Evolution and Ecology of Actinobacteria and Their Bioenergy Applications

Abstract

The ancient phylum Actinobacteria is composed of phylogenetically and physiologically diverse bacteria that help Earth’s ecosystems function. As free-living organisms and symbionts of herbivorous animals, Actinobacteria contribute to the global carbon cycle through the breakdown of plant biomass. In addition, they mediate community dynamics as producers of small molecules with diverse biological activities. Together, the evolution of high cellulolytic ability and diverse chemistry, shaped by their ecological roles in nature, make Actinobacteria a promising group for the bioenergy industry. Specifically, their enzymes can contribute to industrial-scale breakdown of cellulosic plant biomass into simple sugars that can then be converted into biofuels. Furthermore, harnessing their ability to biosynthesize a range of small molecules has potential for the production of specialty biofuels.

INTRODUCTION

Actinobacteria are ubiquitous and one of the most diverse groups of bacteria in nature. Its members range from anaerobic, unicellular organisms to aerobic, filamentous, and spore-forming lineages. Alone, the genus Streptomyces accounts for almost 5% of the ~16,000 described bacterial species (http://www.bacterio.net/). Actinobacteria have made a substantial positive contribution to human health; they are the producers of many compounds that are used as important drugs, including most antibiotics (58). In addition, the phylum includes important pathogens, such as Mycobacterium tuberculosis, the causal agent of one of the most devastating diseases in human history, tuberculosis. Largely because of these biomedical impacts, Actinobacteria have avoided the obscurity relegated to most bacterial phyla, receiving substantial scientific study of their genetics, molecular biology, and biochemistry.

In addition to their important influence on human health, Actinobacteria have key ecological roles. Before focusing on antibiotic discovery, Nobel laureate Selman Waksman’s work on soil bacteria and their impact on agricultural productivity was among the first to implicate Actinobacteria as important contributors to the process of plant biomass decomposition (58, 146). More recently, Actinobacteria have been revealed as widespread symbionts of eukaryotes, helping herbivores gain access to plant biomass as nutritional mutualists and producing natural products as defensive mutualists (e.g., 18, 33, 38, 65). Studies of Actinobacteria as defensive mutualists have led to the discovery of new antibiotics with potential pharmaceutical applications (97, 106, 117), renewing recognition of the value in understanding the ecology of Actinobacteria for drug discovery (e.g., 12, 122) and hearkening back to Waksman’s journey to the discovery of streptomycin.

We suggest, much like in drug discovery, the ecology and evolution of Actinobacteria position this phylum to contribute to the development of a sustainable bioenergy industry. Specifically, as Actinobacteria are implicated as important decomposers of plant material in nature, their cellulolytic enzymes can be used to more efficiently break down plant biomass into simple sugars, which can then be used for the production of biofuels. In addition, the diverse biosynthetic capacity of Actinobacteria, which evolved to mediate their environmental interactions, could be leveraged to produce a range of bioproducts, including specialty biofuel compounds. Here, we review our current understanding of the ecology and evolution of Actinobacteria. Then, within this framework, we discuss the potential application of Actinobacteria in bioenergy.

EVOLUTION AND GENOMICS OF ACTINOBACTERIA

The evolutionary origin of the Actinobacteria is ancient, dating back to ~2,700 million years ago, predating oxygenation of the atmosphere (~2,300 million years ago) (Figure 1) (8, 150). 16S rRNA gene sequencing, multilocus phylogenetics, and phylogenomics support a shared common ancestor for Actinobacteria, Cyanobacteria, and Deinococcus (8, 52). Together, these three lineages share a most recent common ancestor with the phylum Firmicutes, with divergence estimated at more than 3,000 million years ago (8). Consistent with evolving in an anoxic atmosphere, the earliest Actinobacteria were obligate anaerobes, were not yet filamentous or able to form spores, and instead had simple rod/coccus morphology (26). The lineage Actinobacteria subsequently diverged into at least six classes and 53 families (52). The ability to grow as filaments and produce spores evolved in the more recently derived lineages but has been secondarily lost in a few groups (26).

Figure 1.

