Phylogenetic distribution of bromophenols in marine algae and the generation of a comprehensive bromophenol database

Marine algae are the source of a plethora of halogenated compounds, in particular brominated phenols, possessing various bioactivities. Since these natural products are typically unavailable commercially, isolation is usually indispensable for biological activity testing. However, targeted isolation may be challenging due to difficulties in identifying desired compounds via high-resolution LC–MS in crude extracts or fractions. While bromophenols have been extensively reviewed regarding their bioactivities, less attention has been given to their distribution and chemotaxonomic relevance among marine algae. Knowledge of the distribution of bromophenols may aid species identification and also point to species containing potentially novel compounds. To facilitate targeted and untargeted isolation of bromophenols from marine algae, an overview of the distribution and chemotaxonomic relevance of algal bromophenols considering recent phylogenetic findings is presented along with key analytical features of bromophenols relevant for mass spectrometric identification. Additionally, a comprehensive database listing brominated phenols from marine algae and their key analytical properties has been constructed.


Introduction
Marine algae are well known to be a rich source of halogenated and particularly brominated compounds and in recent years a number of reviews focusing on the bioactivities and potential therapeutic applications have been published (Liu et al. 2011;Jesus et al. 2019;Dong et al. 2020;Carpena et al. 2022). In contrast, chemotaxonomic considerations and analytical properties of bromophenols have only been discussed sparingly (Wang et al. 2013). This is of particular interest as the taxonomy and species assignments of a substantial number of the phytochemically investigated algae have changed considerably over the roughly 60 years since the first reports of bromophenols from algae (Hodgkin et al. 1966;Díaz-Tapia et al. 2017a;Cabrera et al. 2021). While the respective algae themselves and their secondary metabolites did not change, chemotaxonomic classifications have to be adapted to recent phylogenetic discoveries.
This review aims at presenting the phylogenetic distribution of bromophenols and at discussing their chemotaxonomic significance in marine algae. A special focus in addition to phylogeny and chemotaxonomy will be given to common analytical features of marine bromophenols, helpful in identifying these natural products via LC-MS in crude extracts or fractions. This is especially relevant since very few of these natural products are commercially available and isolation is almost always necessary for biological activity testing. To aid the identification of bromophenols by mass spectrometry we additionally assembled a comprehensive database (Table S1) comprising all bromophenols detected in marine algae until now. Regarding taxonomic classifications, we verified the current taxonomic status and name of each species and-if needed-updated the names used in the original publication to comply with the currently accepted denotation.

Data acquisition and generation of a comprehensive bromophenol database
The aim of the database generation was to cover every known bromophenolic compound produced by marine algae in Table S1. In order to identify relevant substances from literature, the databases ReaxysÒ and SciFinder n were searched by structural features. Additionally, certain genera were entered in Google Scholar together with the term ''metabolite'' or ''bromophenol''. Inclusion criteria for compounds were either isolation or unambiguous identification via MS and authentic reference compounds from a marine alga. The molecular feature for inclusion into the database was the presence of a hydroxylated and brominated benzene ring and molecular structures were depicted as reported in the original publications if the structural elucidation was convincing. Here, special attention was paid to the methods used for elucidation of stereochemistry. In case substance properties (state, color, absorption maxima, fragmentation or adduct formation in MS) had been reported, they were included in the database.
Structures, exact masses, molecular weight, sum formulae and SMILES were generated using Chem-Draw Ver. 21.0, PerkinElmer Informatics Inc. 2022 (Waltham, MA, USA). Since ESI-MS is one of the most common ionization methods today (Kind et al. 2018) and many bromophenols form salt adducts (see also 4.2 Key identifiers and common fragmentation reactions of bromophenols), the theoretical masses of the most common salt adducts ([M ? NH 4 ] ? , [M ? Na] ? and [M ? K] ? ) were included (Keller et al. 2008). Exact masses of the adduct ions were calculated using the exact adduct masses given by Keller et al. (2008) and the exact mass of each molecule. Sulfates are listed in protonated form.
The current names, families and information on the higher taxonomy of algae were taken from Algaebase (Guiry and Guiry 2022) if not indicated differently. To ensure traceability, those species for which name changes have occurred since the time of publication are given with the name mentioned in the original publication in brackets. The family was also added if the investigated alga is not a member of the Rhodomelaceae family. An overview of the changed species names is given in Table S2.

