Functional diversity and metabolic engineering of plant-specialized metabolites

Abstract

Plants are talented biochemists that produce a broad diversity of small molecules. These so-called specialized metabolites (SMs) play critical roles in the adaptive evolution of plants to defend against biotic and abiotic stresses, attract pollinators, and modulate soil microbiota for their own benefits. Many plant SMs have been used as nutrition and flavor compounds in our daily food, as well as drugs for treatment of human diseases. Current multi-omics tools have significantly accelerated the process of biosynthetic pathway elucidation in plants through correlation analyses, genetic mapping, and de novo biosynthetic gene cluster predictions. Understanding the biosynthesis of plant SMs has enabled reconstitution of naturally occurring specialized metabolic pathways in microbial hosts, providing a sustainable supply of these high-value molecules. In this review, we illustrate the general functions of several typical plant SMs in natural ecosystems and for human societies. We then provide an overview of current methods elucidating the biosynthetic pathways of plant SMs, and synthetic biology strategies that optimize the efficiency of heterologous biosynthetic pathways in microbial hosts. Moving forward, dissection of the functions and application of plant SMs by using current multidiscipline approaches would be greatly benefit to the scientific community and human societies.

Graphical abstract

Keywords

plant specialized metabolites / metabolic engineering

Introduction

Plants are superb biochemists that can produce a large diversity of small molecules by absorbing CO₂, mineral elements, and water from the environment. These diverse metabolites are typically present in lineage-, developmental stage-, and tissue-specific manners, which are hence termed as specialized metabolites (SMs). The total number of structurally distinct SMs found across the plant kingdom has been estimated to be in the range of hundreds of thousands [1]. In contrast to the broadly conserved primary metabolites that are required for plant growth and development under ideal growth conditions, SMs play important roles in the interactions between plants and their environments. SMs add desirable flavors and important health benefits to human nutrition. In the last decades, molecular genetics and physiology studies have revealed biosynthesis of several important plant SMs. These knowledges have enabled synthetic biology approaches to reconstruct biosynthetic pathways of plant SMs in chassis organisms, producing the target compounds on an industrial scale. In this review, we briefly summarize the function of several typical plant SMs in natural ecosystems and for human consumption. We then delineate current methods in elucidating biosynthetic pathways of SMs in plants, and synthetic biology strategies to enhance target compound production in microbial chassis.

SMs mediate plant–environment interactions

As sessile organisms, plants have adapted to a wide array of hostile environmental conditions. To cope with these challenges, plants produce diverse SMs to protect themselves from environmental stressors, attract animal pollinators for successful cross-fertilization, engineer local soil microbiota to support their own growth, and adapt to abiotic stresses. Examples illustrating these ecological functions of SMs are summarized in this section (Fig. 1).

Protection of plants against insect herbivores and microbial pathogens

Plant SMs were initially appreciated for their protective functions against insect herbivores and pathogenic microbes. Classic examples of such compounds include glucosinolate [2], cyanogenic glucosides [3], nicotine [4], and benzoxazinoids [5]. Current mass spectrometry-based high-throughput analytical chemistry platforms and chemo-informatics tools have enabled exploration of more diverse defensive metabolites that present in low abundance in non-model organisms, as recently reviewed by Li and Gaquerel [6]. For example, a natural-history-guided omics approach has been taken to identify a novel compound, caffeoylputrescine-5-(Z)-3-hexenal, from a wild tobacco species, which was directly toxic to Empoasca leafhoppers when fed at the physiologically relevant concentration [7]. Interestingly, plants have also evolved exquisite detoxification mechanisms to avoid autotoxicity caused by the SMs they produce [8].

Other than being directly toxic against the intruding organisms, SMs could also protect plants by scavenging cytotoxic reactive oxygen species (ROS), fortifying the physical barriers of plants, altering the behavior of insect herbivores, and serving as environmental signal molecules. For example, many phenolic compounds could promptly remove excessive ROS in plant cells following the rapid ROS burst in response to environmental stimuli to avoid oxidative damage [9, 10]. Some phenolic compounds could also cross-link the cellulose fibers in plant cell walls to resist physical damage inflicted by insect attack and pathogen penetration [11]. Diverse classes of SMs including methyl salicylates [12], fatty acid derivatives [13], and terpenoids [14] have been demonstrated to either repel insect herbivores, or attract their natural enemies to protect the plants under attack. Some volatile compounds could also be perceived by neighboring plants as warning signals [15–21].

While plants are usually protected by the SMs that they produce, co-evolving pests have sometimes evolved measures to take advantage of plant SMs for their own benefits. For example, the hallmark SMs of maize (Zea mays), benzoxazinoids, are used by the specialized underground herbivore western corn rootworm (Diabrotica virgifera) as a reliable chemical signal to locate maize roots [22]. These specialized herbivores could not only tolerate the toxicity of benzoxazinoids, but also exploit the metal ion-chelating property of these compounds to supplement iron acquisition in their diets [23], and sequester the toxic compounds to protect themselves from natural enemies [24]. The delivery of plant SMs through complex food chains further extends the ecological significance of these natural products. Indeed, a recent experimental ecology study has demonstrated that allelic variation in a plant-specialized metabolic gene could enhance the persistence of the local food web [25].

Attraction of animal pollinators

The evolutionary success of flowering plants has been associated with insect-mediated cross-pollination. While the fundamental rewards provided to insect pollinators are primary metabolites such as carbohydrates (i.e. nectar) and proteins (i.e. pollen), SMs play critical roles in pollinator attraction by adding colors and fragrances to the floral display [26, 27]. From early field ecology observations to recent molecular genetics dissections, studies in plant–pollinator interactions have transcended multiple biological scales. A classic example was illustrated by a series of studies on Mimulus species. Bradshaw [28] identified the YELLOW UPPER locus, which influenced the petal color by regulating tissue-specific carotenoid metabolism, and this variation in floral coloration was associated with either bees or hummingbirds as the predominant pollinators in their natural habitats. The authors then hybridized these species to generate an experimental near iso-genic population with the different YELLOW UPPER alleles introgressed into the reciprocal genetic background, and demonstrated that the native pollinator visitation rate was negatively influenced by this introgression [29, 30]. More examples demonstrating the function of plant SMs in plant–pollinator interactions have been elegantly synthesized in a recent review by Fattorini and Glover [31]. Interestingly, some pollinator-attracting plant SMs could also exert dual functions as a defensive metabolite against florivorous insects [32].

Modulation of rhizospheric microbiota

As early as the first century ad, it was observed that some plants, such as black walnut (Julgans nigra), were capable of inhibiting the growth of neighboring plants, a phenomenon known as allelopathy. It has since been shown that allelopathic interactions between plants are mediated by toxic SMs exuded into the rhizosphere [33]. In the last decades, researchers have demonstrated that root exuded SMs are not limited to toxic compounds, but could rather function as critical chemical messengers that facilitate interactions between plants and soil microbes. Textbook examples on this topic include fabaceous flavones, which induce expression of nodulation genes in Rhizobium bacteria [34], and strigolactones, which stimulate branching of germinating arbuscular mycorrhizae fungi [35]. More recently, technological advances in environmental DNA sequencing and microbe culturing have revealed the robust dynamics of plant–microbiota interactions in the rhizosphere. To date, the rhizospheric microbiota-engineering function has been demonstrated for a long list of exuded SMs including benzoxazinoids [23, 36], phenolics [37, 38], and terpenoids [39, 40]. One of the less studied topics in this area is how the SMs in root exudate are exported on the cellular level. In the model species Arabidopsis thaliana, exudation of coumarins was facilitated by an ATP-binding cassette-containing transporter [41]. Recent evidences from our group suggest that a Multidrug and Toxic Compound Extrusion transporter expressed in melon (Cucumis melo) roots is responsible for the exudation of a triterpene compound, cucurbitacin B, which can suppress soil-borne fungal pathogens by enriching specific taxa of beneficial bacteria [42].