Actinobacterial phylogenomics and relative abundance across ecosystems. Forty-six publicly available 16S rRNA gene datasets, spanning four ecosystems, were analyzed to determine the average relative percentage of reads assigned to each actinobacterial genus (93) (Supplemental Table 1; follow the Supplemental Material link from the Annual Reviews home page at http://www.annualreviews.org). Phylogenetic analysis was performed using concatenated alignments from 94 conserved TIGRFAM protein families in representative complete actinobacterial genomes (Supplemental Table 2), with the outgroup Deinococcus. Sequences were aligned with MAFFT 7.221, and the phylogeny was generated using RAxML 8.1.24 (67, 124). Molecular clock calculations were performed with RelTime using the origin of life (3,500–3,800 million years ago), the origin of Cyanobacteria (2,500–3,500 million years ago), and the Escherichia-Salmonella divergence time (50–150 million years ago) as calibration points (133). Statistical support for the phylogeny, based on a sequence resampling bootstrap analysis, is shown on nodes with less than 100% support.

In general, the genomes of Actinobacteria are large, relative to other phyla of bacteria, with an average genome size of over 5 megabases (Mb). However, genome sizes vary widely, from less than 1 Mb in some Tropheryma species to over 12 Mb in a few Streptomyces species. In general, the genomes have a high guanine-plus-cytosine (G+C) content, with the G+C content of some genera, such as Streptomyces and Pseudonocardia, more than 70%. However, G+C content is less than 50% in Gardnerella vaginalis and Tropheryma whipplei (2, 6). Of those analyzed, all Actinobacteria have a single chromosome, but the presence of large plasmids is not uncommon. The ancestral state is a circular chromosome, but the genera Gordonibacter, Kineococcus, Rhodococcus, and Streptomyces all have linear chromosomes, which appear to have independently evolved in each genus.

Genome content in bacteria is shaped by the interplay between vertical inheritance (i.e., clonal reproduction) with gene loss and acquisition, the latter either involving the genesis of new genes within a lineage or horizontal gene transfer (HGT) from other lineages. Our current understanding of the evolutionary forces shaping gene loss and gene origin in Actinobacteria is limited. Recent genomic analyses of 120 Streptomyces genomes indicate that, at least within this genus, HGT is relatively rare and that long-term retention of horizontally transferred genes occurs several orders of magnitude less frequently than the origin of point mutations (B.R. McDonald & C.R. Currie, unpublished information). Nevertheless, HGT enables adaptive divergence of populations through the introduction of novel genetic diversity. For example, virulence operons in M. tuberculosis and plant pathogenic Streptomyces species appear to have been acquired through HGT (69, 109). Furthermore, HGT has been shown as a major driver of secondary metabolite diversity in the marine actinobacterial genus Salinispora (161).

Coevolution is another evolutionary force involved in shaping the biology of Actinobacteria. Coevolution inherently recognizes the importance of species interactions in the evolution and diversity of lineages (134). Thus, given that Actinobacteria typically occur either within complex communities of interacting microbes or as symbionts of eukaryotic hosts (see “Ecology of Actinobacteria,” below), reciprocal selection is likely to influence the evolution of these lineages. One of the few studied examples of coevolution in Actinobacteria is the Pseudonocardia-ant defensive mutualism (see the sidebar, Coevolution of Actinobacteria in the Fungus-Growing Ant System and Figure 2a). In this example, as well as other defensive mutualisms, Actinobacteria provide antimicrobial protection to a host organism, and species interactions are mediated by secondary metabolites that are likely influenced and shaped by coevolution. Further studies are needed, both on how chemical diversification is shaped by coevolutionary forces and on the nutritional interactions between Actinobacteria and their hosts.

Figure 2.

Actinobacteria in nature. (a) An Acromyrmex leaf-cutter ant in her fungus garden covered in white Pseudonocardia bacteria. (b) A Streptomyces strain growing on decaying plant biomass. (c) Frankia nodules on the roots of a Russian olive tree. (d) X-ray showing a healthy lung on the left and Mycobacterium tuberculosis growth on the right. Illustrations courtesy of Gina Lewin.

COEVOLUTION OF ACTINOBACTERIA IN THE FUNGUS-GROWING ANT SYSTEM.

