Dimeric (Poly)Hydroxynaphthazarins, Metabolites of Echinoderms and Lichens: The History of the Synthesis and Structure Elucidation

This review provides information on the synthesis and revision of the structures of natural dimeric (poly)hydroxynaphthazarins, metabolites of echinoderms and lichens, and on the refinement of the direction and mechanism of reactions in the synthesis of some of these compounds.


Introduction
Compounds based on the structure of 1,4-naphthoquinone are widespread in natural objects [1][2][3][4][5][6][7][8][9]. Among thousands of its derivatives, including dimeric and trimeric products [1][2][3][10][11][12], derivatives of naphthazarin (5,8-dihydroxy-1,4-naphthoquinone) occupy a very prominent place [13][14][15][16][17]. Among them, dimeric (poly)hydroxynaphthazarin metabolites of echinoderms and lichens constitute a relatively small but structurally diverse group of biologically active natural compounds. The first representatives of this series of compounds were isolated in the early 1970s, yet little information on the isolation of new dimeric products of this group, their synthesis, and correction of the structures of previously isolated compounds appears in print to this day. Our research initially concerned the synthesis of compounds of this group, with the aim of obtaining them in quantities required for biotests. However, it soon became clear that the real structures of some of the compounds did not correspond to the declared ones. This is primarily due to the imperfection of the physicochemical equipment that existed at the time of the study and the lack of or incorrect interpretation of the information available. It should be noted that ignorance of the exact structures of substances makes it impossible to create structureactivity correlations and, therefore, a targeted search for substances with desired properties. In this situation, synthesis, in addition to a supplier of substances with a given structure, plays the role of a reliable tool for its analysis. This review provides information on the synthesis of natural dimeric (poly)hydroxynaphthazarins, metabolites of echinoderms and lichens, as well as some information on the revision of their structure and the mechanism of formation. In this way, this review differs from the recently published [18,19], in which significant attention was paid to the isolation and structures of spinochromes, as well as the assessment of their biological activity, the parent organisms, and the methods used for isolation and identification. In addition, attention was paid to the study of the biosynthesis of spinochromes and the ecological function, stability, and chemical synthesis of (poly)hydroxynaphthazarins.
When analyzing the structures of the above compounds, we can conclude that all, formally, are the end products of three types of reactions. The first subgroup consists of dimeric (poly)hydroxynaphthazarins, which are the end products of the aldol condensation  (4) xanthentetraones. In contrast to [20], all quinonoid compounds are depicted as derivatives of 1,4-naphthoquinone.  (4) xanthentetraones. In contrast to [20], all quinonoid compounds are depicted as derivatives of 1,4-naphthoquinone.
Thus, the structure of the dehydrated ethylidene-bis(trihydroxy-naphthazarin) derivative 1 remained not fully elucidated, and therefore it is not mentioned in the fundamental monography by Thomson, R. H. [2]. At the same time, some authors referred to it as proven [21], and only in 2017 was complete spectral information on the structure of dibenzo[b,i]xanthetetraone 2 [22]. Recently, mirabiquinone (1H-dibenzo[b,h]xanthenetetraone, 3) was isolated from the sea urchin Scaphechinus mirabilis ( Figure 2) [23], which was previously considered as an alternative to 5H-dibenzo[b,i]xanthenetetraone 2 isolated from the sea urchin S. purpureus [20]. In the IR spectrum of mirabiquinone (CHCl3), there is one absorption band of carbonyl at 1626 cm −1 . Comparison of the IR spectra of the anhydro derivative 2, mirabiquinone (3), and the cyclization product of ethylidene bis(trihydroxynaphthazarin) 1, under the action of concentrated sulfuric acid [20], made it possible to establish that the latter is a mixture of dibenzo[b,i]xanthetetraones 2 and 3.
Mirabiquinone (3) and two related binaphthazarins 1 and 2 demonstrated excellent scavenging of the 2,2-diphenyl-1-picrylhydrazyl radical [23]. However, these compounds are not easily accessible on a preparative scale for extended bioassays due to their very low natural abundance and separation difficulties. Therefore, the question of the synthesis of these compounds for biotesting has become pertinent.
The key stage in the synthesis of biquinone 1 was the aldol condensation of spinochrome D dimethyl ether 5 and acetaldehyde (Scheme 1) [24]. Demethylation of tetramethyl ether 6a by the action of AlCl3 in nitrobenzene gave ethylidene-bis(thrihydroxynaphthazarin) 1 in good yield [25]. An attempt of demethylation of tetramethyl ester 6а by the action of conc. HBr yielded 5H-dibenzo[b,i]xanthetetraone 2, previously isolated from the sea urchins S. purpureus [20] and S. droebachiensis [21], and mirabiquinone A (3), a metabolite of the sea urchin Scaphechinus mirabilis [23], in a ratio of 1.2:1. It is obvious that the cyclization of ethylidene-bis(thrihydroxynaphthazarin) 1 and its derivatives is a key step for the preparation of 5H-dibenzo[b,i]-(2) and 1H-dibenzo[b,h]-(3) xanthenetetraone. It has been found that the boiling of tetra-(6a), Thus, the structure of the dehydrated ethylidene-bis(trihydroxy-naphthazarin) derivative 1 remained not fully elucidated, and therefore it is not mentioned in the fundamental monography by Thomson, R. H. [2]. At the same time, some authors referred to it as proven [21], and only in 2017 was complete spectral information on the structure of dibenzo[b,i]xanthetetraone 2 [22].
