Frontiers in Natural Product Chemistry: Volume 3 E-Book
156,12 €
Mehr erfahren.
- Herausgeber: Bentham Science Publishers
- Kategorie: Fachliteratur
- Serie: Frontiers in Natural Product Chemistry
- Sprache: Englisch
Frontiers in Natural Product Chemistry is a book series devoted to publishing monographs that highlight important advances in natural product chemistry. The series covers all aspects of research in the chemistry and biochemistry of naturally occurring compounds including coverage of work on natural substances of land and sea and of plants, microbes and animals. Reviews of structure elucidation, biological activity, organic and experimental synthesis of natural products as well as developments of new methods are included.
The third volume of the series brings seven reviews covering natural products from marine plant sources, natural oligosaccharides, topical sesquiterpenes for pain treatment, biological activity of piperidinols and much more.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 645
Veröffentlichungsjahr: 2017
Ähnliche
Frontiers in Natural Product Chemistry
(Volume 3)
Edited by
Atta-ur-Rahman, FRS
BENTHAM SCIENCE PUBLISHERS LTD.
End User License Agreement (for non-institutional, personal use)
This is an agreement between you and Bentham Science Publishers Ltd. Please read this License Agreement carefully before using the ebook/echapter/ejournal (“Work”). Your use of the Work constitutes your agreement to the terms and conditions set forth in this License Agreement. If you do not agree to these terms and conditions then you should not use the Work.
Bentham Science Publishers agrees to grant you a non-exclusive, non-transferable limited license to use the Work subject to and in accordance with the following terms and conditions. This License Agreement is for non-library, personal use only. For a library / institutional / multi user license in respect of the Work, please contact: [email protected].
Usage Rules:
All rights reserved: The Work is the subject of copyright and Bentham Science Publishers either owns the Work (and the copyright in it) or is licensed to distribute the Work. You shall not copy, reproduce, modify, remove, delete, augment, add to, publish, transmit, sell, resell, create derivative works from, or in any way exploit the Work or make the Work available for others to do any of the same, in any form or by any means, in whole or in part, in each case without the prior written permission of Bentham Science Publishers, unless stated otherwise in this License Agreement.You may download a copy of the Work on one occasion to one personal computer (including tablet, laptop, desktop, or other such devices). You may make one back-up copy of the Work to avoid losing it. The following DRM (Digital Rights Management) policy may also be applicable to the Work at Bentham Science Publishers’ election, acting in its sole discretion:25 ‘copy’ commands can be executed every 7 days in respect of the Work. The text selected for copying cannot extend to more than a single page. Each time a text ‘copy’ command is executed, irrespective of whether the text selection is made from within one page or from separate pages, it will be considered as a separate / individual ‘copy’ command.25 pages only from the Work can be printed every 7 days.3. The unauthorised use or distribution of copyrighted or other proprietary content is illegal and could subject you to liability for substantial money damages. You will be liable for any damage resulting from your misuse of the Work or any violation of this License Agreement, including any infringement by you of copyrights or proprietary rights.
Disclaimer:
Bentham Science Publishers does not guarantee that the information in the Work is error-free, or warrant that it will meet your requirements or that access to the Work will be uninterrupted or error-free. The Work is provided "as is" without warranty of any kind, either express or implied or statutory, including, without limitation, implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the results and performance of the Work is assumed by you. No responsibility is assumed by Bentham Science Publishers, its staff, editors and/or authors for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products instruction, advertisements or ideas contained in the Work.
Limitation of Liability:
In no event will Bentham Science Publishers, its staff, editors and/or authors, be liable for any damages, including, without limitation, special, incidental and/or consequential damages and/or damages for lost data and/or profits arising out of (whether directly or indirectly) the use or inability to use the Work. The entire liability of Bentham Science Publishers shall be limited to the amount actually paid by you for the Work.
General:
Any dispute or claim arising out of or in connection with this License Agreement or the Work (including non-contractual disputes or claims) will be governed by and construed in accordance with the laws of the U.A.E. as applied in the Emirate of Dubai. Each party agrees that the courts of the Emirate of Dubai shall have exclusive jurisdiction to settle any dispute or claim arising out of or in connection with this License Agreement or the Work (including non-contractual disputes or claims).Your rights under this License Agreement will automatically terminate without notice and without the need for a court order if at any point you breach any terms of this License Agreement. In no event will any delay or failure by Bentham Science Publishers in enforcing your compliance with this License Agreement constitute a waiver of any of its rights.You acknowledge that you have read this License Agreement, and agree to be bound by its terms and conditions. To the extent that any other terms and conditions presented on any website of Bentham Science Publishers conflict with, or are inconsistent with, the terms and conditions set out in this License Agreement, you acknowledge that the terms and conditions set out in this License Agreement shall prevail.Bentham Science Publishers Ltd. Executive Suite Y - 2 PO Box 7917, Saif Zone Sharjah, U.A.E. Email: [email protected]
PREFACE
Frontiers in Natural Product Chemistry presents recent advances in the chemistry and biochemistry of naturally occurring compounds. It covers a range of topics including important researches on natural substances of plants, microbes and animals. The book is a valuable resource for pharmaceutical scientists and postgraduate students seeking updated and critically important information in natural product chemistry. The chapters are written by authorities in the field. The contents of the present volume represent exciting recent researches on structure elucidation, biological activity, and synthesis of natural products as well as developments of new methods. I hope that the readers will find these reviews valuable and thought provoking so that they may trigger further research in the quest for the new and novel therapies against various diseases.
I am grateful for the timely efforts made by the editorial personnel, especially Mr. Mahmood Alam (Director Publications), and Mr. Shehzad Naqvi (Senior Manager Publications) at Bentham Science Publishers.
Atta-ur-Rahman, FRS Kings College University of Cambridge Cambridge, UKList of Contributors
Microbial Involvement in the Production of Natural Products by Plants, Marine Invertebrates and Other Organisms
Lesley-Ann Giddings1,*,David J. Newman2Abstract
In the early 1980s, there were occasional reports of natural products isolated from marine invertebrates that were either identical to compounds from terrestrial sources, or were close chemical relatives. Since that time period it has become evident that microbes, whether they can currently be fermented under “normal conditions” or require genetic analyses and subsequent elaboration in surrogate hosts etc., are very heavily involved in the production of marine invertebrate secondary metabolites.
