Secondary Metabolites in Soil Ecology
Petr Karlovsky

(A review chapter in Vol. 14 of Soil Biology Series (Ed. in Chief A. Varma), Springer, pp. 1-19)

Download introduction or order reprint from the author.

Content

1. Introduction: Chemical Interactions in Soil
2. Should the Term "Secondary Metabolites" Be Abandoned?
3. Overcoming the Phytochemist’s Approach to Secondary Metabolites
4. Chemical Ecology of Microorgansims Has Been Neglected
5. The Origin of Chemical Diversity in Soil
6. Secondary Metabolites and Fitness: Evolution Meets Ecology
6.1     Chemical Interactions and Coevolution of Soil Species
6.2     Cost of Biosynthesis
6.3     Complexity of Chemical Interactions in Soil
6.4     Regulation of Biosynthesis as a Key to Function
7 Pitfalls in Search for Function
8 Future of Secondary Metabolite Research

Excerpts from selected sections

Section 3: Overcoming the Phytochemist’s Approach to Secondary Metabolites

...While commercial interests generated pressure to purify and run though bioassays more and more compounds each year, little effort has been devoted to questions of primary scientific interest—namely, for what reasons plants and microbes make them and what happens to them in nature. The vast majority of publications on secondary metabolites have been limited to structure elucidation, at best accompanied by arbitrarily selected bioassays. Any randomly selected issue of the Journal of Natural Products will illustrate this practice. This situation is reminiscent of old-time entomology, when scholars were collecting and meticulously describing insects but devoted little effort to the physiology, genetics, ecology or ethology of their subjects...

...The elucidation of biosynthetic pathways by natural product chemists was limited to establishing sequences of intermediates, and it usually failed short of experimentally addressing the enzymatic reactions involved. Enzymes were a domain of biochemistry, which was well isolated from organic chemistry at traditional universities, being affiliated with the faculty of biology rather than that of chemistry. Biochemists were still busy investigating the intricacies of primary metabolism, while natural product chemists were publishing hundreds of weird and beautiful structures each year as on the assembly line.

...Only in the 1990s did research on secondary metabolites began to overcome its limits. On one hand, biologists installed gas chromatography and high performance liquid chromatography systems in their laboratories and learned how to purify secondary metabolites from plant extracts and microbial cultures. On the other hand, chemists learned that apart from growing producing strains in fermenters, they can genetically manipulate biosynthetic pathways and use cellfree extracts or purified enzymes to perform biosynthetic reactions in test tubes. The transition was all but smooth because questions arising in biology traditionally caused little excitement in chemistry."

Section 4: Chemical Ecology of Microorgansims Has Been Neglected

Ecological chemistry of soil is dominated by microbes. Most research activities labeled as chemical ecology worldwide have so far been concerned with interaction between insects and plants. The selection of papers published in the Journal of Chemical Ecology provide a good example. According to its mission statement, the journal is devoted to "promoting an ecological understanding of the origin, function, and significance of natural chemicals that mediate interactions within and between organisms," but the majority of its articles deal with insect–plant interactions. This is just another manifestation of a phenomenon known from systematic biology: the smaller the dimensions of members of a taxonomical group are, the more species the group possesses and the fewer the taxonomists that deal with it. While whole institutes are devoted to ecological studies of insect–plant interactions, only a handful of laboratories seriously investigate chemical communication among microbes in nature. Three systems with a high potential for practical applications are prominent exceptions: quorum sensing in bacteria, biological control of plant diseases, and interaction of plant pathogens with their hosts. A review of advances in ecological chemistry written by the late Jeffrey B. Harborne (1999), one of the most influential doyens of phytochemistry, nicely documents this bias. The review is divided into four sections according to interacting organisms: animal–animal, plant–animal, plant–plant and plant–microbe. A section on microbe–microbe interactions, which would arguably be concerned with chemical interactions more substantial for the survival of their participants than any of the four combinations listed above, just did not occur.

Section 5: The Origin of Chemical Diversity in Soil

...An intuitive concept that the force driving the diversification of secondary metabolites produced by soil-borne or soil-inhabiting microorganisms is competition is widespread. In terms of interference competition, an organism which acquires the ability to produce a new antibiotic will experience a gain in fitness. The efficiency of the antibiotic declines as resistance mechanisms arise and spread, in analogy to the race between the pharmaceutical industry and human-pathogenic bacteria. Intuitively, this situation appears to favor diversity in antimicrobial metabolites. This view has recently been corroborated by the outstanding work by Czaran et al. (2002). The authors simulated an evolutionary arms race which takes place in a spatially structured environment. The basic idea was that the production of a secondary metabolite which blocks competitors either increases or decreases the net fitness of the producer, depending on the presence of the competitor and its resistance towards this particular toxin. The crucial point that led to the generation of diversity was the introduction of costs of resistance. In a spatially segmented, two-dimensional substrate, several strains survived at a stable total density but with periodically fluctuating abundance at local regions.