Recently, mirabiquinone (1H-dibenzo[b,h]xanthenetetraone, 3) was isolated from the sea urchin Scaphechinus mirabilis ( Figure 2) [23], which was previously considered as an alternative to 5H-dibenzo[b,i]xanthenetetraone 2 isolated from the sea urchin S. purpureus [20]. In the IR spectrum of mirabiquinone (CHCl 3 ), there is one absorption band of carbonyl at 1626 cm −1 . Comparison of the IR spectra of the anhydro derivative 2, mirabiquinone (3), and the cyclization product of ethylidene bis(trihydroxynaphthazarin) 1, under the action of concentrated sulfuric acid [20], made it possible to establish that the latter Is a mixture of dibenzo[b,i]xanthetetraones 2 and 3.
Mirabiquinone (3) and two related binaphthazarins 1 and 2 demonstrated excellent scavenging of the 2,2-diphenyl-1-picrylhydrazyl radical [23]. However, these compounds are not easily accessible on a preparative scale for extended bioassays due to their very low natural abundance and separation difficulties. Therefore, the question of the synthesis of these compounds for biotesting has become pertinent.
The key stage in the synthesis of biquinone 1 was the aldol condensation of spinochrome D dimethyl ether 5 and acetaldehyde (Scheme 1) [24]. Demethylation of tetramethyl ether 6a by the action of AlCl 3 in nitrobenzene gave ethylidene-bis(thrihydroxynaphthazarin) 1 in good yield [25]. An attempt of demethylation of tetramethyl ester 6a by the action of conc. HBr yielded 5H-dibenzo[b,i]xanthetetraone 2, previously isolated from the sea urchins S. purpureus [20] and S. droebachiensis [21], and mirabiquinone A (3), a metabolite of the sea urchin Scaphechinus mirabilis [23], in a ratio of 1.2:1. Thus, the structure of the dehydrated ethylidene-bis(trihydroxy-naphthazarin) derivative 1 remained not fully elucidated, and therefore it is not mentioned in the fundamental monography by Thomson, R. H. [2]. At the same time, some authors referred to it as proven [21], and only in 2017 was complete spectral information on the structure of dibenzo[b,i]xanthetetraone 2 [22]. Recently, mirabiquinone (1H-dibenzo[b,h]xanthenetetraone, 3) was isolated from the sea urchin Scaphechinus mirabilis ( Figure 2) [23], which was previously considered as an alternative to 5H-dibenzo[b,i]xanthenetetraone 2 isolated from the sea urchin S. purpureus [20]. In the IR spectrum of mirabiquinone (CHCl3), there is one absorption band of carbonyl at 1626 cm −1 . Comparison of the IR spectra of the anhydro derivative 2, mirabiquinone (3), and the cyclization product of ethylidene bis(trihydroxynaphthazarin) 1, under the action of concentrated sulfuric acid [20], made it possible to establish that the latter is a mixture of dibenzo[b,i]xanthetetraones 2 and 3.
Mirabiquinone (3) and two related binaphthazarins 1 and 2 demonstrated excellent scavenging of the 2,2-diphenyl-1-picrylhydrazyl radical [23]. However, these compounds are not easily accessible on a preparative scale for extended bioassays due to their very low natural abundance and separation difficulties. Therefore, the question of the synthesis of these compounds for biotesting has become pertinent.
As suggested by the authors of [20,21], the cyclization of ethylidene-bis(trihydroxynaphthazarin) 1 in concentrated sulfuric acid produces anhydrous derivative 2. To obtain some amount of this substance for biological testing, a replica of this experiment was conducted. Surprisingly, it was found that, under the described conditions, ethylidene-bis(trihydroxynaphthazarin) 1 undergoes cyclization with the formation of mirabiquinone (3) instead of 5H-dibenzo[b,i] xanthenetraone 2 in good yield [26]. This suggests that UV-, IR-spectroscopy, and TLC are not reliable enough at establishing structures of polyhydroxynaphthazarins, even by the comparison method. For example, these methods cannot distinguish between bisnaphthazarin 1 and related naphthazarins 2 and 3, which were considered in the cited work [20].

Oxidative Coupling Compounds
Other representatives of this group of compounds are dimeric (poly)hydroxynaphthazarins 8-10 [27][28][29], which are products of the oxidative C-C or C-O coupling of ethylmompain (11a) and boryquinone (11b) (Scheme 3).  As suggested by the authors of [20,21], the cyclization of ethylidene-bis(trihydroxynaphthazarin) 1 in concentrated sulfuric acid produces anhydrous derivative 2. To obtain some amount of this substance for biological testing, a replica of this experiment was conducted. Surprisingly, it was found that, under the described conditions, ethylidenebis(trihydroxynaphthazarin) 1 undergoes cyclization with the formation of mirabiquinone (3) instead of 5H-dibenzo[b,i] xanthenetraone 2 in good yield [26]. This suggests that UV-, IR-spectroscopy, and TLC are not reliable enough at establishing structures of polyhydroxynaphthazarins, even by the comparison method. For example, these methods cannot distinguish between bisnaphthazarin 1 and related naphthazarins 2 and 3, which were considered in the cited work [20].