Some lineages of Pseudonocardia are associated with fungus-growing ants (tribe Attini), with the bacterium residing on the ants’ exoskeleton (see Figure 2a) (38). These Pseudonocardia produce antimicrobial compounds that help defend the ants’ food source, a fungal cultivar, from specific pathogens (38, 97, 152). The bacterium is vertically inherited (38), and there is a critical window in which the ants must be exposed to the bacterium for acquisition to occur (87). Specialized glands and structures on the ant exoskeleton supply nutrients to the bacterium, indicating evolutionary selection on the ant for hosting Pseudonocardia (37, 126). Likewise, molecular phylogenetic studies of symbiotic Pseudonocardia reveal congruence with the host ant phylogeny (22, 23). Additional complexity is seen in the population dynamics of Pseudonocardia, which are correlated to that of a system-specific pathogen, Escovopsis (24, 36, 38). Indeed, studies examining pair-wise interactions between Pseudonocardia and Escovopsis strains show varying degrees of pathogen inhibition, indicating diversity in secondary metabolites produced by Pseudonocardia as well as variation in pathogen resistance (22, 23, 38). These findings support coevolution shaping the dynamics and biology of this system, which likely has generated and influenced the diversity of secondary metabolism of the bacterial symbiont. Similar dynamics are likely common across defensive mutualisms, such as in that of solitary digger wasps (tribe Philanthini), which engage in a defensive mutualism with Streptomyces strains to offer antimicrobial protection during larval development (66, 72). Potentially, similar dynamics also occur in nutritional mutualisms.

ECOLOGY OF ACTINOBACTERIA

Historically, Actinobacteria were largely viewed as soil bacteria but are now recognized as being cosmopolitan; they are found in virtually all ecosystems, with a distribution covering most of the planet. As free-living microbes, they are abundant in soil environments, especially those with a higher pH (77), and in marine and freshwater ecosystems (Figures 1 and 2). Actinobacteria are also associated with eukaryotic hosts in diverse niches, such as the exoskeleton of some tropical ants, the lungs and skin of mammals, and the roots and inner tissues of plants (Figures 1 and 2). Some genera, including Streptomyces, Kineococcus, and Mycobacterium, span diverse ecosystems from soil to marine to freshwater environments. Others, such as Atopobium, Bifidobacterium, Kocuria, and Rothia, are mostly found in host associations (Figure 1) (Supplemental Table 1; follow the Supplemental Materials link from the Annual Reviews home page at http://www.annualreviews.org). The diversity of niche utilization in Actinobacteria mirrors their physiological capability to grow in a range of conditions, as supported by culture- and genome-based analyses. Nevertheless, our understanding of the ecological roles of Actinobacteria in many of these systems remains poorly resolved. Here, we explore their ecology, focusing on cases where the functional role of Actinobacteria has been elucidated with the understanding that Actinobacteria are critical for a wide range of ecosystem services and host interactions.

Role of Free-Living Actinobacteria in Carbon Cycling

The largest terrestrial source of organic carbon is the network of energy-rich but recalcitrant polymers in the plant cell wall (156). The primary polymer in the plant cell wall is cellulose, an insoluble crystal of β-1,4-linked glucose molecules, which is surrounded by a complex mesh of hemicellulose and lignin (99). Hemicellulose is a heterogeneous polymer composed of five- and six-carbon sugars, such as xylose, arabinose, mannose, and uronic acids. Lignin is an amorphous polymer formed from variable, highly cross-linked phenolic alcohols. In contrast to most other phyla of bacteria, Actinobacteria are enriched in the full suite of enzymes necessary to deconstruct plant biomass (14). This enrichment in enzymes, combined with the abundance of Actinobacteria in soil and the ability of some isolates to grow on plant biomass or its constituent polymers, supports their contribution to plant biomass degradation in soil in conjunction with fungi and other organisms (41, 77, 90). Several recent in vivo studies provide further validation (54); for example, in Arctic peat soil, Tveit and colleagues (138) found that most cellulase and hemicellulase RNA transcripts were assigned to Actinobacteria.

Bacteria in the genus Streptomyces are among the most abundant organisms in soil (2a, 62), and the ability of some isolates to grow on cellulose was first identified ~100 years ago (145). Subsequently, a variety of Streptomyces strains isolated from soil have been shown to grow on cellulose (16, 120, 144), hemicellulose (45, 61), and potentially even lignin (34, 163) (Figure 2b). However, the majority of strains tested cannot grow on crystalline cellulose or hemicellulose, suggesting that most Streptomyces species in the soil only use the breakdown products of plant cell wall polymers (18). Other genera of Actinobacteria have also been implicated in plant biomass degradation in the soil, including Cellulomonas, Nocardia, and Micromonospora, again on the basis of their high prevalence in soil and the ability of some isolates to grow on plant biomass-derived substrates (57, 90, 163).

