Benefits of Unconventional Methods in the Total Synthesis of Natural Products.

This article provides a survey of four "unconventional" methods employed in the synthesis of natural products in the Hudlicky group. The utility of flash vacuum pyrolysis is highlighted by examples of many natural products attained via vinylcyclopropane-cyclopentene rearrangement and its heterocyclic variants. Preparative organic electrochemistry was used in oxidations and reductions with levels of selectivity unattainable by conventional methods. Yeast reduction of ketoesters was featured in the total synthesis of pyrrolizidine alkaloids. Finally, the use of toluene dioxygenase-mediated dihydroxylations in enantioselective synthesis of natural products concludes this presentation. Recently, synthesized targets in the period 2010-2019 are listed in the accompanying table. The results of research from the Hudlicky group are placed in appropriate context with the work of others, and a detailed guide to the current literature is provided.

Introduction

The practice and the methods of organic synthesis have changed very little in the last (almost) 200 years.‡ The new technology, such as modern analytical instrumentation, made it possible to accrue results at a faster rate and with very small quantities of material. However, the actual physical execution of experiments has changed very little; we still perform reactions in ordinary glassware and follow up with ordinary work up and purification. Compared with the advances in biology, especially molecular biology and genomics, there has not been a detectable paradigm shift in the way we perform organic synthesis of complex molecules. Of course, there exist some departures from “conventional” methods that have appeared in the last 50 years or so. These would include the use of ultrasound, microwave, mechanochemistry, photochemistry, flow technology, development of transition metal catalysis, and other improvements.[1] Major developments in computational chemistry have allowed advances and predictions, especially in catalysis. Electrochemical and enzymatic methods have been used in total synthesis occasionally but not with the frequency of use of the above-mentioned techniques. The use of lipases is well established as a means of resolution of meso compounds, but the employment of oxidoreductases is less common. Even less common is the use of preparative electrochemistry.

In this review, four “unconventional” methods are discussed in the context of their contribution to specific research problems in the Hudlicky group over the last four decades. A brief literature overview and the state of the art are provided also.

Discussion

Flash Vacuum Pyrolysis (FVP)

Pyrolysis (Greek pyr and lysis; separation by fire) has been used since ancient times and middle ages to prepare methanol, sulfuric acid, glass, and other products. The use of high temperature to effect chemical transformations dates back to the 19th century and was frequently used in structure elucidation of natural products by degradation. A classic experiment in this category is, without a doubt, von Gerichten’s pyrolysis of morphine in 1881[2] that produced phenanthrene, as shown in Figure . He mixed 10 g of morphine with a 10-fold excess weight of zinc dust and heated the dry mixture to 300 °C. Purification of the distillate yielded crystalline phenanthrene identified by its melting point (98–100 °C), elemental analysis, smell, and taste! This single experiment established, together with an additional “gentler and kinder” confirmation by Hofmann elimination,[3] that the phenanthrene core of morphine contained 14 of 17 carbon atoms of this alkaloid.

Figure 1

von Gerichten’s degradation of morphine to phenanthrene and morphol.

Besides the use of pyrolysis in degradation studies, various forms of this technique have been applied to rearrangements requiring elevated temperatures. These techniques include heating a solution in a sealed tube or an autoclave, passing a solution through a hot tube under inert gas, gas-phase pyrolysis under a flow of inert gas, and evaporation of a substance through a hot tube under high vacuum, i.e., flash vacuum pyrolysis (FVP). Many other experimental variations exist, and these have been well reviewed.[4] The advantage of FVP rests in the very short time that the substrate spends in the hot zone (milliseconds) under high vacuum (∼10–5 mm/Hg). Under these conditions, molecular collisions are minimized and all of the thermal energy is used for activation.

When I started my independent career at the Illinois Institute of Technology, one of my projects involved the development of a two-step [4 + 1] cyclopentene annulation as a general method of synthesis for triquinane terpenes.[5] The method is comprised of intramolecular cyclopropanation followed by thermal rearrangement of vinylcyclopropanes to cyclopentenes.[6] Eventually, this method evolved into a more efficient [2 + 3] intermolecular annulation,[7] as depicted in Figure .