As suggested by the authors of [20,21], the cyclization of ethylidene-bis(trihydroxynaphthazarin) 1 in concentrated sulfuric acid produces anhydrous derivative 2. To obtain some amount of this substance for biological testing, a replica of this experiment was conducted. Surprisingly, it was found that, under the described conditions, ethylidene-bis(trihydroxynaphthazarin) 1 undergoes cyclization with the formation of mirabiquinone (3) instead of 5H-dibenzo[b,i] xanthenetraone 2 in good yield [26]. This suggests that UV-, IR-spectroscopy, and TLC are not reliable enough at establishing structures of polyhydroxynaphthazarins, even by the comparison method. For example, these methods cannot distinguish between bisnaphthazarin 1 and related naphthazarins 2 and 3, which were considered in the cited work [20].
Mar. Drugs 2023, 21, x 6 of 19 well-studied mechanism of biosynthesis of compounds analogous to spinochromes [32,33] can be considered as circumstantial evidence in favor of structure 8. The structures of all natural naphthoquinone derivatives containing the 2,7 dihydroxynaphthazarin fragment as a subgroup [2,29,[34][35][36][37] are consistent with the above mechanism. Another formal product of oxidative dimerization is islandoquinone, a metabolite isolated from the lichen C. islandica [28]. Structure 18 was proposed for this compound as a result of the comparison of the compound obtained with the lapachol peroxide structure 19 ( Figure 6) [38,39]. However, the IR-spectrum of islandoquinone did not contain ν(C=O) absorption bands at ≈1750 cm −1 that are found for 2,3-dihydro-2-oxo-1,4-naphthoquinones [40].  (19), and its correction in favor of gem-diol 20, considering the data of IR-spectroscopy.
Accordingly, the 2,3-dihydro-2-oxo-1,4-naphthoquinonoid structure of the Q2H fragment of biquinone 18 was revised, and this natural product was identified as 20, i.e., the 2,3-dihydro-2,2-dihydroxy-1,4-naphthoquinonoid structure was assigned to the Q2H subgroup [41]. An argument in favor of structure 20 was based on the comparison of its Another formal product of oxidative dimerization is islandoquinone, a metabolite isolated from the lichen C. islandica [28]. Structure 18 was proposed for this compound as a result of the comparison of the compound obtained with the lapachol peroxide structure 19 ( Figure 6) [38,39]. However, the IR-spectrum of islandoquinone did not contain ν (C=O) absorption bands at ≈1750 cm −1 that are found for 2,3-dihydro-2-oxo-1,4naphthoquinones [40]. well-studied mechanism of biosynthesis of compounds analogous to spinochromes [32,33] can be considered as circumstantial evidence in favor of structure 8. The structures of all natural naphthoquinone derivatives containing the 2,7 dihydroxynaphthazarin fragment as a subgroup [2,29,[34][35][36][37] are consistent with the above mechanism. Another formal product of oxidative dimerization is islandoquinone, a metabolite isolated from the lichen C. islandica [28]. Structure 18 was proposed for this compound as a result of the comparison of the compound obtained with the lapachol peroxide structure 19 ( Figure 6) [38,39]. However, the IR-spectrum of islandoquinone did not contain ν(C=O) absorption bands at ≈1750 cm −1 that are found for 2,3-dihydro-2-oxo-1,4-naphthoquinones [40].  (19), and its correction in favor of gem-diol 20, considering the data of IR-spectroscopy.
Based on the accumulated spectral data, it was concluded that islandoquinone is one of four dioxabenzo[a]tetracenetetraones from the two diastereoisomeric pairs of 9, 9 and 21, 21 ( Figure 7). According to quantum chemical calculations [44], the diastereoisomers 7aS*,13aS*-(9) and 7aR*,13aR*- (21) are more favorable than the corresponding diastereoisomers 7aS*,13aR*-(9 ) and 7aR*,13aS*- (21 ). The difference in Gibbs energy between 9 and 21 is only 0.4 kcal/mol [45]. The conclusive choice in favor of either 9 or 21 may be based on the X-ray diffraction analysis of islandoquinone or structurally similar compounds. spectral data with those of the 2,3-dihydro-2,2-dihydroxy-1,4-naphthoquinones described in the literature [42,43]. However, doubts have emerged regarding the proposed structure of islandoquinone, with the major discrepancy concerning the presence of proton signals of only three α-hydroxy groups in the 1 H NMR spectrum of islandoquinone [28]. Based on the accumulated spectral data, it was concluded that islandoquinone is one of four dioxabenzo[a]tetracenetetraones from the two diastereoisomeric pairs of 9, 9′ and 21, 21′ (Figure 7). According to quantum chemical calculations [44], the diastereoisomers 7aS*,13aS*-(9) and 7aR*,13aR*- (21) are more favorable than the corresponding diastereoisomers 7aS*,13aR*-(9′) and 7aR*,13aS*-(21′). The difference in Gibbs energy between 9 and 21 is only 0.4 kcal/mol [45]. The conclusive choice in favor of either 9 or 21 may be based on the X-ray diffraction analysis of islandoquinone or structurally similar compounds. Within this context, the oxidative coupling products of ethylhydroxynaphthazarins 22a and 22b (Figure 8) [42] were synthesized and analyzed. The oxidative coupling of the chlorinated hydroxynaphthazarin 22a upon treatment with lead dioxide in boiling acetic acid resulted in a product that, judging from the spectral data, was an unsymmetrical biquinone [45]. In the case of cristazarin (22b), a mixture of two biquinones (1:1.8 ratio, 1 H NMR) was produced. The crystallization of the chlorinated biquinone and major product of oxidative dimerization of cristazarin (22b) from acetone afforded crystals that were suitable for single-crystal X-ray diffraction.  