...How do microorganisms generate and maintain chemical diversity on a biochemical level? Firn and Jones (2000, 2003) suggested that a small set of enzymes with relaxed specificities may generate a large set of different but structurally related metabolites. Only some among these products exert effects which enhance the fitness of their producer under current conditions. The other metabolites serve merely as a supply of diversity for future needs. Apart from postulating how relaxed enzyme specificities generate structural diversity, which can easily be accommodated by the current framework of evolutionary theory, a novel and controversial aspect of their metabolic grid concept is the notion that evolution optimized retention of chemical diversity at minimum metabolic cost, including the production of metabolites which do not exert any beneficial effect on their producers...

Section 6.2: Costs of Biosynthesis

...Let us look at the metabolic costs of secondary metabolite synthesis, which can be easily investigated in simple systems. In plant–insect interactions this issue has been extensively addressed (Gershenzon 1994). Determining the cost of biosynthesis of a particular metabolite by a microorganism appears to be a straightforward issue, providing suitable mutants are available. Wilkinson et al. (2004) recently determined the effect of a stepwise deactivation of the sterigmatocystin biosynthesis pathway in Aspergillus nidulans on the fitness of the fungus. Their result was surprising: the number of conidia produced in axenic cultures increased with the progression of sterigmatocystin synthesis. The lowest number of conidia was found in cultures of a mutant in which the complete pathway had been shut off via a regulatory gene aflR; the highest number of conidia was found in the wild-type strain. Because the strains were isogenic, hidden effects of additional mutations can be excluded. The authors showed that the effect cannot be explained by protection against light...

...An alternative explanation to direct benefits to the fungus as postulated by the authors is that the observed effect could have resulted from regulatory phenomena. This hypothesis is corroborated by the fact that both conidia development and sterigmatocystin synthesis are derepressed by a common activator FluG, which counteracts the affect of the repressor SfgA (Seo et al. 2006). The work of Wilkinson et al. (2004) was the first one addressing the effect of a stepwise deactivation of a biosynthetic of a secondary metabolite on fungal fitness, but the observation of a negative rather than a positive effect of the loss of a dispensable pathway on fitness under axenic conditions is not unique. For example, Gaffoor et al. (2005) disrupted all polyketide synthase (PKS) genes of Fusarium graminearum and observed inhibition of mycelial growth in mutants that lost two out of 15 PKS genes. Similarly, Zhou at al. (2000) observed growth inhibition in A. parasiticus after disruption of PKS FLUP. The mechanisms of these effects are unknown.

Section 6.3: Complexity of Chemical Interactions in Soil

Growth inhibition or toxicity in general are not the only effects exerted by metabolites involved in chemical warfare in soil. Microorganisms may avoid harmful effects of antimicrobial compounds produced by their competitors by suppressing their synthesis. The interpretation of such effects from an ecological point of view is straightforward. For example, fusaric acid is a mycotoxin and presumably a virulence factor of F. oxysporum. Plant infection by F. oxysporum can be suppressed by certain strains of Pseudomonas fluorescens which produce the antifungal metabolite 2,4-diacetylphloroglucinol (see Chap. 5). Notz et al. (2002) showed that fusaric acid suppresses the production of 2,4-diacetylphloroglucinol by P. fluorescens. Importantly, this effect was demonstrated not only in vitro, but strains carrying reporter fusions for 2,4-diacetylphloroglucinol synthesis were investigated in the rhizosphere and the effects of F. oxysporum strains producing different amounts of fusaric acids were compared.

Secondary metabolites involved in antagonistic interaction may affect other functions and activities of competitors to benefit their producers. For instance, mycotoxin deoxynivalenol produced by F. graminearum appears to inhibit the expression of a chitinase gene in Trichoderma atroviride (Lutz et al. 2003). Because chitinase activity is a decisive factor determining the efficiency of the biocontrol agent T. atroviride against F. graminearum, the repression of chitinase production by deoxynivalenol may be regarded as a defense mechanism. This results revealed a new ecological role for mycotoxin deoxynivalenol, which was known to act as a virulence factor of F. graminearum in wheat. Deoxynivalenol obviously plays at least two different and unrelated ecological roles. (Because of the induction of vomiting and food refusal by deoxynivalenol in mammals, the mycotoxin might also be involved in interference competition between Fusarium and grain- or seed-consuming animals.)

...Detoxification is a widespread mechanism of defense of target organisms against harmful secondary metabolites (Karlovsky 1999). Antimicrobial plant metabolites are often detoxified by a phytopathogenic microorganism (Pedras and Suchy 2005; Pedras and Hossain 2006; Morrissey and Osbourn 1999; Glenn et al. 2003). These processes have been studied with plant metabolites extracted from leaves and stems, but plant phytoalexins and phytoanticipins also reach soil with root exudates (see Chap. 11) and with plant debris (see Chap. 10). Detoxification of plant defense chemicals is therefore as important in the rhizosphere as it is in aboveground plant organs.

