Preface to the 1st Edition

The two words natural products have different impacts on distinct audiences. On the one hand there is the sustained consumer interest in organic foods, supplements, and nutraceuticals from nature, built around the simple and abiding premise that natural is good. To chemists and pharmacognosists (specialists who study medicinal substances from nature) natural products are jointly about the molecules themselves, their isolation, structural, and initial functional characterization (the chemists) and the effects they have in acute and chronic therapeutic settings in humans (the pharmacognosists). In recent decades specific natural products, such as wortmannin, brefeldin, staurosporin, trapoxin and many others, have been powerful tools for cell biologists to study almost any facet of signal transduction, cell growth and division, protein secretion, and apoptosis.

Historically, natural products have been used in folk medicine for at least 4500 years, stretching back to early Egyptian and Sumerian written records. Starting in the first decades of the 19th century, chemists began to isolate alkaloids in pure form, beginning with morphine in 1817, marking the first major turning point from millennia of use of plant extracts containing multiple mixtures of compounds to purified single molecule preparations. Over the intervening two centuries the characterization of the inventory of natural products, from plants, fungi and bacteria, has spurred many of the conceptual and practical advances in organic chemistry. These have included analytical tools such as mass spectrometry and sophisticated two dimensional high field nuclear magnetic resonance (NMR) and have encompassed the golden age of total synthesis of natural products. For medicinal chemists the scaffolds of different classes of natural products have been starting points and/or inspiration for both semisynthetic and fully synthetic approaches to new drugs (e.g. podophyllotoxin to etoposide, staurosporine to the plethora of synthetic heterocyclic protein kinase inhibitors).

The second major turning point, and one that has sparked a renaissance in the natural products arena, has been the availability of thousands of microbial genomes and the beginnings of a few plant genomes over the past two decades. Bioinformatic analysis of natural product biosynthetic capacity, known as genome mining, has become a core discipline at the intersections of chemistry, life sciences, and medicine in all its manifestations. In addition to generalizing the premise that microbial biosynthetic genes are clustered (and so easier to find, interrogate, and move en bloc), bioinformatic analyses show that many actinomycetes, myxobacteria, and fungi have some 30–50 predicted biosynthetic gene clusters for polyketides, nonribosomal peptides, and terpenes. Such strains typically produce one to five natural products under standard laboratory culturing techniques: the universe of natural products to be explored has thus instantly gone up by an order of magnitude. Given an expected 20 000 microbial genome sequences to be available in the next decade or so, that would give about 10⁶ biosynthetic gene clusters for the indicated three classes of natural products, with >99% of those as yet unexamined. In addition, there are large classes of plant and fungal alkaloids and plant-derived phenylpropanoids not counted in that set of predictions.

In this period of natural product renaissance, there are many strands of investigation being brought to bear on determining what new molecules are left to find in nature. In many therapeutic arenas, from infectious disease (beta-lactam antibiotics, erythromycins, tetracyclines, aminoglycosides), to cancer (vincristine, paclitaxel), to immunosuppression (cyclosporine, rapamycin, FK506), to cholesterol lowering agents (lovastatin), natural products have been key contributors to new and effective medicines. To what extent as yet unknown scaffolds of natural products will continue to illuminate new therapeutic modalities is one of the challenges and new opportunities in the postgenomic era.

This volume takes up the biosynthesis of major classes of natural products – polyketides, peptides, nucleosides, isoprenoids/terpenoids, alkaloids, phenylpropanoids, and glycosides. The approach is not to be encyclopedic, nor to illuminate every subclass or intriguing chemical functional group. Instead the approach is to codify the chemical logic that underlies each natural product structural class as they are assembled from building blocks of primary metabolism. A few simple reaction types are used in each natural product class, some that are common in primary metabolism and some that are much more frequent in secondary pathways than primary pathways.

Of the seven natural product cases noted above, two of them – polyketides (PK) and nonribosomal peptides (NRP) – but none of the others, are built on enzymatic assembly lines where the growing chains are covalently tethered as thioesters to carrier protein domains. Hybrid nonribosomal peptide-polyketide scaffolds comprise some of the most therapeutically interesting molecules from nature – rapamycin, FK506, bleomycin, epothilones – and they are built on convergent PK and NRP assembly lines. The dominant reaction in enzymatic polyketide chemistry is an iterated decarboxylative thioclaisen condensation for C–C bond formation, essentially carbanion chemistry.

