Springer Nature is making Coronavirus research free. View research | View latest news | Sign up for updates

Recent advances in the biochemistry of spinosyns

Abstract

Spinosyn and its analogs, produced by Saccharopolyspora spinosa, are the active ingredients in a family of insect control agents. They are macrolides with a 21-carbon, 12-membered tetracyclic lactones that are attached to two deoxysugars, tri-O-methylrhamnose and forosamine. Labeling studies, analysis of the biosynthetically blocked mutants, and the genetic identification of the spinosyn gene cluster have provided detailed information concerning the mechanism of spinosyn biosynthesis and have enabled combinatorial biosynthesis of a large group of new spinosyns. The following developments have recently impacted the field of spinosyn biology: (1) A second-generation spinosyn called spinetoram (XDE-175) was launched in late 2007; it is a semisynthesized spinosyn derivative produced through the modification of 3′-O-methyl group of rhamnose and the double bond between C5 and C6 of spinosyn J and L. This molecule was shown to have improved insecticidal activity, enhanced duration of control, and an expanded pest spectrum. (2) A new class of spinosyns, the butenyl-spinosyns, was discovered from Saccharopolyspora pogona. The butenyl-spinosyns are similar to spinosyns, but differ in the length of the side chain at C-21. In addition to structural similarities with the spinosyns, the butenyl-spinosyns exhibit a high level of similarity in insecticidal activity to spinetoram. (3) Spinosyn analogs, 21-cyclobutyl-spinosyn A and 21-cyclobutyl-spinosyn D were generated by metabolic engineering of the spinosyn biosynthetic gene cluster. They showed better insecticidal activities against cotton aphid and tobacco budworm than that of spinosyn A and D. Future progress toward the development of more potent spinosad analogs, as well as enhancements in production yields will likely result from these recent advances in the genetics and biochemistry of spinosyns.

Introduction

Spinosyns (Fig. 1) are produced by Saccharopolyspora spinosa originally isolated from soil sample collected in the Caribbean island in 1982 (Mertz and Yao 1990). They were discovered by screening the fermentation broths of S. spinosa for mosquito larvicidal activity (Thompson et al. 1997). Such activity was not found within the class of traditional macrolide insecticides, such as avermectins (MacNeil 1995) and milbemycins (Davies and Green 1986). From fermentation broth extracts of S. spinosa, a series of spinosyn factors were purified and structurally characterized (originally called A83543, and later named as spinosyns; Kirst et al. 1992). Spinosyns are unique macrocyclic lactones, containing a tetracyclic core consisting of a 12-membred macrocyclic lactone that is fused to a 5,6,5-cis-anti-trans tricyclic ring system. Attached to the tetracyclic core are two sugars, the amino sugar forosamine and a neutral sugar, 2,3,4-tri-O-methylated rhamnose (Fig. 1). So far, more than 25 spinosyns have been isolated and identified from S. spinosa that vary in methyl substitution patterns on the forosamine nitrogen, the 2′-, 3′-, 4′-methyl positions of the rhamnose, and at the C6, C16, and C21 positions of the tetracycle (Crouse et al. 2001). Additionally, several hundred spinosoids (synthetic or semisynthetic spinosyn analogs) have been prepared in the laboratory during the past several years (Mergott et al. 2004; Salgado and Sparks 2005). The most abundant spinosyns from S. spinosa fermentation broth are spinosyn A (≈85% of spinosad) and spinosyn D (≈15% of spinosad). They are the main active ingredient in a number of Naturalyte® products called spinosad (Tracer®, Success®, SpinTor®, Conserve®, Entrust®) that is marketed by Dow AgroSciences for insect control, and in the animal health product, Elector® that is marketed by Elanco. Currently, the human health application product, Natrova™, is under development by ParaPRO LLC and is expected to receive registration approval by the U.S. Food and Drug Administration in 2009.