...The effects of secondary metabolites on the biology of soil inhabitants are too numerous to list here exhaustively. Metabolites of plant origin induce germination of fungal spores and microsclerotia, attract and repel nematodes, mediate allelopathy among plants and induce chemotaxis in zoospores and protozoans. Strigolactones (Humphrey and Beale 2006) belong to the most interesting compounds not discussed in this volume. These plant secondary metabolites, which are secreted by roots in extremely low quantities challenging our most sensitive analytical techniques, stimulate the germination of parasitic weeds and mycorrhiza fungi. Siderophores are another group of secreted metabolites involved in complex interactions. They are synthesized to facilitate the uptake of iron by their producers, but many microorganisms hijack foreign siderophores to lower their costs of iron extraction, or even use them decadently as a cheap nutrient. Similarly as in marine ecosystems (Engel et al. 2002), nontoxic concentrations of antimicrobial compounds involved in interference competition may effect microbial behavior, corroborating the view that chemical communication is the primary factor controlling interorganismal interactions in soil.

Section 7: Pitfalls in Search for Function

...antibacterial or antifungal effects may be overlooked when a metabolite is well known in a different context. For instance, a strong toxic effect of mycotoxin zearalenone on filamentous fungi remained unnoticed for decades (Utermark and Karlovsky 2007). Zearalenone is known as a potent estrogen and the ingestion of contaminated food and feeds poses a health risk to humans and farm animals. This prominent biological activity and the label "mycotoxin" apparently prevented people working with zearalenone from subjecting it to a standard antifungal assay.

...Research on secondary metabolites involved in interaction of microbial pathogens with plants suffered from serious setbacks. Gäumann (1954) and his disciples postulated half a century ago that phytotoxins are causally involved in all plant diseases. A generation of phytopathologists generated phytotoxicity data to support their hypothesis, but a convincing proof did not surface even for a single toxin at that time because of the lack of appropriate experimental tools. Referring to this era, Robert Scheffer and Steve Briggs once wrote: "The literature on toxins affecting plants is vast, but much of it is meaningless." Their harsh judgment was embraced by the next generation of phytopathologists, who went to the other extreme and abandoned research into secondary metabolites acting as virulence factors for nearly three decades. (Host-specific toxins were a noticeable exception.) As a consequence, opportunities to design novel resistance mechanisms for crops based on detoxification of fungal toxins were considerably delayed and our understanding of pathogen–plant relationships was deprived of one of its principal facets. A renewed interest of phytopathologists in non-host-specific toxins, as we experience it now, will likely benefit not only plant protection but also basic research on secondary metabolites in general.

Section 8: Future of Secondary Metabolite Research

Thousands of secondary metabolite structures have been published, but educated guesses about biological function are possible only for a negligibly small fraction of them. Besides, they are seldom more than guesses: when a bioassay demonstrates toxic effects upon a competitor, we still do not know whether the substance is produced under relevant conditions in nature, whether its local concentration is sufficient to exert the effects observed in vitro and how adsorption, degradation and interaction with other metabolites modulates its toxicity in situ...

...How is secondary metabolite research advancing beyond its traditionally descriptive approach? Natural product chemistry is extending its scope and embracing techniques and concepts originating from biochemistry and genetics, while ecologists and environmental microbiologists recognize that chemical interactions mediated by secondary metabolites are crucial for our understanding of soil ecosystems. Empirical screening of natural products for biological activities, as well as high-throughput purification and structure elucidation of natural products from arbitrarily selected sources, should be left to the responsibility of the pharmaceutical industry and service laboratories, releasing capacity in academia and basic research to address fundamental questions. The following emerging approaches and technologies are likely to play a role in this transition:

• Application of genetic engineering in systematically controlling the production and/or degradation of secondary metabolites, followed by monitoring how these perturbations affect the system, allows us to assess the effect of secondary metabolites on the fitness of soil organisms.
• Analytical techniques for the quantification of many metabolites in matrices as complex as soil are needed to follow the dynamics of secondary metabolite production, transformation and degradation in soil. In situ detection and nondestructive analysis are needed in order to take into account the heterogeneous structure of soil ecosystems.
• Routine techniques available for monitoring microbial populations in soil are differential gradient gel electrophoresis (DGGE ) of amplified ribosomal RNA genes or reverse-transcribed ribosomal RNA, terminal restriction fragment length polymorphism (T-RFLP ) of ribosomal RNA genes and in situ hybridization of taxon-specific oligonucleotides labeled by fluorescent dyes (FISH ). In future these techniques they will be extended by large-scale metagenome sequencing (Eisen 2007; Rusch et al. 2007).
• Modeling chemical interactions in microbial ecosystems and their evolutionary consequences will be increasingly important. The interplay of factors such as metabolic costs, competition, spatial heterogeneity, synergy of antibiotic effects or many metabolites, adsorption and detoxification can be investigated by computer modeling, while it is difficult to address more than one factor experimentally.
• Soil is arguably the most complex and difficult system to chose for the study of ecological functions of secondary metabolites. However, soil is also the ecosystem in which chemical interactions play the most substantial role, and from where major insights into the evolution of chemical diversity are expected to come in future.

Download introduction from Spring website or order reprint from the author.