In the last few years, the situation with plant-derived natural products is very reminiscent of the early 1980s / marine invertebrate stories, as there are now significant numbers of reports invoking microbes (usually endophytic fungi), in the production of nominally plant-derived natural products. In one particular case, that of maytansine, the production by epiphytic root bacteria in the nominal producing plant is definitive.
Each issue of current journals covering genetic analyses of plants or marine invertebrates, often contains at least one article (basic science or review), that furthers the potential involvement of microbes in the production of even well-known molecules such as taxol, vinca alkaloids, homoharringtonine on the plant side and pederin-related (e.g. onnamide) derivatives on the marine side. We will also give information on bacterial, fungal and algal interactions that together lead to the production of natural products, though the exact involvement may not yet be known. We will broadly discuss the current situation and then hone in on areas where microbial involvement is definitive, and give the evidence for areas where it is still circumstantial.
1. INTRODUCTION
Since Alexander Fleming’s serendipitous discovery of penicillin from the fungus Penicillium notatum in 1928, microbe-derived natural products have been a prolific source of clinically-approved drugs. The clinical use of penicillin marked the beginning of the “Golden Age of Antibiotics” in the 1940s, resulting in the extensive investigation of microorganisms as sources of new therapeutics. Major breakthroughs were made in the area of drug discovery, including the development of blockbuster drug classes, such as the penicillins, cephalosporins, aminoglycosides, tetracyclines, cyclosporines, erythromycins, ivermectins, rapamycins, and cholesterol-lowering statins [1]. However, in the late 1970s, the frequency of finding structurally novel compounds decreased, as chemists had already exploited easily accessible microbes from terrestrial environments.
In the 1980s, pharmaceutical companies began to refocus their drug discovery efforts toward developing synthetic drugs using combinatorial chemistry. Unfortunately, this shift did not lead to a significant increase in the number of clinically-approved small molecule drugs. As a matter-of-fact, from 1981 to 2014, over 50% of all small molecule new chemical entities were classified as natural products, (semi)synthetic derivatives, or based on natural product pharma-cophores [2], demonstrating the constancy of the number of natural product-inspired drugs in the clinical pipeline, though during this time frame, pharmaceutical companies were abandoning natural products altogether. Not surprisingly, the structural complexity and diversity, high selectivity, and biological activity of these molecules from marine or terrestrial microbes, plants, and other organisms make them invaluable pharmaceuticals. Researchers now have an even stronger argument for revisiting natural products for drug discovery.
Interestingly, microbes from all three domains of life, Archaea, Prokarya, and Eukarya, have been either identified as the producer of natural products or speculated to be involved in their production via symbiotic associations. Over the past twenty or so years, evidence of microbes from all sources being involved in the production of bioactive agents, has grown from being a slight possibility, to now at least in the area of marine natural products, being a major productive source(s) of bioactive agents reported from Porifera and perhaps other marine invertebrates. Confirmation of the “true producer” of these invertebrate-linked molecules has been rather difficult, because <1% of microbes that can be visualized in seawater and invertebrates / sediments by direct staining, could be cultivated under standard laboratory conditions. We should point out at this stage that the methodologies often used in past studies, were based on media that were “just too rich in carbon-containing components” when compared to the levels in seawater. However, advances in culture-dependent and -independent techniques, such as (meta)-genomic and single cell sequencing, cell sorting, and other molecular approaches, have provided insight into genomes and microbiomes, enabling researchers to track, isolate, and validate the true producer(s) of these metabolites. We are now realizing the nominal “i.e., collected source” of marine natural products may not necessarily be the producing organism.
In the case of plant-derived bioactive compounds, more reports of production by mainly epi- and endophytic microbes, usually fungi and actinobacteria have surfaced, though in some cases, production may possibly be due to horizontal gene transfer or genetic recombination occurring between plants and associated symbionts. These reports demonstrated that microbes can produce low levels of a “plant metabolite” upon fermentation but upon subsequent sub-culturing, the microbe was reported to lose its ability to produce this metabolite. However, in the last few years, academic researchers have started to use (rediscovered?) techniques that were commonly used in the pharmaceutical industry to find antibiotic producers but never formally published. For example, the supplementation of fermentation broths with extracts of parts of the “nominal producing source” can be used to induce or maintain the production of meta-bolites of interest.
In recent years, there have been reports of new bioactive compounds produced by microbial consortia, which contain specialized mutualistic and / or parasitic relationships. In mutualism, the interactions between two or more species are beneficial to all parties (e.g., nutrition or protection), whereas microbes exploit each other in parasitism (e.g., competition for resources). These interactions are mediated by chemical signals transmitted between the host and its microbiome. Researchers have been accustomed to the idea of a single microbe producing a given compound, when in reality microbes in nature rarely grow under axenic conditions. There are an infinite number of complex microbial interactions found in nature, and now mixed cultures are being recognized for playing a role in the “production” of new secondary metabolites, usually via silencing or activating the expression of biosynthetic gene clusters (BGCs) in one or more of organisms.
In this chapter, we describe the isolation and characterization of bioactive compounds in which the host was thought to be the producing organism, plus compounds found from mutual interactions between microbes in or around a host or hosts. Identification of the true producers of bioactive natural products, may ultimately aid in the production of these agents by use of techniques such as controllable activators of BGCs in surrogate hosts. These examples are chosen to demonstrate the metabolic and chemical diversity that arises from the unique environments created from microbial interactions with other organisms, and suggest ways in which these rich sources can be used in the future.