Figure 2

Evolution of FVP-based annulation strategies for the synthesis of natural product frameworks containing five-membered rings.

Both methods relied on FVP at ∼550 °C for the rearrangement to annulated cyclopentenes. Initial experiments were met with failure, and pyrolysis of the vinylcyclopropane provided mixtures of various dienes and decomposition products.

Reviewing the background of the vinylcyclopropane–cyclopentene rearrangement yielded an interesting piece of information: In Corey’s 1975 paper,[8] a footnote described performing the rearrangement by FVP through a tube containing chips of “lead potash glass”. Such material was no longer available, because of FDA safety concerns, and we have instead performed the pyrolysis by using a Vycor tube pretreated with a slurry of PbCO3.§ This protocol provided excellent results and reasonable yields of cyclopentenes.

It is not clear what exactly the role of lead would be in the high-temperature rearrangement. After the pyrolysis, we observed that the Vycor tube was coated with metallic lead. This, of course, would require a reduction. Some speculation has been advanced but has not been substantiated.§

A possible explanation would involve the pyrolysis of lead carbonate to lead(II) oxide and CO2, dissociation of the vinylcyclopropane to a carbene species, trapping of it with PbO (producing lead carbenoid and N2O), metallo-Diels–Alder cycloaddition, and reductive elimination. These methods later evolved into [4 + 1] azide-diene annulation and were applied to the synthesis of pyrrolizidine alkaloids.[9]

Since its discovery by Neureiter in 1959,[10] many applications of the vinylcyclopropane–cyclopentene rearrangement and its heteroatom variants have been reported. Note that the corresponding cylopropylimine–pyrroline rearrangement had been reported in 1929 by Cloke.[11] In Figure , we highlight some of the historical milestones of these rearrangements. The rearrangement of the parent system to cyclopentene was reported independently in 1960–1961 by the group of Vogel, Overberger, and Frey.[12] Vinyloxirane–dihydrofuran rearrangement followed in 1971, published by Paladini and Chuche.[13] Atkinson and Rees,[14] and Lwowski[15] reported the vinylaziridine–pyrroline rearrangement in 1967–1968. Wilson provided the conditions for cyclopropyl aldehyde–dihydrofuran rearrangement as early as 1947.[16] It is also likely that some of these rearrangements occurred much earlier but were not identified as such. For example, the vinylcyclopropane–cyclopentene rearrangement may have taken place during the preparation of vinylcyclopropane itself by Hofmann elimination at high temperatures.[17] In all synthetic applications, some form of high-temperature thermolysis was employed to provide the required activation energy.

Figure 3

Some of the milestones in the rearrangements of vinylcyclopropanes and their heteroatom analogues.

In the applications to natural product syntheses, most of the high-temperature rearrangements proceeded in respectable yields. The success depends on the volatility of the substrate and/or the quality of the vacuum attainable. Figure shows selected examples of terpene synthesis that featured the vinylcyclopropane–cyclopentene rearrangement. In 1979, Piers used it in his synthesis of zizaene (26)[18] and Trost employed his cyclopentene annulation method in the total synthesis of aphidicolin (29).[19] In the 1980s, the rearrangement was featured in many syntheses of cyclopentane-containing terpenes especially during the so-called “triquinane era”. Paquette’s synthesis of α-vetispirene (35)[20] reportedly proceeded in 100% yield! [Unfortunately, the treatise on the practical limits of chemical yields was not available until 2010.[21]]. In 1978, our group initiated a program aimed at a general synthesis of natural products containing five-membered rings. The program was successful, as evidenced by the attainment of many triquinane terpenes, such as hirsutene (32),[22] isocomene (38),[23] pentalenic acid (41),[24] and most notably, a 14-step synthesis of the racemate of retigeranic acid (44),[25] still standing as the shortest one to date. The preparation of terpenoids the size of retigeranic acid (44) or aphidicolin (29) may very well define the limits of the utility of FVP, as substrates of higher molecular weight may not be fully vaporized, even under very high vacuum.