Phylogenetical distribution of bromophenols and their chemotaxonomic significance
Bromophenols are common metabolites in marine organisms and especially the most elementary brominated phenols (2-bromophenol, 4-bromophenol, 2,4dibromophenol, 2,6-dibromophenol and 2,4,6-tribromophenol; Fig. 1, type O) are found throughout green, brown and red algae . Due to the widespread distribution of these compounds, they represent no suitable phylogenetical identifiers for these clades. We summarized their distribution as reported in literature in Table S3 (Flodin et al. 1999;Whitfield et al. 1999;Flodin and Whitfield 2000;Chung et al. 2003;Kim et al. 2008) and their reported molecular characteristics are outlined in an extra spreadsheet in In contrast to these simple molecules, more complex bromophenols show intricate distribution patterns in regard of the species they have been isolated from. Intriguingly, certain compounds are found in almost all species of a genus (e.g., laurinterol, 5), some are found in different genera (lanosol, 6) but most compounds (233 of 279) have only been reported to be present in distinct species. The latter might to some extent also be attributed to the varying chemical composition of the seaweeds depending on the time of the year and collection site which has been reported for several algae (Flodin et al. 1999;Chung et al. 2003;Shoeib et al. 2006;Hofer et al. 2019). Another potential explanation for reports on varying contents of secondary metabolites was formulated by Howard et al. (1980) as they found samples of morphological identical Laurencia species from different collection sites to differ in their phytochemical composition. As these species were unable to reproduce, they suggested them to be sister species.
Red algae (Rhodophyta)-Rhodomelaceae Red algae (Rhodophyta) produce the largest diversity of brominated compounds comprising a great variety of scaffolds, as all structural skeletons shown in Fig. 1 except for the 4-bromo-2,5-dihydroxybenzyl (L) and the chrysophaentin type (M) can be found. Notably, every red alga reported to contain bromophenols belongs to the class Florideophyceae, one of the largest classes within the Rhodophyta (Guiry and Guiry 2022). However, it should be taken into account that other classes might be phytochemically understudied due to different availabilities. Since the Rhodomelaceae family is the largest family of red algae (Díaz-Tapia et al. 2017a) and it is reported to contain most bromophenols, we will discuss the bromophenols of its members with special attention. Furthermore, the phylogeny of several taxa within this family has been overridden in recent years (Díaz-Tapia et al. 2017a, b;Cabrera et al. 2021;Popolizio et al. 2022), revealing previously unanticipated relationships of chemotaxonomic relevance. The currently accepted tribes within the Rhodomelaceae and their subordinated species for which bromophenols have been reported are summarized in Table 1 together with the respective basic scaffolds (Díaz-Tapia et al. 2017a;Cabrera et al. 2021;Guiry and Guiry 2022).

Bostrychieae
Bostrychia calliptera For the genus Bostrychia, only three compounds from one study have been reported. All of the bromophenols isolated represented the 3-bromo-4-hydroxybenzyl type (A) with sulfated phenolic functions (Orfanoudaki et al. 2020) and have only been reported for B. calliptera. The simplest derivative is 2-bromo-4-(hydroxymethyl)phenyl hydrogen sulfate (7, Fig. 2) and the other two compounds contain a 2,3-dihydroxycyclopent-2-en-1-one side chain (8,9). Interestingly, the three investigated lineages of B. calliptera contained different mycosporine-like amino acid (MAA) and bromophenol patterns, with one lineage even completely lacking bromophenols (Orfanoudaki et al. 2020). Since the lineages were determined to be genetically different (Orfanoudaki et al. 2020), they could potentially be regarded as sibling species differentiated by their phytochemistry.