Adaptation toward abiotic stresses

In addition to protecting plants against biotic stressors, SMs also participate in the plant adaptive processes toward various abiotic stresses. For example, enhanced accumulation of flavonoids in Tibetan barley (Hordeum vulgare, qinke) was proposed to be an adaptive mechanism to protect against exacerbated ultraviolet-B irradiation of the Tibetan plateau [43]. Concentrations of various SMs were significantly altered in maize plants grown under low-phosphorus conditions [44]. Most recently, genome-wide association studies followed by candidate validation with genetic mutant analyses have revealed the functionality of a number of maize SMs in drought and salt resistance [45, 46].

SMs used as flavor, nutrition, and medicine

Plants are the primary producers of our ecosphere. While the primary metabolites in plants provide essential nutrients to humans, plant SMs are valued for their flavoring, nutraceutical, and medicinal properties. Current metabolomics methods have enabled researchers to pinpoint and quantify the keystone plant metabolites that are associated with consumer preference or specific therapeutic functions. Some of these examples are delineated in this section (Fig. 2).

Flavors

Sweetness is perhaps one of the most desirable traits for freshly consumed fruits and vegetables (Fig. 2a). Recent studies in tomato (Solanum lycopersicum) and blueberry (Vaccinium corymbosum) have demonstrated that the sense of sweetness is not only associated with sugar content, but also promoted by various volatile compounds [47, 48]. In contrast, organic acid contents have been reported to be negatively correlated with perceived levels of sweetness in tomato [48], apple (Malus spp.) [49–51], and kiwifruit (Actinidia arguta) [52, 53]. The complex biochemical composition associated with the sense of sweetness provides the opportunity to improve this important quality trait by targeting the SMs of these crops rather than directly manipulating their sugar metabolism, so as to avoid any potential yield penalty. Furthermore, some plant SMs, such as stevia glycosides and mogrosides, are inherently sweet, and are widely exploited as natural low-caloric sweeteners. Besides sweetness, different fruits and vegetables often have their characteristic tastes and aroma, which are also endowed by their SMs [54]. Distinctive aroma of cinnamon (Cinnamomum verum) [55], melon [56], cucumber (Cucumis sativa) [57, 58], pumpkin (Cucurbita spp.) [59], and kiwifruit [60, 61] are determined by their characteristic volatile phenolics and fatty acid derivatives. A variety of volatile terpenoids and uncommon nitrogen-containing volatiles have also been shown to contribute to the aroma of tomato [62–65].

Since SMs mainly function as defensive compounds for plants in nature, most SMs are toxic and/or have unpleasant taste to humans [66, 67] (Fig. 2a). The presence of these SMs in the edible parts of crop plants has often been negatively selected during domestication. In cucurbits, a mutation in a fruit-specific transcription factor abolished the biosynthesis of bitter-tasting triterpenoids in fruits [68]. Similar changes in tissue-specific transcription regulation have recently been demonstrated to lower the level of toxic cyanogenic glycosides in domesticated almond kernels [69]. In tomato fruits, a tissue- and developmental stage-specific nitrate/peptide family transporter could export the bitter α-tomatine from vacuole to cytosol during fruit ripening for prompt degradation [70]. On the other hand, some SMs that are nontoxic but have pungent tastes to human are preserved during crop domestication. For example, assorted phenolic acids contribute to bitterness and astringency in tea leaves (Camellia sinensis), which contribute to the complex flavor of tea [71, 72]. In olive fruits (Olea spp.), the content of bitter-tasting oleuropein needs to be properly controlled during olive breeding for a balanced taste in the fruits [73, 74].

While SMs can directly affect the taste and aroma of freshly consumed fruits and vegetables, food processing for most staple crops has a significant impact on the biochemical composition, and hence the flavor of food. Maillard reaction is one of the most common chemical reactions that occurs during food processing, which produces flavorful volatiles [75]. For example, furans contribute the caramel-like odor of heated carbohydrates, and pyrazines and methional provide the characteristic aroma of baked potatoes [76].

Nutraceuticals and medicine

Besides serving as flavor molecules, some plant SMs, such as vitamins, are essential components of human nutrition. Vitamin C (l-ascorbic acid) is the most classic example of essential nutrient that is exclusively supplied by fresh plants since humans and other primates have lost their own capacity to biosynthesize this vital molecule [77]. Similarly, plants are an important source of diverse B vitamins including B1 (thiamin from grains), B5 (pantothenic acid from sunflower seeds and potatoes), and B9 (folate from spinach) [78–80]. Recent technological advances in genetic engineering and genome editing have inspired a series of bio-fortification efforts to enhance the nutritional quality of staple crops, for more accessible nutrition in under-developed regions. For example, five carotenoid biosynthetic genes have been specifically expressed in maize kernels to promote accumulation of a variety of carotenoids, which can be converted to vitamin A in animals after ingestion [81]. Chickens raised on these bio-fortified feed produced high-carotenoid poultry meat and were more resistant against a protozoan parasite [82]. Most recently, a specific isoform of 7-dehydrocholesterol reductase in tomato was knocked-out to allow accumulation of 7-dehydrocholesterol, which would be simultaneously converted into vitamin D upon ultraviolet irradiation [83].

Plant SMs have been the active ingredients of many traditional medicines for millennia. In the last 2 centuries, detailed analytical chemistry studies have revealed the principal active compounds associated with various medicinal plants (Fig. 2b). For example, the dried aerial parts of the herb Artemisia annua have been used in China for centuries to treat fever and malaria [84]. The main active compound, artemisinin, is highly effective against malaria parasites Plasmodium falciparum, without being toxic to humans or animals at the same concentration [85]. Cucurbitacins, a class of triterpenoids widely distributed in cucurbit plants, have been exploited for their anti-inflammatory, hepatoprotective, and antiproliferative effects [86]. Paclitaxel, a diterpene from Taxus species, is a well-known chemotherapy agent against various cancers [87]. Other classic chemotherapeutic compounds, vinblastine and vincristine, are naturally sourced from Vinca rosea for treatment of Hodgkin’s lymphoma and childhood lymphoblastic leukemia [88]. More of the current progress and future perspectives of the use of plant SMs as a source of medicine have been recently reviewed by Jacobowitz and Weng [89].

Elucidating biosynthetic pathways of SMs in plants

Since plant SMs are appreciated for their flavoring and therapeutic functions for human, understanding the biosynthesis of SMs has always been one of the top priorities in plant biology research. Early efforts to clarify plant-specialized metabolism relied on traditional biochemistry techniques such as stable isotope feeding and protein chromatography coupled with in vitro enzymatic assays. In the last decade, rapid development in multi-omics tools have enabled a tremendous leap in specialized metabolic pathway elucidation in plants. This section summarizes studies exemplifying these current methods (Fig. 3).

Metabolite-gene expression correlation

Plant SMs are well-known for their tissue-specific distribution and inducible accumulation. This spatiotemporal heterogeneity allows researchers to identify potential biosynthetic genes by looking for statistical correlation between experimentally determined gene expression and target metabolite abundance (Fig. 3a). Since the 2010s, this simple yet effective approach has been successfully applied to identify key biosynthetic enzymes of a number of renowned plant SMs such as podophyllotoxin [90], vinblastine and vincristine [88], and colchicine [91, 92]. This method also contributed to the recent characterization of root-specific triterpenoid biosynthetic pathways in Arabidopsis [39], and proto-limonoid biosynthesis in rutaceaeous plants [93]. One of the key advantages of this approach is that it only requires a few samples for transcriptomics and targeted metabolomics analyses to identify potential candidate genes. For example, it took 12 RNAseq samples to identify the candidate biosynthetic genes of podophyllotoxin through a co-expression analysis [90]. Also, transcriptome data can be de novo assembled for quantitative analyses, which bypasses the requirement to sequence and assemble large plant genomes. These advantages have significantly enhanced the practical viability of correlation analyses in genetic dissection of plant-specialized metabolism, especially in non-model medicinal plant species that lack diverse germplasms and reliable reference genomes. However, it should be noted that some SMs are transported away from their main tissues of biosynthesis and stored elsewhere (e.g. nicotine, cardenolide), which would hence nullify the presumed cross-tissue correlation of the abundance of the SM with the expression of its biosynthetic genes.

