Stemarane Diterpenes and Diterpenoids †

In this article the scientific activity carried out on stemarane diterpenes and diterpenoids, isolated over the world from various natural sources, was reviewed. The structure elucidation of stemarane diterpenes and diterpenoids was reported, in addition to their biogenesis and biosynthesis. Stemarane diterpenes and diterpenoids biotransformations and biological activity was also taken into account. Finally the work leading to the synthesis and enantiosynthesis of stemarane diterpenes and diterpenoids was described.

In the following years Tamogami and co-workers at Central Research Laboratories, Idemitsu Kosan Co., Ltd., Chiba and at Ibaraki University, Ibara, isolated from Pyricularia oryzae Cav. attacked Oryza sativa L. (Poaceae) (Figure 4) (+)-oryzalexin S [9] 11, a stemarane diterpenoid which displays phytoalexin properties. The production by the plant of (+)-11 is also induced by UV irradiation [9]. The production of diterpenoid phytoalexins after induction by UV irradiation was also studied in five rice genotypes of different susceptibility to the rice blast fungus Pyricularia oryzae at Reading University by Harborne and his group [10]. Finally in 2001 Oikawa, Sassa, and co-workers, at Hokkaido and Yamagata Universities respectively, reported the isolation of (+)-13-stemarene 12 [11] from the phytopathogenic fungus Phoma betae.

Structure
The stemarane diterpenes carbon skeleton is reported in Figure 5, and its main structural features are reported below: the bicyclic system C/D is constituted by a bicyclo[3.2.1]octane fused to the bicyclic A/B system in a different fashion with respect to other tetracyclic diterpenes possessing the bicyclo[3.2.1]octane system; two contiguous quaternary carbon atoms, the C(9) and C(10), are present, the former being a spirocyclic atom; oxygenated functions can be present at C(2), C(7), C(13), C(17), C (18), and C (19); the A/B ring junction is trans; the H-C(8) is syn to the CH 3 -C(10), and to the two carbon bridge connecting C-(9) and C-(12) (In this review only compounds with such features are reported.). The structure and absolute stereochemistry of (+)-stemarin 1 was assigned by Manchand and Blount after a chemical study and an X-ray crystallographic analysis on tosylate (+)-13 [1].
The structure and relative configurations of stemarane diterpenoids 2-10 were assigned by Garbarino and co-workers by means of 1 H-and 13 C-NMR (Nuclear Magnetic Resonance), mass spectroscopy and by comparison with the spectral data of (+)-1. On the basis of biogenetic considerations, the ent absolute configuration was also assigned. No chemical correlations and/or X-ray structure determinations were ever made.
The structure and relative configuration of (+)-oryzalexin S 11 was established by a series of 2D-NMR experiments, which proved the relative configuration [13]. The structure and relative configuration of (+)-13-stemarene 12 was established by 1 H-NMR [11]. The comparison with an authentic sample, obtained by our group from commercially available (+)-podocarpic acid, confirmed the proposed structure as well as its absolute configuration [14].