Amansieae
Halopithys incurva Halopithys incurva was shown to produce 3-bromo-4-hydroxybenzyl type (A) bromophenols (Glombitza et al. 1974;Wagner et al. 1981) by means of MS and isolation. While Glombitza et al. (1974) identified 2-bromo-4-(hydroxymethyl)phenol (10, Fig. 2) and 3-bromo-5-(hydroxymethyl)benzene-1,2-diol (11) via GC-MS, Wagner et al. (1981) could isolate 5-bromoprotocatechualdehyde (12) from an extract of H. incurva. In addition to these compounds, a substance showing an unusual substitution pattern (2-(2,6-dibromo-3,5-dihydroxyphenyl)acetic acid, 13) ). Finally, another study presented the isolation of two 3,5dibrominated bromophenols, however these compounds were not included in the database since the authors methylated the phenolic fraction and thus the genuine state of methylation cannot be inferred from the isolated compounds (Chantraine et al. 1973 As there are only a few reports of single substances found in H. incurva, further studies on this alga are needed to corroborate these findings, especially regarding the irregularly substituted 13. Rytiphlaea tinctoria Another member of the Amansieae, Rytiphlaea tinctoria, produces lanosol (B) and phloroglucinol type (K) bromophenols, however the isolated compounds were not included into the database since they were methylated before isolation (Chevolot-Magueur et al. 1976). As methylation also commonly occurs naturally in red algae (e.g., in Rhodomela converfoides, Li et al. 2011Li et al. , 2012, permethylation does not allow to detect the genuine methylation pattern of the natural product anymore. In case the compounds are easily degraded, acetylation might be a more feasible alternative (Glombitza and Gerstberger 1985). Interestingly, the lanosol type compounds in this study were isolated from mediterranean algae and the bromophloroglucinols were found in a specimen of the same species collected on the Breton coast (Chevolot-Magueur et al. 1976) which might either be an example for the dependency of the phytochemical composition on the collection site or for the aforementioned concept of sibling species (Howard et al. 1980).
Osmundaria obtusiloba Osmundaria obtusiloba (syn. Vidalia obtusiloba) produces lanosol type (B) bromophenols, and in total seven bromophenols have been isolated from this red alga (Table S1) (Wiemer et al. 1991;de Carvalho et al. 2006). Among those, a number of sulfated phenols including 4,7disulfolanosol (14, Fig. 2), two sulfated dimers (de Carvalho et al. 2006) and Vidalols A and B (15, 16) comprised of a lanosol (B) and a bromophloroglucinol (K) moiety have been reported (Wiemer et al. 1991). The presence of the latter might be a potential chemotaxonomic marker to distinguish Osmundaria and Vidalia species as Vidalia does not seem to contain bromophloroglucinols, although a systematic phytochemical evaluation of a larger subset of species from these genera would be needed to answer this question.
Vidalia colensoi As the genera Osmundaria and Vidalia are close relatives, it is not surprising that they contain very similar bromophenolic metabolites, for instance lanosol type (B) compounds that are otherwise uncommon in other Amansieae. In fact, the two bromophenol-containing species have both once been assigned to the respective other genus (Guiry and Guiry 2022). Six lanosol derivatives have been isolated from Vidalia colensoi (syn. Osmundaria colensoi) among them Colensolide A (17, Fig. 2) which bears a nitrogen-containing aliphatic cycle in its side chain. The authors proposed this side chain to be originating from an oxygenated histidine (Popplewell and Northcote 2009). Additionally, rhodomelol (18), another alicyclic compound with a highly oxidized hexahydrofuro[3,2-b]furan moiety was found in V. colensoi and also had been isolated from other Rhodomelaceae Lever et al. 2019) before. Since these uncommon side chains do not occur in Osmundaria they might also be potential chemotaxonomic markers to distinguish Vidalia and Osmundaria species.
Apart from these three major bromophenol producers, several other Laurencia species have been reported to contain bromophenolic metabolites. L. filiformis has been found to produce lauranes (H), among them the chromane type compound filiforminol (50) (Kazlauskas et al. 1976). Similarly, L. glandulifera also produces chromane type lauranes (H c ) in addition to cyclolauranes (Suzuki and Kurosawa 1979b), however no unique bromophenolic metabolite has been reported for this species. This is also the case for L. pacifica, L. johnstonii and L. nidifica that only produce dihydrobenzofuran type lauranes (H b ), dihydrobenzofuran type lauranes and cyclolauranes (H b , I) or regular lauranes and cyclolauranes (H, I) respectively (Crews and Selover 1986;Vairappan et al. 2001;García-Davis et al. 2018). Further halogenated aliphatic compounds have been isolated from L. pacifica (Selover and Crews 1980) but these were not covered by the inclusion criteria (aromatic brominated phenols) of this review. Comparable compounds to those occurring in L. pacifica have been found in L. obtusa which is also the only algal source of 2-bromo-3-hydroxy-15-nor-cuparan-10b-ol (51, Fig. 4) (König and Wright 1997;Kladi et al. 2006;Bawakid et al. 2017).
Despite the apparent diversity of sesquiterpenoid metabolites in the Laurencia genus, two species were reported to contain untypical bromophenol scaffolds: Tribrominated phenols (1,2,5-tribromo-3-(2-bromo-4,6-dimethylphenoxy)benzene and 2,3-dibromo-4,5dimethoxyphenyl)(2,3,6-tribromo-4,5-dimethoxyphenyl)methanone, 53, 54, Fig. 4) were found in L. similis together with highly brominated indoles (Qin et al. 2010;Sun et al. 2013). A possible explanation for the remarkable disparity between the occurrence of these scaffolds in L. similis and the more common scaffolds G-J in other Laurencia species might be a changed bromoperoxidase selectivity in this species with preference to indoles and polyphenols instead of sesquiterpene scaffolds. Sun et al. (2013) pointed out that a brominated tryptamine derivative might be a plausible precursor of the indole derivatives. However, as this is mere conjecture, further research on the precise biosynthetic pathways leading to the bromophenols found in L. similis is needed.
Additionally, Mikami and colleagues (2013) reported the isolation of another Laurencia-atypic bromophenol (3,5-dibromo-4-hydroxybenzaldehyde, 55, Fig. 4) from L. nipponica. But as this is the only report on this substance from L. nipponica and this alga has been reported to otherwise contain completely different non-bromophenolic metabolites of terpenoid origin (Abe et al. 1999), it remains uncertain whether 55 is a genuine metabolite of L. nipponica. An explanation for its occurrence could also be a carryover from Polysiphonia stricta, as this alga was investigated within the same study (Mikami et al. 2013) and it commonly contains 55 (Li et al. 2008b).
It seems worth mentioning that other Laurencia species have also been phytochemically evaluated, but as they do not contain any bromophenols as defined (molecules including a brominated phenol), they were not included in this review. Examples for such species are L. distichophylla (cuparane and laurane-like derivatives, halogenated in the side chain), L. mariannensis (halogenated aliphatic heterocycles) and L. dendroidea (syn. L. majuscula; non-aromatic chamigrane derivatives) (Blunt et al. 1984;Vairappan et al. 2001). As these three species do not belong to a common clade of non-bromophenol-producing species and considering the above-mentioned taxonomic uncertainty concerning the exact placement of each Laurencia species in the Laurencia complex (Cabrera et al. 2021;Popolizio et al. 2022), it does not seem appropriate to infer chemotaxonomic principles from these phenomenological findings.
Chondrophycus sp. Some of the species originally classified as Laurencia have today been reevaluated to be part of independent genera (Cabrera et al. 2021;Popolizio et al. 2022). One example is Chondrophycus intermedius (syn. Laurencia intermedia) which was reported to contain some laurane (H) and cyclolaurane (I) type bromophenols (Irie et al. 1970) (Fig. 1).
However, these findings were later corrected by the same group as they reinvestigated the used algal material and found that it actually was a mixture of Laurencia okamurai, Chondrophycus intermedius (syn. L. intermedia) and Palisada capituliformis (syn. L. capituliformis) (Suzuki and Kurosawa 1979a). Only the thalli of L. okamurai contained bromophenols, which is in line with other reports of Palisada and Chondrophycus species lacking bromophenols (Vairappan et al. 2001;Fujii et al. 2011;Sun et al. 2011) and emphasizes the unique metabolic abilities of the genus Laurencia.
Osmundea pinnatifida The only Osmundea species reported to contain bromophenols was previously regarded as a member of the genus Laurencia (Guiry and Guiry 2022). Osmundea pinnatifida (syn. Laurencia pinnatifida) is reported to contain a brominated filiformin (7-bromo-2-(bromomethyl)-5,8,10-trimethyl-2,3,4,5-tetrahydro-2,5-methanobenzo [b]oxepine, 56, Fig. 3) (Gonzalez et al. 1984), which also occurs in Laurencia caraibica and L. glandulifera (Izac and Sims 1979;Suzuki and Kurosawa 1979b). As described above, the closely related genus Chondrophycus does not contain any similar structures, which poses the interesting question if Osmundea developed the capability of synthesizing the laurane scaffold independently or-more likely-Chondrophycus lost the ability to synthesize such compounds. On the other hand, this is the only report of such a compound being isolated from an Osmundea species, making it difficult to assess the chemotaxonomic significance of this finding. In addition, a number of halogenated aliphatic Oheterocycles not covered in this review, have been isolated in the same study (Gonzalez et al. 1984).