Forward genetics

In addition to studies of rare and medicinal SMs in non-model plant species, examination of SMs in staple and economic crops is another fast-developing area in plant research. Here, thanks to the diverse germplasm and genotypic information accumulated from generations of crop breeders, the forward genetics approach is preferred to identify the biosynthetic and/or regulatory genes associated with the target SMs (Fig. 3b). In maize, e.g. classic bi-parental mapping populations are now routinely adopted to locate the genetic loci associated with SMs [94–98]. Though genetic mapping with bi-parental populations alone rarely gets down to single-gene resolution, addition of a clear biochemical hypothesis and gene expression data are usually sufficient to narrow down the list of candidate genes to a range that can be experimentally validated.

Compared to bi-parental populations, the more diverse genome-wide association panels offer much more frequent recombination, and hence much higher mapping resolution. Indeed, studies in rice (Oryza sativa) [99], maize [100, 101], and tomato [48, 102] have demonstrated that single-gene level mapping can be achieved with sufficient molecular marker coverage. On the contrary to what was once commonly believed, high-resolution association genetics revealed that the genetic architecture of plant SMs is not necessarily simpler than that of composite traits such as grain yield [101]. This is probably reflective of the complex cis and trans regulation of specialized metabolism in plants. Nevertheless, current genetic dissection of SM biosynthetic pathways in crop species have demonstrated the potential to enhance crop quality and to protect crop plants against environmental stresses through genetic engineering [102, 103].

Genome-mining based on operon-like gene clusters

Since the first report of a cluster of five genes (Bx1 through Bx5) that are required for synthesis of antimicrobial compound, 2,4-dihydroxy-1,4-benzoaxin-3-one (DIBOA), at a narrow region on a maize chromosome [104], another metabolic gene cluster for biosynthesis of antimicrobial compounds, avenacins, was defined by genetic and bacterial artificial chromosme analysis in oats [105, 106]. Since then, similar biosynthetic gene clusters (BGCs) have been reported across diverse plant species for the production of many plant SMs [107] (Fig. 3c; Table 1). A potential evolutionary advantage of these BGCs in plant genomes is that the physical proximity between these enzyme-encoding genes could facilitate shared epigenomic modifications and hence coordinated expression regulation [67, 68, 131–133]. For example, tissue-specific trans regulators can directly bind to the conserved E-box elements in the promoters of nine clustered cucurbitacin biosynthetic genes, and coordinately activate their transcriptions to produce the target compound [67, 68]. Interestingly, the later studies have demonstrated that BGCs can include not only enzyme-encoding genes, but also genes that encode specific transporters of the SMs produced [134, 135].

As the idea of plant BGCs receives accumulating experimental support, bioinformatic tools that predict BGCs solely based on genomic information have also been developed to annotate previously unknown SM biosynthetic genes [136]. A recent study on wheat (Triticum aestivum) and a related model grass species exemplified this reverse chemical genetics approach, where the authors started by annotating potential BGCs in recently published genomes and ended up identifying previously unknown SMs in these species by combining pathogen-inducible gene expression data and functional validation experiments [137].

It is noteworthy that the functionally related but structurally non-homologous plant BGCs bear close resemblance to prokaryotic operons, since members of both BGCs and operons are tightly linked in the genomes and tend to be co-expressed [138]. That being said, BGCs and operons most likely arise independently during evolution, given the distant phylogenetic separation between plants and bacteria. While prokaryotic operons are well-known to horizontally transfer between co-habituating microbes, plant BGCs tend to arise independently in each phylogenetic lineage which then go on to differential evolutionary trajectories [67, 139–141]. However, recent studies of the classic benzoxazinoid BGCs in maize and wheat suggested that this cluster may have been horizontally transferred between their ancestors, followed by independent evolution of a functionally homologous downstream modification enzyme [142, 143]. It will take further studies to determine whether this proposed horizontal BGC transfer is an isolated case or a harbinger of a more common pattern in plant evolution.

Synthetic biology for sustainable supply of plant SMs

Though plant SMs are widely utilized by human beings, the low abundance of valuable SMs in planta and the low efficiency in the extraction process limit their supply from native sources. The expanding inventory of functionally characterized plant enzymes, coupled with advancing synthetic biology tools, provides opportunities to sustainably produce plant SMs in engineered chassis microbes such as Escherichia coli and Saccharomyces cerevisiae [109, 144–148]. However, heterologous plant SM biosynthetic pathways in microbes often perturb the innate metabolic balance of the host, resulting in lowered target productivity or even host fatality [149]. Therefore, fine-tuning of the heterologous pathways is essential to unlock the full potential of microbial cell factories for plant SM production. Here we summarize several representative synthetic biology strategies in optimizing the native metabolic pathways of chassis microbes for enhanced production of heterologous compounds (Fig. 4).

Enzyme engineering (design and generation of highly efficient enzymes)

On the most basic molecular level, the efficiency of a metabolic pathway depends on the catalyzing efficiency of the enzymes. Assembly of naturally occurring plant enzymes into heterologous biosynthetic pathways in chassis organisms often suffers from lowered catalytic activity, narrowed substrate specificity, poor protein structure stability, and unexpected allosteric inhibition [150]. Traditionally, these problems had been addressed by increasing the titer of the heterologous enzymes with optimized gene expression levels or gene copy numbers, as well as utilization of superior alternative enzymes from other natural sources [151]. More recently, protein engineering methods such as directed evolution and site-directed mutagenesis have emerged as more efficient approaches to enhance the performance of rate-limiting enzymes [152] (Fig. 4a).

Directed evolution is one of the earliest protein engineering approaches as it does not require detailed knowledge of protein structure. By reiteratively selecting for the most efficient structural variant of the enzyme under study, this process enhances overall enzyme performance through evolution by selection. This approach has been adopted to enhance astaxanthin and various isoprenoids production in yeast and E. coli [153–156]. However, it should be noted that successful cases of enzyme engineering through directed evolution are almost exclusively reported for carotenoid-related pathways, as the distinctive pigmentation of the end product provides an easy proxy for high-throughput selection [157]. More recently, researchers have attempted to circumvent this limitation by incorporating machine-learning algorithms into directed evolution schemes to reduce the sequence space that had to be experimentally tested [158, 159].

The rapid development of crystallography-based structural biology has enabled structure-guided protein engineering by computer-assist molecular simulation. Through this so-called site-directed mutagenesis approach, key amino acid residues that affect enzyme activities can be identified by resolving the crystal structure of the target protein and simulating possible molecular interactions. The protein structure can then be optimized by designing proper amino acid substitution schemes [150]. For example, modification of a glycosyltransferase through structure-based rational design achieved a 1800-fold increase in catalytic efficiency for the production of ginsenoside Rh2, a potential anticancer compound [160]. This approach was also applied to improve substrate selectivity and protein stability of enzymes [161]. One of the main bottlenecks for broader application of site-directed mutagenesis of enzymes is the difficulty to obtain precise protein structures experimentally. Recent progress in computational protein structure prediction algorithms (e.g. AlphaFold and RoseTTA-Fold) has significantly promoted the prediction accuracy, reaching the same level of accuracy as the experimentally resolved protein structures in many cases [162, 163]. This combination of prediction accuracy and practical flexibility could enable site-directed improvement of more biosynthetic enzymes of plant SMs.