Within this context, the oxidative coupling products of ethylhydroxynaphthazarins 22a and 22b (Figure 8) [42] were synthesized and analyzed. The oxidative coupling of the chlorinated hydroxynaphthazarin 22a upon treatment with lead dioxide in boiling acetic acid resulted in a product that, judging from the spectral data, was an unsymmetrical biquinone [45]. In the case of cristazarin (22b), a mixture of two biquinones (1:1.8 ratio, 1 H NMR) was produced. The crystallization of the chlorinated biquinone and major product of oxidative dimerization of cristazarin (22b) from acetone afforded crystals that were suitable for single-crystal X-ray diffraction. spectral data with those of the 2,3-dihydro-2,2-dihydroxy-1,4-naphthoquinones described in the literature [42,43]. However, doubts have emerged regarding the proposed structure of islandoquinone, with the major discrepancy concerning the presence of proton signals of only three α-hydroxy groups in the 1 H NMR spectrum of islandoquinone [28]. Based on the accumulated spectral data, it was concluded that islandoquinone is one of four dioxabenzo[a]tetracenetetraones from the two diastereoisomeric pairs of 9, 9′ and 21, 21′ (Figure 7). According to quantum chemical calculations [44], the diastereoisomers 7aS*,13aS*-(9) and 7aR*,13aR*- (21) are more favorable than the corresponding diastereoisomers 7aS*,13aR*-(9′) and 7aR*,13aS*-(21′). The difference in Gibbs energy between 9 and 21 is only 0.4 kcal/mol [45]. The conclusive choice in favor of either 9 or 21 may be based on the X-ray diffraction analysis of islandoquinone or structurally similar compounds. Within this context, the oxidative coupling products of ethylhydroxynaphthazarins 22a and 22b (Figure 8) [42] were synthesized and analyzed. The oxidative coupling of the chlorinated hydroxynaphthazarin 22a upon treatment with lead dioxide in boiling acetic acid resulted in a product that, judging from the spectral data, was an unsymmetrical biquinone [45]. In the case of cristazarin (22b), a mixture of two biquinones (1:1.8 ratio, 1 H NMR) was produced. The crystallization of the chlorinated biquinone and major product of oxidative dimerization of cristazarin (22b) from acetone afforded crystals that were suitable for single-crystal X-ray diffraction.  The molecular structures of the obtained products and their corresponding structural formulas (23a,b) are shown in Figure 9 [45].
The upfield signals of the ethyl group protons at C(15), C(16), C(17), and C(18) and carbon atoms C(6a), C(7a), C(13a), and C(14a) ( Table 1) of the dioxane ring of the biquinones 23a,b were in very good agreement with the corresponding signals of islandoquinone [28,42]. Therefore, the connection of rings B and C and the position of the substituents in these rings in these biquinones and islandoquinone are identical. These data indicate that the early proposed structure of islandoquinone should be revised in favor of dioxabenzo[a]tetracenetetraone 9 and compound 23b is its dimethyl ether. All attempts to convert 23b into islandoquinone were unsuccessful at yielding a complex mixture of compounds. The upfield signals of the ethyl group protons at C(15), C(16), C(17), and C(18) and carbon atoms C(6a), C(7a), C(13a), and C(14a) ( Table 1) of the dioxane ring of the biquinones 23a,b were in very good agreement with the corresponding signals of islandoquinone [28,42]. Therefore, the connection of rings B and C and the position of the substituents in these rings in these biquinones and islandoquinone are identical. These data indicate that the early proposed structure of islandoquinone should be revised in favor of dioxabenzo[a]tetracenetetraone 9 and compound 23b is its dimethyl ether. All attempts to convert 23b into islandoquinone were unsuccessful at yielding a complex mixture of compounds.
By analogy with hybocarpone, the related binaphtho[2,3-b; 2,3-d]furantetraones 24a,b were synthesized from the corresponding derivatives of 1,4-naphthoquinone 25b,c [47] (Figure 13). In another synthetic approach to hybocarpone (10) and its analogs, the use of the direct oxidative dimerization of the 2-hydroxynaphthazarin precursors 31a,b was explored ( Figure 14) [53]. These compounds are more readily available [54,55] than the 1,4-naphthoquinone precursors 25a-c. However, all attempts to construct the appropriate binaphtho[2,3-b; 2,3-d]furantetraone skeleton by the action of CAN in MeCN [47] led to degradation of the starting structures 31a,b. Upon screening a number of reagents and conditions, success was finally achieved with the use of Pb(OAc)4 as an oxidant in benzene.