Rhodomeleae
Rhodomela confervoides Among the non-parasitic genera of the Rhodomeleae (Rhodomela, Odonthalia, Neorhodomela and Beringiella) only the latter one has not been studied phytochemically and for the other three, a plethora of 3-bromo-4-hydroxybenzyl type (A) and lanosol (B) derivatives have been reported. Certain species also contain the scaffolds C-E (see Table 1), however these are usually represented to a lesser extent (Suzuki et al. 1980). Along with the tribes Streblocladieae, Polysiphonieae, Pterosiphonieae, Dipterosiphonieae, Pleurostichidieae, Herposiphonieae and Thaumatelleae, the Rhodomeleae belong to one of the two major subgroups of the Rhodomelaceae, the other one containing the above covered tribes Bostrychieae, Amansieae and Laurencieae (Díaz-Tapia et al. 2017a). Apart from the occurrence of a limited number of lanosol derivatives in the Amansieae, lanosols are mainly concentrated in the Rhodomeleae and the Streblocladieae.
A possible explanation for the considerable amount and diversity of bromophenols in R. confervoides might be the secondary modification of adding one or several lanosyl moieties to basic structures like amino acids (e.g. 236, Fig. 6), nucleosides (e.g. 258, Fig. 6) (Zhao et al. 2005) or one of the bromophenol scaffolds (e.g. 131, Fig. 5  ). However, the exact biosynthetic pathways leading to these compounds still need elucidation. This should especially be emphasized regarding the involved enzymes and mechanisms, as a majority of biosynthetic routes are based on hypothetical propositions made by the original authors despite a lack of experimental data (Wang et al. 2013). One example for an at least partially characterized biosynthetic route is the biogenesis of the 3-bromo-4-hydroxybenzyl type bromophenols starting from tyrosine (Manley and Chapman 1978). Additionally, the authors proposed that this pathway might also be a potential source of lanosol (Manley and Chapman 1978).

Isovanillin derived
Dimers elements, sulfur is incorporated into a marine organosulfur cycle where it can be found in several intermediate metabolites (Thume et al. 2018). Dimethyl sulfoxide (DMSO), one of these metabolites, is very common in sea water and microalgae (Lee and de Mora 1999). Its formation is facilitated by oxidation of zwitterionic organic acids with a cationic sulfonium or sulfoxonium moiety (Lee and de Mora 1999;Thume et al. 2018). Similar to these cationic metabolites, isolation of a dimethylsulfonium lanosol derivative (70) from V. lanosa was recently reported (Jacobtorweihen et al. 2022) which might be a possible precursor of the sulfoxide derivative (68) found in R. confervoides (Zhao et al. 2004). To confirm this hypothesis, the respective sulfonium or sulfoxide derivative along with the enzymatic pathways leading to these molecules had to be identified in R. confervoides or in V. lanosa, respectively. However, the occurrence of each of the sulfur containing bromophenols is a hint for the existence of pathways attaching sulfur to organic molecules in red algae similar to those reported for microalgae (Thume et al. 2018).