Synthetic orthogonal pathways (modulation of microbial chassis)

While targeted protein engineering could enhance the metabolic flux within the engineered SM pathways, the metabolic bottlenecks in microbial chassis could also be positioned further upstream, limiting the supply of chemical precursors. In these cases, synthetic orthogonal pathways are built to provide an additional route of precursor supply without perturbing the native pathway to avoid undesirable side-effects [164] (Fig. 4b). This strategy has been widely adopted in the case of heterologous terpenoid production, where the canonical terpene precursors are tightly coupled to sterol biosynthesis essential for the survival chassis organisms. Although heterologous introduction of the entire mevalonate pathway into E. coli to replace the native methylerythritol 4-phosphate pathway is now commonly adopted [165–167], both pathways are strongly coupled with the central carbon metabolism and are tightly regulated, thus limiting the availability of the terpenoid precursors, isopentenyl pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP). Recently, a synthetic orthogonal pathway was developed to produce IPP and DMAPP from isoprenol or prenol through two steps of phosphorylation [168]. This synthetic Isopentenol Utilization Pathway (IUP) is decoupled from the central carbon metabolism of E. coli, and hence can sustain a high-carbon flux toward terpene precursors without disturbing the normal metabolism of the chassis microbe. In addition, IUP only uses ATP as the sole co-factor, further reducing the potential impact of this synthetic pathway on the host metabolism through co-factor competition. A similar approach of orthogonal IPP and DMAPP production from glycerol has also been reported, which eliminated the need to feed isopentenols in the culturing medium [169]. These orthogonal pathways have managed to enhance heterologous isoprenoid production in various microbial systems [168–171]. In another example, introduction of a heterologous geranyl diphosphate (GPP) synthase using neryl diphosphate as the predominant substrate has enhanced GPP production compared to the native yeast prenyltransferase, which in turn promoted the heterologous monoterpene accumulation [164].

Co-compartmentalization of biosynthetic pathway

In plants, specialized metabolism occurs in specific subcellular locations to promote metabolic efficiency and avoid autotoxicity by bioactive intermediates [89]. Inappropriate localization of a heterologous biosynthetic pathway in the chassis organism could result in a poor target compound yield, side-products accumulation, and/or toxicity to the host [172]. Therefore, current synthetic biology schemes emphasize the proper compartmentalization of heterologous biosynthetic pathways to ensure their proper function in microbial hosts (Fig. 4c).

Localization-aware metabolic engineering in eukaryotic host microbes has enhanced production of isoprenoids [173–177], alkaloids [178, 179], and fatty acid derivatives [180, 181] in various organelles. The value of proper pathway compartmentalization in synthetic biology is probably best exemplified by the recent reconstitution of the tropane alkaloid pathway in baker’s yeast [182]. In this systematic engineering scheme, the authors heterologously expressed >20 carefully selected biosynthetic genes from a number of plant and microbe genomes to rebuild the tropane alkaloid pathway in different cellular compartments. During this process, the authors converted the littorine synthase (LS) into an artificial transmembrane protein to bypass the impediment in maturation and trafficking of the natural plant LS protein through yeast trans-Golgi network, a process that is required for the final production of the target compounds. To further facilitate the function of this engineered LS in yeasts, the matured protein was targeted to yeast vacuoles, which better mimics the plant tonoplast environment, where this enzyme naturally functions. The tropine substrates were transported into yeast vacuoles by an additional transporter protein encoded by a tobacco gene. In another recent example, the astaxanthin pathway was targeted to three compartments in Y. lipolytica to bring astaxanthin biosynthetic enzymes into close proximity with their precursor compound β-carotene [183]. This compartmentalization-facilitated physical proximity accelerated the conversion of β-carotene into astaxanthin, while significantly decreased the accumulation of metabolic intermediates during this process.

Concluding remarks and prospectives

From mediating plant–environment interactions to bringing flavor and health to humans, plant SMs play pivotal roles in our ecosphere and societies. Continued genetic dissection of the biosynthetic pathways and optimization of heterologous production systems are required for sustainable exploration and supply of high-value plant SMs to fulfill the ever-growing demand. Current technological advances in genome editing and high-resolution mass spectrometry-based metabolomics have in turn promoted functional studies of plant SMs in ecological contexts. For instance, accelerated biosynthetic gene identification and subsequent production of targeted genetic mutants have facilitated examination of the ecological function of specific plant SMs on a single-pathway or single-compound resolution. On a more systematic scale, integration of mass spectra-based molecular network and information theory statistics can be adapted to test classic ecological theories, as recently been demonstrated with wild tobacco populations [184].

With the powerful tools of current multi-omics analytics and synthetic biology, plant biologists are marching into the uncharted biochemical diversity that remains in the enormous number of non-model species, and developing a variety of chassis organisms for efficient production of different plant SMs. We expect these modern technologies to bring new light into the utilization of plant-specialized metabolism for human health and well-beings.

References

[1] Mithöfer A, Boland W. Plant defense against herbivores: chemical aspects. Annu Rev Plant Biol 2012; 63: 431–50.

[2] Clay NK, Adio AM, Denoux Cet al. Glucosinolate metabolites required for an Arabidopsis innate immune response. Science 2009; 323: 95–101.

[3] Gleadow RM, Moller BL. Cyanogenic glycosides: synthesis, physiology, and phenotypic plasticity. Annu Rev Plant Biol 2014; 65: 155–85.

[4] Xu SQ, Brockmoller T, Navarro-Quezada Aet al. Wild tobacco genomes reveal the evolution of nicotine biosynthesis. Proc Natl Acad Sci USA 2017; 114: 6133–8.

[5] Zhou S, Richter A, Jander G. Beyond defense: multiple functions of benzoxazinoids in maize metabolism. Plant Cell Physiology 2018; 59: 1528–37.

[6] Li DP, Gaquerel E. Next-generation mass spectrometry metabolomics revives the functional analysis of plant metabolic diversity. Annu Rev Plant Biol 2021; 72: 867–91.

[7] Bai Y, Yang C, Halitschke Ret al. Natural history-guided omics reveals plant defensive chemistry against leafhopper pests. Science 2022; 375: eabm2948.

[8] Li JC, Halitschke R, Li DPet al. Controlled hydroxylations of diterpenoids allow for plant chemical defense without autotoxicity. Science 2021; 371: 255–60.

[9] Apel K, Hirt H. Reactive oxygen species: metabolism, oxidative stress, and signal transduction. Annu Rev Plant Biol 2004; 55: 373–99.

[10] Waszczak C, Carmody M, Kangasjarvi J. Reactive oxygen species in plant signaling. Annu Rev Plant Biol 2018; 69: 209–36.

[11] Lee MH, Jeon HS, Kim SHet al. Lignin-based barrier restricts pathogens to the infection site and confers resistance in plants. EMBO J 2019; 38: e101948.

[12] Hardie J, Isaacs R, Pickett JAet al. Methyl salicylate and (-)-(1R,5S)-myrtenal are plant-derived repellents for black bean aphid,Aphis fabae Scop. (Homoptera: Aphididae). J Chem Ecol 1994; 20: 2847–55.

[13] Ameye M, Allmann S, Verwaeren Jet al. Green leaf volatile production by plants: a meta-analysis. New Phytol 2018; 220: 666–83.

[14] Wang H, Guo WF, Zhang PJet al. Experience-induced habituation and preference towards non-host plant odors in ovipositing females of a moth. J Chem Ecol 2008; 34: 330–8.

[15] Rosenkranz M, Chen Y, Zhu Pet al. Volatile terpenes - mediators of plant-to-plant communication. Plant J 2021; 108: 617–31.

[16] Zhou S, Jander G. Molecular ecology of plant volatiles in interactions with insect herbivores. J Exp Bot 2022; 73: 449–62.

[17] Hu L. Integration of multiple volatile cues into plant defense responses. New Phytol 2022; 233: 618–23.

[18] Erb M. Volatiles as inducers and suppressors of plant defense and immunity—origins, specificity, perception and signaling. Curr Opin Plant Biol 2018; 44: 117–21.

[19] Hu L, Zhang K, Wu Zet al. Plant volatiles as regulators of plant defense and herbivore immunity: molecular mechanisms and unanswered questions. Curr Opin Insect Sci 2021; 44: 82–8.

[20] Jing T, Du W, Gao Tet al. Herbivore-induced DMNT catalyzed by CYP82D47 plays an important role in the induction of JA-dependent herbivore resistance of neighboring tea plants. Plant Cell Environ 2021; 44: 1178–91.