Figure 4

Vinylcyclopropane–cyclopentene rearrangement in terpene synthesis.

Heteroatom variants of the rearrangement (oxa-, aza-, and thia-) were also employed in total synthesis of natural products, some of which are shown in Figure .

Figure 5

Examples of natural products attained from the heteroatom variants of the vinylcyclopropane–cyclopentene rearrangement.

The syntheses of mesembrine (45)[26] and aspidospermine (46)[27] by Stevens featured the cylopropylimine–pyrroline rearrangement, as did Pinnick’s synthesis of isoretronecanol (47).[28] Vinylaziridine–pyrroline rearrangement was featured in our syntheses of supinidine (48)[29] and hastanecine (49),[30] among others. In addition to FVP, a low-temperature rearrangement by nucleophilic opening and reclosure was also employed.[31] Vinyloxirane–dihydrofuran rearrangement was employed in our synthesis of ipomeamarone (50).[32] It would appear that the various forms of the rearrangement resurfaced during the first decade of the 21st century. Somfai’s synthesis of anisomycin (51)[33] featured a vinylaziridine–pyrroline rearrangement. Njardarson’s synthesis of biotin (52)[34] featured a vinylthiirane–dihydrothiophene rearrangement, whereas salviasperanol (53)[35] was prepared by Majetich via vinyloxirane–dihydrofuran rearrangement.

From the foregoing discussion, it is clear that the use of FVP had an enormous impact on the total synthesis of natural products via the vinylcyclopropane–cyclopentene rearrangement and its heterocyclic variants. Only a few examples of this technology have been discussed in this chapter, and for a more detailed listing, recent reviews can be consulted.[6f,6g,7]

Preparative Organic Electrochemistry

The use of electrochemistry in synthetic organic chemistry is not very common. It is widely used by physical organic chemists and, of course, in industry for the manufacturing of many commodity chemicals. In preparative organic chemistry, electrochemistry represents an ideal technology for oxidations and reductions by removal or addition of electrons by Faraday units. For this reason, we have referred to it as a “no reagent approach” to synthesis.[36] In the United States, this technology has been widely used by several groups in methodology development as well as in application to natural product synthesis. Incorporation of electrochemistry into daily usage requires some investment into instrumentation and a bit of a learning curve on the part of the chemist. When an electrochemical oxidation or reduction is appropriate for a solution of a problem and when such a transformation is successful, it is far superior to any traditional method. The applications of electrochemistry to organic synthesis have been reviewed and appear to enjoy a recent renaissance.[37]

A few beautiful examples are shown in Figures and 7. Little employed cathodic reduction followed by oxidation in his synthesis of hirsutene to generate the azo intermediate 57 as a direct precursor to the diyl 58 whose cyclization produced the linear triquinane 59, which was converted to hirsutene (32), as shown in Figure .[38] Little also applied electrochemical reductive cyclizations in his synthesis of quadrone (64),[39] as shown in Figure .

Figure 6

Little’s syntheses of hirsutene (32) and quadrone (64).

Figure 7

Electrochemical furan annulations by Wright and Moeller.

Wright applied electrochemical furan annulations and [4 + 3] cycloadditions to the synthesis of tricyclic systems such as 67,[40] envisioned to be useful in an approach to the nerve growth factors such as erinacine (68), as shown in Figure . The electrochemical furan annulations were developed into a general method of synthesis.[41] Moeller applied this methodology to a creative synthesis of alliacol (71) in 2004.[42] Other notable examples of electrochemical transformations can be found in the cited reviews.