Biogenesis
The biogenesis of stemarane diterpenes and diterpenoids, outlined in Scheme 1, was proposed by Tamogami and co-workers in 1993 [13]. The Authors proposed the biogenesis of this compound from GGPP (I) and compared it with the biogenesis of the other rice labdane-related diterpenoids. Since neither a proton nor a vinyl group is present at C(9) in (+)-oryzalexin S (11), the Authors assumed that its biogenesis begins with a hydride rearrangement from C(9) to C(8) by which II is converted into III in which the CH 3 -C(13) and the vinyl-C(13) group are α and β oriented, respectively. The attack to the C(9) carbocation by the vinyl group at C(13) gives then intermediate IV. A further rearrangement leads to intermediate V from which (+)-13-stemarene 12 is formed by the loss of H -. Oxidative processes at C(2) and C (19) give then (+)-oryzalexin S 11. Scheme 1. Biogenetic pathway to (+)-13-stemarene 12 as proposed by Tamogami and co-workers [13]. The black curved arrow represents the electrons'movement.
In 1996, Mohan and co-workers from the University of Illinois reported the biosynthesis of cyclic diterpenes with enzyme extracts from rice cell suspension cultures to verify proposed pathways and intermediates in the production of momilactone and oryzalexin phytoalexins [15]. In the course of this work, the authors could confirm the role of 9,10-syn-copalyl diphosphate (syn-CDP) as a precursor of, inter alia, (+)-13-stemarene 12, from which (+)-oryzalexin S 11 is presumed to be formed.
In 2004, Yamane and co-workers from the University of Tokyo reported that two species of mRNA encoding OsDTC1 and OsDTC2, two putative diterpene cyclases, were expressed in rice cell suspensions in response to chitosan-elicitation [30]. The same authors obtained OsDTC2 cDNA from chitin-elicited suspension-cultured rice cells [16]. They overexpressed OsDTC2 cDNA in Escherichia coli as a fusion protein with glutathione S-transferase and demonstrated that the recombinant protein function as stemar-13-ene synthase, the enzyme that catalyse in the conversion of syn-CDP into (+)-13-stemarene 12, the putative precursor of (+)-oryzalexin S 11 (Scheme 2). They also observed (+)-13-stemarene 12 accumulation in both chitin-elicited suspension-cultured rice cells and in UV-irradiated rice leaves.
The carbocation reaction network for the formation of C(9)-ethano-bridged diterpenes, including stemarane diterpenes was described in 2002 by Oikawa and co-workers who integrated chemical and computational methods [31]. Finally, in 2018 Young and Tantillo from the University of California shed new light on the mechanisms of formation of the stemarene, stemodene, betaerdene, aphidicolene, and scopadulanol diterpenes from syn-CDP. The Authors demonstrated that the compounds of interest are interconnected by a complex network of reaction pathways, and that the interconnection of these paths leads to multiple routes for formation of each diterpene, which could lead to different origins for some carbon atoms in a given diterpenes under different conditions [32].

Biological Activity
In the folk medicine of Dutch Antilles an infusion of leafy branches of sea lavander (Lavandula) and Stemodia maritima L., mixed with Epsom salts, is used by men against venereal diseases [41]. Plants of Calceolaria genus are used in Central and South America popular medicine as stomach tonics, bactericidal agents, and sweeteners [42]. Nevertheless, to our best knowledge, the biological activity of pure isolated stemarane diterpenes was not investigated with the exception (+)-oryzalexin S 11 which, as stated above, was found to possess phytoalexin activity [9].
Essential oils obtained by a Brazilian research group, leaded by Arriaga at Universidade Federal do Ceará, Fortaleza, from Stemodia maritima L. leaves and stems, collected in the state of Ceará, showed larvicidal properties against the larvae of the mosquito Aedes aegypti, responsible for the transmission of yellow fever in Central and South America and in west Africa and a vector of dengue hemorrhagic fever [43]. Nevertheless, the major components found in the leaf oil were β-caryophyllene and 14-hydroxy-9-epi-β-caryophyllene, while in the stem oil β-caryophyllene and caryophyllene oxide were the most abundant. This biological activity cannot therefore be attributed to stemarane diterpenes and diterpenoids.
Besides, the same research group evaluated the antioxidant and antibacterial activity of some Stemodia maritima L. isolated metabolites, but stemarane diterpenoids [44], and could also observe that Stemodia maritima L. extracts decrease inflammation, oxidative stress, and alveolar bone loss in an experimental periodontitis rat model [45].
It appears, therefore, that the biological activity of pure isolated stemarane diterpenes and diterpenoids, but (+)-oryzalexin S 11, has not been evaluated yet. The biological activity ascertained so far appears due to Stemodia maritima L. metabolites is different from that of stemarane diterpenes and diterpenoids. Comparing the content of stemarane diterpenes and diterpenoids among Stemodia maritima L. plants, collected in different geographical areas and extracted with the same procedure, seems a due task.