[21] Chen S, Zhang L, Cai Xet al. (E)-Nerolidol is a volatile signal that induces defenses against insects and pathogens in tea plants. Hortic Res 2020; 7: 52.

[22] Robert CA, Veyrat N, Glauser Get al. A specialist root herbivore exploits defensive metabolites to locate nutritious tissues. Ecology Lett 2012; 15: 55–64.

[23] Hu L, Mateo P, Ye Met al. Plant iron acquisition strategy exploited by an insect herbivore. Science 2018; 361: 694–7.

[24] Zhang X, van Doan C, Arce CCMet al. Plant defense resistance in natural enemies of a specialist insect herbivore. Proc Natl Acad Sci USA 2019; 116: 23174–81.

[25] Barbour MA, Kliebenstein DJ, Bascompte J. A keystone gene underlies the persistence of an experimental food web. Science 2022; 376: 70–3.

[26] Gang DR. Evolution of flavors and scents. Annu Rev Plant Biol 2005; 56: 301–25.

[27] Grotewold E. The genetics and biochemistry of floral pigments. Annu Rev Plant Biol 2006; 57: 761–80.

[28] Bradshaw HD, Wilbert SM, Otto KGet al. Genetic mapping of floral traits associated with reproductive isolation in monkeyflowers (Mimulus). Nature 1995; 376: 762–5.

[29] Bradshaw HD, Schemske DW. Allele substitution at a flower colour locus produces a pollinator shift in monkeyflowers. Nature 2003; 426: 176–8.

[30] Schemske DW, Bradshaw HD. Pollinator preference and the evolution of floral traits in monkeyflowers (Mimulus). Proc Natl Acad Sci USA 1999; 96: 11910–5.

[31] Fattorini R, Glover BJ. Molecular mechanisms of pollination biology. Annu Rev Plant Biol 2020; 71: 487–515.

[32] Kessler D, Bing J, Haverkamp Aet al. The defensive function of a pollinator-attracting floral volatile. Funct Ecol 2019; 33: 1223–32.

[33] Bais HP, Weir TL, Perry LGet al. The role of root exudates in rhizosphere interactions with plants and other organisms. Annu Rev Plant Biol 2006; 57: 233–66.

[34] Redmond JW, Batley M, Djordjevic MAet al. Flavones induce expression of nodulation genes in Rhizobium. Nature 1986; 323: 632–5.

[35] Akiyama K, Matsuzaki K, Hayashi H. Plant sesquiterpenes induce hyphal branching in arbuscular mycorrhizal fungi. Nature 2005; 435: 824–7.

[36] Cotton TEA, Petriacq P, Cameron DDet al. Metabolic regulation of the maize rhizobiome by benzoxazinoids. ISME J 2019; 13: 1647–58.

[37] Voges M, Bai Y, Schulze-Lefert Pet al. Plant-derived coumarins shape the composition of an Arabidopsis synthetic root microbiome. Proc Natl Acad Sci USA 2019; 116: 12558–65.

[38] Harbort CJ, Hashimoto M, Inoue Het al. Root-secreted coumarins and the microbiota interact to improve iron nutrition in Arabidopsis. Cell Host & Microbe 2020; 28: 825–37.e6.

[39] Huang AC, Jiang T, Liu YXet al. A specialized metabolic network selectively modulates Arabidopsis root microbiota. Science 2019; 364: eaau6389.

[40] Chen Q, Jiang T, Liu YXet al. Recently duplicated sesterterpene (C25) gene clusters in Arabidopsis thaliana modulate root microbiota. Sci China Life Sci 2019; 62: 947–58.

[41] Fourcroy P, Siso-Terraza P, Sudre Det al. Involvement of the ABCG37 transporter in secretion of scopoletin and derivatives by Arabidopsis roots in response to iron deficiency. New Phytol 2014; 201: 155–67.

[42] Zhong Y, Xun W, Wang Xet al. Root-secreted bitter triterpene modulates the rhizosphere microbiota to improve plant fitness. Nat Plants 2022; 8: 887–96.

[43] Zeng X, Yuan H, Dong Xet al. Genome-wide dissection of co-selected UV-B responsive pathways in the UV-B adaptation of Qingke. Mol Plant 2020; 13: 112–27.

[44] Luo BW, Ma P, Nie Zet al. Metabolite profiling and genome-wide association studies reveal response mechanisms of phosphorus deficiency in maize seedling. Plant J 2019; 97: 947–69.

[45] Zhang F, Wu JF, Sade Net al. Genomic basis underlying the metabolome-mediated drought adaptation of maize. Genome Biol 2021; 22: 260.

[46] Liang XY, Liu SY, Wang Tet al. Metabolomics-driven gene mining and genetic improvement of tolerance to salt-induced osmotic stress in maize. New Phytol 2021; 230: 2355–70.

[47] Gilbert JL, Guthart MJ, Gezan SAet al. Identifying breeding priorities for blueberry flavor using biochemical, sensory, and ­genotype by environment analyses. PLoS One 2015; 10: e0138494.

[48] Tieman D, Zhu G, Resende MFet al. A chemical genetic roadmap to improved tomato flavor. Science 2017; 355: 391–4.

[49] Wege S. Sweet or Sour? Important link between nitrate signaling and malate accumulation identified in Apple. Plant Physiol 2020; 183: 439–40.

[50] Hu DG, Sun CH, Ma QJet al. MdMYB1 regulates anthocyanin and malate accumulation by directly facilitating their transport into vacuoles in apples. Plant Physiol 2016; 170: 1315–30.

[51] Zhang QY, Gu KD, Cheng LLet al. BTB-TAZ domain protein MdBT2 modulates malate accumulation and vacuolar acidification in response to nitrate. Plant Physiol 2020; 183: 750–64.

[52] Nishyama L, Fukuda T, Shimohashi Aet al. Sugar and organic acid composition in the fruit juice of different Actinidia varieties. Food Sci Technol Res 2008; 14: 67–73.

[53] Wang RC, Shu P, Zhang Cet al. Integrative analyses of metabolome and genome-wide transcriptome reveal the regulatory network governing flavor formation in kiwifruit (Actinidia chinensis). New Phytol 2022; 233: 373–89.

[54] Schwab W, Davidovich-Rikanati R, Lewinsohn E. Biosynthesis of plant-derived flavor compounds. Plant J 2008; 54: 712–32.

[55] Atkinson RG. Phenylpropenes: occurrence, distribution, and biosynthesis in fruit. J Agric Food Chem 2016; 66: 2259–72.

[56] Freilich S, Lev S, Gonda Iet al. Systems approach for exploring the intricate associations between sweetness, color and aroma in melon fruits. BMC Plant Biol 2015; 15: 71.

[57] Buescher RH, Buescher RW. Production and stability of (E,Z)-2,6-nonadienal, the major flavor volatile of cucumbers. J Food Sci 2001; 66: 357–61.

[58] Wei G, Tian P, Zhang FXet al. Integrative analyses of nontargeted volatile profiling and transcriptome data provide molecular insight into VOC diversity in cucumber plants (Cucumis sativus). Plant Physiol 2016; 172: 603–18.

[59] Maarse H. Volatile Compounds in Foods and Beverages. New York: Routledge, 2017, 198–277.

[60] Gunther CS, Marsh KB, Winz RAet al. The impact of cold storage and ethylene on volatile ester production and aroma perception in ‘Hort16A’ kiwifruit. Food Chem 2015; 169: 5–12.

[61] Zhang A, Zhang Q, Li Jet al. Transcriptome co-expression network analysis identifies key genes and regulators of ripening kiwifruit ester biosynthesis. BMC Plant Biol 2020; 20: 103.

[62] Buttery R. Quantitative and sensory aspects of flavor of tomato and other vegertables and fruits. Flavor Science: Sensible Principles and Techniques. Washington DC: American Chemical Society, 1993, 259–286.

[63] Baldwin EA, Scott JW, Shewmaker CKet al. Flavor trivia and tomato aroma: biochemistry and possible mechanisms for control of important aroma components. HortScience 2000; 35: 1013–22.