2. MARINE-DERIVED BIOACTIVE AGENTS AND MICROBES
The ocean is one of the largest unexplored sources of specialized metabolites due to its inaccessibility, as more than 70% of the Earth is covered by water, mainly in oceans, with an average depth of 3800 metres. The oceans contain 105–106 bacteria per milliliter of seawater, which is the equivalent of 1012 tons of bacterial weight. The rich chemical diversity found in the world’s oceans has provided a plethora of specialized metabolites with unique carbon skeletons and varying degrees of halogenation. Although numerous compounds have been reported with diverse biological activities, very few have been definitively proven to originate from invertebrates. There is sufficient evidence invoking microbial production of a significant number of these bioactive metabolites with for example, the number of compounds isolated from blue-green algae (cyanobacteria, which are prokaryotes though often described as microalgae in literature even as late as the early 1980s), as well as other bacteria and fungi isolated from sediments (both shallow and abyssal) or from invertebrates [3]. There is no question of the “nominal actual producer” in these cases, since fermentation of the isolated microorganism produced the compounds of interest. Even in such a case, the original source(s) of the bioactive gene clusters (BGCs) may not be known with any certainty due to gene transfer between microbes. Several bioactive compounds and / or their derivatives from marine sources that are currently in clinical trials have been reported, or speculated to be, microbe-derived. The web site curated by Professor Alejandro Mayer at Midwestern University’s Department of Pharmacology should be checked for the most up to date information in this regard (http://marinepharmacology.midwestern.edu/)
2.1. Pederin, Mycalamides, Onnamides And Similar Molecules
The story of pederin and structurally related compounds is quite remarkable, as it shows how the BGC that was the source of a toxin found in a Brazilian blister beetle, was 50 years later identified as the producer of metabolites found in marine sponges collected in different parts of the world, though the bacterial host differed. The effects of this toxin on humans were first noticed in Brazil in 1912 [4], and it took 37 years for the active principle to finally be isolated from the rove beetle Paederis fuscipes and a partial structure to be assigned [5]. In 1965, Cardani et al. [6] proposed an initial structure of the toxic principle, now named pederin; however, this structure was later revised in 1968 by Matsumoto et al. [7] to the structure shown Fig. (1-1). This very interesting chemical structure led to a number of different chemical syntheses of pederin and analogs being published [8, 9], though not until after the reports of the marine-sourced compounds discussed below.
Fig. (1)) Pederin-Related Structures; 1 – 11.In the middle to late 1980s, the Blunt and Munro group at the University of Canterbury in New Zealand reported the isolation and identification of mycalamides A Fig. (1-2) and B Fig . (1-3). These were extracted from a Mycale sp., (Porifera; marine sponge) collected at approximately 40 metre depth in cold water (2 °C) off Dunedin in South Island, New Zealand [10, 11]. Inspection of the structures of these two compounds shows that only relatively minor changes (i.e., methylation or lack of methylation of hydroxyl groups and ring closure), occurred when compared with the pederin nucleus Fig. (1; 1). Almost simultaneously, an international group from the University of the Ryukus in Japan and SeaPharm, Inc., in Florida published the structure of onnamide A Fig. (1-4), which was isolated from a Theonella sponge species collected off Okinawa in warm (+30 °C) water [12]. Not only did all of these compounds possess antiviral and cytotoxic biological activities, but they also contained a core structure defined by two tetrahydropyran moieties and an exomethylene group and like pederin Fig. (1-1), these molecules were also powerful vesicants. Since these initial reports, more than 30 related compounds have been reported from a variety of sponge genera collected all over the Pacific. An excellent review giving details of the chemistry of these and related compounds / structures, together with data on biological structure-activity relationships was published in 2012 by Mosey and Floreancig [9]. This paper should be read by interested parties, particularly in conjunction with the report below that shows that an as yet unculturable microbe is the actual producer of onnamide, not the sponge from which this compound was originally isolated.
2.1.1. Actual Producers of Pederin-related Molecules
Now one could ask, what is the relationship between Paederus beetles and deep-water marine sponges from different locales? The following reports identifying the true producer of pederin-related molecules reveal the connection. In 1999, German entomologist Rupert Kellner published a very interesting paper entitled “What is the basis of pederin polymorphism in the Paederus riparus rove beetle? The endosymbiont hypothesis” [13]. In it, he presented data suggesting an endosymbiont may be the actual producer of the toxin. Two years later, he reported the suppression of pederin biosynthesis in the closely related species, Paederus sabaeus, when antibiotics were used to eliminate endosymbionts, implying a common bacterial component in the production of pederin in two different species of the beetle [14].
However, to bring the story to its climax, one needs to return to the marine environment. From 1988 to early 2000, there were reports that a significant number of sponge extracts contained more pederin-related molecules, such as others in the onnamide class, with onnamide F Fig. (1-5) being an example [15] together with the theopederins, examples being the structurally similar theopederin K Fig. (1-6) and L Fig. (1-7) [16]. Then in 2002, Kellner identified an endosymbiont from Paederus beetles that was related to the well-known Gram negative bacterium Pseudomonas aeruginosa, and then demonstrated that the interspecific transmission of the endosymbiont was related to the genetic makeup of individual isolates from beetles [17, 18].
In the period 2002–2005, Piel’s research group in Germany reported genetic analyses of these Paederus-related microbes. In these experiments, genetic probes were used to identify ketosynthase domains in the polyketide synthase (PKS) gene clusters that encoded pederin in this symbiotic pseudomonad [19, 20]. These genetic probes were later used by Piel, in collaboration with a Japanese group led by Fusetani and Matsunaga at the University of Tokyo, to investigate the production of the closely related onnamides, which as mentioned earlier, were originally isolated from the Japanese sponge Theonella swinhoei (yellow variant), collected in warm, shallow waters off of Okinawa [12]. The “pseudomonal-based genes” were detected in the sponge metagenome, thus the investigators were able to locate the nexus of the biosynthesis to an as-yet-uncultured symbiont in the sponge [21]. Preliminary details of those studies were then published in a short review in the Journal of Natural Products in 2005 [22]. Piel also demonstrated evidence for what is now known as a “symbiosis island” that permitted horizontal acquisition of the pederin biosynthetic capabilities in Paderus fuscipes [23]. Six years later, in 2011, Kador et al. published specific oligonucleotide probes that could be used to detect pederin producers in Paederus beetles thus effectively closing the “genetic circle” [24].