Fig. 1
figure1

The structures of spinosyn and its analogs

Spinosad kills susceptible insects by causing rapid excitation of the insect nervous system, probably through the interaction and binding at the nicotinacetylcholine (NACh) and δ-amino butyric acid (GABA) receptors (Millar and Denholm 2007). This new class of insecticides showed a high level of selectivity by effectively eliminating several crop pests such as tobacco budworm (Heliothis virescens) and southern armyworm (Spodoptera eridania), but they showed little or no effect on a broad range of nontarget insects and mammals (Kirst et al. 1992). This feature, coupled with its excellent environment profile, has driven significant growth in the use of spinosyn-based insecticides for the management of insect pests in agriculture and has been a driving force for completion of a number of studies that have evaluated new applications for this class of insecticides.

The biosynthetic pathway of spinosyns has been elucidated by several studies that have focused on precursor-labeling, identification of intermediates produced by block mutants, conversion of intermediates by blocked mutants, and in vitro analysis of enzymes involved in the spinosyn biosynthesis (Kirst et al. 1993). Early genetic studies and the nucleotide sequence of the spinosyn biosynthetic gene cluster have been reviewed (Waldron et al. 2000); however, many details concerning the order for coupling deoxysugars to the macrocycle, the regulation of spinosyn synthesis, and the mechanism of macrocycle cross-linkage were not elucidated in these early studies. In this review, we describe the discovery of novel spinosyn derivatives, current research status of the biosynthetic pathway of spinosyns, metabolic engineering of the pathway, the regulation of biosynthesis, and approaches to improve spinosyn productivities from fermentation processes.

New spinosyn analogs

Spinetoram (also called XDE-175), which was launched at late 2007, was discovered by applying neural network-based quantitative structure–activity relationship (QSAR) to identify new synthetic directions for spinosyn chemistry (Sparks et al. 2008). When a mixture of spinosyn J and L (Fig. 1) was modified to the 5,6-dihydro and 3′-O-ethyl analog of spinosyn, a new semisynthetic insecticide given the generic name, spinetoram, was produced, which consisted of the 3′-O-ethyl-5,6-dihydro-spinosyn J (major component) and 3′-O-ethyl spinosyn L (minor component) (Fig. 1). Spinetoram has a more complicated impurity profile due to the alkylation and hydroxylation of six to eight low-level (<1% w/w) spinosyns that are present in the final technical. Overall, spinetoram is more active than spinosad and has expanded spectrum while maintaining the exceptional environment and toxicological profile that is associated with the first-generation spinosad products.

The S. spinosa culture used for the production of the spinetoram precursor, spinosyns J and L, was derived through random mutagenesis of the spinosad production strain using the chemical mutagen, N-methyl-N′-nitro-N-nitrosoguanidine. The mutant phenotype was extremely rare among isolates in the mutant population and was found to occur at a frequency of approximately 1 in over 20,000 mutants screened. Sequence analysis of the deoxysugar and PKS genes in the mutant strains showed the presence of a point mutation in a highly conserved region of the 3′-O-methyltransferase that is encoded by the spnK gene (Huang et al., unpublished results). As a semisynthetic natural product, spinetoram cannot be included in the Naturalye® class of insecticides, and therefore, is amenable to a greater diversity of strain development approaches, including genetic engineering as a means to advance production yields.

Butenyl-spinosyns (also called pogonins) (Fig. 1) were discovered in 1990 (Lewer et al. 2002) from the soil sample collected in Indiana (USA) by using a number of target insect pests such as beet armyworm (Spodoptera exigua) in a 96-well plate format as a primary natural product extracts screening tools. From the screening effort, an extract was not only active in the primary screening, but also exhibited interesting contact activity coupled with symptomology that was very similar with the spinosyns. Further isolation, separation, and characterization by nuclear magnetic resonance (NMR) spectroscopy and liquid chromatography mass spectroscopy (LC-MS) showed that the activity was due to a spinosyn analog that was named as butenyl-spinosyn. The distinctive feature was the presence of a 2-butenyl group at the C-21 position of the macrocyclic ring system in place of the typical ethyl group in spinosyns. To date, more than 30 factors have been isolated and characterized (Lewer et al. 2002) with the primary factors 21-butenyl-spinosyn A and D. The organism that produces butenyl-spinosyns was identified as Saccharopolyspora pogona (Hahn et al. 2006), which had a number of significant differences from S. spinosa, including a hairy rather than spiny spore coat, bacteriophage sensitivity, and a different 16 sRNA secondary structure. The 21-butenyl-spinosyn A and D compounds show a high degree of similarity to the insecticidal activity of spinetoram.