Oxidative coupling of hydroxynaphthazarin (31a) on treatment with Pb(OAc)4 in benzene gave two compounds in a ratio of ca. 1:1. One of them exhibited a simple 1 H NMR spectrum reminiscent of that of 10, and judging from the spectral data, was binaphtho[2,3-b; 2,3-d]furantetraone (32a) ( Figure 14). As such, the upfield 1 H NMR signals of the ethyl group protons and 13 C NMR signals of carbon atoms of the tetrahydrofuran ring of 32a were in very good agreement with the corresponding signals of hybocarpone (10) [29]. The other isomeric biquinone exhibits the correct mass (by mass spectrometry) and simple 1 H and 13 C NMR spectra, but the upfield signals of the ethyl group protons and In another synthetic approach to hybocarpone (10) and its analogs, the use of the direct oxidative dimerization of the 2-hydroxynaphthazarin precursors 31a,b was explored ( Figure 14) [53]. These compounds are more readily available [54,55] than the 1,4-naphthoquinone precursors 25a-c. However, all attempts to construct the appropriate binaphtho[2,3-b; 2,3-d]furantetraone skeleton by the action of CAN in MeCN [47] led to degradation of the starting structures 31a,b. Upon screening a number of reagents and conditions, success was finally achieved with the use of Pb(Oac) 4 as an oxidant in benzene. In another synthetic approach to hybocarpone (10) and its analogs, the use of the direct oxidative dimerization of the 2-hydroxynaphthazarin precursors 31a,b was explored ( Figure 14) [53]. These compounds are more readily available [54,55] than the 1,4-naphthoquinone precursors 25a-c. However, all attempts to construct the appropriate binaphtho[2,3-b; 2,3-d]furantetraone skeleton by the action of CAN in MeCN [47] led to degradation of the starting structures 31a,b. Upon screening a number of reagents and conditions, success was finally achieved with the use of Pb(OAc)4 as an oxidant in benzene.
Oxidative coupling of hydroxynaphthazarin (31a) on treatment with Pb(OAc)4 in benzene gave two compounds in a ratio of ca. 1:1. One of them exhibited a simple 1 H NMR spectrum reminiscent of that of 10, and judging from the spectral data, was binaphtho[2,3-b; 2,3-d]furantetraone (32a) ( Figure 14). As such, the upfield 1 H NMR signals of the ethyl group protons and 13 C NMR signals of carbon atoms of the tetrahydrofuran ring of 32a were in very good agreement with the corresponding signals of hybocarpone (10) [29]. The other isomeric biquinone exhibits the correct mass (by mass spectrometry) and simple 1 H and 13 C NMR spectra, but the upfield signals of the ethyl group protons and Oxidative coupling of hydroxynaphthazarin (31a) on treatment with Pb(Oac) 4 in benzene gave two compounds in a ratio of ca. 1:1. One of them exhibited a simple 1 H NMR spectrum reminiscent of that of 10, and judging from the spectral data, was binaphtho[2,3-b; 2,3-d]furantetraone (32a) ( Figure 14). As such, the upfield 1 H NMR signals of the ethyl group protons and 13 C NMR signals of carbon atoms of the tetrahydrofuran ring of 32a were in very good agreement with the corresponding signals of hybocarpone (10) [29].
The other isomeric biquinone exhibits the correct mass (by mass spectrometry) and simple 1 H and 13 C NMR spectra, but the upfield signals of the ethyl group protons and carbon atoms of the tetrahydrofuran ring of that compound do not fit those of 10. The crystallization of this biquinone from hexane-acetone afforded crystals that were suitable for single-crystal X-ray diffraction. The molecular structure of the product obtained (33a) is shown in Figure 15. carbon atoms of the tetrahydrofuran ring of that compound do not fit those of 10. The crystallization of this biquinone from hexane-acetone afforded crystals that were suitable for single-crystal X-ray diffraction. The molecular structure of the product obtained (33a) is shown in Figure 15. Thus, the oxidative coupling of hydroxynaphthazarin (31a), on treatment with Pb(OAc)4 in benzene, gave the diastereomeric mixture of 5aS*,6aS*,12aS*,12bS* (32a) and 5aS*,6aR*,12aR*,12bS* (33a), analogs of hybocarpone (10). The ratio of the arising compounds 32a and 33a (1:1) was determined by the equally possible formation of the in- Thus, the oxidative coupling of hydroxynaphthazarin (31a), on treatment with Pb(Oac) 4 in benzene, gave the diastereomeric mixture of 5aS*,6aS*,12aS*,12bS* (32a) and 5aS*,6aR*, 12aR*,12bS* (33a), analogs of hybocarpone (10). The ratio of the arising compounds 32a and 33a (1:1) was determined by the equally possible formation of the intermediary S*S* (34a) and R*S* (35a) diastereomers (Scheme 5). So, according to the quantum chemical calculation [44], the difference between Gibbs energy of diastereomers 32a and 33a is less than 0.4 kcal/mol) [53]. Thus, the oxidative coupling of hydroxynaphthazarin (31a), on treatment with Pb(OAc)4 in benzene, gave the diastereomeric mixture of 5aS*,6aS*,12aS*,12bS* (32a) and 5aS*,6aR*,12aR*,12bS* (33a), analogs of hybocarpone (10). The ratio of the arising compounds 32a and 33a (1:1) was determined by the equally possible formation of the intermediary S*S* (34a) and R*S* (35a) diastereomers (Scheme 5). So, according to the quantum chemical calculation [44], the difference between Gibbs energy of diastereomers 32a and 33a is less than 0.4 kcal/mol) [53]. The hydration/cyclization of intermediary S*S* (34a) and R*S* (35a) diastereomers would potentially lead to the formation of up to six diastereomeric furan systems arranged in two rows of three ( Figure 16). Molecular modeling and computational studies indicated that, among the diastereoisomeric compounds in each row, the isomers 32a and 33a appeared to be clearly favored in terms of relative Gibbs energy [44,56]. Since the calculated energy differences among compounds in each row are large (more than 9 kcal/mol), and because the central dihydroxyfuran systems of them can exist in equilibrium with their open chain counterparts, diastereoisomers 32a and 33a are the only imaginable products in this reaction. Scheme 5. S*S*-(34a) and R*S*-(35a) hexaketone intermediates, the precursors of the S*,S*,S*,S*-(32a) and S*,R*,R*,S*-(33a) binaphthofurantetraones.