We entered this field by accident (literally) and out of dire necessity because the project we were working on presented serious hazards. In the mid-1990s, we were asked by Novartis Crop Protection, Inc. to synthesize some presumed metabolites of the insecticide pymetrozine (72) isolated from several soil samples. The structures of the two metabolites were assigned by mass spectrometry, and synthetic samples were required for matching. Eventually, a large amount of both metabolites would be required and a reliable procedure provided to prepare 14C-labeled samples for soil fate studies. The synthesis of both 73 and 74 was extremely arduous and required anionic hydroxylation, as shown in Figure . Several attempts at the synthesis of 73 and 74 by various condensation approaches failed. The anionic hydroxylation shown in Figure did provide small amounts of the desired metabolite after multiple chromatographic separations. We adjusted the synthesis to a 20 g scale and successfully ran 16 reactions, producing ∼2.6 grams of 73 per run. The 17th attempt exploded during the warm-up phase completely destroying the fume hood [the lithium peroxide intermediate, small peroxides derived from s-BuLi, oxygen, and tetrahydrofuran formed a very hazardous mixture]. Because >150 grams of product was eventually required, we had to get creative [it was estimated that to complete this task by the current method, at least 45 new fume hoods would have to be acquired!].

Figure 8

Anionic hydroxylation of pymetrozine.

The solution to the problem was provided by the anodic oxidation of pymetrozine, as shown in Figure , and proved to be remarkably effective. We were able to eventually produce >150 g of the desired metabolite 73 as well as 74, by performing the electrochemical oxidation on different derivatives of pymetrozine.[43] The advantage of using the electrochemical method was clear: no hazard, clean conversion, effective solution to the problem, and great savings in time [to produce the initial small samples took >6 months of effort. To prepare 150 g of 73 by electrochemical oxidation took 3 days!].

Figure 9

Electrochemical oxidation of pymetrozine on a medium scale.

Encouraged by these results, we have initiated our own forays into electrochemistry with the intent to develop selective methods of oxidation and reduction. Some examples of such selective transformations are shown in Figure .

Figure 10

Examples of electrochemical oxidations and reductions.

For example, anodic oxidation of 77 under the conditions similar to those used for pymetrozine produced a mixture of hemiaminals 78 that were then cyclized under Lewis acidic conditions to octahydroisoquinoline 79, an intermediate in one of our approaches to morphine.[44] Electrochemical reduction of vinyl bromide in 80 was accomplished with mercury pool cathode and proved the method of choice for generation of many intermediates used for the synthesis of inositols.[45] A highly selective reduction of vinyl iodide over vinyl bromide in 81 was also achieved, as was oxidation of the diene diol 82 to the syn epoxide 84 (via the intermediate bromohydrin 83).[45] Electrochemical reduction of the alkenyl bromide 85 derived chemoenzymatically from p-bromothioanisole was accomplished selectively using a mercury pool cathode to furnish diene diol 86,[46] as shown in Figure . Such selectivity might not be attainable by using conventional reducing agents. It should be noted that diol 86 was not available by enzymatic dihydroxylation of thioanisole (only the corresponding sulfoxide was obtained).[46]

To expand the utility of electrochemical reductions, we became interested in the selectivity of reduction of allyl and cinnamyl groups as a means of differential protection in the synthesis of conduritols. Some examples of highly selective reductions are shown in Figure . The selective deprotection of the cinnamyl group over allyl group was accomplished in the bis-ethers 87 and 90 at −2.8 V (Hg pool).[47] Note the comparison with the classic dissolved-metal reduction, which was not selective at all. It is interesting that the exo-cinnamyl group underwent reduction preferentially to the endo-cinnamyl unit in 93, furnishing a good yield of the allylic alcohol 94 (itself not reduced further), whereas the dissolved-metal reduction produced only the fully reduced alcohol 95.[47] Substrate 96 contains two cinnamyl ethers and three allylic C–O bonds. At −2.91 V (Hg pool), only the exocyclic cinnamyl ether is reduced to afford alcohol 97 in 78% yield.[47]

Figure 11

Selective reductions of cinnamyl over allylic ethers.