The insecticidal compounds, 21-cyclobutyl-spinosyn A and D (Fig. 1), were produced by the genetically engineered strain, S. spinosa BIOT-1066 (Sheehan et al. 2006). The spinosyn polyketide synthase (PKS) loading domain in this strain was replaced with the loading module of the avermectin PKS from S. avermitilis. When S. spinosa BIOT-1066 was grown in the standard spinosyn production medium, spinosyn A and spinosyn E were identified as the major compounds along with a minor amount of spinosyn D. However, when exogenous carboxylic acid was fed to the fermentation culture, several new compounds were produced. Among these compounds was 21-cyclobutyl-spinosyn A and D, which showed improved insecticidal activity against the cotton aphid and tobacco budworm when compared to spinosyn A and D.

Biosynthetic pathway of spinosyns

Based on the feeding studies using 14C-labeled acetate, propionate, butyrate, isobutyrate, and methionine, it has been established that spinosyns are assembled by a polyketide pathway from acetate and propionate, which ultimately leads to the introduction of three intramolecular C–C bonds to form spinosyn tetracycle (Fig. 2). To the tetracycle, a neutral sugar (rhamnose) and an amino sugar (forosamine) are coupled at C9 and C17, respectively (Fig. 2). Rhamnose is subsequently methylated by three O-methyltransferases and forosamine is methylated at the amine position prior to being attached to the spinosyn tetracycle (Fig. 2). Labeling studies with [14C]methionine showed that the three O-methyl groups of tri-O-methylrhamnose and the two N-methyl groups of forosamine are derived from S-adenosyl-methionine.

Fig. 2
figure2

Predicted modular organization of spinosyn polyketide synthase and biosynthetic pathway of spinosyn

Recently, the entire spinosyn biosynthetic gene cluster was determined through gene sequencing and functional analysis of the gene products in S. spinosa (Waldron et al. 2001). The sequencing analysis revealed the organization of the spinosyn biosynthetic genes that spanned a distance of 80 kb and provided insight into their deduced functions, as predicted by gene homology (Fig. 3). Five large genes (spnA, B, C, D, E) encoded a type I polyketide synthase. These genes spanned a 56-kb region and included a loading module and ten extender modules. Other genes in this gene cluster included four genes involved in intramolecular C–C bond formation (spnF, J, L, M), four genes involved in rhamnose attachment and methylation (spnG, I, K, H), and six genes involved in forosamine biosynthesis (spnP, O, N, Q, R, S). The genes for synthesis of rhamnose (gtt, gdh, epi, and kre) were not closely linked to the spinosyn gene cluster, apparently because of their bimodal functions in primary and secondary metabolism (Madduri et al. 2001a, b). No key regulatory genes were found to be linked to the spinosyn gene cluster; this lack of regulatory genes was also observed in the erythromycin gene cluster (Mironov et al. 2004). Overall, the spinosyn biosynthetic pathway can be viewed as: (1) formation of aglycone from acetate (nine) and propionates (two) by polyketide synthase and bridging enzymes, (2) formation of pseudoaglycone by glycosylation of aglycone using deoxythymidine diphosphate (dTDP)-rhamnose, and then methylation of the rhamnose, and (3) production of spinosyn by glycosylation of pseudoaglycone using dTDP-forosamine (Waldron et al. 2000, 2001).