The hydration/cyclization of intermediary S*S* (34a) and R*S* (35a) diastereomers would potentially lead to the formation of up to six diastereomeric furan systems arranged in two rows of three ( Figure 16). Molecular modeling and computational studies indicated that, among the diastereoisomeric compounds in each row, the isomers 32a and 33a appeared to be clearly favored in terms of relative Gibbs energy [44,56]. Since the calculated energy differences among compounds in each row are large (more than 9 kcal/mol), and because the central dihydroxyfuran systems of them can exist in equilibrium with their open chain counterparts, diastereoisomers 32a and 33a are the only imaginable products in this reaction. These observations were used as the basis for the synthesis of hybocarpone (10). Methylcristazarin (31b) is a more available substrate for this purpose. As in the case of 31a, the oxidative coupling of hydroxynaphthazarin (31b), on treatment with Pb(OAc)4 in benzene, gave two products. One of them, according to the spectral data, was hybocarpone dimethyl ether 32b (Figure 14). The other product was the isomer 33b bearing a sin relationship of the two ethyl groups at the junction joining the two monomeric units.
These observations were used as the basis for the synthesis of hybocarpone (10). Methylcristazarin (31b) is a more available substrate for this purpose. As in the case of 31a, the oxidative coupling of hydroxynaphthazarin (31b), on treatment with Pb(Oac) 4 in benzene, gave two products. One of them, according to the spectral data, was hybocarpone dimethyl ether 32b (Figure 14). The other product was the isomer 33b bearing a sin relationship of the two ethyl groups at the junction joining the two monomeric units.
These observations were used as the basis for the synthesis of hybocarpone (10). Methylcristazarin (31b) is a more available substrate for this purpose. As in the case of 31a, the oxidative coupling of hydroxynaphthazarin (31b), on treatment with Pb(OAc)4 in benzene, gave two products. One of them, according to the spectral data, was hybocarpone dimethyl ether 32b (Figure 14). The other product was the isomer 33b bearing a sin relationship of the two ethyl groups at the junction joining the two monomeric units.
Dimethyl ethers 32b and 33b were deprotected with AlCl3 in EtSH-CH2Cl2 to afford 10 and 36 ( Figure 17). Synthetic 10 exhibited spectral data ( 1 H and 13 C NMR, mass spectrometry) identical to those reported for natural hybocarpone, a cytotoxic metabolite isolated from L. hybocarpa lichen [29]. It should be noted that in the previous report [47], the course of the reaction through the S*,S* hexaketone intermediate 27 (Scheme 4) was only postulated, and the possibility of R*,S* diastereomer formation was not discussed; thus, we compared our results for the oxidative dimerization of dihydrolapachole 25c with Pb(OAc)4 in benzene [57] to those previously reported using CAN in MeCN [47].
Oxidative dimerization of dihydrolapachole 25c upon treatment with Pb(OAc)4 in benzene yielded three products following chromatographic purification [57]. A yellow product established as 24b by Nicolaou [47]was determined to be 3-(naphthoquinone-2-yloxy)naphthalenetrione 37 (42%) based on its spectral data ( 1 H, 13 C NMR and mass spectrometry) and comparison with the authentic sample [58] ( Figure  18). The formation of a high percentage of 37 resulted due to the steric bulk around the reacting carbon in starting substrate 25c; thus, С-О coupling was more probable than C-C coupling. It should be noted that in the previous report [47], the course of the reaction through the S*,S* hexaketone intermediate 27 (Scheme 4) was only postulated, and the possibility of R*,S* diastereomer formation was not discussed; thus, we compared our results for the oxidative dimerization of dihydrolapachole 25c with Pb(Oac) 4 in benzene [57] to those previously reported using CAN in MeCN [47].
Oxidative dimerization of dihydrolapachole 25c upon treatment with Pb(OAc) 4 in benzene yielded three products following chromatographic purification [57]. A yellow product established as 24b by Nicolaou [47]was determined to be 3-(naphthoquinone-2-yloxy)naphthalenetrione 37 (42%) based on its spectral data ( 1 H, 13 C NMR and mass spectrometry) and comparison with the authentic sample [58] (Figure 18). The formation of a high percentage of 37 resulted due to the steric bulk around the reacting carbon in starting substrate 25c; thus, C-O coupling was more probable than C-C coupling. Two colorless products were determined to be pentacyclic compounds 24b (24%) and 24c (21%) based on the spectral data and comparison of its 1 H and 13 C NMR data with spectral data of derivatives 32a,b and 33a,b. We found that the same mixture was obtained upon exposure of the monomeric unit 25c to CAN in MeCN under the previously reported conditions [47].

Diene Condenstion Compounds
Among natural partially methylated derivatives of echinochrome, there are no examples of structures that simultaneously contain 2-hydroxy and 3-(1-hydroxyethyl) groups in one core [1][2][3]. This is obviously due to the instability of such compounds. In- Two colorless products were determined to be pentacyclic compounds 24b (24%) and 24c (21%) based on the spectral data and comparison of its 1 H and 13 C NMR data with spectral data of derivatives 32a,b and 33a,b. We found that the same mixture was obtained upon exposure of the monomeric unit 25c to CAN in MeCN under the previously reported conditions [47].