3-bromo-4-hydroxybenzyl type (A)
Neorhodomela sp. Neorhodomela species are very closely related to the genus Rhodomela which is reflected by their morphology, geographical distribution and also by their phytochemical composition Park et al. 2012;Díaz-Tapia et al. 2017a). Two species (N. aculeata and N. larix, formerly Rhodomela larix) have been investigated regarding their secondary metabolites (Masuda 1982;Guiry and Guiry 2022 3-bromo-4-hydroxybenzyl (A) and lanosol derivatives (B) and occasionally 3,5-dibromo-4-hydroxybenzyl type (C) or 3,6-dibromo-4,5-dihydroxybenzyl type (D, only N. aculeata) bromophenols (Kurata and Amiya 1977;Suzuki et al. 1980;Park et al. 2012;Mikami et al. 2016) (Fig. 1). Potential species-specific markers for N. aculeata and N. larix are 3,6-dibromo-4-(methoxymethyl)benzene-1,2-diol (71) and 2-bromo-4-(methoxymethyl)phenol (72), respectively (Suzuki et al. 1980;Park et al. 2012) (Fig. 7). Notably, 4,7disulfolanosol (14, Fig. 2) has only been isolated from N. larix and might be considered a potential discriminative marker between the two Neorhodomela species. However, for N. aculeata only organic solvent partitions of the respective methanol extracts have been examined so far (Park et al. 2012;Mikami et al. 2016), but not the composition of the respective polar phases. In this context, investigations of the more polar constituents in general have the potential to yield novel compounds and might also shed light on the hypothesis formulated by Weinstein et al. (1975) who suggested the bromophenols genuine to the algae to be sulfated, and desulfation occurring only during extraction or isolation. Moreover, the last report on phytochemical investigations of N. larix dates back to the 1980s (Carlson et al. 1989) and reinvestigating this alga using state-of-the-art analytical methods could provide further insights to its secondary metabolites. Since the close relative Rhodomela confervoides is the source of fourfold as many metabolites, further phytochemical investigations on N. larix seem promising.
All of these structural analogies suggest that both species possess similar ways of modifying the basic bromophenol scaffold to generate a variety of secondary metabolites. The effectiveness of these variations is well illustrated by the enormous numbers of bromophenols both species produce (48 bromophenols for S. latiuscula and 59 for R. confervoides), the majority of which is unique. From a biogenetic point of view, the major factors for variation in bromophenols in these species seem to be either dimerization or linkage to primary metabolites like amino acids, nucleosides or organic acids as suggested for the aconitic acid residue in the symphyocladins (Xu et al. 2017). However, as previously mentioned, the exact biosynthetic routes leading to the plethora of bromophenols from R. confervoides and also S. latiuscula are unknown and further research is needed to fully understand the function and biosynthesis of these compounds.
Pterosiphonia complanata While S. latiuscula is very well studied and a large number of compounds have been isolated from it, only one other member of the Pterosiphonieae has been reported to contain bromophenols. Pterosiphonia complanata was also found to produce a 2,3,6-tribrominated compound (3,4,6-tribromo-5-(methoxymethyl)benzene-1,2-diol, 79, Fig. 8) (Etahiri et al. 2007) which shows that the occurrence of this bromination pattern might be a characteristic of the entire tribe Pterosiphonieae. However, further research on additional species from this particular tribe is needed to verify this hypothesis.
One of the most striking features in the phytochemistry of Polysiphonia species is the complete absence of 2,3-dibrominated lanosol derivatives. This is especially surprising as most of the closer related tribes (Streblocladieae, Rhodomeleae and Pterosiphonieae) produce 2,3-dibrominated (B) or 2,3,6tribrominated 3,4-dihydroxybenzyl (E) type compounds Li et al. 2012;Xu et al. 2017;Hofer et al. 2019). However upon closer inspection, the lack of lanosol derivatives supports the recent regrouping of this section of the Rhodomelaceae quite well (Díaz-Tapia et al. 2017a;Savoie and Saunders 2019) and it might-in addition to other identifying features (Savoie and Saunders 2019)serve as a characteristic for P. stricta and P. morrowii. On the other hand, only a small portion of Polysiphonia species has been studied regarding their phytochemistry and it might therefore be too early to infer universal chemotaxonomic rules from the phytochemistry of just two species.
Melanothamnus sp. As with Carradoriella and Leptosiphonia, both phytochemically investigated Melanothamnus species had been classified as Polysiphonia species until recently (Díaz-Tapia et al. 2017b). For Melanothamnus ferulaceus (syn. Polysiphonia ferulacea), lanosol-type bromophenols (B) are described, whereas M. sphaerocarpus (syn. Polysiphonia sphaerocarpa) was found to contain simple brominated phenols (O). For members of the Streblocladieae, the presence of the lanosol-type bromophenols is not surprising as these are also present in other members of this tribe (e.g., Vertebrata, Leptosiphonia). The compounds described for M. ferulaceus are lanosol (6), its methyl ether (58) and a unique 3-bromo-4hydroxybenzyl type (A) compound (polysiphenol, 122, Fig. 10) (Aknin et al. 1992). Interestingly, polysiphenol represents a regioisomer of 4,7dibromo-9,10-dihydrophenanthrene-2,3,5,6-tetraol (114, Fig. 10) found in Polysiphonia stricta (Li et al. 2008b). As both molecules are probably formed via linkage of the benzylic carbons of two 3-bromo-4hydroxybenzyl alcohol derivatives with an additional b Fig. 10 Selected characteristic bromophenols found in the Polysiphonieae and the Streblocladieae. Compounds are grouped according to the respective genus they occur in. Numbering according to occurrence in Table S1. Compound names are only displayed if trivial names were reported in literature C-C bond formed between the two arenes, it seems likely that they share the same biogenetic origin and only differ in the rotation of one of the benzene rings.
Since this is the only study focusing on the phytochemistry of M. ferulaceus, further investigations are needed to elucidate the phylogenetical implications of the occurrence of these metabolites.
Another peculiar phenomenon in this genus is the occurrence of brominated basic phenols, cresols and anisoles in M. sphaerocarpus (Flodin and Whitfield 2000) as these types of structures are very different from the bromophenols found in other members of the Rhodomelaceae. Since these compounds were however only detected in smaller amounts compared to other species, they could also be a sign of contamination with bacteria or fungi, as the authors of the original publication already pointed out in regard of the O-methylated anisoles (Flodin and Whitfield 2000).
As the term cresol is ambiguous and the original publication does not include any structural formulas, the reported cresols were not included in the database. Additionally, their assignment of 3-bromo-4-hydroxybenzaldehyde (1) and 3,5-dibromo-4-hydroxybenzaldehyde (55) in the extract solely based on MS data is tentative and the detected compounds might represent isomers of the assigned compounds (e.g. a 2,3dibrominated derivative for 55). A biogenetic assignment of the substitution pattern of these compounds does not seem to fit either as M. sphaerocarpus is not classified as a member of the Polysiphonia genus any longer and the members of the Streblocladieae produce both 3,5-dibrominated and 2,3-dibrominated phenols (Suzuki et al. 1980;Jacobtorweihen et al. 2022). The anisole derivatives however were included since they were identified by GC-MS and comparison to authentic reference compounds (Flodin and Whitfield 2000). Once again, further studies on this species are needed to corroborate the findings reported by Flodin and Whitfield (2000), especially regarding the bromination pattern of the aldehydes.
Vertebrata sp. Just as the other members of the Streblocladieae discussed before, most Vertebrata species were previously classified as Polysiphonia sp. until recent molecular genetic findings defined them to be members of a separate clade within the Rhodomelaceae (Díaz-Tapia et al. 2017a, b). In total, six Vertebrata species have been investigated regarding their bromophenols with V. lanosa being the most extensively studied species. For the remaining species, only single reports on their phytochemistry exist (V. decipiens) and some were even only screened by GC-MS (V. thuyoides, V. fruticulosa, V. fucoides and V. nigra) (Glombitza et al. 1974;Lever et al. 2019). Most compounds found in Vertebrata species are lanosol-derived (B) with varying side chains (Stoffelen et al. 1972;Hofer et al. 2019;Jacobtorweihen et al. 2022). Some side chains are of bromophenolic origin (dimers 72 and 73), some amino acid derived (e.g., vertebratol, 123, Fig. 10) and other ones comprise unsusual heterocycles (e.g., rhodomelol, 18) Hofer et al. 2019). Additionally, a number of sulfated derivatives has been reported (14, 70, 124 and 125) (Jacobtorweihen et al. 2022) (Fig. 10). Only certain compounds deviate from this general bromination pattern (2,6-dibromo-4-(hydroxymethyl)phenol and 3,5-dibromotyrosine, 121, 126, Fig. 10) and they are only found in V. lanosa and V. fucoides (Glombitza et al. 1974;Jacobtorweihen et al. 2022). However, as none of the species has been investigated to the same extent as V. lanosa it is likely that closer analysis of their brominated metabolites might lead to the discovery of further 3,5-dibrominated compounds.
V. lanosa itself contains a variety of unique compounds of which rhodomelol (18) and methylrhodomelol (127) have been known longest   (Fig. 10). Both of these substances comprise a lanosyl moiety linked to a cyclic ascorbyl residue and only differ in the methylation of a single hydroxyl function. A similar non-brominated scaffold, delesserine, has been isolated from another red alga, Delesseria sanguinea (Yvin et al. 1982) and rhodomelol and its lanosylated derivative polysiphonol (128) have been found in V. decipiens (Lever et al. 2019). It should be noted that methylrhodomelol does not represent an artifact from methanolic extraction as it was also detected in non-methanolic extracts .
More recently, the isolation of several sulfated bromophenols from the aqueous phase of a methanolic V. lanosa extract has been described (Jacobtorweihen et al. 2022). Some of these represent sulfated forms of previously reported molecules (5-sulfovertebratol, 124), some might be precursors in the biosynthesis of the lanosol or 3-bromo-3,4-dihydroxybenzyl alcohol derivatives (3-bromo-5-sulfo-L-dihydroxyphenylalanine, 125) and others contain the aforementioned sulfonium moiety (4-sulfo-7dimethylsulfonium lanosol, 70; see paragraphs on Rhodomela confervoides and Symphyocladia latiuscula) (Fig. 10). As mentioned in the section about Neorhodomela, the investigation of the polar compounds of algal extracts might corroborate the ''genuine-sulfate-hypothesis'' by Weinstein et al. (1975) as suggested by the presence of the sulfated vertebratol. On the other hand, sulfation or desulfation could potentially be influenced by other factors like seasonal variation or the site of collection. Notably, sulfurcontaining and ureido side chains are solely occurring in the tribes Rhodomeleae, Pterosiphonieae and Streblocladieae which all belong to a well-defined clade within the Rhodomelaceae.