Fig. 3
figure3

Comparison of spinosyn pathway genes (top) and butenyl-spinosyn pathway genes (bottom). The additional PKS module for the two carbons in the C-21 tail of butenyl-spinosyn was indicated as KS AT DH KR ACP

Most of the present data for the functional assignment of genes in the spinosad biosynthetic cluster have been based on sequence homology to other well-defined PKS systems including erythromycin and tylosin; thus, only a small number of genes from the spinosad gene cluster have functions that have been confirmed by functional analysis. Due to the transcriptional effect on downstream genes, single cross-over knockout experiments could not make conclusive assignments for the gene products. Double cross-over knockouts, using the appropriate resistance cassettes without transcriptional terminator or by making in-frame deletions could make conclusive assignments for these genes; however, this method has not been successfully applied to S. spinosa. Recently, Huang et al. (2008) combined protein expression and sequencing of the spinosyn K mutant to prove that spnH encodes for a 4′-rhamnose-methyltransferase. This approach was also applied to characterize the function of proteins encoded by spnI and spnK (Huang et al., unpublished results), respectively. This study also demonstrated the successful expression of spnI, spnK, and spnH genes individually in the Saccharopolyspora erythraea host strain SGT2 (Gaisser et al. 2001), which is blocked both in endogenous erythromycin biosynthesis and glycosyltransferases eryBV and EryIII. Results of this study showed that the three heterologous O-methyltransferases from S. spinosa were capable of methylating the alternative substrate, 3-O-rhamnosyl-erythronolide, or 3-O-rhamnosyl-erythomycin D when fed the unmethylated substrate.

The pathway for forosamine biosynthesis in S. spinosa has been the focus of several recent investigations by Liu and coworkers (Zhao et al. 2005, Hong et al. 2006, 2008). These studies successfully elucidated the function of several heterologously expressed proteins encoded by the genes, spnO, spnN, spnQ, spnR, and spnS. All proteins in this pathway were heterologously expressed, purified to homogeneity, and their activities were examined in vitro. The results showed that spnQ functioned as a pyridoxamine 5′-monophosphate (PMP)-dependent 3-dehydratase that required a cellular reductant such as ferredoxin reductase or flavodoxin reductase to catalyze C-3 deoxygenation of the chemically synthesized TDP-4-keto-2,6-dideoxy-d-glucose. The gene products, spnO and spnN, functioned as a 2,3-dehydratase and a 3-ketoreductase, respectively, and spnR functioned as a transaminase that converted the spnQ product, TDP-4-2,3,6-trideoxy-d-glucose to TDP-4-amino-2,3,4,6-tetradeoxy-d-glucose. Finally, spnS was found to function as a 4-dimethyltransferase that converted the spnR product to TDP-d-forosamine (Fig. 4). These investigators also explored the substrate specificities of spnQ and spnR and proposed the possibility for modification of forosamine through metabolic engineering to produce new spinosyn analogs.

Fig. 4
figure4

Proposed pathway for the biosynthesis of forosamine. The first two steps in the pathway are common to rhamnose and forosamine pathways, suggested to be catalyzed by enzymes Gtt (NDP-glucose synthase) and Gdh (NDP-glucose dehydratase), while other steps are catalyzed by forosamine pathway enzymes (spnO, N, Q, R, S)

Polyketides with cross-links in the macrocyclic ring are rare in nature: Examples of natural products with cross-linked macrocycles include the fungicide maltophilin, which has a 6,5,5 cyclic substructure (Jakobi et al. 1996); the antibacterial ikarugamycin, which has a 5,6,5 structure (Ito and Hirata 1972); and the anticancer marine polyketide discodermide, which has 6,5,5 structure (Shaw 2008). Interestingly, these natural products contain an ornithine residue that is not present in the spinosyns. Based on the organization of the functional domains in the spinosyn PKS, the spinosyn intermediate “spilactone” (Fig. 2) is predicted to be an unbridged monocyclic lactone, which has not yet been isolated. This intermediate is then bridged to produce the spinosyn aglycone, which is generated by the thioesterase domain in module 10 of the spinosyn polyketide synthase. Gene homology and disruption results indicate that four genes (spnJ, spnF, spnM, and spnL) may be involved in the aglycone synthesis; however, no biochemical data is available to support this conclusion.