Free-living Actinobacteria contribute to carbon cycling in other environments enriched in decaying plant material, including leaf litter, compost, and manure (91, 116). In many composting systems, genera of Actinobacteria, including Thermobifida, Saccharomonospora, and Thermobispora, are abundant community members during the thermophilic stage (56, 127, 159). For example, metagenomic analysis of rice straw within composting cow manure showed that Actinobacteria were abundant and encoded more cellulases than any other phylum (149). Similar to compost, leaf litter is also host to a high abundance of Actinobacteria, and metagenomic sequencing identified cellulases from Streptomyces and Thermobispora genera, as well as from many other Actinobacteria (13, 70, 89, 162). Micromonospora strains are common in lake sediment, another environment rich in decaying plant material, and isolates in this genus have been shown to degrade cellulose and hemicellulose (35, 48).

Actinobacteria are also implicated in the cycling of a variety of other carbon sources. This includes chitin, the second most abundant organic carbon source in nature, which is a structural polymer in the cell wall of fungi and in the exoskeletons of invertebrates. As with cellulose, few organisms can degrade chitin, but some Actinobacteria, including most Streptomyces and Arthrobacter species, have this ability (68, 92). The Ac1 Actinobacteria, dominant organisms in freshwater, are also thought to degrade chitin or uptake degradation by-products of chitin from fungi, crustaceans, and diatoms (9, 46). Additionally, some Actinobacteria are able to degrade hydrocarbons and organic contaminants (5, 85). For example, Rhodococcus can degrade diverse natural and synthetic molecules, including chlorinated or aromatic hydrocarbons, nitroaromatics, sulfonated azo dyes, and pesticides (10).

Role of Actinobacteria in Host Associations

Although Actinobacteria are still generally viewed as free-living bacteria, the recent increased focus on symbiotic associations has revealed their common association with eukaryotic hosts (18, 49, 84). They have broadly been identified in association with animals, both in their guts and on external surfaces. For example, the genera Bifidobacterium and Corynebacterium are associated with animals ranging from insects to mammals, including humans (137, 143). Additionally, Actinobacteria are associated with plants, including in the rhizosphere and as endophytes (32, 84, 115). Efforts are underway to determine the ecological roles of these bacteria, from aiding hosts with nutrient acquisition to serving as defensive mutualists.

A range of animals and plants rely on Actinobacteria to supplement their diets or help digest complex food sources. Surveys of insect guts, using 16S rRNA gene sequencing, indicate that Actinobacteria are more abundant in the guts of insects that consume detritus or decaying wood, especially termites (31, 64). Strains that can degrade plant biomass have been isolated from termites and from a range of other insects associated with wood niches, including Sirex woodwasps and mountain pine beetles (Figure 2b) (1, 18, 98, 151). Other insects, especially some true bugs, rely on Actinobacteria to provide nutritional supplementation (65). For example, Rhodococcus rhodnii provides B vitamins to Rhodnius prolixus, the insect vector of Chagas disease (20), and likewise, Coriobacteriaceae strains provide B vitamins to Pyrrhocoridae (Hemiptera), such as firebugs (110). In the mammalian gut, Bifidobacterium species aid in digestion of complex carbohydrates, including plant-derived oligosaccharides and gastric mucin (94, 105). The most clearly defined nutritional association between plants and Actinobacteria is the association between nitrogen-fixing Frankia strains and actinorhizal plants (Figure 2c) (30), which enables plants to colonize early during primary succession (78, 118, 141).

Actinobacteria also function as defensive mutualists for a number of eukaryotic hosts. A well-studied example of this relationship is in fungus-growing ants (see the sidebar, Coevolution of Actinobacteria in the Fungus-Growing Ant System and Figure 2). Similarly, solitary digger wasps, commonly known as beewolves, inoculate Streptomyces symbionts on their larval brood cells to protect cocoons from microbial infection (66), and southern pine beetles harbor a symbiotic Streptomyces species to protect their fungal food source (117). Plants are protected by the genera Streptomyces and Micromonospora, as well as by other Actinobacteria in the rhizosphere, which can help inhibit plant pathogens through the production of secondary metabolites (57, 81, 115). Also, endophytic Actinobacteria suppress pathogens and activate the plant immune system (32, 50). In addition to the nutritional roles discussed above, Bifidobacterium species in the human gut function as defensive mutualists through antimicrobial activity against pathogens and potentially through aiding in immune system development and function (86, 121, 131). Finally, in marine environments, Actinobacteria with diverse secondary metabolism potential are found in association with sponges, but their ecological functions in these environments have not been fully elucidated (119, 153). Although the diversity of Actinobacteria-host defensive relationships has only been explored in a small subset of organisms, it is likely that host defense by Actinobacteria is widespread.