Encouraged by these remarkably selective reductions, we examined reductions of cinnamyl groups in the presence of other functionalities, as shown in Figure . Thus, a selective reduction of a cinnamyl over benzyl ether was accomplished with 98.[48] The cinnamyl group was reduced preferentially from oxygen versus nitrogen in ether 99 and from the ester moiety in ester-amide 100.[48] The use of a reticulated vitreous carbon electrode allowed the selective reduction of the vinyl bromide in 101, whereas both the bromine and the cinnamyl ether were reduced at Hg-pool cathode.[48] Cinnamyl carbamate was selectively deprotected over allyl carbamate in 102 in 79% yield.[48] Finally, selective reduction of the cinnamyl group was accomplished from carbonates versus carbamates, as shown for 103,[49]Figure .

Figure 12

Selectivity in reductions of the cinnamyl group from various substrates: halide vs ether and oxygen vs nitrogen selectivity.

The examples shown above demonstrate that a remarkable level of selectivity is attainable by the use of electrochemical methods. There is, of course, a learning curve associated with their usage and a modest investment in equipment (potentiostat and electrodes) is required. The practitioner needs to learn how to obtain either cyclic or linear sweep voltammograms to gather information about the potential at which various electrochemical events may occur [this is the electrochemist’s equivalent of running NMR spectra of starting materials before committing them to a reaction].

Once the response of the substrate to the applied current or voltage is known, preparative scale electrolysis can be performed. There are also more sophisticated methods, such as the use of divided cells, but the examples shown in this section have all been performed under very simple conditions. Despite the simplicity of operation and the excellent chemoselectivity obtained in oxidations and reductions, electrochemistry remains on the fringes of synthetic organic chemistry. We hope that the foregoing discussion will help convince the traditionally minded organic chemists of the utility of electrochemical methods.

There has been some renaissance of interest in electrochemical methods. Two recent reviews have been published by Baran, in Angew. Chem. Int. Ed.[50] and in Chem. Rev.[51] listing various examples as well as benefits of electrochemical transformations. The latter review contains the word “renaissance” in the title and is truly exhaustive, with 914(!!) references. The sentiments expressed above related to the “activation energy” on part of the practitioner to be willing to engage in new technologies as well as the cost of operation are also echoed in Baran’s reviews. His recent work on the electrochemical oxidation in the synthesis of subglutinols A and B[52] and dixiamycin B[53] represents truly exceptional examples of creativity and efficiency in synthesis. Perhaps these disclosures will stimulate wider interest in organic electrochemistry.

Examples of the Use of Oxidoreductase Enzymes in Synthesis

The use of oxidoreductases in synthesis is not as common as the use of various lipases for desymmetrization of meso compounds or resolutions. This is likely because of two factors: first, oxidoreductases require cofactors, and therefore working with isolated enzymes would be prohibitively expensive, unless recycling loops are introduced into the experimental protocol. Second, the whole-cell transformations, in which the cofactors are provided by the life cycle of the cell, require specialized equipment [fermentors, autoclave, high-speed centrifuge, incubators, etc.], not readily available in a synthetic chemistry laboratory. For these reasons, whole-cell biocatalysis is not commonly employed in academia but is, of course, quite prevalent in the pharmaceutical industry. There are, however, many manuals available for the “uninitiated” organic chemist who is interested in biocatalysis as a method of choice for asymmetric synthesis.[54]

Baker’s Yeast Reductions

The reduction of carbonyl groups in ketoesters and/or diketones represents a widely used method in synthesis and is easily performed with conventional glassware available in a synthetic laboratory.[55]

In the 1970s, when I was in graduate school at Rice University, R. V. Stevens taught several “intense” courses in synthesis (both mechanisms and design of complex molecules). He always said to look at both sides of a functional group and/or draw synthetic targets in several different orientations because different ideas may transpire.

This advice constituted a major departure from the usual teaching of synthetic chemistry or even general organic chemistry [for example, any chapter on esters focuses primarily on the chemistry of the carbonyl group and not on that of the attendant alkoxide]. I have frequently used this principle and found it extremely beneficial in both synthetic design and the generation of new ideas.

One such idea is shown in Figure and to my knowledge was unprecedented at the time. Of the thousands of reported baker’s yeast reductions of ketoesters such as 104, none were performed with substrates that would contain a chiral center on the alkoxy portion. This was surprising and at the same time could be exploited as a means of resolution of the alkoxide(s), as depicted in Figure .