Insight into the biochemical mechanism for cross-linkage of the macrocycle was recently generated by fusing the spinosyn PKS genes, spnA, spnB, and spnC, to the erythromycin thioesterase domain and then the fused gene product was expressed in S. erythraea. Fermentations of the recombinant S. erythraea strain contained a new product, pentaketide lactone (Table 1) that was purified and structurally characterized by two-dimensional NMR and LC-MS. The data for this study indicated that the KR domain in module 4 was very likely an “A-type” KR domain and further suggested that the cryptic step of C15 oxidation must occur during elongation of the ketide during synthesis by the spinosyn PKS. Martin et al. (2003) proposed that following the release of the spinosyn intermediate from the PKS, the C15 position is oxidized to a keto group by the protein encoded by spnJ to enable formation of the final cross-link in the lactone ring. This hypothesis was further explored by Kim et al. (2007) by expressing and purifying spnJ from Escherichia coli and then characterizing the flavin-dependent oxidase activity of this enzyme. This enzyme, however, was found to convert the macrolactone to the corresponding ketone, not the linear mature polyketide precursor. This result clearly indicated that the macrolactone formation is preceded by 15-OH oxidation since the linear polyketide is not a substrate for spnJ. Future data generated using this approach will be important in characterizing the function of the remaining spinosyn bridge enzymes.

Table 1 Selected new spinosyn analogs generated by combinational biosynthesis of the spinosyn pathway

The butenyl-spinosyn biosynthetic pathway genes were cloned recently from S. pogona using spinosyn pathway genes as probes (Hahn et al. 2006). The coding sequences of the butenyl-spinosyn biosynthetic genes showed 94% identity to the DNA sequence of the spinosyn biosynthetic genes (Fig. 3). The only notable difference in the sequence comparisons was the presence of one additional PKS module in the butenyl-spinosyn gene sequence (Fig. 3) that was responsible for the two additional carbons in the C-21 tail of the spinosyn (KS, AT, DH, KR, and ACP domain). This suggested that the spinosyn biosynthetic cluster might have been derived as an in-frame deletion from butenyl-spinosyn biosynthetic cluster. The gene order and orientation of forosamine pathway genes, the cross-bridging genes, the glycosyltransferases, and O-methyltransferases all exhibited a high level of homology with the spinosyn biosynthetic pathway genes (Fig. 3). As with the spinosyn gene cluster, neither the regulatory genes nor genes for biosynthesis of rhamnose were directly linked to the butenyl-spinosyn PKS gene cluster. However, the genes flanking the butenyl-spinosyn gene cluster were found to be completely divergent from the spinosyn gene cluster (Fig. 3). In addition to producing butenyl-spinosyns, which have a 12-membered lactone ring, a minor 14-membered lactone, tridecenolactone spinosyn A was also isolated from fermentation broths of S. pogona (Hahn et al. 2006). While the detailed mechanism for synthesis of the 14-membered lactone has remained unclear, this capability of the butenyl-spinosyn PKS to produce a greater variety of products may underline the distinct differences in substrate specificity and flexibility for this enzyme.

Spinosyn pathway engineering

In recent years, the combinatorial biosynthesis technique has been applied to produce new molecules that are otherwise difficult to obtain by traditional chemical synthesis (Baltz 2006). A variety of tools and techniques to genetically manipulate S. spinosa have been developed and tested by adapting them from other actinomycetes, especially from Streptomyces coelicolor and S. erythreae. Protoplast and conjugation transformation systems for S. spinosa were developed in the early 1990s and were successfully applied to the genetic engineering of S. spinosa, even though this strain was reported to be highly recalcitrant to transformation efforts (Matsushima et al. 1994, Matsushima and Baltz 1994). The first example of new spinosyn analogs produced through the biotransformation of S. erythraea was the replacement of the forosamine moiety at C17 of spinosyn by a l-mycarose sugar and d-glucose (Gaisser et al. 2001; Table 1). The spinosyn pseudoaglycone was fed to a fermentation containing a genetically engineered S. erythraea strain in which the erythromycin polyketide synthase and both glycosyltransferase genes were deleted, while the spinosyn pathway gene, spnP, was heterologously expressed in S. erythreae.