Synthesis
The C/D ring mojety constituted by a bicyclo[3.2.1]octane system fused in a novel way to the ring B, the presence of various stereocenters, two of which are the adjacent quaternary carbons (C-9 and C-10) and the interesting biological activity of some terms of this class of compounds, make stemarane diterpenoids a worthy synthetic challenge.
Kelly and co-workers at the University of New Brunswick (St. John N.B., Canada) and our group, both belonging to the Wiesner [46] school, got engaged with the synthesis of these very interesting compounds by the approach had been developed by their Mentor for the construction of the C/D ring system of diterpene alkaloids [47][48][49][50][51][52][53].
The approach adopted at first for obtaining this class of diterpenoids was based on the following steps: (a) Allene photoaddition to a suitably substituted α,β-unsaturated carbonyl intermediate [54][55][56][57][58][59][60] 7.1. The (±)-Stemarin 1 Total Synthesis by Allene Photoaddition to a 9(11)-Podocarpen-12-one Intermediate The first synthesis of a stemarane diterpene was disclosed in 1980 by Kelly and co-workers [67]. The St. John N.B. group obtained (±)-stemarin 1 from known racemic tricyclic intermediate 30 [68] (Scheme 4). Tricyclic enone 30, was submitted to allene photoaddition at −78 • C. The resulting photoadduct 31, whose stereochemistry followed from its conversion into the final compound, was methylated at C(13) to give 32. After protection of the carbonyl group (33), the exocyclic methylene was oxidatively cleaved to give 34. The latter was than reduced to the masked ketol 35. Thus, under acidic conditions the ketal group was removed and the resulting ketol underwent a retroaldol-aldol reaction to give ketols 37a and 37b in an about 3:1 ratio. Only the aldol 37b has the correct hydroxyl configuration for the subsequent rearrangement to the stemarin skeleton. The synthesis was therefore continued with 37b. Deoxygenation led to alcohol 39 which was converted into the tosylate 40. The latter upon rearrangement gave the olefinic ester 41. The conversion of this compound into (±)-stemarin 1 was then accomplished by stereoselective epoxidation to give 42 and hydride reduction of the latter. The synthetic material was found to be identical to an authentic sample by comparing the IR (Infrared radiation) and NMR spectra and by the identical behavior on TLC (Thin-layer chromatography) in a variety of solvents.  While Kelly and his group were engaged in the synthesis of (±)-stemarin 1, our group was involved in the synthesis of stemodane [69][70][71] diterpenoids and aphidicolin [69,72,73]. After the successful conclusion of this work, a diastereoselective route to the key 6-hydroxy-1-methylbicyclo[2.2.2]octane intermediate (missing in the Kelly approach) appeared to us a worthwhile synthetic challenge. (+)-13-stemarene 12 and (+)-18-deoxystemarin 2, the simplest terms in the class were chosen as targets.
As mentioned before, under thermodynamically controlled conditions, 3-oxocyclohexaneethanals of type 43 give by intramolecular aldol reaction an about 85/15 endo/exo 6-hydroxybicyclo[2.2.2]octan-2-one 44 mixture (Scheme 5). It was shown by our group that this equilibrium distribution is due to an unfavourable 1,3 boat-axial interaction experimented in the exo epimer by the pseudo-axially oriented hydroxy group [74]. It follows that, in a substituted system, the location of the carbonyl influences the orientation of the hydroxyl group. We decided to solve the problem by the Wiesner two carbons annellation methodology. Two approaches were suggested by retrosynthetic analysis (Scheme 6): the first one (A), starting from a 9(11)-podocarpen-12-one, as in the Kelly approach, would have required the HO-C(12) configuration inversion; the second one (B) would have requested the photoaddition to be performed on a 8(9)-podocarpen-14-one, which, on the basis of the Wiesner empirical rule [57][58][59], should have ensured the same stereofacial selectivity and produced a 6-hydroxy-bicyclo[2.2.2]octan-2-one epimeric mixture whose major epimer should have the HO-(C12) group properly oriented (α) for the rearrangement to the stemarane system. In this approach, in the hydroxybiciclo[2.2.2]octanone intermediates XII the H-C(8) is adjacent to the C(14) carbonyl group. In principle, two epimers (XIIa and XIIb) at C(8) could be formed as result of the experimental conditions necessary for the intramolecular aldol reaction. Comparing the structures of XIIa and XIIb the epimer having the H-C(8) α configurated (XIIb) appears less stable because of the presence of a number of unfavourable steric interactions: while ring A is in a chair conformation ring B is in a boat conformation and the C(9)-C(10) and CH 3 -C(10) bonds are eclipsed (Figure 7, right). On the contrary in XIIa in which the H-C(8) is β configurated both rings A and B are in the chair conformation and the C(9)-C(10) and the CH 3 -C(10) bonds are staggered (Figure 7, left). After equilibration the C(8) stereogenic center should therefore materialize in the desired configuration in which the H-C(8) is β oriented. This approach-which has been recently reviewed [75]-has been successfully accomplished and resulted in the synthesis from (+)-podocarpic acid of (+)-13-stemarene 12 and (+)-18-deoxystemarin 2 (Scheme 7) [14,76,77]. It was based on the inversion of configuration of the HO-C (12). Thus the major ketol epimer 47a was converted into the corresponding tosylate 48 and the latter treated with Et 4 N(PhCOO) in acetone at reflux affording the exo-benzoate 49. This methodology, described in the past by Streitweiser and co-workers for the inversion of configuration of acyclic secondary alcohols [78], is quite convenient in that produces a locked exo-ketol which cannot therefore re-equilibrate to the more stable endo epimer. Thus an expeditious preparation of 50b was also elaborated by equilibrating under acidic conditions the endo rich 6-hydroxybicyclo[2.2.2]octan-2-one ethylene dithioacetal mixture 50. It was found that, after equilibration, the exo epimer 50b is the major one [79]. The approach to stemarane diterpenes via a 8(9)-podocarpen-14-one appeared quite attractive since the allene photoaddition was expected, on the basis of the Wiesner empirical rule (vide supra), to proceed from the α-side, as in the case of 9(11)-podocarpen-12-ones, thus ensuring the correct stereochemistry at C (9). Besides, the HO-C(12) and H-C(8) should have both emerged from the aldol reaction in the desired orientation (vide supra).

