Semisynthesis and Cytotoxic Evaluation of an Ether Analogue Library Based on a Polyhalogenated Diphenyl Ether Scaffold Isolated from a Lamellodysidea Sponge

The known oxygenated polyhalogenated diphenyl ether, 2-(2′,4′-dibromophenoxy)-3,5-dibromophenol (1), with previously reported activity in multiple cytotoxicity assays was isolated from the sponge Lamellodysidea sp. and proved to be an amenable scaffold for semisynthetic library generation. The phenol group of 1 was targeted to generate 12 ether analogues in low-to-excellent yields, and the new library was fully characterized by NMR, UV, and MS analyses. The chemical structures for 2, 8, and 9 were additionally determined via single-crystal X-ray diffraction analysis. All natural and semisynthetic compounds were evaluated for their ability to inhibit the growth of DU145, LNCaP, MCF-7, and MDA-MB-231 cancer cell lines. Compound 3 was shown to have near-equivalent activity compared to scaffold 1 in two in vitro assays, and the activity of the compounds with an additional benzyl ring appeared to be reliant on the presence and position of additional halogens.


Introduction
Dysideidae sponges, such as those in the genus Lamellodysidea, are known to contain a wide variety of compounds, which generally are categorized into two distinct chemotypes: those containing highly functionalized peptides and terpenoids, and those solely containing oxygenated polyhalogenated diphenyl ethers (O-PHDEs) [1,2].It has long been understood that while the latter chemotypes are structurally similar to anthropogenic fire retardants, there is extensive evidence for their presence in sponges to be due to their antimicrobial and antifeedant activity [3][4][5][6][7].It is now accepted knowledge that O-PHDEs are produced by symbiotic cyanobacteria within marine sponges, with the first evidence of this claim being published in 1994, where it was found that O-PHDEs from Lamellodysidea herbacea were produced by the cyanobacterial symbiont Hormoscilla spongeliae and excreted into the aqueous sponge mesophyll [2].
Since this initial research, much of the biosynthetic pathway for these compounds has been described following research into various proteobacteria associated with marine eukaryotes [1,8,9].The brominated marine pyrroles/phenols (bmp) bacterial gene locus eukaryotes [1,[8][9].The brominated marine pyrroles/phenols (bmp) bacterial gene locus contains two key enzymes: phenol brominase Bmp5, which is able to incorporate bromine or iodine into phenol and catechol radicals; and cytochrome P450 Bmp7, which mediates the homo or heterocoupling of bromophenol and bromocatechol (Figure 1a) [8,9].Promiscuous "off-pathway" chlorinases and brominases can selectively chlorinate, brominate, or iodinate the structure post-coupling (Figure 1b), with the scarceness of iodinated O-PHDEs being explained by the greater abundance of chloride and bromide in seawater [1,9].While diversity in halogenating enzymes, differential coupling of bromophenol and bromocatechol, and methylation of phenolic groups allows for an extraordinary diversity of substitution patterns in these compounds, only one iodinated O-PHDE has been reported to date, and it appears that the halogen substitution patterns observed in known O-PHDEs are dictated by biosynthetic rules which have not been defined [1,3,[8][9][10].(a) Generalized process for the biosynthesis of O-PHDEs.Phenol (i) and catechol (ii) are brominated with Bmp5 (halogens omitted from iii-ix for clarity) prior to Bmp7 bromophenol homocoupling (iii-iv), bromophenol-bromocatechol heterocoupling (v-vii), or bromocatechol homocoupling (viii-ix) [8,9].Note that no O-PHDE with the oxygenation pattern described as ix has been reported to date.(b) Substitution patterns of all PHDEs isolated from Dysideidae sponges to date.Note that compounds with the oxygenation pattern III-3 cannot be generated solely through the biosynthetic process described in Figure 1a, suggesting that post-translational oxidases or hydroxylases may play a role in this case [9,10].[8,9].Note that no O-PHDE with the oxygenation pattern described as ix has been reported to date.(b) Substitution patterns of all PHDEs isolated from Dysideidae sponges to date.Note that compounds with the oxygenation pattern III-3 cannot be generated solely through the biosynthetic process described in Figure 1a, suggesting that post-translational oxidases or hydroxylases may play a role in this case [9,10].Due to the Davis group's interest in the use of new or under-utilized natural products as scaffolds for the semisynthesis of unique biodiscovery screening libraries, a new species of Lamellodysidea sponge (that is in the process of being fully taxonomically described) was prioritized for chemical investigations [11,12].The CH 2 Cl 2 :MeOH extract from this specimen showed only one major polybrominated UV-active peak via UHPLC-MS (Supplementary Figure S1) that was subsequently identified as the known O-PHDE, 2-(2 ′ ,4 ′ -dibromophenoxy)-3,5-dibromophenol (1), a bioactive scaffold amenable to semisynthesis [13].In previous studies, 1 had been isolated from Dysidea and Lamellodysidea sponges and reported by other researchers to not only have antibacterial and antifeedant activity but also broad cytotoxicity, including moderate activity (IC 50 ≤ 10 µM) in PANC-1 (epithe- lioid carcinoma), MCF-7 (triple-positive breast cancer), and BS-C-1 (continuous epithelial kidney cell) cancer cell lines and moderate activity against the cancer-relevant proteins Tie2 kinase, PTP1B, and IMPDH [3][4][5][6][14][15][16][17][18].In the current studies reported here, the reactive phenol group of 1 was exploited to generate a novel O-PHDE library for continued investigations into this structure class's activity against MCF-7, in addition to activity against MDA-MB-231 (triple-negative breast cancer), DU145 (prostate carcinoma) and LNCaP (androgen-sensitive human prostate adenocarcinoma) cell lines [19,20].
Scaffold 1 (10 mg) was subsequently reacted with a series of commercially available alkyl halides (R-X) in dry acetone for 1 h at 50 • C, using K 2 CO 3 as a base (Scheme 1 and Figure 2).The reactions were initially tested in laboratory-grade and deuterated acetone, where it was found that deuterated acetone was sufficiently dry for the reactions to take place with low-to-excellent yields.Following a workup with CH 2 Cl 2 /H 2 O partitioning and HPLC purification (MeOH/H 2 O/0.1% TFA), a total of 12 analogues (2-13), including one naturally occurring O-PHDE (2) [22], were obtained in yields ranging from 17 to 99%, and purities of >95%, as determined by UHPLC-MS analysis.Furthermore, all O-PHDE derivatives were fully characterized via 1D/2D NMR, UV, and MS data analyses.