Red algae other than Rhodomelaceae
In contrast to the bromophenol-rich Rhodomelaceae, other families of red algae have only rarely been reported to contain more complex bromophenolic compounds. Just like the Rhodomelaceae, all of them (except the Corallinales) belong to the subclass Rhodymeniophycidae in the class Florideophyceae (Yang et al. 2015;Guiry and Guiry 2022). However, no chemotaxonomic principles can be inferred from the available data as most species only produce few characteristic compounds (Callophycus), investigations are scarce (Rhabdonia, Grateloupia and Polyopes) and certain species might not even be the actual producers of the bromophenols (Gracilaria coronopifolia, Ceramium tenuicorne) (Nagai et al. 1997). An overview of the reported skeletons is provided in Table 2 and selected characteristic compounds are shown in Fig. 11.
Numbering according to occurrence in Table S1. Compound names are only displayed if trivial names were reported in literature iodocallophycols A-C (140-142) were isolated from a non-identified species of the genus Callophycus (Lavoie et al. 2017), while all other compounds were reported from C. serratus (Lane et al. 2007;Woolner et al. 2018).

Ceramiales
Ceramium tenuicorne Malmvärn et al. (2005Malmvärn et al. ( , 2008 reported the occurrence of polybrominated diphenyl ethers in the red alga Ceramium tenuicorne (Ceramiaceae). As they methylated the phenolic fraction before analysis via MS and thus the genuine methylation state could not be inferred from the available data, these compounds were not included in the database. Additionally, in a subsequent mass spectrometric study, the respective substances were eventually detected in cyanobacteria, supporting their initial hypothesis that these compounds did not originate from algae (Malmvärn et al. 2008).