Regio-and Diastereoselective Synthesis of (+)-2-Deoxyoryzalexin S 2 from (+)-Podocarpic Acid
Finally, having in hand an efficient methodology for the construction of the C/D ring system of stemarane diterpenoids, we decided to apply it to the synthesis of (+)-2-deoxyoryzalexin S 2 the structure and absolute configuration of which had been established only on the basis of the 1 H and 13 C spectra [2,3,5]. The strategy adopted is described in Scheme 11. The starting material was (−)-19-hydroxypodocarp-9-en-12-one 64, available in four steps from (+)-podocarpic acid. Photoaddition of allene to (−)-64 in THF at 78 • C gave quantitatively the photoadduct (+)-65 the structure of which was established by 2D NMR experiments. Prior to methylation at C(13), the HO-C (19) in (+)-65 was protected to give (+)-66. Compound (+)-66 was then methylated to give 67. The latter was then converted into the acetal 68, and the exocyclic methylene cleaved with OsO 4 /NaIO 4 to give the cyclobutanone 69 along with some unprotected keto-alcohol 70. Compound 69 was then reduced to the cyclobutanol 71. Treatment of 71 with a 2:1 THF/2N HCl mixture gave 74 as an about 80:20 endo/exo C(12) epimeric mixture. The latter dissolved in toluene was heated at 85 • C in the presence of TsOH, giving (+)-75. Thioacetalization of (+)-75 by standard methods followed by deoxygenation gave then (+)-2-deoxyoryzalexin S 2 which was then acetylated to (+)-77 [12]. Remarkably, in this work all chiral centers present in (+)-podocarpic acid are maintained. The relative configuration of (+)-77 was confirmed by an X-Ray structure determination. This work allowed us to demonstrate that the structure of (+)-2-deoxyoryzalexin S 2 could not be attributed to a Chilean Calceolaria isolated diterpenoid to which this structure had been assigned.