Just to bring the onnamide story up to early 2017, we will mention the seminal paper from the Piel group in 2014 [25], in which they identifed the as yet uncultured microbe from the onnamide producing sponge, genetically amplified the DNA from “one microbe” and then proceeded to prove that this one microbe was from a new phylum, provisionally named as “Tectomicrobia”. At that time they demonstrated that the original isolate appeared to be two very closely related bacteria from genetic analyses on single cells as BGCs appeared to be different. In 2017, the Piel group published further evidence that this was the case, with the original isolate proven to be two very closely related filamentous bacteria, “Candidatus Entotheonella factor” and “Candidatus Entotheonella gemina” [26], with both being producers of onnamide in the sponge Theonella swinhoei Y (yellow variant).
Returning to other pederin-related compounds from the marine environment whose “true producers are not yet identified”, in 2004, the Pettit group reported the discovery of irciniastatin A Fig. (1-8) and B Fig. (1-9) and other cytotoxic pederin derivatives [27] from the Indo-Pacific marine sponge Ircinia ramosa. Irciniastatin A was subsequently reported as psymberin by the Crews’ group the same year [28] from another Pacific sponge, Psammocinia sp. Careful inspection of the supporting information in the paper from the Crews’ group revealed they knew that the same compound under a different name, and from a different sponge genus, was in the process of publication by the Pettit group. Since the Pettit group had an earlier submission date than their paper, Pettit has priority for this finding.
The difficulty both groups had with isolating these compounds from a sponge extract (extremely low levels in the extracts), may well be further evidence of a symbiont being the producer of these molecules. In 2009, Fisch and coworkers were able to amplify a ketosynthase domain of a trans-AT PKS gene cluster involved in irciniastatin A biosynthesis, from the metagenome of the sponge Psammocinia aff. bulbosa [29]. This potentially provided a way to study the uncultivated sponge bacteria, which is most likely its true producer (cf the story on onnamide above). Although the function of this gene cluster has not been experimentally validated, a symbiont is most likely involved in the biosynthesis of these compounds, as the sequence is identical to counterparts in the metagenomes of sponges from distant locations containing polyketide-producing bacterial symbionts [30]. Due to the very interesting chemistry of these compounds, numbers of irciniastatin A/ psymberin syntheses have been published in the last few years [31, 32].
The two latest additions to the pederin family are diaphorin Fig. (1-10) and nosperin Fig. (1-11) but as will be seen, these marked a return to terrestrial environments. In 2013, diaphorin Fig. (1-10) was isolated from a β-proteobacterium symbiont Candidatus profftella armatura dwelling in the Asian citrus psyllid Diaphorina citrid [33, 34]. Once the D. citri bacteriome was sequenced, the Ca. profftella armatura genome was reassembled from 59 reads from the syncytial cytoplasm. Analyses showed the presence of large portions of PKS BGCs remarkably similar to those involved in pederin biosynthesis. Only the upstream half of the diaphorin biosynthetic multidomain PKS gene resembled the pederin counterpart, as orthologs of the two O-methyl transferases were missing. It would not be surprising if these genes were acquired via horizontal gene transfer from a predator-prey relationship, as the Paederus beetle feeds on hemipteran insects. Diaphorin has cytotoxic activity against human HeLa and rat neuroblastoma cells. Nosperin Fig. (1-11), another pederin-like compound, was reported in lichens by Kampa and coworkers in 2013 [35]. This is one of the rare cases in which a lichen-derived polyketide is made by a bacterial photobiont rather than a fungus. Using metagenomics, the true producer of this compound was determined to be a lichen-associated Nostoc sp. cyanobacterium, suggesting a role for these compounds in symbiosis, and that these biosynthetic genes are in fact widespread from the marine to the terrestrial sphere.
What began as a discussion of the toxin produced by the blister beetle found in Brazilian forests / jungle and in other parts of the World, led to the following. The ability to identify and express genetic loci related to the biosynthesis of pederin-related molecules, and finding these genes in locations not even thought to be possible. The beetle toxin was, in fact, used by Nature to generate molecules in organisms as diverse as shallow and deep-water marine sponges, in warm (close to 30 °C) and cold (2 °C) water environments and even in terrestrial lichens. None of these were thought of in the wildest dreams of the original researchers working on beetle toxins.
2.2. Trabectedin, A Naphthyridinomycin / Tetrahydroisoquinoline Derivative
As mentioned earlier, marine microbes have been receiving a lot more attention in recent years, as several compounds found in marine environments have led to approved drugs and / or clinical candidates, some of which may be produced by symbiotic microbes. One of the clinically approved drugs, the tetrahydro-isoquinoline alkaloid trabectedin (ecteinascidin-743, ET-743, Yondelis®: Fig. (2-12), was the first compound “directly from the sea” (i.e. unmodified structure) to be approved for the treatment of cancer, and is an excellent example of a compound originally isolated from a marine tunicate that is now almost certainly produced by symbiotic bacteria.
As background to the trabectedin story, some earlier history is necessary. In 1982, the Faulkner group at the Scripps Institute of Oceanography reported the isolation of renieramycin A Fig. (2-13) from the Eastern Pacific sponge Reniera sp [36]. This material had antibiotic properties with a structure similar to those of known antitumor agents of the saframycin class. The saframycins had been reported five years earlier by Takahashi and Kubo from the terrestrial microbe, S. lavendulae [37]. Two later papers gave the structures of saframycins B Fig. (2-14) and C Fig. (2-15) [38], followed by the structure of saframycin A Fig. (2-16) the next year [39]. Then in 1988, the isolation of saframycin Mx1 Fig. (2-17) from the myxobacterium Myxococcus xanthus strain Mx48 was reported by Irschik et al. [40]. Thus, in just over 10 years, closely related antibacterial and antitumor compounds had been isolated from terrestrial streptomycetes and myxobacteria, and from a marine sponge. However, these were only the “later tips of the iceberg”, as the base molecule for all these agents, naphthyridinomycin, Fig. (2-18) was initially reported by Canadian scientists in 1974 [41] and 1975 [42], from the terrestrial streptomycete Streptomyces lusitanus AY B-1026.