The isoprenoid/terpenoid class of natural products, currently the largest natural product set, at over 50 000 known molecules, instead are assembled by enzymes that use diffusible substrates, intermediates, and products. The chemical logic in this class is overwhelmingly carbocation chemistry. This modality starts with initial allylic carbocation formation from the Δ²-prenyl-PP building blocks, through alkylative condensations to scaffolds where cations induce some dramatic scaffold rearrangements. The cyclizations of triterpene squalene and 2,3-oxidosqualene to tetracyclic and pentacyclic product scaffolds continue the cation-driven reaction manifolds and emphasize the nucleophilic role of the p electrons of double bonds in C–C forming reactions.

In contrast to the polyketides, isoprenoids, and phenylpropanoids where the product scaffolds are essentially devoid of nitrogen atoms, the peptides, nucleosides, and alkaloids contain nitrogens that are central to the chemistry of these classes. The largest and most structurally diverse class comprises the alkaloids, in part because they were defined experimentally by the minimal criterion of at least one basic nitrogen in a heterocycle, driven by isolation schemes that went back and forth between free base and salt forms. The alkaloids range from simple monocyclic amines to amazingly complex frameworks in opioids, the poison strychnine, the antimalarial quinine, and the dimeric antitumor vinca alkaloids vinblastine and vincristine.

One of the pervasive features of the maturation of natural product scaffolds across all these classes is the tailoring of nascent product frameworks by oxygenase enzymes. Oxygenases are used sparingly in primary metabolic pathways, steroids being the major exception (and given the plethora of plant steroidal scaffolds one might put steroids in a bridging category between primary and secondary metabolism). By contrast, oxygenases proliferate in plant and microbial maturations of secondary natural products. A particularly clear example is the introduction of eight oxygen atoms around the periphery of the C20 isoprenoid hydrocarbon taxadiene to get to the tubulin inhibitor and anti-ovarian-cancer drug paclitaxel.

We take up oxygenases in detail as key elements of natural product enzymatic machinery but also because they bring carbon radicals and homolytic C–C bond formation manifolds into play. We note many cases where molecular oxygen reduction generates an enzyme-bound high valent oxoiron reagent that initiates substrate C–H bond homolysis. In several contexts intramolecular C–C or C–X bond formation competes with net hydroxylation. In those cases the oxygen input is cryptic, but the reaction manifold has been radical-based, in contrast to the heterolytic C–C bond-forming flux in polyketides and isoprenoids. The phenylpropanoid lignan scaffolds are also built by phenoxy radical dimerization enzymology.

Two additional chapters are devoted to enzyme classes typically not broken out as features in natural product biosynthesis in other treatments of the subject. One of them deals with S-adenosylmethionine (SAM), not in its familiar role as a donor of [CH₃⁺] equivalents to cosubstrate nucleophiles but as a radical initiator, continuing the theme of C–C bond formation via radical intermediates. At two extremes, aerobically with O₂ and anaerobically with SAM and four iron/four sulfur cluster enzymes, radical chemistry is in play for C–C bond formation in a diverse set of natural product frameworks.

The second focused chapter deals with natural product glycosides. Often treated as an afterthought in natural product maturation, the sugars serve many purposes in the localization, solubility, and functional activity of essentially every natural product class. The biosynthesis of dedicated deoxy- and aminodeoxyhexoses by enzymes encoded within biosynthetic gene clusters testifies to the importance of these sugar moieties in those pathways. The chemical logic of activation of glucose or ribose as electrophiles at C_1ʹ reveals why all glycosylated natural products are linked though C_1ʹ.

The final section of the book takes up strategies for isolation of natural products, from the arc of medicinally important products historically, to the challenges in the postgenomic era of how to prioritize hundreds of thousands of unexamined gene clusters for novel frameworks and new activities. Given the intense coalescence of new methodologies, from genetics, genomics, synthetic biology, hetero- logous expression, and gene cluster activation, we do not delve deeply into any one set of current technologies, on the assumption they may be rapidly outdated. Instead we set an overview perspective and raise the strategic questions of what criteria could be assembled to increase the probability of finding members of the natural product inventory with novel structural and functional properties.