[64] Liscombe DK, Kamiyoshihara Y, Ghironzi Jet al. A flavin-dependent monooxygenase produces nitrogenous tomato aroma volatiles using cysteine as a nitrogen source. Proc Natl Acad Sci USA 2022; 119: e2118676119.

[65] Vogel JT, Tieman DM, Sims CAet al. Carotenoid content impacts flavor acceptability in tomato (Solanum lycopersicum). J Sci Food Agric 2010; 90: 2233–40.

[66] Friedman M. Potato glycoalkaloids and metabolites: roles in the plant and in the diet. J Agric Food Chem 2006; 54: 8655–81.

[67] Zhou Y, Ma YS, Zeng JGet al. Convergence and divergence of bitterness biosynthesis and regulation in Cucurbitaceae. Nat Plants 2016; 2: 16183.

[68] Shang Y, Ma YS, Zhou Yet al. Biosynthesis, regulation, and domestication of bitterness in cucumber. Science 2014; 346: 1084–8.

[69] Sanchez-Perez R, Pavan S, Mazzeo Ret al. Mutation of a bHLH transcription factor allowed almond domestication. Science 2019; 364: 1095–8.

[70] Kazachkova Y, Zemach I, Panda Set al. The GORKY glycoalkaloid transporter is indispensable for preventing tomato bitterness. Nat Plants 2021; 7: 468–80.

[71] Xu YQ, Zhang YN, Chen JXet al. Quantitative analyses of the bitterness and astringency of catechins from green tea. Food Chem 2018; 258: 16–24.

[72] Yang C, Hu ZY, Lu MLet al. Application of metabolomics profiling in the analysis of metabolites and taste quality in different subtypes of white tea. Food Res Int 2018; 106: 909–19.

[73] Perez AG, Leon L, Sanz Cet al. Fruit phenolic profiling: a new selection criterion in olive breeding programs. Front Plant Sci 2018; 9: 241.

[74] Rodríguez MM, Ópez C, Hong Bet al. Two bi‐functional cytochrome P450 CYP72 enzymes from olive (Olea europaea) catalyze the oxidative C-C bond cleavage in the biosynthesis of secoxy-iridoids – flavor and quality determinants in olive oil. New Phytol 2021; 229: 2288–301.

[75] Everts S. The Maillard reaction turns 100. Chem Eng News 2012; 90: 58–60.

[76] Duckham SC, Dodson AT, Bakker Jet al. Effect of cultivar and storage time on the volatile flavor components of baked potato. J Agric Food Chem 2002; 50: 5640–8.

[77] Office of Dietary Supplements. Vitamin C - Health Professional Fact Sheet. Dietary Summplement Fact Sheets 2021 2021-03-26 [cited 2022 2022-06-01].

[78] Office of Dietary Supplements. Folate - Health Professional Fact Sheet. Dietary Supplement Fact Sheets 2021 2021-03-29 [cited 2022 2022-06-01].

[79] Office of Dietary Supplements. Thiamin - Health Professional Fact Sheet. Dietary Supplement Fact Sheets 2021 2021-03-26 [cited 2022 2022-06-01].

[80] Office of Dietary Supplements. Pantothenic Acid - Health Professional Fact Sheet. Dietary Supplement Fact Sheets 2021 2021-03-26 [cited 2022 2022-06-01].

[81] Zhu C, Naqvi S, Breitenbach Jet al. Combinatorial genetic transformation generates a library of metabolic phenotypes for the carotenoid pathway in maize. Proc Natl Acad Sci USA 2008; 105: 18232–7.

[82] Nogareda C, Moreno JA, Angulo Eet al. Carotenoid-enriched transgenic corn delivers bioavailable carotenoids to poultry and protects them against coccidiosis. Plant Biotechnol J 2016; 14: 160–8.

[83] Li J, Scarano A, Gonzalez NMet al. Biofortified tomatoes provide a new route to vitamin D sufficiency. Nat Plants 2022; 8: 611–6.

[84] Bhakuni RS, Jain DC, Sharma RPet al. Secondary metabolites of Artemisia annua and their biological activity. Current Sci 2001; 80: 35–48.

[85] Fidock DA, Rosenthal PJ, Croft SLet al. Antimalarial drug discovery: efficacy models for compound screening. Nat Rev Drug Discovery 2004; 3: 509–20.

[86] Chen XP, Bao JL, Guo JJet al. Biological activities and potential molecular targets of cucurbitacins: a focus on cancer. Anticancer Drugs 2012; 23: 777–87.

[87] Xiong X, Gou J, Liao Qet al. The Taxus genome provides insights into paclitaxel biosynthesis. Nat Plants 2021; 7: 1026–36.

[88] Caputi L, Franke J, Farrow SCet al. Missing enzymes in the biosynthesis of the anticancer drug vinblastine in Madagascar periwinkle. Science 2018; 360: 1235–9.

[89] Jacobowitz JR, Weng JK. Exploring uncharted territories of plant specialized metabolism in the postgenomic era. Annu Rev Plant Biol 2020; 71: 631–58.

[90] Lau W, Sattely ES. Six enzymes from mayapple that complete the biosynthetic pathway to the etoposide aglycone. Science 2015; 349: 1224–8.

[91] Nett RS, Lau W, Sattely ES. Discovery and engineering of colchicine alkaloid biosynthesis. Nature 2020; 584: 148–53.

[92] Nett RS, Sattely ES. Total biosynthesis of the tubulin-binding alkaloid colchicine. J Am Chem Soc 2021; 143: 19454–65.

[93] Hodgson H, De La Pena R, Stephenson MJet al. Identification of key enzymes responsible for protolimonoid biosynthesis in plants: opening the door to azadirachtin production. Proc Natl Acad Sci USA 2019; 116: 17096–104.

[94] Meihls LN, Handrick V, Glauser Get al. Natural variation in maize aphid resistance is associated with 2,4-dihydroxy-7-methoxy-1,4-benzoxazin-3-one glucoside methyltransferase activity. Plant Cell 2013; 25: 2341–55.

[95] Handrick V, Robert CA, Ahern KRet al. Biosynthesis of 8-O-methylated benzoxazinoid defense compounds in maize. Plant Cell 2016; 28: 1682–700.

[96] Zhou S, Kremling KA, Bandillo Net al. Metabolome-scale genome-wide association studies reveal chemical diversity and genetic control of maize specialized metabolites. Plant Cell 2019; 31: 937–55.

[97] Ding Y, Weckwerth PR, Poretsky Eet al. Genetic elucidation of interconnected antibiotic pathways mediating maize innate immunity. Nat Plants 2020; 6: 1375–88.

[98] Förster C, Handrick V, Ding Yet al. Biosynthesis and antifungal activity of fungus-induced O-methylated flavonoids in maize. Plant Physiol 2022; 188: 167–90.

[99] Chen W, Gao Y, Xie Wet al. Genome-wide association analyses provide genetic and biochemical insights into natural variation in rice metabolism. Nat Genet 2014; 46: 714–21.

[100] Wen W, Li D, Li Xet al. Metabolome-based genome-wide association study of maize kernel leads to novel biochemical insights. Nat Commun 2014; 5: 3438.

[101] Zhou S, Zhang YK, Kremling KAet al. Ethylene signaling regulates natural variation in the abundance of antifungal acetylated diferuloylsucroses and Fusarium graminearum resistance in maize seedling roots. New Phytol 2019; 221: 2096–111.

[102] Zhu G, Wang S, Huang Zet al. Rewiring of the fruit metabolome in tomato breeding. Cell 2018; 172: 249–61.e12.

[103] Zhou S, Jander G. Engineering insect resistance using plant specialized metabolites. Curr Opin Biotechnol 2021; 70: 115–21.

[104] Frey M, Chomet P, Glawischnig Eet al. Analysis of a chemical plant defense mechanism in grasses. Science 1997; 277: 696–9.