Fig. (2)) Trabectedin-Related Structures; 12–20.In the middle 1980s to early 1990s, the Rinehart group at the University of Illinois at Champaign-Urbana, in conjunction with the Wright group at Harbor Branch Oceanographic Institution in Florida, published two back to back papers in the Journal of Organic Chemistry showing the structures of the cytotoxic agent ET743 Fig. (2-18) and its congeners, isolated from the Caribbean tunicate Ecteinascidia turbinata [43, 44]. These reports were an extension of the work reported by Holt in 1986 in his PhD thesis completed while in the Rinehart group [45]. That this organism “produced” a cytotoxic compound or compounds was originally reported in 1969 at a scientific meeting by Sigel et al., and then formally published in book format in 1970 [46]. These “marine compounds” were obviously built on the same basic chemical structure reported for naphthyridinomycin, saframycins, and renieramycin. Therefore, one now had multiple bioactive compounds that must have been produced by a similar set of biosynthetic clusters, though it was unknown at the time what the organism or organisms might be, but due to the multiplicity of “nominal sources” microbes were prime candidates.
ET743 became an approved antitumor drug under the aegis of the Spanish company PharmaMar and the methods used in its production ranged from massive large-scale collections, aquaculture of the tunicate in sea and in lakes, which gave enough material for initial clinical trials. In order to be able to continue clinical trials beyond Phase II, PharmaMar then moved to large-scale fermentation of the marine bacterial product, cyanosafracin B Fig. (2-19) followed by semi-synthesis to produce ET743. The story leading to the production of ET743 has been presented by the PharmaMar team in a significant number of publications, and these should be consulted to see the manner by which the various problems were successfully overcome to finally produce a “current Good Manufacturing Practices (cGMP)” quality product [47-50].
In addition to the publications from the PharmaMar group on the semisynthetic processes they used, two other highly relevant reviews are the one in 2002 by Scott and Williams covering the chemistry and biology of the tetrahydroquinoline antibiotics [51], which was then followed in 2015, by another very thorough review from the Williams group on the ecteinascidins [52].
2.2.1. Actual Source of Trabectedin
From a microbial aspect, there were suggestions that the yet uncultured bacterium, Candidatus Endoecteinacidia frumentenis (AY054370), was involved in the production of these molecules. This organism had been found in E. turbinata samples that produced ET743 collected in both the Caribbean and the Mediterranean seas [53, 54]. Using the suggestions made by Piel on how to utilize symbionts from invertebrates [55], and then using knowledge of the organization of the BGCs of the saframycins [56] and safracin B [57] Fig. (2-20) as markers, in 2011 the Sherman group at the University of Michigan were able to identify the “contig” that encoded the NRPS biosynthetic enzymes involved in the ET743 complex. They were also able to identify the probable producing bacterium, as the yet uncultured microbe Candidatus Endoecteinascidia frumentensis, which was present in both the Caribbean and Mediterranean E. turbinata organisms [58]. Then four years later, the same research group directly confirmed the initial report [59]. In the process, they also demonstrated that the “producing bacterium”, Ca. E. frumentensis, may well represent a member of a new family of γ-proteobacteria and has an extensively streamlined genome similar to those of other symbiotic microbes [60], with most of the genetic machinery being devoted to this complex of compounds [61].
Due to assembling the complete genome, the Sherman group provided insight as to why trabectedin is not produced when attempts were made to cultivate this symbiont under “standard fermentation” conditions. Apparently, some of the genes involved in trabectedin are either missing, or somehow distributed throughout the Ca. E. frumentensis genome. Further analyses of the complete genome showed that genes involved in trabectedin biosynthesis appeared to be dispersed over 173 kb of the 631-kb genome.
Thus, the gene encoding the acyl carrier protein (ACP) is typically clustered together with other BGCs involved in producing natural products. However, this gene clusters with other fatty acid biosynthetic genes 61 kb downstream of the trabectedin gene cluster, suggesting that primary and secondary metabolism may work together to make this compound. A gene encoding the E3 component of the pyruvate dehydrogenase complex, which plays an important role in providing acetyl-CoA from the citric acid cycle, was found to be in close proximity to other genes involved in trabectedin biosynthesis. In addition, a number of key genes are also missing within this gene cluster that should produce other previously isolated precursors, suggesting Ca. E. frumentensis may work “symbiotically” with its host to produce this chemotherapeutic.
In addition, the trabectedin gene cluster is the only natural product gene cluster found in the microbe’s genome, suggesting that it may have an important ecological role in protecting marine invertebrates against predators. Ca. E. frumentensis as mentioned above, appears to have undergone a drastic genome reduction, as it has lost genes involved in DNA replication and repair mechanisms. These findings suggest that this microbe may have as its only function, production of this therapeutic class. We are bound to see reports of new routes to producing trabectedin now that we know Ca. E. frumentensis could not survive independently of its host.
2.3. Didemnins and Aplidine
The didemnins, a family of ascidian-derived cyclic depsipeptides, were the first marine natural products to enter Phase I clinical trials for the treatment of cancer. In the 1980s, the Rinehart group published the first reports on didemnins, including didemnin B Fig. (3-21), isolated from the Caribbean tunicate Trididemnum solidum and reported to have antitumor and antiviral properties. Aplidine Fig. (3-22), a derivative of didemnin B, was later isolated from the Mediterranean ascidian Aplidium albicans. The only difference between the structures of aplidine and didemnin B is the presence of a lactyl hydroxyl group on the terminal side chain of didemnin B, which in aplidine is replaced by a ketone. Interestingly, this small structural difference increases the potency of aplidine as anticancer agent and lowers its cardiotoxicity compared to didemnin B. Didemnin B was not developed beyond Phase II clinical trials due to a lack of response, acute cardiotoxicity, and neurotoxicity in observed patients. The detailed history of the isolation, biological activity, and clinical development of the didemnin family as well as aplidine (though only up to 2011), is very well covered in a 2012 review by Lee and co-workers [62].