Figure 13

Principle of resolution of alkoxides containing a chiral center.

The principle is quite simple: in the reductions of substrates such as 107, containing additional chiral centers on the side chain, the level of induction diminishes with increasing distance from the reduction site. On the other hand, the distance from the alkoxide center to the reduction site in 109 is always constant and there was a good reason to believe that the reduction of compounds such as 109 would provide a means of resolving the alkoxides. Compounds 110 and 111 are NOT diastereomers; they are functionally different and are more easily separated than stereoisomers. Following their separation, a simple hydrolysis provides the resolved enantiomers of the alcohols 112. The results of the resolutions are shown in Figure . It became clear that reasonable levels of enantiomeric excess would be available only with alkoxides containing larger groups at the chiral carbon. Nevertheless, these results were useful and represent, to our knowledge, the only case where baker’s yeast reduction was used to resolve the alkoxide portion of the ketoester.[56] The reductions were also conducted with ketoamides, as shown in Figure . The level of resolution was found to depend somewhat on the steric bulk of the substituents at the chiral center.[57,58] Finally, the reduction protocol was applied to the ketoester 131 containing the azidodiene functionality, as shown in Figure .[59]

Figure 14

Resolution of ketoesters and ketoamides via baker’s yeast reductions.

Figure 15

Resolution of ketoester 131 and applications to total synthesis of pyrrolizidine alkaloids.

Reasonable levels of resolution were attained in the intermediates 132 and 133, which furnished, after separation and hydrolysis, the enantiomerically enriched alcohols 134 and 135. These alcohols then served as intermediates for enantiodivergent synthesis of pyrrolizidine alkaloids.[59] Finally, it was observed that at longer reaction times, the resolved ketoester 132, containing an acidic proton at the stereogenic center, and an enolized ketoester that can abstract this proton intramolecularly underwent racemization and were eventually completely converted to the single enantiomer 133, thus allowing all of the substrate mass to be converted to the hydroxyester.[59] The hydroxyester 133 does not have the internal base (enolate) and cannot racemize. This was an important observation that made application of this methodology useful in total synthesis.

I believe the above discussion underscores the value of the advice R. V. Stevens provided to his class at Rice University. No doubt there will be future applications of the microbial processes as these are always conducted in aqueous media and therefore are considered environmentally friendly.

Use of Toluene Dioxygenase in Synthesis of Natural Products

The last section of this article will highlight the use of arene cis-dihydrodiols in the synthesis of natural products. I was very fortunate to be able to enter this field with the help of my former Rice University classmate, Dr Larry Kwart, who joined my group in the late 1980s and brought the required know-how about fermentation chemistry. The concept of this then new program was simple and is outlined in Figure .

Figure 16

Enzymatic dihydroxylation of aromatic compounds by bacterial dioxygenase(s).

The transformations depicted in Figures and 17 are unique and represent one of the very few reactions [if not the only one] for which a synthetic equivalent has not been invented.[60] The combination of the enzymatic oxidation with traditional synthetic methods led to unprecedented levels of efficiency in many applications to the total synthesis of complex molecules. The discovery of this unique transformation is credited to David Gibson who isolated the first dihydrodiol[61] [although this transformation may have gone unobserved in the early 1900s[62]] and who later provided the robust Escherichia coli-based recombinant organisms for large-scale whole-cell fermentation[63] to either dihydrodiols 140 or catechols 141.

Figure 17

Enzymatic dihydroxylation of aromatic compounds.