Scaffold 1 (10 mg) was subsequently reacted with a series of commerciall alkyl halides (R-X) in dry acetone for 1 h at 50 °C, using K2CO3 as a base (Sch Figure 2).The reactions were initially tested in laboratory-grade and deuterat where it was found that deuterated acetone was sufficiently dry for the reacti place with low-to-excellent yields.Following a workup with CH2Cl2/H2O partit HPLC purification (MeOH/H2O/0.1% TFA), a total of 12 analogues (2-13), inc naturally occurring O-PHDE (2) [22], were obtained in yields ranging from 17 t purities of >95%, as determined by UHPLC-MS analysis.Furthermore, all O-PH atives were fully characterized via 1D/2D NMR, UV, and MS data analyses.For example, the HRESIMS of O-PHDE analogue 8 revealed an ion at m/z 688.6566 [M + Na] + that enabled a molecular formula of C19H11 79 Br5O2 to be assigned to the new semisynthetic.The 1 H NMR spectrum (Table 1 The methylene resonating at δH 5.12 (H-1") showed strong ROESY and HMBC correlations to H-6 and C-1 respectively, whichconnected ring B of the O-PHDE scaffold to ring C (Figure 3).The remaining NMR signals for 8 were assigned following extensive NMR data analysis and a comparison of the chemical shifts for the previously reported scaffold 1 (Table 1 and Figure 3) Furthermore, crystals obtained for compound 8 were analyzed via X-ray crystallography and confirmed the NMR-based structure assignment; the ORTEP for 8 is shown below in Figure 3. Furthermore, crystals obtained for compounds 2 and 9 were also analyzed via X-ray crystallography, which confirmed their NMR-based structure assignments; the ORTEPs for 2 and 9 are shown below in Figure 4.   3).The remaining NMR signals for 8 were assigned following extensive NMR data analysis and a comparison of the chemical shifts for the previously reported scaffold 1 (Table 1 and Figure 3).Furthermore, crystals obtained for compound 8 were analyzed via X-ray crystallography and confirmed the NMR-based structure assignment; the ORTEP for 8 is shown below in Figure 3. Additionally, crystals obtained for compounds 2 and 9 were also analyzed via X-ray crystallography, which confirmed their NMR-based structure assignments; the ORTEPs for 2 and 9 are shown below in Figure 4.  Based on previous cytotoxic activity reported for scaffold 1, the natural products 1 and 2 and synthetic analogues 3−13 were screened for activity against four cancer cell lines (Table 2).While scaffold 1 and its derivatives displayed low activity in the DU145 assay, the activity reported in the other cell lines was sufficient to allow for some structure-activity relationships (SARs) to be determined.While most compounds with benzylated additions (6−13) displayed minimal activity in all assays, some preliminary SARs were identified based on the type and positioning of substituents on ring C. For example, compounds 7 and 8 that are methylated and brominated in the para-position of ring C respectively, displayed no activity, while 9−11 of the benzylated series, which all contain metapositioned electron-withdrawing groups on ring C demonstrated some activity with the brominated analogue being the most toxic of the three compounds.Based on previous cytotoxic activity reported for scaffold 1, the natural products 1 and 2 and synthetic analogues 3-13 were screened for activity against four cancer cell lines (Table 2).While scaffold 1 and its derivatives displayed low activity in the DU145 assay, the activity reported in the other cell lines was sufficient to allow for some structure-activity relationships (SARs) to be determined.While most compounds with benzylated additions (6-13) displayed minimal activity in all assays, some preliminary SARs were identified based on the type and positioning of substituents on ring C. For example, compounds 7 and 8 that are methylated and brominated in the para-position of ring C respectively, displayed no activity, while 9-11 of the benzylated series, which all contain meta-positioned electron-withdrawing groups on ring C demonstrated some activity with the brominated analogue being the most toxic of the three compounds.Compounds 12 and 13 each have ortho-substitutions on ring C and did not display appreciable activity thus indicating that this substitution pattern is detrimental to toxici-tyFuture libraries with varied positions and species of halogens and different functional groups are required in order to shed more light on SAR.Compound 3 displayed the best activity of all semisynthetic analogues tested, displaying indistinguishable cytotoxicity to scaffold 1 against MCF-7 and MDA-MB-231 breast cancer cell lines, which suggests that the addition of smaller alkyl groups was better for retaining bioactivity than the addition of benzyl rings.In conclusion, 12 O-PHDE ether analogues (including one known naturally occurring compound) were synthesized in low-to-excellent yields, and the new was characterized via NMR, US, and MS analyses.Whilst cytotoxicity evaluations identified no significant toxicity against four human cancer cell lines, this new library will be added to the Davis Open Access Natural Product-Based Library and screened in other bioassays in the future [23,24].Compounds 12 and 13 each have ortho-substitutions on ring C and did not display appreciable activity thus indicating that this substitution pattern is detrimental to toxicity.Future libraries with varied positions and species of halogens and different functional groups are required in order to shed more light on SAR.Compound 3 displayed the best activity of all semisynthetic analogues tested, displaying indistinguishable cytotoxicity to scaffold 1 against MCF-7 and MDA-MB-231 breast cancer cell lines, which suggests that the addition of smaller alkyl groups was better for retaining bioactivity than the addition of benzyl rings.
In conclusion, 12 O-PHDE ether analogues (including one known naturally occurring compound) were synthesized in low-to-excellent yields, and the new library was characterized via NMR, US, and MS analyses.Whilst cytotoxicity evaluations identified no significant toxicity against four human cancer cell lines, this new library will be added to the Davis Open Access Natural Product-Based Library and screened in other bioassays in the future [23,24].