Fig. (3)) Didemnin and Aplidine Structures 21 & 22.Aplidine has become PharmaMar’s second most advanced compound, being evaluated in a Phase II study for the treatment of aggressive non-Hodgkin lymphoma (PharmaMar; NCT00884286) and currently, in conjunction with dexamethasone, is the only non-approved marine-derived agent in Phase III clinical trials for multiple myeloma (NCT01102426; the ADMYRE trial). This combination has had its MAA, the EU equivalent of an NDA application to the US FDA, accepted by the EU for approval as a drug.
Not surprisingly, supply problems have hindered the development of this bioactive agent. Total synthesis has been used to produce aplidine and related compounds [62] for clinical studies, but now several reports have suggested that a microbial symbiont might be involved in the production of this class of bioactive secondary metabolites, especially due to their structural similarity to the didemnin metabolites produced by a free-living microbe from Japanese waters reported by Tsukimoto et al. [63]. In 2012, Xu and coworkers sequenced the genome of the marine α-proteobacterium Tistrella mobilis, a microbe very similar to one isolated by Tsukimoto et al. but in this case, isolated from the Red Sea instead of Japanese waters, and identified the didemnin gene cluster. Moreover, using imaging mass spectrometry, for the first time, the real-time conversion of didemnin X and Y precursors to didemnin B was observed in these experiments [64].
While the didemnins are very bioactive metabolites, the ecological role they play within their hosts remain to be determined, as they are toxic to the host, raising questions of how the host survives in the presence of these metabolites. More ecological studies need to be done to understand the host-symbiont relationship and identify where these molecules are localized. Such information may help investigators to gain insight as to how to improve their production. Furthermore, more didemnins are still being found, as demonstrated by the report in 2013 by Ankisetty et al., of the isolation of two new chlorinated didemnins with cytotoxic and anti-inflammatory activities from the tunicate Trididemnum solidum [65].
Metagenomic analyses of Aplidium albicans using Tistrella mobilis gene clusters as markers, may lead to the identification of the actual bacterial gene cluster involved in aplidine biosynthesis, because the production of both aplidine and its reduced congener, didemnin B, by the same free-living microbe has not yet been proven. In addition, following the identification of the didemnin B gene cluster, genetic engineering of the BGC to produce the ketone derivative might prove feasible. If so, this may also create a renewable supply of aplidine and related molecules via microbial fermentation.
2.4. Bryostatins
It is rather difficult to determine the true role of the metabolites produced by symbionts due to technical problems in manipulating obligate host-symbiont relationships. Compounds moving between symbionts and the host organism often can obfuscate the process of identifying the true producer. However, the bryostatins represent one of the few cases in which there is direct experimental evidence of symbiont-produced compounds used to defend the host. Bryostatins are a family of more than 20 bioactive macrocyclic lactones [66] that originate from the invasive marine bryozoan Bugula neritina but almost certainly have microbial origins. All metabolites in this family generally share a 20-member macrolactone core and three remotely functionalized polyhydropyran rings. Bryostatins structurally differ from one another by substitutions at C7 and C20 and the placement of a γ-lactone at either C19 or C23 in the polyhydropyran ring.
Bryostatins have a high binding affinity for protein kinase C (PKC) isozymes. These proteins play a major role in learning and memory, and animal models treated with bryostatin 1 have been reported to show improvements in these areas, demonstrating that bryostatins and analogs might be used to treat cognitive diseases. In addition, bryostatin 1 Fig . (4-23) has also been shown to restore hippocampal synapses and spatial learning and memory in adult fragile X mice. Bryostatin I was used as a test compound in over 80 Phase I or II clinical trials with or without cytotoxic agents for the treatment of various cancers, but none of these trials have had results warranting their continuation. Currently, the NIH clinical trials database (URL: clinicaltrials.gov) shows one trial at the Phase II level in Alzheimer’s disease (NCT02431468) under the aegis of Neurotrope, Inc.
Fig. (4)) Bryostatin 1 Structure; 23.The isolation of bryostatin 1 required heroic efforts to obtain enough material from B. neritina for initial clinical trials. Obtaining a sufficient supply from natural sources of this compound and other bryostatins, remains a major challenge, precluding significant additional studies. Other synthetically accessible bryostatins have been investigated and observed to exhibit similar bioactivities in cancer and PKC-related assays. Several bryostatins and analogs have been synthesized using methods, such as function-oriented synthesis [67] to make simplified analogs with comparable or improved biological activities, but the economical production of bryostatins via synthesis requires further investigation. In due course, it might be easier and more cost-effective to figure out how to culture the microbe or express putative gene cluster(s) involved in its production in a surrogate host.
In 1997, Haywood and Davidson used microscopy and genomic techniques to reveal that the true producer of bryostatin 1 was the yet uncultured, rod-shaped bacterium, Candidatus Endobugula sertula, located in the pallial sinus of larvae of bryostatin-producing bryozoans [68]. To confirm the role of this symbiont, one would reintroduce Ca. E. sertula to the cured bryozoan and look for the restoration of the host’s chemical defense. Since most marine symbionts are currently unculturable, reintroduction would be difficult to say the least. At the time, the most promising piece of evidence was the reduction in the amount of bryostatin 1, the most abundant bryostatin, in B. neritina colonies treated with antibiotic-treated larvae, suggesting these compounds are protective agents. Subsequently, in 1999, different strains of the symbiont associated with the production of different bryostatins were reported [69]. These results were further confirmed in 2004 when Lopanik et al. reported the levels of bryostatins in larvae and adult colonies. They then demonstrated that without bryostatin production, such larvae were food for predators, thus proving a role for the symbiont’s product [70].
Stronger evidence for the microbial production of bryostatin 1 came in 2007 from the Sherman group at the University of Michigan, working in conjunction with the Haygood and Lopanik groups, when they identified the putative bryostatin gene cluster from Ca. E. sertula. In vitro biochemical assays with heterologously expressed portions of a putative bryostatin PKS gene cluster confirmed the role of these genes in bryostatin biosynthesis, but the difficulty in expressing large, trans-AT PKSs have deterred their full characterization [71-73]. Over the next few years, this microbe was reported in other examples of B. neritina but appeared to have a latitudinal restriction and strain variation with depth, and various “sub-sets” of B. neritina were also described [74].