In contrast to these beneficial symbioses, Actinobacteria can also be virulent pathogens of plants and animals. The diverse genus Mycobacterium causes tuberculosis (M. tuberculosis), leprosy (M. leprae), and additional diseases in humans and other mammals (Figure 2d) (111). Actinobacteria also cause a range of other diseases in humans, including diphtheria (Corynebacterium diphtheria), Whipple’s disease (Tropheryma whipplei), and bacterial vaginosis (Gardnerella). In plants, Streptomyces scabies and other plant pathogenic Streptomyces strains share a pathogenicity island, allowing them to infect higher plants and cause diseases including, most famously, potato scab (69, 82). Other common plant pathogens fall within the Microbacteriaceae family, including Leifsonia xyli, which is the cause of ratoon stunting disease in sugarcane, and Clavibacter michiganensis, which infects xylem and causes plant wilting in alfalfa, corn, tomato, and potato (21, 47). Finally, leafy gall syndrome, which infects a wide range of dicotyledonous herbaceous plants, is caused by Rhodococcus fascians (128, 129).

ACTINOBACTERIA IN BIOENERGY

As of 2013, fossil fuels (petroleum, coal, and natural gas) supplied 80% of the world’s energy (60). The negative climate impacts of burning fossil fuels has driven worldwide interest in sustainable, economical production of liquid fuels and other chemicals from renewable sources (40). Use of renewable electric sources, such as solar, wind, and hydroelectric, is expanding, but high-energy density liquid fuels are needed to move ourselves and our goods across continents and oceans (29, 40). Plants and some microbes capture the energy of the sun via photosynthetic carbon dioxide fixation. Efforts to engineer photosynthetic plants, algae, or cyanobacteria to produce fuel compounds are underway and have long-term potential (79, 108, 142, 148). However, near-term advances require tapping into stored energy in the plant cell wall. To develop a sustainable biofuel industry, microbes are needed to convert plant biomass into soluble sugars and to transform these sugars into a wide range of desirable compounds (51). Here, we argue, in large part because of their ecology and evolutionary history, Actinobacteria have potential to contribute to both of these challenges.

Actinobacteria as a Source of Enzymes in Biofuels

Industrial production of fuels from plant biomass represented less than 5% of energy used for transportation in the United States in 2014 (139). This energy is largely derived from the starch in corn kernels, which is readily converted into ethanol to supplement gasoline (154). However, the use of edible carbohydrates for fuel directly competes with their use as food (135). In contrast, using the abundant, nonedible polysaccharide and lignin fractions of corn plants, other grasses, and trees to make fuels and chemicals is a promising alternative and could remove more carbon from the atmosphere than is released during production and use of the fuel (53). Currently, industrial lignocellulose hydrolysis is carried out using enzyme mixtures from engineered fungi, such as Trichoderma reesei and Aspergillus niger (101). However, the enzymatic release of intact clean sugar from plant cell walls remains an important bottleneck to efficient biofuel production (55). As detailed below, the diversity and biochemistry of actinobacterial cellulases, evolved within plant-degradation niches, have promise for integration into a successful cellulosic biofuels industry (also see the sidebar, Integrating Actinobacterial Enzymes in Industrial Cellulosic Biofuel Production).

INTEGRATING ACTINOBACTERIAL ENZYMES IN INDUSTRIAL CELLULOSIC BIOFUEL PRODUCTION.

A number of aspects of the evolution, ecology, and physiology of Actinobacteria make them well suited for integration with cellulosic biofuel production. First, Actinobacteria evolved to break down plant biomass under diverse environmental conditions, including under high temperatures, varying pH levels, and anoxic or aerobic states. Similarly, industrial biofuel production requires enzymes that function under diverse conditions, depending on factors such as the biomass composition and pretreatment protocol. The identification of Actinobacteria that have evolved within particular niches can lead to cellulases suitable for specific industrial settings. For example, thermostable enzymes are found in Thermobifida fusca and Thermomonospora curvata from heated composting environments and in Acidothermus cellulolyticus from acidic hot springs (107, 130, 136, 155). Furthermore, given that Actinobacteria freely secrete their enzymes and have evolved in some niches in close association with fungi, actinobacterial cellulases are promising candidates to complement fungal enzymes for plant biomass deconstruction. Finally, Actinobacteria are attractive for industrial use because of their history as successful workhorse organisms in industrial processes (42), the ability to engineer strains that constitutively secrete cellulases at high levels (18, 88), and the resistance of their cell walls to processes, such as milling, that may be important in sustainable cellulosic biofuel production (100).

