C-3,$\mathsf{\delta }$_C 78.7; C-4, $\mathsf{\delta }$_C 94.0; C-11, $\mathsf{\delta }$_C 79.0 and C-12 $\mathsf{\delta }$_C 95.5) and one quaternary carbon signal (C-18, $\mathsf{\delta }$_C 176.0) (

Table 1. Experimental NMR data (600/150 MHz, CD₃OD) for compound 1.

Experiment							DFT Calculation
Atom	Mult	$\mathsf{\delta }$C	$\mathsf{\delta }$H	Mult., J in Hz	COSY	HMBC H → C	$\mathsf{\delta }$C	$\mathsf{\delta }$H
1	CH	116.0	6.70	d (14.5)	2	2, 3	126.5	6.9
2	CH	119.0	6.20	dt (14.5, 2.4)	1, 5	1, 4	121.6	6.5
3	C	78.7					76.8
4	C	94.0					98.6
5	CH2	20.0	2.30	dd (6.9, 2.4)	2, 6	2, 3, 4	18.9	2.3
6	CH2	29.0	1.44	m	5, 7	5, 7	28.1	1.45
7	CH2	29.6	1.52	m	6, 8	8	30.3	1.36
8	CH2	30.0	2.26	dd (7.1)	7, 9	10	33	2.16
9	CH	142.0	5.81	dt (10.9, 7.1)	8, 10	8, 11	138.3	5.6
10	CH	111.0	5.45	dq (10.9, 2.1)	9, 13	12	115.2	4.9
11	C	79.0					79.6
12	C	95.5					98.2
13	CH2	20.0	2.33	dd (7.0, 2.1)	10, 14	10, 11, 12	18.9	2.4
14	CH2	30.0	1.55	m	13, 15	13, 15, 16	31.8	1.55
15	CH2	29.3	1.46	m	14, 16	13, 14, 16	29.03	1.42
16	CH2	25.8	1.64	q (7.5)	15, 17	14, 17, 18	26.3	1.8
17	CH2	35.5	2.38	t (7.5)	16	14, 15, 16, 18	34.7	2.4
18	C	176.0					178.3
19	OCH2	66.0	a 4.11 b 4.18	dd (11.5, 6.2) dd (11.5, 4.4)	20	18, 20	56.4	3.6 4.2
20	OCH	69.8	4.00	q (5.3)	19a, 19b, 21	21	69.3	4.3
21	OCH2	67.4	3.91	m	22	22	67.8	4.2
22	OCH2	67.8	3.65	m	23	23, 24	66.9	3.9
23	N-CH2	60.0	4.30	d (6)	22		60.1	4.2
24	(CH3)3	54.0	3.23	s		23	54.5	3.1

and Figures S3–S7).

¹H NMR data show four downfield signals (H-1, $\mathsf{\delta }$_H 6.7; H-2, $\mathsf{\delta }$_H 6.2; H-9, $\mathsf{\delta }$_H 5.8; H-10, $\mathsf{\delta }$_H 5.45). Each signal integrates for one proton, consistent with two olefinic pairs, with COSY correlations between H-1/H-2 and H-9/H-10, with J-constants of 10.9 and 14.5 Hz, corresponding to Z (cis) and E (trans) geometry, respectively. HMBC data indicated the presence of an enyne system through cross-peaks from respective olefinic protons (H-1 and H-2) to acetylenic quaternary carbons C-3 ($\mathsf{\delta }$_C 79.0) and C-4 ($\mathsf{\delta }$_C 94.0). The position of the second enyne system is confirmed by the HMBC correlation from H-9 to C-11 ($\mathsf{\delta }$_C 79.0), and from H-10 to C-12 ($\mathsf{\delta }$_C 95.5).

A four-methylene spin system (H-5, δ_H 2.3; H-6, δ_H 1.44; H-7, δ_H 1.52; and H-8, δ_H 2.26) connected two acetylenic units (Figure S6). Another methylene chain (H-13, δ_H 2.33; H-14, δ_H 1.55; H-15, δ_H 1.46; H-16, δ_H 1.64; H-17, δ_H 2.4) joined the second acetylenic carbon (C-12) to carbonyl group C-18 ($\mathsf{\delta }$_C 176), supported by HMBC correlation from H-13 to C-11 and from H-16 to C-18.

