*2.3. Nitropyrroles*

A new subclass of pyrroleterpene MNPs is represented by 2-nitro-substituted pyrroles carrying a diversely functionalized farnesyl chain attached to the 4-position of the pyrrole core. The nitropyrrolin and heronapyrrole families known to date are formed biosynthetically by means of an electrophilic aromatic substitution of the pyrrole core by a farnesyl pyrophosphate. Subsequent nitration, oxidation to epoxides and alcohols, as well as cascade cyclization reactions then produce a variety of different substituted metabolites.

The first MNP from this subclass was isolated back in 2006, however, the structural characterization appears to be incomplete and no information about the stereochemistry was given [113]. In 2010, the group of Fenical reported the isolation of five farnesyl-2- nitropyrroles **104**–**108** from the marine actinomycete strain CNQ-509 and referred to them as nitropyrrolins A–E (**104**–**108**) (Figure 14) [114]. The authors performed several chemical

modifications, including an acetonide formation from epoxide **105**, and the Mosher method was applied to unequivocally identify the full stereochemistry of nitropyrrolins A–E (**104**– **108**). Among compounds **104**–**108**, nitropyrrolin D (**107**) displayed the most promising IC50 value of 5.7 μM in biological assays against HCT-116 colon carcinoma cells, whereas a lower antibacterial activity against MRSA was observed for all nitropyrrolins **104**–**108** (MIC values >20 μg/mL). Some of the synthetic derivatives synthesized in the course of the structure elucidation process showed strong to moderate cytotoxic (IC50 values between 9.2 μM and 24.4 μM) and promising antibacterial properties (MIC value of 2.8 μg/mL) [114].

**Figure 14.** Nitropyrrolins A–E (**104**–**108**) represent the family of 4-farnesylated 2-nitropyrroles.

In 2016, the Morimoto group reported the first total synthesis of nitropyrrolins A (**104**), B (**105**), and D (**107**) in a sequential fashion (Scheme 9) [115]. As a key step, the authors performed a lithium–halogen exchange on bromopyrrole **109** and reacted the intermediary lithium species with epoxybromide **110**, which was prepared from a known epoxy alcohol. Subsequent deprotection and α-nitration of the pyrrole core then furnished nitropyrrolin B (**105**) in 7% over two steps. Treatment of the epoxide **105** with BF3·OEt2 and acetone produced the *cis*-acetonide, the stereochemistry of which could be investigated by NOE spectroscopy. Cleavage of the acetonide under acidic conditions then generated nitropyrrolin A (**104**) in 76% over two steps. When nitropyrrolin B (**105**) was reacted with TMSOTf, a regio- and stereoselective epoxide ring-opening occurred. In a one-pot approach, the intermediary allylic TMS-ether was cleaved under the addition of TBAF producing nitropyrrolin D (**107**) in 90% yield (Scheme 9) [115].

Only a few days after disclosure of nitropyrrolins A–E (**104**–**108**) as natural products, the group of Capon reported the extraction of three further 2-nitropyrroles, the heronapyrroles A–C (**111**–**113**) (Figure 15) [116]. These compounds share the same 4-farnesyl-2- nitropyrrole scaffold and are closely related to the nitropyrrolins **104**–**108** (Figure 14). The heronapyrroles **111**–**113** were isolated from a microbial culture of *Streptomyces* sp. strain CMB-M0423 in only minor quantities, which prevented a meaningful analysis of the full stereochemistries. However, on the basis of biosynthetic considerations, the absolute configurations were tentatively assigned as 7*S* and 15*R*. Although heronapyrroles A–C (**111**–**113**) neither displayed cytotoxicity against several cell lines (HeLa, HT-29, AGS) nor showed any activity towards Gram-negative bacteria such as *Pseudomonas aeruginosa* (ATCC 10145) and *Escherichia coli* (ATCC 11775), promising activity against Gram-positive bacteria such as *Staphylococcus aureus* (ATCC 9144, IC50 values between 0.6 μM and 0.8 μM) and *Bacillus subtilis* (ATCC 6633, IC50 values between 0.8 μM and 4.2 μM) could be observed [116].

**Figure 15.** The heronapyrroles A–D (**111**–**114**) only differ in their oxidation state in the farnesyl side chain.

Since the stereochemistries of heronapyrroles A–C (**111**–**113**) were only based on a biosynthetic assumption, several total syntheses of members belonging to the heronapyrrole family have been undertaken in the last decade. In 2012, Stark and co-workers focused on biosynthetic considerations and published a bioinspired synthesis attempting to synthesize heronapyrrole C (**113**) [117]. Starting with a lithium–halogen exchange-mediated coupling of 3-bromopyrrole **109** and farnesyl bromide **115** followed by nitration of the pyrrole core and Boc-protection, farnesylpyrrole **116** was generated in 13% over five steps. Asymmetric dihydroxylation of compound **116**, followed by a key double organocatalytic epoxidation using the (+)-Shi catalyst enabled a biomimetic polyepoxide cyclization cascade under acidic conditions, yielding pyrrole *ent*-**113b**. However, the product *ent*-**113b** showed an opposite optical rotation compared to the isolated natural product, prompting the authors to propose the corresponding enantiomer (+)-**113a** to be the true natural structure (Scheme 10) [117].