In 2014, Lopanik et al. published a paper speculating about the role of these compounds within B. neritina. These ideas were based on the differential gene expression in colonies with or without the symbiont. Interestingly, once bryostatin production is high, the symbiont induces the expression of glycosyl hydrolase family 9 and family 20 proteins, actin, and a Rho-GDP dissociation inhibitor within the host. Thus, these compounds appear to be ecologically relevant as they may regulate the distribution of the symbiont within B. neritina as a signal of defense capabilities, protecting larvae developing in the reproductive zooid from fish and other predators [75]. In a very recent paper, the same group demonstrated the “holobiont fitness” via specific interactions with the host organism’s proteinkinase C enzymes. Thus, even if the organism’s symbiont cannot yet be cultivated, its effect and product can be measured by modern techniques [76].
Even today, researchers are still reporting more derivatives from natural sources. Thus in 2015, Yu et al. identified new bryostatin derivatives from a bacterial symbiont in B. neritina from bryozoan colonies collected from the South China Sea [66].
In 2016, a very interesting paper was published by Wender et al. demonstrating the inhibition of Chikungunya virus-induced cell death by synthetically accessible bryostatin analogues. In this publication which involved “bryologs”, synthetic molecules based on the bryostatin skeleton, the authors demonstrated that this effect does not appear to be mediated by a PKC pathway, a dramatic contrast to the typical mechanism(s) of action of bryostatins and bryologs. In light of these findings, there are possibly new targets to be explored for the treatment of the Chikungunya virus by inhibition of its reproduction [67].
Thus, even after close to 50 years from the first reports of bryostatins, this family of compounds is still an active structural class for natural product chemists to find novel relatives from various sources, for synthetic chemists to modify, continuously expand the skeleton, and then for all to utilize in a search to find new biological targets.
2.5. Kahalalides
Mollusca is the largest marine phylum and these invertebrates are associated with numerous diverse bacteria, as bacterial density is typically 106 microbes per ml of sea water [77]. Interestingly, these marine invertebrates lack immunological memory, and have thus developed chemical defense strategies for protection against pathogens. These defensive chemicals are not only found in molluscs, but also in their mucus which is also thought to play a role in defense. Molluscs have been reported to synthesize or incorporate diverse secondary metabolites that may play a role in their predatory behavior, communication, and defense based on their “algal diet”. An example of these metabolites are the kahalalides, a family of depsipeptides of variable size and peptide sequences, ranging from tripeptides to tridecapeptides, and decorated with fatty acid chains of varying lengths. The cyclic depsipeptide kahalalide F Fig. (5-24), a fish deterrent and one of the most active antitumor metabolites of the kahalalide family, was first isolated from the herbivorous sacoglossan mollusc, Elysia rufescens, which grazes on the green macroalga Bryopsis sp.
Fig. (5)) Kahalalide F Structure; 24.Although algal prey has been reported to affect sacoglossan metabolomes, algae do not appear to be the source of the kahalalides. Following its isolation and structural characterization, kahalalide F was found in lower concentrations in the algae, compared to those of molluscs, when based on wet weight, suggesting this compound is likely a specialized metabolite produced by a symbiont. Hill et al. later filed patents in which they described the isolation of kahalalide F and derivatives from V. mediterranei / shilonii isolated from Bryopsis and E. rufesens. Using liquid chromatography-mass spectrometry (LCMS) and nuclear magnetic resonance (NMR), the Hill group isolated kahalalide F from symbionts. However, this metabolite was not consistently produced by these Vibrio strains, thus more studies were needed to determine the true producing organism(s).
In an effort to understand the microbial diversity within E. rufescens and its mucus, Davis et al. [77] performed deep sequencing on this organism and its mucus, looking at the distribution of bacterial symbionts among the samples. Interestingly, mucus samples were richer in bacteria, including the rare Vibrio species, than the mollusc itself, suggesting the mucus may be involved in the recruitment and selection of certain bacteria. Epifluorescence and MALDI-MS imaging enabled the visualization of autofluorescent chloroplasts as well as the presence of kahalalide F in the outer region of E. rufescens, possibly in the mucus. Once the dynamics and composition of the microbial communities associated with the mucus are better understood, large-scale fermentation of the true producer(s), can be used to generate renewable supplies of this depsipeptide. For more details on the isolation, structural elucidation, and biological activity of kahalalide F and analogs, the 2011 review by Gao and Hamann should be consulted [78].
Related molecules have been isolated from other Indo-Pacific molluscs, such as E. grandifolia and E. ornata, and possibly from more Elysia and related genera / species, due to the ambiguity in a number of these taxa [79]. Information giving further insight(s) into the kahalalide-producing BGCs and clues on how to cultivate the symbionts should be obtained once the kahalalides are properly surveyed, and the metagenomes of a variety of molluscs are characterized. In the meantime, efficient solid-phase peptide syntheses can be used to obtain large quantities of kahalalide F and derivatives.
Kahalalide F (PM-92102) was licensed to PharmaMar in the 1990s by the University of Hawaii and entered clinical trials for the treatment of cancer, but there have been no recent developments reported for this compound in relevant clinical trials databases. Isokahalalide F was developed at PharmaMar and entered a number of clinical trials for the treatment of cancer, but in 2012, the company decided to redirect its resources towards other drugs in the pipeline. Nevertheless, the search for new kahalalides and analogs is ongoing due to interest in their cytotoxic properties.