An ABX spin system composed of oxygenated diastereotopic methylenes and a methine (H-19, δ_H 4.12, 4.18; H-20, δ_H 4.0; H-21, δ_H 3.9), characteristic of a glycerol moiety, also showed an HMBC correlation from H-19 to the carbonyl carbon C-18 (see Figure 3).

Long-range ⁴J coupling across the triple bond was observed between the H-2 and H-5 (2.4 Hz) and between H-10 and H-13 (2.1 Hz). Long-range proton coupling across a carbon–carbon triple bond occurs via four-bond (⁴J) coupling, typically 2–3 Hz, due to efficient transmission through the linear π-system, and weaker five-bond (⁵J) coupling with coupling constants generally below 2 Hz [18,19]. A systematic review of 20 articles from 1985 to the present found that the ⁴J range is 1.7–2.4 Hz. At the same time, for about ⁵J, only 4 studies mentioned the range, which is between 0.6 and 0.7 Hz (see Table S2).

The A₂B₂ spin system of the two methylene groups (H₂-22, δ_H 4.12; H₂-23, δ_H 4.18) together with the HMBC cross-peak from the symmetrical trimethylammonium group H₃-24 (δ_C/δ_H; 54/3.22) to C-23 (δ_C 60.0), and the presence of the phosphorus atom and two additional oxygens indicated by the molecular formula (C₂₆H₄₂BrNO₇P) collectively confirmed the presence of the phosphocholine substituent in compound 1. Detailed MS² analysis of 3 yielded fragment ions at m/z 258.1094, 184.0732, 124.9993, and 104.1074, thereby corroborating the proposed structure (Figure 4).

The specific rotation of 1 was observed as negative, with a specific rotation −24.6°. And the stereochemistry of the glycerol unit of 1 was tentatively assigned as an R configuration at C-20 analogous to reported natural products containing glycerol units from marine sponges [20].

¹H and ¹³C NMR chemical shifts were calculated using DFT calculations. Compound 1 was initially subjected to a conformational search using the Merck molecular force field. Thirty conformers were found within a 5 kcal/mol energy threshold from a global minimum. All these conformers were geometrically optimised, and frequency calculations were performed using density functional theory (DFT) at the B3LYP/6-31G(d) level with PCM (see computational details in Section 3.5. Calculated ¹H and ¹³C NMR chemical shifts were compared with the experimental NMR data using statistical correlation (R²: 0.9803 and R²: 0.9937, respectively); finally, we obtained a result in perfect agreement with the experimental data (Figure 5 and Figure 6).

2.2. Conclusions

In this study, we isolated a new brominated compound from the marine sponge Aiolochroia crassa, collected from the Saudi Arabian Sea, using a combination of metabolomic and chromatographic analyses. This study highlights the continued potential of marine sponges as a reservoir of unexplored compounds.

3. Materials and Methods

3.1. General Experimental Procedures

Optical rotation was measured on an ADP 410 digital polarimeter (Bellingham + Stanley Ltd., Tunbridge Wells, Kent, UK). NMR spectra were recorded on Bruker AVANCE III spectrometers at 600 MHz (Bruker Corporation, Billerica, MA, USA). Solid Phase Extraction (SPE) with reverse-phase Phenomenex Strata^® C18-E (55 μm, 70 Å) (Phenomenex LTD, Torrance, CA, USA) cartridges, and 2 g/12 mL Giga Tubes cartridges were used to fractionate the sample. The isolation and purification of the sample were performed using High-Pressure Liquid Chromatography (HPLC) connected to a Sunfire semipreparative C18 column (5 μm, 100 Å, 10 × 250 mm). The Bruker MAXIS II mass spectrometer (Bruker Corporation, Billerica, MA, USA)equipped with a Quadrupole-Time-of-Flight mass analyser joined to an HPLC (Agilent 1290 Infinity equipped with a diode array detector) (Agilent Technologies, Santa Clara, CA, USA) on a Kinetex EVO l C18 column (2.6 µm, 100 Å, 2.1 × 100 mm) was used to analyse the sample. A mobile phase gradient of 0.1% formic acid in LC-MS-grade H₂O:acetonitrile from 5 to 95% in 10 min, and isocratic elution at 95% for 2 min at a flow rate of 1 µL/min with MS acquisition for 12 min were employed in the UPLC-MS system.