Gracilariales
Gracilaria coronopifolia The edible red alga Gracilaria coronopifolia (Gracilariaceae) has been reported to cause cases of food poisoning in Hawaii which has been attributed to manauealides A, B and aplysiatoxin (Nagai et al. 1997). These compounds are macrolides and the bromophenols manauealide B (145, Fig. 11) and aplysiatoxin (146) were included in the database. However, it should be obvious that the bromophenol moiety in these molecules is not their most striking feature. Additionally, Nagai and colleagues (1997) suggested the isolated compounds to be of epiphytic origin, which could explain why they show only little structural resemblance to secondary metabolites from other algae.
C. aberrans on the other hand has been reported to contain laurane (H) and cyclolaurane type (I) bromophenols, namely laurinterol, isolaurinterol, aplysin, aplysinol and aplysinal (5, 22-24 and 43) (Ohta and Takagi 1977). As previously discussed, these compound types are typical for Laurencia species and the authors of the original publication pointed out that several Laurencia species grew in close vicinity to the collection site of their algal material and the isolated compounds might therefore be ''derived'' from them (Ohta and Takagi 1977).

Brown algae (Phaeophyceae)
Despite the major occurrence of bromophenols in red algae, many brown algae also contain simple bromophenols (O, see Tables S1 and S3). Some are reported to possess phloroglucinol derivatives (K, Cystophora retroflexa and Eisenia arborea) and one species (Leathesia marina) has even been reported to contain 3-bromo-4-hydroxybenzyl (A), lanosol (B) and 3,5-dibromo-4-hydroxybenzyl derivatives  (Sailler and Glombitza 1999;Xu et al. 2004a. A potential chemotaxonomic marker for the investigated brown algae might be a C-O linkage between the respective phenolic subunits (see C. retroflexa and L. marina). For an overview of the structures from brown algae mentioned in the text, see Fig. 12.

Sargassaceae
Cystophora retroflexa Cystophora retroflexa (syn. Cystophora congesta, Sargassaceae) produces phloroglucinol derived bromophenols (K) that are composed of two to four monobrominated, monochlorinated or regular phloroglucinol monomers (148-157) (Sailler and Glombitza 1999). The simplest representative (2-(2-bromo-3,5dihydroxyphenoxy)benzene-1,3,5-triol, 148, Fig. 12) has first been isolated by Koch and Gregson (1984) and the exact position of the bromine substituent was refined later by Sailler and Glombitza (1999). One remarkable feature of these phloroglucinol oligomers is the linkage between the subunits as the phenolic function of one subunit connects to a previously nonphenolic aromatic carbon of the next subunit, leading to a pyrogallol-like substitution pattern. This structural element is uncommon in other algal bromophenols except for Odonthalia corymbifera (76, Fig. 7) and might be a potential chemotaxonomic marker for C. retroflexa. Moreover, all ten bromophenols isolated from C. retroflexa seem to be unique to this alga.
The discussed bromophenols have originally been isolated after acetylation but since acetylation is rather uncommon in algal secondary metabolites, we assume the genuine compounds to be phenolic and included them in the database as the respective non-acetylated phenols. In addition to the included bromophenols, further chlorinated compounds have been isolated (Sailler and Glombitza 1999).

Lessoniaceae
Eisenia arborea Comparable to C. retroflexa, Eisenia arborea (Lessoniaceae) produces phloroglucinol derivatives (K). Apart from the basic bromophloroglucinol (134, Fig. 12), a brominated phloroglucinol trimer with a dibenzodioxin substructure (4'-bromoeckol, 158) has been reported (Glombitza and Gerstberger 1985). The dibenzodioxin substructure is probably formed secondarily via cyclisation of the phenolic phloroglucinol function. As for C. retroflexa, also these compounds were isolated after acetylation. Since this is the only report on bromophenols from E. arborea, further studies are needed to evaluate the chemotaxonomic relevance of the isolated compounds.
Compounds are grouped according to the respective genus they occur in and numbered according to occurrence in Table S1.
To fully clarify the occurrence of lanosol and 3,5dibromo-4-hydroxybenzyl type bromophenols in L. marina further studies on its phytochemistry and the specific epiphyte population of the investigated specimen are needed.

Green algae (Chlorophyta)
Just as the other two classes of algae, a number of green algae have been reported to produce simple bromophenols (O) (Flodin et al. 1999;Whitfield et al. 1999), whereas some genera tend to produce more complex bromophenols (Avrainvillea and Cymopolia). Avrainvillea species are the source of a few unique 3-bromo-hydroxybenzyl type (A) compounds, while Cymopolia barbata produces a variety of unique 4-bromo-2,5-dihydroxybenzyl type (L) bromophenols (Carte et al. 1989;Dorta et al. 2002). Since most of the bromophenols from these genera are exclusive to them, they might serve as potential chemotaxonomic markers. For an overview of the structures from green algae mentioned in the text, see Fig. 12.

Miscellaneous marine algae
Pelagomonadaceae Chrysophaeum taylorii Related to the large class Phaeophyceae (brown algae) a small clade called Pelagophyceae exists, which contains the microalga Chrysophaeum taylorii (Pelagomonadaceae) (Guiry and Guiry 2022). This organism is the source of a number of exceptional halogenated compounds, eight of which represent bromophenols (Plaza et al. 2010;Davison and Bewley 2019). While the hemichrysophaentins B-D (176-178, Fig. 12 and Table S1) possess a diaryl butene scaffold, which is hydroxylated and halogenated in several positions (chrysophaentin type, M), chrysophaentins B-D, G and H (179-183, Fig. 12 and Table S1) represent dimerized hemichrysophaentins. Since all of these compounds are unique to C. taylorii and it is the only species of its genus, the phytochemical findings underline the taxonomic position of this alga.