General Experimental
Melting points were measured using a Cole-Parmer melting point apparatus and are uncorrected.UV spectra were recorded using an Ocean Optics USB-ISS-UV/Vis spectrometer.NMR spectra were recorded at 25 • C on a Bruker AVANCE III HD 800 MHz NMR spectrometer (Billerica, MA, USA) equipped with a cryoprobe.The 1 H and 13 C chemical shifts were referenced to solvent peaks for DMSO-d 6 at δ H 2.50 and δ C 39.52.HRESIMS data were acquired on a Bruker maXis II ETD ESI-qTOF.Silica gel 60 (Merck, 40-63 µm, 60 Å) was packed into an open glass column (38 × 90 mm) for flash column chromatography.Davisil C 8 bonded silica (30-40 µM, 60 Å) was used for pre-adsorption work before HPLC separations, and the pre-absorbed sample was packed into a Grace stainless steel guard cartridge (10 × 30 mm).A Phenomenex Luna C 18 column (5 µm, 90-110 Å, 10 mm × 250 mm) attached to a Thermo Fisher Scientific Dionex Ultimate 3000 UHPLC (Waltham, MA, USA) was used for semipreparative HPLC separations.A All chemical reagents used throughout the experiments were purchased from Sigma-Aldrich, and all solvents used for chromatography, UV, and MS were Honeywell Burdick & Jackson or Lab-Scan HPLC grade.NMR spectra were processed using MestReNova version 11.0.3 (Mestrelab Research, Santiago de Compostela, Spain).Chemical structures were drawn using ChemDraw Ultra 12.0.2.HPLC, and LC-MS results were analyzed via Thermo Scientific TM Dionex TM Chromeleon TM 7.2.UV data were analyzed using Logger Pro 3 (Vernier Software & Technology, Beaverton, OR, USA).