**Scheme 10.** First total synthesis of (+)-heronapyrrole C (**113a**) by Brimble in 2014 and its enantiomer (−)-heronapyrrole C (*ent*-**113b**) by Stark.

Just as heronapyrroles A–C (**111**–**113**), heronapyrrole D (**114**) could be isolated by Stark and co-workers from a microbial culture of *Streptomyces* sp. (strain CMB-M0423) in 2014 and showed significant inhibition of Gram-positive bacteria *Staphylococcus aureus* subsp. (ATCC 25923, IC50 value 1.8 μM), *Staphylococcus epidermis* (ATCC 12228, IC50 value 0.9 μM) and *Bacillus subtilis* (ATCC 6633, IC50 value 1.8 μM), but was inactive against Gramnegative bacteria *Pseudomonas aeruginosa* (ATCC 10145), *Escherichia coli* (ATCC 25922) and *Candida albicans* (ATCC 90028) [118]. Along with its isolation, the authors also published the total synthesis of (+)-heronapyrrole D (**114**), using the same strategy as in their previous synthesis of 2012. The only exception is represented by the Shi-epoxidation, in which substoichiometric amounts of the oxidant (Oxone®) were applied to generate *mono*-epoxides. Cyclization furnished the desired (+)-heronapyrrole D (**114**) (Scheme 10) [118].

Although the Stark laboratory further elaborated their studies on the nitration step and improved the entire synthesis in 2014 [119], the group of Brimble published the first total synthesis of the naturally occurring (+)-heronapyrrole C (**113a**) almost at the same time [120]. Based on their key intermediates **117** and **118**, synthesized in 4 and 11 steps, respectively, a Julia–Kocienski olefination merged the pyrrole subunit and the terpenoid side chain. A subsequent Shi-epoxidation then furnished compound **119** in 25% over two steps. The authors mentioned that the use of *N*-benzoyloxymethyl (Boz) as a protecting group was crucial to perform the final cyclization and deprotection under mild conditions. In this way, (+)-heronapyrrole C (**113a**) could be obtained in 80% yield over two steps (Scheme 10) [120]. The spectroscopic data of the (+)-isomer **113a** match those of the natural product and confirm the proposed reassignment by Stark et al. in 2012.

In 2015, the Morimoto group published the total synthesis of the remaining (+)- heronapyrroles A (**111**) and B (**112**) [121]. Taking into account the reported syntheses of (−)-heronapyrrole C (*ent*-**113b**) by Stark (2012) and (+)-heronapyrrole C (**113a**) by Brimble (2014) together with the biogenetic relationship of heronapyrroles A–C (**111**–**113**), a stereochemical reassignment of pyrroles **111** and **112** was proposed. Morimoto's group established a strategy similar to the approaches published by Stark and Brimble by installing the farnesylated chain through alkylation of pyrrole **109** with epoxy bromides **120** or **121**. In the case of (+)-heronapyrrole A **111**, the generated epoxide **122** was opened regioselectively by BF3·OEt2, yielding a masked C7–C8 *anti*-diol, which, after sodium-mediated ring-opening of the THF moiety and several further transformations, led to the formation (+)-heronapyrrole A (**111**) in 3% yield over seven steps (Scheme 11). Just as (+)-**111**, (+)- heronapyrrole B (**112**) was synthesized in a corresponding manner by opening the epoxide **123** via the same sequence to give a *cis*-acetonide, which, after nitration and acid-mediated cleavage of the acetonide functional groups, gave (+)-heronapyrrole B (**112**) in 18% yield over five steps (Scheme 11). In both cases, the absolute configuration was determined by the Mosher method which confirmed the proposed structure. As a consequence, the initially proposed stereochemistries for heronapyrroles A (**111**) and B (**112**) from the Stark laboratory in 2012 were reassigned [121].

**Scheme 11.** Total synthesis of (+)-heronapyrrole A (**111**) and (+)-heronapyrrole B (**112**) by a convergen<sup>t</sup> approach leading to stereochemical reassignments.

This rare class of nitropyrroles has attracted some attention from synthetic chemists in recent years. Not least because of previous synthetic work and the promising effects against Gram-positive bacteria, nitropyrroles may represent interesting targets for further drug design [115,117,118,120–123].