Methods
Over the roughly 60 years of research on halogenated marine secondary metabolites, the means to analyze small molecules have developed tremendously which is also reflected in the methods used in publications reporting bromophenols from algae. While older investigations mainly used EI as the ionization method for the identification of bromophenols (Glombitza et al. 1974Blackman and Matthews 1982;Kurata et al. 1997), in more recent reports, the predominant ionization method is ESI (Popplewell and Northcote 2009;Xu et al. 2012a;García-Davis et al. 2018;Hofer et al. 2019). This is especially true for high resolution mass spectrometry (HR-MS) and reflects a general trend in small molecule mass spectrometry (Kind et al. 2018). Based on the methods summarized in the database appended to this review, to this day, both methods have been used to a similar extent in the analysis of marine bromophenols (EI 116 reports, ESI 123 reports). However, it is probable that the share of publications using ESI in this field will soon surpass the ones using EI.

Key identifiers and common fragmentation reactions of bromophenols
Identification of small molecules via MS generally relies on HR-MS data and subsequent generation of sum formulae (Léon et al. 2019) and for this to be successful, the existence of heteroatoms has to be considered. In case of bromine, its presence can easily be inferred from the isotopic pattern of the compound as the two stable bromine isotopes are approximately distributed evenly (Br 79 /Br 81 ca. 50.7/49.3%) (Rosman and Taylor 1998). Several bromine atoms in a molecule lead to more complex isotopic patterns, for instance a molecule containing four bromine atoms will show a cluster of five signals with an intensity ratio of 1:4:6:4:1 (Bright and Chen 1983). The theoretical ratios for any number of bromine atoms can be inferred from Pascals triangle (Meija 2005). Additionally, many compounds tend to easily form debrominated fragments, which can be exemplified by (E)-4-(2-bromo-4,5-dihydroxyphenyl)but-3-en-2-one (184, EI, m/z 177 [M-Br] ? ) and 2'-lanosyl-3'bromo-5',6'-dihydroxybenzyl alcohol (131, (?)ESI, m/z 403/401/399 [M-H 2 O-HBr] ? ) (Li et al. 2011;Jacobtorweihen et al. 2022) (Fig. 13a). This phenomenon can also be observed for halogens like iodine (iodoallolaurinterol, 19, ionization method presumably EI, m/z 295/293 [M-I]) (Izac and Sims 1979). To detect the chlorine atoms in chrysophaentins, Plaza et al. (2010) utilized an increased cone voltage in (-)ESI-MS to realize in-source fragmentation of the chlorinated compounds and therefore confirmed the respective number of chlorine atoms. This approach might be helpful in the analysis of compounds substituted with different halogens since the exact number of each halogen may be difficult to assign from the overlapping isotopic patterns.
The presence of other heteroatoms (e.g., sulfur or phosphorus) usually has to be presumed from fragments and may in case of unknown substances prove to be challenging. In bromophenols-apart from a few exceptions (68)(69)(70)(111)(112)Figs. 6,7 and 10)sulfur mostly occurs in sulfates and may consequently be detected by the exact mass difference of the [M-SO 3 ] fragment (Dm/z 79.9568) to the parent ion, or the sulfate ion itself (m/z 97.9674) (Ma et al. 2006;Keller et al. 2008). Sulfate ions have been reported for a series of dihydroxyphenethyl hydrogen sulfates (185-187, Fig. 13a) in negative mode ESI-MS (Ma et al. 2006 Table S1. Compound names are only displayed if trivial names were reported in literature. b Prominent benzylic cation fragments of Vidalol A (15) in EI-MS (Wiemer et al. 1991). The dibrominated fragment at the bottom is also commonly formed by other lanosol derivatives the respective fragments and neutral losses during analysis of aqueous extracts and fractions since sulfated bromophenols are mainly extracted by polar solvents.
In addition to fragments, most organic chemicals form a number of salt adducts and sometimes multimers which can make up the most intensive signals in a mass spectrum, and the analysis of such spectra may even pose a challenge for experienced scientists (Jaeger et al. 2017). For bromophenols, the most frequent reported salt adducts are sodium adducts (32 reports), followed by multimeric adducts comprising two molecules (10 reports). Ammonium and potassium adducts have also been reported occasionally (9 and 3 reports, resp.). As adduct formation also offers structural information, their assignment is critical for the determination of the correct sum formula, especially in unknown metabolites and this may be achieved either manually via the exact mass difference or by using modern software tools (Keller et al. 2008;Jaeger et al. 2017).
Unrelated to mass spectrometry, another shared analytical feature of bromophenols is their UV absorption maximum between 280 and 300 nm.

Conclusion
Aim of this review was to present the distribution of bromophenols among marine algae considering their role as taxonomic markers. Within the phytochemically well-studied family Rhodomelaceae which was subject to recent taxonomic reclassifications, the assignment of new phyla in particular correlates well with an accumulation of specific bromophenol subtypes. For example, the absence of lanosol derivatives (scaffold B), but frequent occurrence of 3,5-dibrominated compounds (scaffold C) in the genus Polysiphonia, in contrast to other tribes such as Rhodomeleae and Streblocladieae, support the phylogenetic classification. Similarly, the predominant occurrence of cuparane, laurane, cyclolaurane and laurokamurane type bromophenols in the genus Laurencia further corroborates the distinction of this genus from the other members of the Laurencieae and from other tribes of the Rhodomelaceae. However, for families of red, brown and green algae that are less frequently studied, it remains difficult to make supported statements about their chemotaxonomic relation due to the low number of bromophenolrelated reports for the respective species.
In summary, the review is intended to contribute to an improved perception of the chemodiversity in algae while the bromophenol database is provided as a tool to facilitate further phytochemical studies on marine algae, as many genera await further exploration.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.