[105] Qi X, Bakht S, Leggett Met al. A gene cluster for secondary metabolism in oat: implications for the evolution of metabolic diversity in plants. Proc Natl Acad Sci USA 2004; 101: 8233–8.

[106] Qi X, Bakht S, Qin Bet al. A different function for a member of an ancient and highly conserved cytochrome P450 family: From essential sterols to plant defense. Proc Natl Acad Sci USA 2006; 103: 18848–53.

[107] Polturak G, Osbourn A. The emerging role of biosynthetic gene clusters in plant defense and plant interactions. PLoS Pathog 2021; 17: e1009698.

[108] Field B, Fiston-Lavier AS, Kemen Aet al. Formation of plant metabolic gene clusters within dynamic chromosomal regions. Proc Natl Acad Sci USA 2011; 108: 16116–21.

[109] Liu Z, Suarez Duran HG, Harnvanichvech Yet al. Drivers of metabolic diversification: how dynamic genomic neighbourhoods generate new biosynthetic pathways in the Brassicaceae. New Phytol 2020; 227: 1109–23.

[110] Zhou Y, Ma Y, Zeng Jet al. Convergence and divergence of bitterness biosynthesis and regulation in Cucurbitaceae. Nat Plants 2016; 2: 16183.

[111] Jozwiak A, Sonawane PD, Panda Set al. Plant terpenoid metabolism co-opts a component of the cell wall biosynthesis machinery. Nat Chem Biol 2020; 16: 740–8.

[112] Krokida A, Delis C, Geisler Ket al. A metabolic gene cluster in Lotus japonicus discloses novel enzyme functions and products in triterpene biosynthesis. New Phytol 2013; 200: 675–90.

[113] Shimura K, Okada A, Okada Ket al. Identification of a biosynthetic gene cluster in rice for momilactones. J Biol Chem 2007; 282: 34013–8.

[114] Mao L, Kawaide H, Higuchi Tet al. Genomic evidence for convergent evolution of gene clusters for momilactone biosynthesis in land plants. Proc Natl Acad Sci USA 2020; 117: 12472–80.

[115] Swaminathan S, Morrone D, Wang Qet al. CYP76M7 is an ent-cassadiene C11 α-hydroxylase defining a second multifunctional diterpenoid biosynthetic gene cluster in rice. Plant Cell 2009; 21: 3315–25.

[116] King AJ, Brown GD, Gilday ADet al. Production of bioactive diterpenoids in the euphorbiaceae depends on evolutionarily conserved gene clusters. Plant Cell 2014; 26: 3286–98.

[117] Zhan C, Lei L, Liu Zet al. Selection of a subspecies-specific diterpene gene cluster implicated in rice disease resistance. Nat Plants 2020; 6: 1447–54.

[118] Liu Z, Cheema J, Vigouroux Met al. Formation and diversification of a paradigm biosynthetic gene cluster in plants. Nat Commun 2020; 11: 5354.

[119] Matsuba Y, Zi J, Jones ADet al. Biosynthesis of the diterpenoid lycosantalonol via nerylneryl diphosphate in Solanum lycopersicum. PLoS One 2015; 10: e0119302.

[120] Liang J, An T, Zhu JXet al. Mining of the Catharanthus roseus genome leads to identification of a biosynthetic gene cluster for fungicidal sesquiterpenes. J Nat Prod 2021; 84: 2709–16.

[121] Itkin M, Heinig U, Tzfadia Oet al. Biosynthesis of antinutritional alkaloids in solanaceous crops is mediated by clustered genes. Science 2013; 341: 175–9.

[122] Winzer T, Gazda V, He Zet al. A Papaver somniferum 10-gene cluster for synthesis of the anticancer alkaloid noscapine. Science 2012; 336: 1704–8.

[123] Chen X, Hagel JM, Chang Let al. A pathogenesis-related 10 protein catalyzes the final step in thebaine biosynthesis. Nat Chem Biol 2018; 14: 738–43.

[124] Shen SQ, Peng M, Fang Het al. An Oryza-specific hydroxycinnamoyl tyramine gene cluster contributes to enhanced disease resistance. Sci Bull 2021; 66: 2369–80.

[125] Fang H, Shen SQ, Wang Det al. A monocot-specific hydroxycinnamoylputrescine gene cluster contributes to immunity and cell death in rice. Sci Bull 2021; 66: 2381–93.

[126] Takos AM, Knudsen C, Lai Det al. Genomic clustering of cyanogenic glucoside biosynthetic genes aids their identification in Lotus japonicus and suggests the repeated evolution of this chemical defence pathway. Plant J 2011; 68: 273–86.

[127] Knoch E, Motawie MS, Olsen CEet al. Biosynthesis of the leucine derived α-, β- and γ-hydroxynitrile glucosides in barley (Hordeum vulgare L.). Plant J 2016; 88: 247–56.

[128] Jeon JE, Kim JG, Fischer CRet al. A pathogen-responsive gene cluster for highly modified fatty acids in tomato. Cell 2020; 180: 176–87.e19.

[129] Hen-Avivi S, Savin O, Racovita RCet al. A metabolic gene cluster in the wheat W1 and the barley Cer-cqu loci determines beta-diketone biosynthesis and glaucousness. Plant Cell 2016; 28: 1440–60.

[130] Fan P, Wang P, Lou YRet al. Evolution of a plant gene cluster in Solanaceae and emergence of metabolic diversity. Elife 2020; 9: e56717.

[131] Nützmann HW, Doerr D, Ramírez-Colmenero Aet al. Active and repressed biosynthetic gene clusters have spatially distinct chromosome states. Proc Natl Acad Sci USA 2020; 117: 13800–9.

[132] Yu N, Nützmann HW, MacDonald JTet al. Delineation of metabolic gene clusters in plant genomes by chromatin signatures. Nucleic Acids Res 2016; 44: 2255–65.

[133] Cárdenas PD, Sonawane PD, Pollier Jet al. GAME9 regulates the biosynthesis of steroidal alkaloids and upstream isoprenoids in the plant mevalonate pathway. Nat Commun 2016; 7: 10654.

[134] Kellner F, Kim J, Clavijo BJet al. Genome‐guided investigation of plant natural product biosynthesis. Plant J 2015; 82: 680–92.

[135] Darbani B, Motawia MS, Olsen CEet al. The biosynthetic gene cluster for the cyanogenic glucoside dhurrin in Sorghum bicolor contains its co-expressed vacuolar MATE transporter. Sci Rep 2016; 6: 37079.

[136] Kautsar SA, Suarez Duran HG, Blin Ket al. plantiSMASH: automated identification, annotation and expression analysis of plant biosynthetic gene clusters. Nucleic Acids Res 2017; 45: W55–63.

[137] Polturak G, Dippe M, Stephenson MJet al. Pathogen-induced biosynthetic pathways encode defense-related molecules in bread wheat. Proc Natl Acad Sci USA 2022; 119: e2123299119.

[138] Chae L, Kim T, Nilo-Poyanco Ret al. Genomic signatures of specialized metabolism in plants. Science 2014; 344: 510–3.

[139] Nützmann HW, Huang A, Osbourn A. Plant metabolic clusters–from genetics to genomics. New Phytol 2016; 211: 771–89.

[140] Kitaoka N, Zhang J, Oyagbenro RKet al. Interdependent evolution of biosynthetic gene clusters for momilactone production in rice. Plant Cell 2021; 33: 290–305.

[141] Li R, Zhang J, Li Zet al. Dissecting the labdane-related diterpenoid biosynthetic gene clusters in rice reveals directional cross-cluster phytotoxicity. New Phytol 2022; 233: 878–89.

[142] Li B, Förster C, Robert CAMet al. Convergent evolution of a metabolic switch between aphid and caterpillar resistance in cereals. Sci Adv 2018; 4: eaat6797.

[143] Wu D, Jiang B, Ye C-Yet al. Horizontal transfer and evolution of the biosynthetic gene cluster for benzoxazinoids in plants. Plant Commun 2022; 3: 100320.