Animal Material
The undescribed Lamellodysidea sp.OTU 2054 specimen [12] was an olive green to light brown color underwater and turned brown to orange upon exposure to air, while producing copious black mucous.It formed thickly encrusted mats, with thick erect microconulose lamellae forming rounded labyrinthian meshes.The texture was rubbery, compressible, and soft, but it was firmly attached to the hard pavement directly behind the reef crest on Ribbon Reef (Great Barrier Reef, Queensland, Australia).The skeleton consisted of similarly sized cored primary and secondary fibers.A voucher specimen of this sponge, Lamellodysidea sp.OTU 2054 (QM G325118), has been deposited at the Queensland Museum.

X-ray Crystallography Analysis of Compounds 2, 8, and 9
Intensity data for compound 8 were collected with an Oxford Diffraction Synergy diffractometer with Mo Kα radiation, while data for 2 and 9 were collected at the MX2 beamline at the Australian Synchrotron [25].The temperature during data collection was maintained at 100.0(1) K, using an Oxford Cryosystems cooling device.The structure of each compound was solved by using direct methods and difference Fourier synthesis [26].Hydrogen atoms were placed in their idealized positions and included in subsequent refinement cycles.Thermal ellipsoid plots were generated in Mercury within the WINGX suite of programs [27,28].Crystallographic data for 2, 8, and 9 were deposited with the Cambridge Crystallographic Data Centre and assigned the CCDC deposit codes 2307709, 2294089, and 2307710, respectively.These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via http://www.ccdc.cam.ac.uk/data_request/ cif, accessed on 15 June 2023. Crystal

Cancer Cell Cytotoxicity Assays
Evaluation of compound cytotoxicity was performed as previously described, with minor modifications [29,30].MCF-7 and MDA-MB-231 cells were cultured in DMEM media supplemented with 10% heat-inactivated FBS (2500 cells and 2000 cells/50 µL/well seeded, respectively).DU145 cells were cultured in DMEM media supplemented with 10% heat-inactivated FBS (1000 cell/50 µL/well seeded).LNCaP cells were cultured in RPMI media supplemented with 10% heat-inactivated FBS (2000 cells/well seeded).Compounds (prepared at 20 mM stock concentration in DMSO) were evaluated using a 11-point assay concentration range from 0 to 50 µM.Assay controls included 0.4% DMSO negative control) and either 10% DMSO or 50 µM puromycin (final assay concentrations) (positive controls).Compounds were added 24 h after cell seeding into Greiner black-wall, clear-bottom 384-well cell culture plates and incubated for 72 h.After 66 h, resazurin was added to a final concentration of 60 µM, and samples were incubated for another 6 h.Fluorescence was monitored (excitation and emission wavelengths, 530 and 590 nm, respectively) using a

Figure 1 .
Figure 1.(a) Generalized process for the biosynthesis of O-PHDEs.Phenol (i) and catechol (ii) are brominated with Bmp5 (halogens omitted from iii-ix for clarity) prior to Bmp7 bromophenol homocoupling (iii-iv), bromophenol-bromocatechol heterocoupling (v-vii), or bromocatechol homocoupling (viii-ix)[8,9].Note that no O-PHDE with the oxygenation pattern described as ix has been reported to date.(b) Substitution patterns of all PHDEs isolated from Dysideidae sponges to date.Note that compounds with the oxygenation pattern III-3 cannot be generated solely through the biosynthetic process described in Figure1a, suggesting that post-translational oxidases or hydroxylases may play a role in this case[9,10].

Figure 1 .
Figure 1.(a) Generalized process for the biosynthesis of O-PHDEs.Phenol (i) and catechol (ii) are brominated with Bmp5 (halogens omitted from iii-ix for clarity) prior to Bmp7 bromophenol homocoupling (iii-iv), bromophenol-bromocatechol heterocoupling (v-vii), or bromocatechol homocoupling (viii-ix)[8,9].Note that no O-PHDE with the oxygenation pattern described as ix has been reported to date.(b) Substitution patterns of all PHDEs isolated from Dysideidae sponges to date.Note that compounds with the oxygenation pattern III-3 cannot be generated solely through the biosynthetic process described in Figure1a, suggesting that post-translational oxidases or hydroxylases may play a role in this case[9,10].

Scheme 1 . 1 .Figure 2 .
Scheme 1. Reaction conditions and reagents used to generate the semisynthetic library Scheme 1. Reaction conditions and reagents used to generate the semisynthetic library (2-13).

Table 1 .
NMR data of compound 8 in DMSO-d6 a .