World patent, WO 2005044979, further demonstrated that rhamnose and forosamine could be replaced by other deoxysugars using heterologous glycosylation systems generated by the expression of sugar (6-deoxysugar, 2,6-deoxysugar, 2,3,6-deoxysugar) pathway genes in the heterologous host, S. diversa together with spnG or spnP, then spinosyn aglycone and 17-pseudoaglycone were fed to S. diversa. A few new spinosyn analogs (Table 1) were isolated from fermentations of S. diversa constructs containing sugar synthetic pathway genes and spnG or spnP that were fed the aglycone or the 17-pseudoaglycone, but the insecticidal activities of these compounds were shown to be inferior to spinosyn A and D.

The spinosyn PKS loading module and the first four extension modules (spnA, spnB, and spnC) were fused into the erythromycin thioesterase and then expressed in S. erythreae by Martin et al. (2003). A novel pentaketide lactone was isolated from the fermentation broths (Table 1). This result indicated that the stereochemical type of KR domain in the spinosyn PKS module 4 was an “A-type” and that is what was observed in the isolated pentaketide lactone. Recently, Sheehan et al. (2006) engineered the loading module of the spinosyn PKS by replacing it with the avermectin and erythromycin loading modules and then feeding the recombinants with a variety of carboxylic acid starter analogs. This approach generated many new spinosyns, among them, 21-cyclobutyl-spinosyn A and D, which showed improved insecticidal activities and expanded spectrum for a diverse group of insects ranging from the lepidopterans to the sucking insects, such as aphids. This was the first successful example for the use of a combinatorial biosynthesis strategy to produce more potent insecticidal analogs of spinosyn A and D. During the last decade, a number of new spinosyns have been produced using the combinatorial biosynthesis strategy; however, the insecticidal activities of these compounds were shown to be inferior to unmodified natural products. For example, the expression of the Streptomyces cinnamonensis crotonyl CoA reductase in S. spinosa led to the production of the novel insecticidal products, 21-desethyl-21-n-propyl spinosyn A and D, 6-ethyl spinosyn A (Burns et al. 2003). When the first AT domain (methyl malonyl CoA specific) at module 4 (spnD) was replaced by rapamycin module 2 AT domain (malonyl CoA specific) in S. spinosa, 16-desmethyl spinosyn D was isolated (Burns et al. 2003). A future continuation of this strategy may be to replace an AT domain with a heterologous AT domain that is selective for a modified malonyl CoA that could generate an entirely new group of spinosyns.

Strain improvement and fermentation process development

Classic random mutagenesis is still an effective approach to improve spinosyn yield; however, it is often very challenging to screen the large number of mutants that are necessary to meet the statistical requirements for nonselected mutant populations (Parekh et al. 2000). While many reviews have been published on the design of screening systems employed in random mutagenesis, specific details concerning the improvement rate for this systems is lacking (Vinci and Byng 1999; Parekh et al. 2000). Rational strain improvement strategies overlap with classical random mutagenesis in the area of generating a mutant population; however, a secondary mutant attrition process based on an understanding of the metabolic restrictions in the spinosyn pathway could be applied to significantly reduce the number of mutants that are screened for yield improvements. Using this latter approach, Madduri et al. (2001a, b) determined that the concentration of forosamine was crucial for the conversion of pseudoaglycone to spinosyn A and D. When the forosamine pathway genes were duplicated, more than 90% of the pseudoaglycone was converted to spinosyn A and D. While the final spinosyn yield was increased significantly, the content of total macrolide present in the fermentation was not significantly changed. Jin et al. (2006a, b) screened a small number of the spinosyn A, rhamnose, and 2-deoxy-d-glucose (2-DOG) resistant mutants following UV mutagenesis and isolated a mutant that exhibited a 121% improvement in the yield of spinosyn A and D. After fermentation optimization, the yield reached 458 mg/L, which was 71% higher than the yield achieved by shake flask fermentations. Liang et al. (2008) isolated sodium propionate and rhamnose mutants and improved the spinosyn yield from 32.5 to 125.3 mg/L, which represented a 286% improvement. However, the final yields for these studies were still well below 2 g/L, which is considered the economic threshold for solvent extraction and recovery of spinosyns from fermentation broth. It will be interesting to investigate whether these approaches can be applied to the high-yield spinosyn-producing strain currently utilized by Dow AgroSciences for spinosad and spinetoram manufacturing. Consistent with the previous work of Ochi (2007), our unpublished results show that the selection of antibiotic-resistant mutants to polyether antibiotics can also be an effective strategy for generating significant improvements in the yield of spinosad from S. spinosa.