Diene Condenstion Compounds
Among natural partially methylated derivatives of echinochrome, there are no examples of structures that simultaneously contain 2-hydroxy and 3-(1-hydroxyethyl) groups in one core [1][2][3]. This is obviously due to the instability of such compounds. Indeed, attempts to synthesize 2-hydroxy-3-(1-hydroxyethyl) naphthazarin 38 by alkaline hydrolysis of lomazarin (39a), its 1′-bromo-(39b) and 1′-acetoxy-(39c) derivatives are invariably resulted in spinochrome D dimethyl ether 5 ( Figure 19). Most likely, 1′,2-dihydroxy-3-ethylnaphthazarin 38, formed from starting compounds 39a-c under basic conditions, is converted to dimethyl ether 5 via the mechanism of retroaldol decomposition of intermediate keto form 40 [24,25].  The formation of benzo[g]chromene-5,10-dione 43 likely occurs via the mechanism of heterodiene condensation (Scheme 6). The 1′-bromoethyl derivative 42 formed during the reaction loses HBr, giving the enone 44 (heterodiene), which are isomerized to the corresponding vinylquinone 45 (dienophile). The isomerization of 44→45 is reversible, since at the end of the reaction the starting diene and dienophile are not found in the mixture. It should be noted that 1′-bromoethyl derivative 42 are labile compounds. In acidified solutions of chloroform, acetone, or on the surface of H + -silica gel, they are rapidly converted into the corresponding benzo[g]chromene-5,10-dione 43. The structure of product 43 and stereochemistry of its asymmetric centers are unambiguously determined by analysis of their 1 H and 13 C NMR spectra [59]. Hydrolysis of benzo[g]chromedione 43 gave product 46 ( Figure 20). The formation of benzo[g]chromene-5,10-dione 43 likely occurs via the mechanism of heterodiene condensation (Scheme 6). The 1 -bromoethyl derivative 42 formed during the reaction loses HBr, giving the enone 44 (heterodiene), which are isomerized to the corresponding vinylquinone 45 (dienophile). The isomerization of 44→45 is reversible, since at the end of the reaction the starting diene and dienophile are not found in the mixture. It should be noted that 1 -bromoethyl derivative 42 are labile compounds. In acidified solutions of chloroform, acetone, or on the surface of H + -silica gel, they are rapidly converted into the corresponding benzo[g]chromene-5,10-dione 43. The structure of product 43 and stereochemistry of its asymmetric centers are unambiguously determined by analysis of their 1 H and 13 C NMR spectra [59]. Hydrolysis of benzo[g]chromedione 43 gave product 46 ( Figure 20).
When this work was in progress, a previously unknown pigment was detected in the extracts of the sea urchins Mesocentrotus nudus and Strongylocentrotus intermedius by HPLC-MS method ( Figure 21) [60]. The retention time, UV, and mass spectra of the detected product coincided with those of synthesized compound 46 [59]. since at the end of the reaction the starting diene and dienophile are not found in the mixture. It should be noted that 1′-bromoethyl derivative 42 are labile compounds. In acidified solutions of chloroform, acetone, or on the surface of H + -silica gel, they are rapidly converted into the corresponding benzo[g]chromene-5,10-dione 43. The structure of product 43 and stereochemistry of its asymmetric centers are unambiguously determined by analysis of their 1 H and 13 C NMR spectra [59]. Hydrolysis of benzo[g]chromedione 43 gave product 46 ( Figure 20). When this work was in progress, a previously unknown pigment was detected in the extracts of the sea urchins Mesocentrotus nudus and Strongylocentrotus intermedius by HPLC-MS method ( Figure 21) [60]. The retention time, UV, and mass spectra of the detected product coincided with those of synthesized compound 46 [59].  mined by analysis of their 1 H and 13 C NMR spectra [59]. Hydrolysis of benzo[g]chromedione 43 gave product 46 ( Figure 20). When this work was in progress, a previously unknown pigment was detected in the extracts of the sea urchins Mesocentrotus nudus and Strongylocentrotus intermedius by HPLC-MS method ( Figure 21) [60]. The retention time, UV, and mass spectra of the detected product coincided with those of synthesized compound 46 [59].  Thus, the discovered product, which was named mesocentroquinone, has the structure 6,7,8,9-tetrahydroxy-4-methyl-2-(3,5,6,7,8-pentahydroxy-1,4-dioxo-1,4-dihydronaphthalen-2-yl)-3,4-dihydro-2H-benzo[g]chromene-5,10-dione. In fact, it can be considered as a dimer of dehydro derivative of echinochrome (47) obtained by diene condensation.