The ecological roles of Actinobacteria in plant biomass degradation, in soil, in compost, and in association with herbivorous animals have selected for the evolution of high cellulolytic abilities (18). Indeed, a number of model cellulolytic bacteria occur in this phylum, such as Cellulomonas spp. and Thermobifida fusca, whose study has helped advance our understanding of the enzymology of cellulose degradation (28, 59, 155). More broadly, the cellulolytic abilities of this phylum are supported by genomic and proteomic analyses, which indicate that actinobacterial genomes have an amazing diversity of enzyme fold families and organizational architectures used to hydrolyze plant biomass compared to most other phyla of bacteria (14). In particular, of the genera analyzed for this review, the genomes of Streptomyces and Micromonospora, some strains of Mycobacterium, and one strain of Propionibacterium are particularly enriched in carbohydrate-active enzyme (CAZy) genes (Figure 3a) (80). Within these enriched genera, the diversity of CAZy classes varies: Eighty-six glycoside hydrolase (GH) classes are found across 42 Streptomyces genomes, whereas only 35 GH classes are found across 149 Mycobacterium genomes (Supplemental Table 3).

Figure 3.

Enrichment of (a) CAZy (carbohydrate-active enzyme) domains and (b) biosynthetic gene clusters in select actinobacterial genera (Supplemental Table 3). Each median value is indicated by the line within the box, each mean value is indicated by the black dot, and outliers are indicated with gray plus signs. CAZy domains were identified using the CAZy database (80). Putative biosynthetic gene clusters were identified through genome searches for conserved elements of polyketide, terpene, and nonribosomal peptide biosynthesis enzymes. Hits <30 kb apart were called the same cluster. The number of clusters may be overestimated in draft genomes.

Here, we focus on the properties of Streptomyces cellulases. The cellulolytic system of Streptomyces was defined in S. reticuli by Schrempf and colleagues (112–114, 144, 147), whose investigations included secreted cellulolytic enzymes, carbohydrate binding proteins, sugar transporters, and gene regulation. Recent studies have expanded on this work by studying host-associated Streptomyces, such as Streptomyces sp. SirexAA-E, isolated from a Sirex woodwasp, which secretes enzymes with comparable specific activity to an enzyme cocktail from Trichoderma reesei (18, 132).

As in many Actinobacteria, cellulolytic Streptomyces species rely on a simple, freely secreted catalytic assembly to efficiently deconstruct cellulose, including endoglucanases (GH5, GH9, and GH12) and reducing- and nonreducing-end exoglucanases (GH48 and GH6, respectively). The cellobiose produced by these enzymes is cleaved into glucose by β-glucosidases (GH1 or GH3) in the cytoplasm (Figure 4a) (17, 18, 132). In addition, the dominant carbohydrate-binding module in these organisms is carbohydrate-binding module family 2. Parallel minimal catalytic assemblies deconstruct other plant cell wall polysaccharides, e.g., xylan (GH10 and GH11), mannan (GH5), and chitin (GH18, GH19, and GH46) (132). Carbohydrate esterases (CEs) that remove sugar modifications and break linkages between hemicellulose and lignin include families CE4 and CE7 in Streptomyces sp. SirexAA-E (132).

Figure 4.

Molecular details of the proposed role of Actinobacteria in biofuels. (a) Use of Actinobacteria for the deconstruction of plant biomass. ❶ Lytic polysaccharide monooxygenases and ❷ endoglucanases create internal cuts in the cellulose structure, creating free ends. ❸ Reducing- and nonreducing-end exoglucanases break cellulose chains into glucose and cellobiose. Additionally, ❹ peroxidases may break apart lignin, and ❺ hemicellulases hydrolyze xylan, mannan, and other hemicelluloses into constituent sugars. ❻ Cellobiose is imported into the cell, where it ❼ binds to the CebR repressor, releasing the repressor from the DNA and ❽ allowing RNA polymerase to transcribe cellulase genes. ❾ These cellulases are secreted from the cell using the Tat or Sec pathways. (b) Use of Actinobacteria for the production of biofuels. ❶ Cellobiose also is broken by β-glucosidases into glucose, which enters central metabolism. ❷ These sugars are processed in the cell and ❸ can be used for production of secreted fuel compounds, such as bisabolene, 1-propanol, or various fatty acid alkyl ester (FAAE) species where R1 is a small alkyl group and R2 is an aliphatic hydrocarbon. Currently, there are limited examples of cells that can both degrade plant biomass and produce fuel compounds, but we propose this scheme as one possible future direction in actinobacterial biofuels research.