3.2. Collection and Identification

The sponge Aiolochroia crassa was collected from the Red Sea, near the city of Alqunfudh in the southwest of the Kingdom of Saudi Arabia. The GPS coordinates for the sample collection are 19°10′40.5″ N 41°02′22.9″ E.

3.3. Extraction and Isolation

The sample was collected from the Red Sea in a mixture of 50% seawater and 50% ethanol. The preservation was changed in the lab to methanol for two days; then three extractions were conducted using 500 mL of 100% methanol and 100% dichloromethane, resulting in a total mass of 1.4 g. It was chromatographed on SPE. It was equilibrated with 3 × 40 mL of the starting 100% water as a solvent. The sample was dissolved in 80% water/20% methanol. After the sample was inserted, five solutions of mobile phases (100% H₂O, 25% MeOH, 50% MeOH, 75% MeOH, and 100% MeOH) were employed to elute the components of the extract. Subsequently, F5 (235.4 mg) was selected for purification upon LC-MS/MS analysis that suggested it was an interesting fraction. A reversed-phase HPLC connected to a Sunfire C18 OBD prep column (100 Å, 5 µm, 10 × 250 mm) was used with a gradient starting with 15% H₂O to 100% MeOH over 50 min, ending with 100% MeOH isocratic elution for 20 min at a flow rate of 1 mL/min, yielding compound 1 (3.4 mg, t_R 28 min).

Compound 1: oily brown colour; [α]D₂₅ −24.6 (c 0.65, MeOH). HRESIMS at m/z (590.1873) [M + H]⁺ (Δ −1.3 ppm) (calcd. for C₂₆H₄₂BrNO₇P⁺, 590.1877); NMR data, see Table 1 and Supporting Information; Figures S1–S7.

3.4. Data Pre-Processing and Feature-Based Molecular Networking

LC-MS data of the fraction and the blank were converted to mzML format using MS Convert software, followed by creating a quantitative MS1 file and MGF file using MZmine version 4.8.0 [21]. Using the GNPS platform, the MZmine files were submitted to create molecular networking clusters and to search the GNPS natural product database for the presence of known compound clusters.

The mzML data underwent pre-processing, with mass ion detection facilitated by the elimination of MS1 and MS2 baseline noise levels established at 1.0 × 10⁴ and 1.0 × 10², respectively. The construction of chromatograms was executed for each centroided MS1 detected using the ADAP chromatogram builder, with parameters set for minimum consecutive scans, minimum intensity for consecutive scans, minimum absolute height, and m/z mass tolerance at 4, 1.0 × 10², 1.0 × 10⁴, and 5 ppm, respectively. The peak detection employed a local minimum resolver as a deconvolution algorithm to generate a chromatogram for each mass ion. The chromatographic threshold for this approach was established at 85%, the minimum search range in the RT absolute range at 0.05 min, the minimum relative height at 0.0%, the minimum absolute height at 2.8 × 10⁴, and the minimum peak top/edge ratio at 1.70. The detected peaks were deisotoped utilising the isotopic peak grouper, with an m/z tolerance (intra-sample) established at 5 ppm, an RT tolerance of 0.02 absolute minutes, a maximum charge of 3 absolute minutes, and the most intense isotope designated as the representative isotope. The deisotoped peaks were subsequently aligned utilising a join aligner to rectify any discrepancies in the retention time of each mass peak: the ion m/z tolerance for alignment was established at 10 ppm, the retention time tolerance at 0.06 absolute minutes, with weights for m/z and RT assigned as 75 and 25, respectively, and the mobility weight set to 1. The peak list generated post-alignment was refined using a feature list row filter to exclude mass ions within the specified range of 100 to 2000 m/z, retaining only peak rows that satisfy all criteria. The GNPS result was visualised using Cytoscape software version 3.10.1.

3.5. Computational Calculations

Computational calculations were performed in the Gaussian 16 program package. Conformational searches for all possible isomers were conducted via molecular mechanics using the Merck molecular force field (MMFF) with a 5 kcal/mol threshold. After this, the B3LYP/6-31 G(d) PCM level in the respective solvent was used to optimise the most abundant conformers. Then, the NMR shielding tensors of all optimised conformers were computed at the wb97xd/6-311+g(2d,p) (IEFPCM=CD₃OD) and then averaged based on the Boltzmann distribution.