[144] Ajikumar PK, Xiao WH, Tyo KEet al. Isoprenoid pathway optimization for Taxol precursor overproduction in Escherichia coli. Science 2010; 330: 70–4.

[145] Galanie S, Thodey K, Trenchard IJet al. Complete biosynthesis of opioids in yeast. Science 2015; 349: 1095–100.

[146] Luo X, Reiter MA, d’Espaux Let al. Complete biosynthesis of cannabinoids and their unnatural analogues in yeast. Nature 2019; 567: 123–6.

[147] Paddon CJ, Westfall PJ, Pitera DJet al. High-level semi-synthetic production of the potent antimalarial artemisinin. Nature 2013; 496: 528–32.

[148] Westfall PJ, Pitera DJ, Lenihan JRet al. Production of amorphadiene in yeast, and its conversion to dihydroartemisinic acid, precursor to the antimalarial agent artemisinin. Proc Natl Acad Sci USA 2012; 109: E111–8.

[149] Lu H, Villada JC, Lee PK. Modular metabolic engineering for biobased chemical production. Trends Biotechnol 2019; 37: 152–66.

[150] Li C, Zhang R, Wang Jet al. Protein engineering for improving and diversifying natural product biosynthesis. Trends Biotechnol 2020; 38: 729–44.

[151] Mak WS, Tran S, Marcheschi Ret al. Integrative genomic mining for enzyme function to enable engineering of a non-natural biosynthetic pathway. Nat Commun 2015; 6: 10005.

[152] Sinha R, Shukla P. Current trends in protein engineering: updates and progress. Curr Protein Pept Sci 2019; 20: 398–407.

[153] Zhou P, Li M, Shen Bet al. Directed coevolution of β-carotene ketolase and hydroxylase and its application in temperature-regulated biosynthesis of astaxanthin. J Agric Food Chem 2019; 67: 1072–80.

[154] Lv X, Gu J, Wang Fet al. Combinatorial pathway optimization in Escherichia coli by directed co‐evolution of rate‐limiting enzymes and modular pathway engineering. Biotechnol Bioeng 2016; 113: 2661–9.

[155] Chen H, Liu C, Li Met al. Directed evolution of mevalonate kinase in Escherichia coli by random mutagenesis for improved lycopene. RSC Adv 2018; 8: 15021–8.

[156] Jakočiūnas T, Pedersen LE, Lis AVet al. CasPER, a method for directed evolution in genomic contexts using mutagenesis and CRISPR/Cas9. Metab Eng 2018; 48: 288–96.

[157] Packer MS, Liu DR. Methods for the directed evolution of proteins. Nat Rev Genet 2015; 16: 379–94.

[158] Yang KK, Wu Z, Arnold FH. Machine-learning-guided directed evolution for protein engineering. Nat Methods 2019; 16: 687–94.

[159] Wu Z, Kan SJ, Lewis RDet al. Machine learning-assisted directed protein evolution with combinatorial libraries. Proc Natl Acad Sci USA 2019; 116: 8852–8.

[160] Zhuang Y, Yang GY, Chen Xet al. Biosynthesis of plant-derived ginsenoside Rh2 in yeast via repurposing a key promiscuous microbial enzyme. Metab Eng 2017; 42: 25–32.

[161] Li D, Ma Y, Zhou Yet al. A structural and data-driven approach to engineering a plant cytochrome P450 enzyme. Sci China Life Sci 2019; 62: 873–82.

[162] Jumper J, Evans R, Pritzel Aet al. Highly accurate protein structure prediction with AlphaFold. Nature 2021; 596: 583–9.

[163] Baek M, DiMaio F, Anishchenko Iet al. Accurate prediction of protein structures and interactions using a three-track neural network. Science 2021; 373: 871–6.

[164] Ignea C, Raadam MH, Motawia MSet al. Orthogonal monoterpenoid biosynthesis in yeast constructed on an isomeric substrate. Nat Commun 2019; 10: 3799.

[165] Liu CL, Bi HR, Bai Zet al. Engineering and manipulation of a mevalonate pathway in Escherichia coli for isoprene production. Appl Microbiol Biotechnol 2019; 103: 239–50.

[166] Zada B, Wang C, Park JBet al. Metabolic engineering of Escherichia coli for production of mixed isoprenoid alcohols and their derivatives. Biotechnol Biofuels 2018; 11: 210.

[167] Yoshida R, Yoshimura T, Hemmi H. Reconstruction of the “archaeal” mevalonate pathway from the methanogenic archaeon Methanosarcina mazei in Escherichia coli cells. Appl Environ Microbiol 2020; 86: e02889–19.

[168] Chatzivasileiou AO, Ward V, Edgar SMet al. Two-step pathway for isoprenoid synthesis. Proc Natl Acad Sci USA 2019; 116: 506–11.

[169] Clomburg JM, Qian S, Tan Zet al. The isoprenoid alcohol pathway, a synthetic route for isoprenoid biosynthesis. Proc Natl Acad Sci USA 2019; 116: 12810–5.

[170] Lund S, Hall R, Williams GJ. An artificial pathway for isoprenoid biosynthesis decoupled from native hemiterpene metabolism. ACS Synth Biol 2019; 8: 232–8.

[171] Rico J, Duquesne K, Petit JLet al. Exploring natural biodiversity to expand access to microbial terpene synthesis. Microb Cell Fact 2019; 18: 23.

[172] Dusséaux S, Wajn WT, Liu Yet al. Transforming yeast peroxisomes into microfactories for the efficient production of high-value isoprenoids. Proc Natl Acad Sci USA 2020; 117: 31789–99.

[173] Yee DA, DeNicola AB, Billingsley JMet al. Engineered mitochondrial production of monoterpenes in Saccharomyces cerevisiae. Metab Eng 2019; 55: 76–84.

[174] Liu GS, Li T, Zhou Wet al. The yeast peroxisome: a dynamic storage depot and subcellular factory for squalene overproduction. Metab Eng 2020; 57: 151–61.

[175] Zhang C, Li M, Zhao GRet al. Harnessing yeast peroxisomes and cytosol acetyl-CoA for sesquiterpene α-humulene production. J Agric Food Chem 2020; 68: 1382–9.

[176] Arendt P, Miettinen K, Pollier Jet al. An endoplasmic reticulum-engineered yeast platform for overproduction of triterpenoids. Metab Eng 2017; 40: 165–75.

[177] Kim JE, Jang IS, Son SHet al. Tailoring the Saccharomyces cerevisiae endoplasmic reticulum for functional assembly of terpene synthesis pathway. Metab Eng 2019; 56: 50–9.

[178] Grewal PS, Samson JA, Baker JJet al. Peroxisome compartmentalization of a toxic enzyme improves alkaloid production. Nat Chem Biol 2021; 17: 96–103.

[179] Thodey K, Galanie S, Smolke CD. A microbial biomanufacturing platform for natural and semisynthetic opioids. Nat Chem Biol 2014; 10: 837–44.

[180] Zhou YJ, Buijs NA, Zhu Zet al. Harnessing yeast peroxisomes for biosynthesis of fatty-acid-derived biofuels and chemicals with relieved side-pathway competition. J Am Chem Soc 2016; 138: 15368–77.

[181] Sheng J, Stevens J, Feng X. Pathway compartmentalization in peroxisome of Saccharomyces cerevisiae to produce versatile medium chain fatty alcohols. Sci Rep 2016; 6: 26884.

[182] Srinivasan P, Smolke CD. Biosynthesis of medicinal tropane alkaloids in yeast. Nature 2020; 585: 614–9.

[183] Ma Y, Liu N, Greisen Pet al. Removal of lycopene substrate inhibition enables high carotenoid productivity in Yarrowia lipolytica. Nat Commun 2022; 13: 572.

[184] Li D, Halitschke R, Baldwin ITet al. Information theory tests critical predictions of plant defense theory for specialized metabolism. Sci Adv, 6: eaaz0381.

RIGHTS & PERMISSIONS

The Author(s) 2022. Published by Oxford University Press on behalf of Higher Education Press.