Fermentation process development has not been studied well since all reported yields for spinosyns are below 2 g/L. Media optimization is often cited as an initial step in the improvement of product yields from fermentation products. Strobel and Nakatsukasa (1993) used response surface methodology to optimize the media for spinosyn production. Four media components, glucose, cottonseed flour, peptonized milk nutrient, and corn steep liquor, were optimized and the results showed that the volumetric titer of spinosyns A and D could be improved by fivefold through simple changes in the ratio of these key nutrient sources. The model for the fitted data accounted for more than 80% of the variability in volumetric titer of the spinosyns and showed that the C/N ratio was important in maximizing spinosad titer. Glucose and phosphate concentration have been determined to be critical for the growth and production of spinosad by S. spinosa. A high concentration of glucose (>79.6 g/L) and phosphate (>44.12 mM) was observed to inhibit the mycelium growth and spinosyn production (Jin et al. 2006a, b), which indicated that the concentration of these nutrients should be limited during initial growth of the culture. Oil is a critical component in the fermentative production of macrolide antibiotics. This raw material functions as surfactant, nutrient, and metabolic precursor. Consistent with the result of Jin et al. (2006a, b), our studies have shown that the addition of vegetable oils and fatty acid esters to spinosad fermentations significantly improves the spinosad yield (Zahn et al., unpublished data). The beneficial effects of oils have also been observed for the erythromycin fermentation process (Brünker et al. 1999).

Prospects

The biosynthetic pathway and the genes involved in the pathway steps of spinosyn biosynthesis have been investigated through genetic and biochemical studies; however, information concerning the regulation of these genes still remains limited. The recent completion of the genome sequencing of S. erythreae, S. avermitillis, and S. colelicolor, in combination with gene expression (microarray, proteomics) and metabolite (metabolomics) data, may provide insight into the regulatory network for spinosyn biosynthesis. The whole genome sequencing of S. spinosa will also yield important further information concerning the molecular biology of this organism and will likely enable more sophisticated strategies to improve the production of spinosyns using classic mutagenesis and rational selection strategies.

The discovery of the second-generation spinosyn, spinetoram, along with the newly discovered butenyl-spinosyns and cyclobutyl-spinosyns, provides evidence that more potent analogs of spinosad can be discovered through a combination of approaches including combinatorial chemistry, structure–activity relationship analysis, natural product discovery, and combinatorial biosynthesis. The combinatorial biosynthesis approach has been proven to be an effective method to generate new spinosyn analogs, but the impact of this technology has been minimized by difficulties associated with transformation of S. spinosa. The spinosyn-producing microorganism, S. spinosa, has been characterized in several genetic studies and has been shown to be recalcitrant to genetic manipulation and gene transfer processes (Matsushima and Baltz 1994). So far, the extremely low efficiencies for protoplast and conjugative gene transfer in S. spinosa have necessitated the use of alternate hosts for the expression of the spinosyn biosynthetic cluster; however, a gene transfer system based on an autonomously replicating plasmid, such as pKC1218 (Bierman et al. 1992), might also be an effective strategy for the expression of recombinant spinosyn genes in S. spinosa. Another important aspect is to establish the gene transfer system for high spinosyn-producing strains, since many industrial strains became more difficult to transform after multiple cycles of random mutagenesis.