Cellulose-oxidizing lytic polysaccharide monooxygenases from the auxiliary activity (AA) family AA10 enzymes primarily are found in Actinobacteria (19). These enzymes likely contribute to the high cellulolytic ability of Actinobacteria and provide an alternative to hydrolytic endoglucanases (Figure 4a) (63, 140). Interestingly, although Streptomyces are enriched in AA10 enzymes, Streptomyces do not appear to secrete an electron donor (e.g., cellobiose dehydrogenase or exogenous donors, such as ascorbate, gallate, or quinol), which is required for lytic polysaccharide monooxygenase function (17, 18, 132). Possibly instead, Streptomyces species depend on community association with fungal partners to provide this electron donor (104).

Nonreducing-end exoglucanase (GH6) genes are also expanded in Actinobacteria relative to many other groups of bacteria and fungi (14). Genetic and biochemical analyses have indicated that GH6 enzymes are necessary but not sufficient for cellulose degradation (7, 18, 157). Evolutionary analyses indicate that cellulolytic Streptomyces strains evolved to degrade cellulose by maintaining GH6 enzymes, which are highly expressed and secreted during growth on plant biomass (17, 18, 132).

Actinobacteria, Specialty Biofuels, and Bioproducts

The enzymatic capabilities described above allow Actinobacteria to deconstruct plant biomass into simple sugars that can be converted into biofuels and other bioproducts. Here, we focus on how Actinobacteria, having evolved to produce diverse chemicals, may contribute to the production of these bioproducts.

Although bioethanol from corn and cane sugars and biodiesel from plant oils have already seen widespread adoption, increasing their use as transportation fuels faces substantial hurdles, including storage, engine infrastructure, hygroscopy, and land-use concerns (3, 4, 25, 102). New bioproducts, including microbially produced specialty biofuels, more closely match the properties of traditional fuels and thus are highly desirable (4). Specialty biofuels include higher-alcohol biofuels, fatty acid alkyl esters (FAAEs or biodiesel), triacylglycerols (TAGs) that can serve as fuel precursors, and a diverse family of energy-dense isoprenoid compounds that can be used as diesel or jet-fuel replacements (11, 27, 39, 73, 102). Also, bio-based production of fuel additives and commodity chemicals, especially those currently obtained from petroleum, would help support a bioenergy industry. Actinobacteria are a particularly promising group to contribute to the development of these classes of chemicals or their precursors, especially those developed through alternative biosynthesis routes, given their well-known capacity to produce and secrete a diversity of compounds, the ability to engineer their biosynthetic pathways, and their long history of use in industrial processes (Figure 4b).

Several promising studies support the potential utility of Actinobacteria for biocommodity production (Supplemental Table 4). For example, Corynebacterium glutamicum has been targeted to produce specialty biofuels because of its innate isobutanol tolerance and ability to produce large amounts of glutamate and various 2-keto acids, which are precursors to branched-chain alcohols. Genetic engineering and optimization of C. glutamicum led to strains that could produce isobutanol at similar yields as in optimized Escherichia coli strains (15, 123, 158). The oleaginous Rhodococcus opacus was shown to accumulate high yields of high-energy density TAG (74). Additionally, native long-chain FAAE (biodiesel) production has been demonstrated in a Streptomyces strain isolated from sheep feces (83), and an isoprenoid biosynthesis pathway in Streptomyces venezuelae was combined with a cyclase gene from Abies grandis (grand fir) to produce bisabolene, another alternative to diesel fuel (103). 1-Propanol, an industrially relevant solvent with good fuel properties, was produced in T. fusca at a titer of 0.5 g/L after the addition of an aldehyde/alcohol dehydrogenase, approaching the yields possible in E. coli (44). Transcriptomic and metabolomic analyses suggested that additional pathway engineering of the intermediates would result in increased yields (43).