Other Strategies
An approach to stemodane and stemarane diterpenoids was also described in 1985 by Ghatak and co-workers at the Indian Association for the Cultivation of Science, Jadavpur, Calcutta [87]. This approach (Scheme 12) was based on the formation of a bicyclo [ A short, expedient, though not diastereoselective route, inter alia, to a potentially key intermediate for their synthesis of stemarane diterpenoids was also realized by Subba Rao and Kaliappan at the Indian Institute of Science, Bangalore, India (Scheme 13) [88]. Also this approach, to our best knowledge, was not implemented with a total synthesis. In turn the decalone intermediate such as 87 can be obtained from the Wieland-Miescher ketone. Since Wieland-Miescher ketone 89 and its C(4) homologue 90 can be obtained enantioselectively [90][91][92][93][94], this approach is quite convenient for the preparation of optically active 9(11)podocarpen-12-ones and hence of the target compounds.
Recently, the synthesis of the bicyclic intermediate 94, necessary for the enantioselective obtaining of (+)-oryzalexin S 11 was described by our group (Scheme 15) [96]. To this end, it was necessary to elaborate the ad hoc side chain 95 to be used in the annulation process leading to ring A. The exocyclic double bond at C(2) would have easied the alkylation of 2-methyl-1,3-cyclohexanedione 91 and allowed its conversion at proper time into the α configurated HO-C (2).
The halide at the end of the side chain would have allowed the introduction of the formyl group necessary for the cyclization. Later the latter group could be converted in the HO-C (18).
Carrying out the aldol reaction in the presence of d-tyrosine at 10 • C the enantioselective cyclization of 97 to 94 was accomplished in very good yield and e.e. The obtaining of 94 followed an X-ray structure determination on 98. (Initially the work was carried out with the cheapest l-amino acid series.)

Conclusions
In this paper, the work by various research groups on stemarane diterpenes and diterpenoids, described in over forthy papers covering isolation, structure elucidation, biogenesis, biosynthesis, biotranformations, synthesis, and enantiosynthesis has been reviewed. From this review, it appears that further work is necessary to establish unambiguously the structure and absolute configuration of stemarane diterpenes and diterpenoids from the Chilean flora. In fact, while the structure and absolute configuration of (+)-stemarin 1 was established by X-ray diffraction, the structure and absolute configuration of stemarane diterpenes and diterpenoids from the Chilean flora was not confirmed by chemical correlation nor by a X-ray structure determination. The biogenesis proposed for this class of compounds was confirmed by several biosynthetic studies. Biotransformations under the activity of a number of fungi leading to interesting metabolites were also carried out. Besides, it appears that the biological activity of pure isolated stemarane diterpenes and diterpenoids, but (+)-oryzalexin S 11, has not been evaluated yet. The biological activity ascertained so far was attributed to Stemodia maritima L. metabolites, which differ from stemarane diterpenes and diterpenoids. Finally, a comparison about the content of stemarane diterpenes and diterpenoids among Stemodia maritima L. plants, collected in different geographical areas, seems also an interesting task. The extensive work towards the synthesis of this class of compounds resulted in a very efficient approach while recent studies allowed the enantiosynthesis of a key intermediate for the synthesis of (+)-oryzalexin S 11, and should pave the way to this interesting and bioactive compound. We hope this review will be useful to those who are interested in this field.