The Tautomerism of Hydroxynaphthazarins
NMR spectroscopy is among the most used methods for the structural study of hydroxynaphthazarins [1][2][3]. The phenomenon of tautomerism inherent in the naphthazarin system leaves an imprint on the nature of the spectra of substituted naphthazarins, including its hydroxy derivatives [61]. Due to the rapid (on the NMR time scale) tautomerism, in the spectra of naphthazarin and its derivatives, the signals of protons and carbon atoms entering the quinoid and benzenoid cycles are indistinguishable in pairs. Thus, in the 1 H NMR spectrum of mompain monomethyl ether 48 (CDCl 3 ), the signals of protons adjacent to the hydroxy and methoxy groups of tautomers Q and B are indistinguishable ( Figure 22) [54]. Thus, on the NMR scale, mompain monomethyl ether is an individual compound. IR spectroscopy is a much faster method when compared to nuclear magnetic resonance spectroscopy, in which there is usually no temporal averaging of spectral parameters. Therefore, the IR spectrum of mompain monomethyl ether 48 taken in CDCl 3 showed that this compound is a mixture of 1,4-naphthoquinoid tautomers 48(Q) and 48(B), and in commensurate proportions (70% and 30%, respectively) [62,63]. In addition, quantum chemical calculations, using the example of 1 -hydroxyalkyl naphthazarin 6 [64], showed that the energy barrier for a process of type 49(Q 1.4 ) carbon atoms entering the quinoid and benzenoid cycles are indistinguishable in pairs. Thus, in the 1 H NMR spectrum of mompain monomethyl ether 48 (CDCl3), the signals of protons adjacent to the hydroxy and methoxy groups of tautomers Q and B are indistinguishable ( Figure 22) [54]. Thus, on the NMR scale, mompain monomethyl ether is an individual compound.
IR spectroscopy is a much faster method when compared to nuclear magnetic resonance spectroscopy, in which there is usually no temporal averaging of spectral parameters. Therefore, the IR spectrum of mompain monomethyl ether 48 taken in CDCl3 showed that this compound is a mixture of 1,4-naphthoquinoid tautomers 48(Q) and 48(B), and in commensurate proportions (70% and 30%, respectively) [62,63]. In addition, quantum chemical calculations, using the example of 1′-hydroxyalkyl naphthazarin 6 [64], showed that the energy barrier for a process of type 49(Q1.4) ⇆ 49(Q1.5) (Figure 22) is less than 5 kcal/mol [64], which makes possible the existence of corresponding 1,5-naphthoquinoid forms. Later, it was shown by IR spectroscopy that in aprotic organic solvents, 1′-hydroxyalkyl naphthazarins are in the form of a mixture of 1,4-and 1,5-naphthoquinoid tautomers [65]. Thus, due to the easily reversible conversion of specified tautomers, it is impossible to isolate any component of the mixture in an individual form. This is also true for dimeric (poly)hydroxynaphthazarins, which are in no way different from monomers in this respect. At the same time, recent studies have reported the isolation of several tautomers in individual form by the HPLC method and the establishment of their structure by NMR spectroscopy [18,19,66]. The most likely reason for this misunderstanding was the ability of (poly)hydroxynaphthazarins to give stable crystal solvates and chelate-type derivatives, which ultimately led to erroneous conclusions about the structure of the isolated products. This misconception has a long history, the beginning of which lies in the first messages about the allocation of (poly)hydroxynaphthazarins from natural objects [67][68][69]. 49(Q 1.5 ) (Figure 22) is less than 5 kcal/mol [64], which makes possible the existence of corresponding 1,5-naphthoquinoid forms. Later, it was shown by IR spectroscopy that in aprotic organic solvents, 1 -hydroxyalkyl naphthazarins are in the form of a mixture of 1,4-and 1,5-naphthoquinoid tautomers [65].
Thus, due to the easily reversible conversion of specified tautomers, it is impossible to isolate any component of the mixture in an individual form. This is also true for dimeric (poly)hydroxynaphthazarins, which are in no way different from monomers in this respect. At the same time, recent studies have reported the isolation of several tautomers in individual form by the HPLC method and the establishment of their structure by NMR spectroscopy [18,19,66]. The most likely reason for this misunderstanding was the ability of (poly)hydroxynaphthazarins to give stable crystal solvates and chelate-type derivatives, which ultimately led to erroneous conclusions about the structure of the isolated products. This misconception has a long history, the beginning of which lies in the first messages about the allocation of (poly)hydroxynaphthazarins from natural objects [67][68][69].

Conclusions
This review provides information on the establishment of the structure of natural dimeric (poly)hydroxynaphthazarins, metabolites of echinoderms and lichens. Due to the relatively low content of these products in natural objects, and for several other reasons, the establishment of their structure has encountered certain difficulties. Success in overcoming this issue was achieved by using modern physico-chemical research methods and counter synthesis. The results of these studies were the revisions of the structures and syntheses of metabolites of lichens Cetraria 'olothuri, C. islandica, deep-sea holothuria Psychropotes longicauda, and a representative of the genus Benthodytes. The structure of islandoquinone, a metabolite of the lichen C. islandica, the backbone of which is dioxabenzo[a]tetracenetetraone, underwent a serious correction. Mesocentroquinone, the structure of which is based on benzo[g]chromedione, was synthesized earlier than it was isolated from the sea urchins Mesocentrotus nudus and Strongylocentrotus intermedius. In addition, the review provides information on clarifying the direction and mechanism of reactions in the synthesis of some natural dimeric (poly)hydroxynaphthazarins. This refers to the conversion of ethylidene-bis(trihydroxynaphthazarin) to linear dibenzo[b,i]xanthenetetraone, both of which are metabolites of the sea urchin Spatangus purpureus. Relatively recently, it was found that, as a result of this reaction, the angular dibenzo[b,h]xanthenetetraone, mirabiquinone was also formed, which was isolated from the sea urchin Scaphechinus mirabilis and synthesized later. Another example is also the clarification of the mechanism of the key synthesis reaction of hybocarpone, a metabolite of the lichen Lecanora hybocarpa.