2.6. Dolastatins
The dolastatins are diet-derived cytotoxic metabolites originally found in very low yields (10-6–10-7%) in herbivorous sea hares. These linear and cyclic peptides are the bioactive components of the sea hare Dolabella auricularia, extracts of which were reported in 1972 to have antineoplastic activity. The most active principle, dolastatin 10 Fig. (6-25), was isolated from D. auricularia in 1987, and years later, the same compound was reported to be produced by a known Palauan cyanophyte of the genus Symploca [80]. Several other cytotoxic dolastatin analogues had been reported from Symploca and Lyngbya cyanobacteria [81, 82], but dolastatin 10 was the most potent. Dolastatin 10 inhibited the polymerization of microtubules and was evaluated in Phase I and then in Phase II clinical trials for the treatment of several solid tumors, including pancreatic (M.D. Anderson Cancer Center; NCT00003677) and kidney (Mayo Clinic and National Cancer Institute; NCT00003914) cancers. However, dolastatin 10 was quite toxic and had minimal responses in cancer patients in many of these trials, and no further development of this compound has been reported for over 10 years.
This peptidic drug skeleton was then “recycled” into modified peptide structures as standalone agents, and subsequently into antibody-directed warheads for the treatment of cancer. The synthetic derivative of dolastatin 10, auristatin PE. completed Phase II clinical trials for the treatment of non-small cell lung cancer (M.D. Anderson Cancer Center; NCT00061854) and metastatic soft tissue sarcoma (Daiichi Sankyo, Inc.; NCT00064220) but as with the dolastatin 10 trials, this agent did not progress any further.
Fig. (6)) Dolastatin 10-Derived Structures; 25–27.However, as a “warhead” the dolastatin 10-related structures have enjoyed significant success. In 2011, an anti-CD30 antibody-conjugated monomethyl auristatin E (brentuximab vedotin or SGN-35; Seattle Genetics) was approved by the FDA for the treatment of various lymphomas. The structure of monomethyl auristatin E (vedotin) is shown above Fig. (6-26) and extensive details of the development of brentuximab vedotin can be found in the 2012 review by Senter and Sievers from Seattle Genetics, the inventors of the antibody drug conjugate (ADC) [83]. This modification of dolastatin and a close relative known as auristatin F Fig. (6-27) are in various clinical trials as warheads on ADCs directed at specific cancer cell epitopes, with a 2017 review by Newman and Cragg covering their current status [84].
No studies have been published on using cyanophyte fermentations for the economical, large-scale production of precursors, or dolastatin derivatives, en route to producing these ADCs instead of using chemical synthesis. However, dolastatin derivatives and other cytotoxic compounds continue to be isolated from Symploca sp. and Phormidium / Lyngbya sp [85, 86], in addition to related compounds from Caldora sp. from different locations [87]. We should insert a note of caution here, as this genus is from a renaming of some Symploca species. It will be interesting to see how similar the dolastatin-producing BGCs are once these cyanophytes are sequenced.
With the advances made in (meta)-genomics, cloning methods for metabolic pathway assembly, and culture-dependent methods [88], novel cryptic metabolites, together with new molecules from analyses of DNA extracted from uncultured soil-born microbes [89-92] and aquatic microbes are being identified. For example, metagenomic sequencing enabled the Piel group to determine that the marine-sponge-derived peptides containing both D- and L-amino acids, the polytheonamides (structures not shown due to size), were post-translationally modified ribosomal peptides from an endosymbiont [93]. Initially only six putative enzymes were found to be involved in making the polytheonamide precursor, with 48 posttranslational modifications to account for. The number of enzymes has now increased by one to seven in a very recent paper by the same group [94]. Furthermore, these bacterial genes do not resemble those from ribosomal pathways, and they also must encode epimerases in order to produce peptides with D-amino acids, creating new opportunities for protein engineering. The polytheonamides now represent members of a new natural products family named proteusins [95]. This study is one of several examples of symbionts being rich sources of unusual pharmacophores. By understanding the chemical ecology of their largely unexplored environments, in due course we will gain more insight into how to culture the currently unculturable microbes.
3. Plant-derived Bioactive Agents and Microbes
Dating back to ancient times, plants have been some of the first organisms used to treat disease due to their bioactivities. Following the isolation and identification of a number of anticancer agents from plant sources in the 1950s, Dr. Jonathan Hartwell at the National Cancer Institute (NCI) initiated an extensive plant collection program in the 1960s. This was performed by using the mechanism of an interagency agreement between the NCI and the US Department of Agriculture’s (USDA) Economic Botany section to obtain plant samples for further testing. As shown later, several novel bioactive chemotypes, such as the taxanes and camptothecins, were isolated and clinically developed. At that time, the middle 1960s, there was no recognition of the potential for endophytic / epiphytic microbes being “a source”, and even today, there are still many arguments and / or debates over these microbes being recognized as “another source, or perhaps the source” of these and other plant-associated compounds.
Though we will be discussing the involvement of microbes in the “production of plant-sourced secondary metabolites” under individual agents in the following sections, we will mention the initial results that led to a massive amount of work in this scientific area. In 1993, Stierle et al. published one of the first papers on the “production of phytochemicals by endophytes”, demonstrating the production of taxol by endophytes, which we define as being bacteria and fungi that colonize intercellular and / or intracellular plant tissue and apoplastic compartments [96]. Due to the purchase of the name “Taxol” from a German company and subsequent use as a trademark by Bristol Myers Squibb, this drug was given the generic name of paclitaxel on approval for ovarian cancer treatment by the FDA in 1992. However, we will use the original name that was given to the compound by the discoverers from now onwards.
Though not noted for a number of years, Li et al. demonstrated that the production of taxol by an endophyte isolated from Torreya grandifolia was increased by treatment with benzoic acid, even though this particular plant was not quoted as being a producer of the compound. The fungus, a Pericona species did produce taxol and then lost the ability to do so with time [97]. As mentioned earlier in this chapter, basic “supplementation methodology” was very common in the 1950s to late 1970s / early 1980s when pharmaceutical companies were searching for microbial secondary metabolites as sources of antibiotics. Since the “concept” was not formally published as it was “common knowledge” amongst these industrial researchers, the method was “reinvented” when academic laboratories started “re-investigating” microbes in the late 1990s to date. The concepts underlying what is now known as “OSMAC” (One Strain, Many Compounds) were first discussed by Zahner [98] in the late 1970s before being renamed by later researchers, though the underlying principles were the same, with the analytical systems being updated. We should note that even Zahner did not formally recognize that he was utilizing systems developed in industry prior to his publication.