These studies represent a few of the many possible options to engineer Actinobacteria to produce bioproducts. The evolution of diverse natural product pathways in Actinobacteria, especially in genera such as Pseudonocardia, Rhodococcus, and Streptomyces (Figure 3b), make these organisms attractive as sources for discovery of biosynthetic enzymes and as hosts for pathway engineering and heterologous expression (45a, 103). The use of terpenoid biosynthetic machinery for bisabolene production in Streptomyces represents just one possible adaptation of natural product pathways for biofuel production. Polyketide natural products represent another source of industrially relevant molecules from Actinobacteria (160). Potentially, rational engineering of modular polyketide synthases can lead to molecules with extremely diverse structural and functional characteristics (71, 102).

Combining the natural abilities of Actinobacteria both to efficiently degrade plant biomass and to produce diverse chemicals make them well suited to contribute to the long-term goal of deploying consolidated bioprocessing (CBP) (Figure 4). The potential of Actinobacteria to contribute to CBP was illustrated by the FAAE-producing Streptomyces species, which grew readily on cellulose and xylan and produced higher biodiesel titers on complex carbohydrates than CBP-engineered E. coli (83, 125). The TAG-accumulating R. opacus was engineered with Streptomyces xylose metabolism genes (76) and subjected to directed evolution for high production of TAG with lignocellulosic biomass as a carbon source (75). Use of CBP approaches with Actinobacteria may also be possible for commodity chemicals. For example, the strain of T. fusca engineered for 1-propanol production produced higher titers when grown on a complex biomass, including cellulose and switchgrass, than on refined sugars. Similarly, Streptomyces maritimus produced benzoate, a food preservative and plastics additive, at higher yields when engineered to grow on cellulose as compared to growth on simple sugars (96). Furthermore, Streptomyces lividans modified with a T. fusca endoglucanase (for cellulose consumption) and two biosynthesis genes could produce 4-vinylphenol, a compound used in the plastics and electronics industries (95).

CONCLUSIONS

As our main source of antibiotics, Actinobacteria became famous for helping contribute to one of the greatest achievements of the twentieth century—vastly reducing the impact of infectious diseases. This century, we have an urgent and critical need for bio-based approaches that help meet our energy requirements while reducing carbon emissions. As described in this review, efforts are underway to leverage the diverse ecology of Actinobacteria to identify enzymes, pathways, and biosynthetic products that can contribute to biofuel and chemical production. We believe that, in the twenty-first century, Actinobacteria have potential to further increase in prominence as they help us meet our global needs for renewable biofuels and other bioproducts while mitigating the dangers of climate change.

Glossary

Nutritional mutualist

a symbiont that provides one or more key dietary benefits to another organism

Natural product

any chemical compound produced by an organism, colloquially used for secondary metabolites that have uses in medicine and biotechnology

Defensive mutualist

a symbiont that provides protection, usually physical or chemical, to another organism

Biofuel

fuel produced through biotechnology, as opposed to fossil fuel

Horizontal gene transfer (HGT)

the acquisition of genetic material into a genome other than through direct vertical inheritance

Coevolution

reciprocal evolutionary change in interacting species

Rhizosphere

the interface between plant roots and soil, often colonized by a specific microbial community, mediated by the plant’s chemical secretions

Endophyte

a symbiont that lives within a plant

Carbohydrate-active enzyme (CAZy)

an enzyme that acts on polysaccharides, curated at http://www.cazy.org and elsewhere

Glycoside hydrolase (GH)

a diverse group of enzymes that cleave glycosidic bonds, for example, between glucose molecules in cellulose

Endoglucanase

a GH enzyme that reacts with internal glycosidic bonds of the cellulose chain

Exoglucanase

a GH enzyme that acts at the reducing or nonreducing end of the cellulose chain, often releasing cellobiose

β-Glucosidase

a GH enzyme that cleaves cellobiose into glucose monomers

Carbohydrate-binding module

proteins that target a specific carbohydrate, often tethered to GH domains to increase catalytic efficiency

Lytic polysaccharide monooxygenase

a copper-dependent enzyme that hydroxylates the glycosidic bond in cellulose, xylan, starch, and chitin

Bioproduct

a chemical, fuel, or other goods produced using biotechnology

Consolidated bioprocessing (CBP)

the combination of enzyme production, biomass degradation, and fuel production in one reactor, often by one organism