This unique methodology has gone unnoticed by organic chemists for almost 20 years but did enter the field of synthesis in the early 1980s, and since then, many applications to enantioselective (and enantiodivergent) synthesis of complex natural products have been published.[64]

Like electrochemistry, the whole-cell fermentation requires a moderate investment into equipment. Sterile conditions and fermentors are required for work with the recombinant strains on scales of 10–15L.[65] However, the work with the blocked mutant strain, Pseudomonas putida 39/D, can be performed in ordinary glassware and has been described in an Org. Synth. publication.[66] The yields of metabolites derived from good substrates (halo- and alkylbenzenes) are as high as >30 g/L when carried out with the recombinant strains, whereas those obtained with P. putida 39/D yield only ∼200 mg/L. The diol derived from bromobenzene is available from Aldrich, although the price may discourage chemists from this commercial source [5 g of suspension in phosphate buffer, 423.5 CAD, catalog number 489492-5G]. The diols provide excellent starting materials for enantioselective synthesis and possess rich functional content, allowing for a variety of transformations, as shown in Figure .

Figure 18

Functional content of arene dihydrodiols and access to their enantiomers.

A frequent criticism of chemoenzymatic synthesis points out that only one enantiomer is accessible by the use of enzymes. The diols provide an excellent opportunity also for enantiodivergent synthesis. This is accomplished in several ways. According to the proposed model for the enzymatic dihydroxylation,[67] because the larger group directs the dihydroxylation, p-iodobromobenzene provides diol 144, in which the more reactive halogen is removed by reduction to furnish ent-146.[68] Enantiodivergent synthesis is also possible by considering symmetry of operations and the order of application of reagents, as has been amply demonstrated on many occasions.[69]

Several major reviews[64c,64d,64f,64g] published in the last decade provided compilations of natural products attained from the arene metabolites and/or summaries of the development of new methodologies by groups working in this area, primarily those of Lewis (Bath), Hudlicky (St. Catharines), Banwell (Canberra), Boyd (Belfast), and Gonzalez (Montevideo). Some of the recent accomplishments also take advantage of the ipso-diols derived from benzoic acid by mutant strains,[70] as shown in Figure . These metabolites have not been used as frequently as those derived from arenes by toluene (and naphthalene or biphenyl) dioxygenase(s) but a fair number of total syntheses originating in diols such as 148 have been reported. As in the case of toluene dioxygenase-mediated dihydroxylations, the precise mechanism of the benzoate dioxygenase-catalyzed production of diols of type 148 remains speculative, although some details have recently been disclosed.[71]

Figure 19

Metabolism of benzoic acids by Ralstonia eutrophus.

The structures shown in Table illustrate the diversity of targets attained from various metabolites by total synthesis. A selection of natural products that were synthesized since ∼2010 is provided. For natural product targets synthesized between 1987–2009, consult the compilations published in several major reviews.[64d,64f,64l,64m]

Table 1

Selected Total Syntheses from Arene cis-Dihydrodiols for the Period 2010–2019[72−109]

The observant reader may notice that all of the targets shown in Table have been attained from only 16 different diene diol metabolites. The detailed compilation of these diols conducted in 2004[64i] listed more than 400 such metabolites, obtained by dihydroxylation of arenes by various strains. Thus, more than 96% of these compounds have yet to find use in enantioselective synthesis! With such a large number of homochiral compounds available and with new metabolites being isolated, one would expect that many more applications will be forthcoming.

Outlook

This brief survey of four, somewhat unconventional, methods of organic synthesis should convince the reader of the enormous diversity in both targets and techniques that become available when the practitioner chooses to depart from the use of conventional methods of operation. The techniques discussed allowed for the synthesis of hundreds of targets that would otherwise not be attainable. With new technologies being introduced into the mainstream portfolio of a synthetic organic chemist, there may be few limits to the complexity of structures that can be synthesized, especially if organic chemists learn to extend their collaborations with biologists. Of special significance must be the recognition of value that biology and biological techniques offer to a synthetic chemist. The table of products that were synthesized from the enzymatically derived diene diols illustrates well the fact that none of these targets could have been easily reached by conventional synthetic methodology. In agreement with the spirit of Marc Tius’s quote,‡ it would not be an exaggeration to expect that the future of synthetic chemistry lies almost solely in the exploitation of biological methods. This premise will be validated (or not) by future generations of practitioners who do not fear venturing outside of their comfort zones and are willing to learn new things.