Another opportunity to better characterize the metabolic steps in the biosynthesis of spinosyns is to apply the techniques of metabolic network analysis to characterize metabolic limitations in these pathways (Wiechert 2001; Christensen and Nielsen 1999, 2000). This approach could provide quantitative metabolic fluxes of key metabolites in the central metabolism during the spinosyn fermentation process for high- and low-yield spinosyn strains, and furthermore, could identify the genes, limiting nutrients, or optimum fermentation conditions that support higher spinosad yields. Metabolic network analysis quantifies metabolic fluxes using 13C labels in key carbon substrates, such as glucose, and provides measurements of central metabolites that are precursors to the amino acids (Maaheimo et al. 2001). The biosynthetic fractional 13C labeling of common amino acids from their respective central metabolic precursors permits the calculation of cellular carbon fluxes with high accuracy. This approach has been successful applied in the investigation of Penicillium chrysogenum (Christensen and Nielsen 2000), a penicillin producer. The metabolic network analysis will assist in identifying biochemical targets for modification through genetic engineering. The desired result would be to redirect carbon flux in the central primary metabolic network toward the spinosyn metabolism.

References

  1. Baltz RH (2006) Combinatorial biosynthesis of novel antibiotics. Society for Industrial Microbiology News 56:148–160

  2. Bierman M, Logan R, O’Brien K, Seno ET, Rao RN, Schoner BE (1992) Plasmid cloning vectors for the conjugal transfer of DNA from Escherichia coli to Streptomyces spp. Gene 116:43–49

  3. Brünker P, Minas W, Kallio P, Bailey J (1999) Methods and compositions for increasing production of erythromycin. US patent 5,908,764

  4. Burns L, Graupner PR, Lewer P, Martin C, Vousden W, Waldron C, Wilkinson B (2003) Novel spinosyn-producing polyketide synthases. US patent 0,304,998

  5. Christensen B, Nielsen J (1999) Isotopomer analysis using GC-MS. Metab Eng 1:282–290

  6. Christensen B, Nielsen J (2000) Metabolic network analysis of Penicillium chrysogenum using 13C-labeled glucose. Biotechnol Bioeng 68:652–659

  7. Crouse GD, Sparks TC, Schoonover J, Gifford J, Dripps J, Bruce T, Larson LL, Garlich J, Hatton C, Hill RL, Worden TV, Martynow JG (2001) Recent advances in the chemistry of spinosyns. Pest Manag Sci 57:177–185

  8. Davies HG, Green RH (1986) Avermectins and milbemycins. Nat Prod Rep 3:87–121

  9. Gaisser S, Lill R, Wirtz G, Grolle F, Staunton J, Leadlay PF (2001) New erythromycin derivatives from Saccharopolyspora erythraea using sugar O-methyltransferases from the spinosyn biosynthetic gene cluster. Mol Microbiol 41:1223–1231

  10. Hahn DR, Gustafson G, Waldron C, Bullard B, Jackson JD, Mitchell J (2006) Butenyl-spinosyns, a natural example of genetic engineering of antibiotic biosynthetic genes. J Ind Microbiol Biotechnol 33:94–104

  11. Hong L, Zhao Z, Liu HW (2006) Characterization of SpnQ from the spinosyn biosynthetic pathway of Saccharopolyspora spinosa: mechanistic and evolutionary implications for C-3 deoxygenation in deoxysugar biosynthesis. J Am Chem Soc 128:14262–14263

  12. Hong L, Zhao Z, Melançon CE 3rd, Zhang H, Liu HW (2008) In vitro characterization of the enzymes involved in TDP-D-forosamine biosynthesis in the spinosyn pathway of Saccharopolyspora spinosa. J Am Chem Soc 130:4954–4967

  13. Huang KX, Zahn J, Han L (2008) SpnH from Saccharopolyspora spinosa encodes a rhamnosyl 4′-O-methyltransferase for biosynthesis of the insecticidal macrolide, spinosyn A. J Ind Microbiol Biotechnol 35:1669–1676

  14. Ito S, Hirata Y (1972) Ikarugamycin II. Structure of ikarugamycin. Tetrahedron Lett 12:1185–1188

  15. Jakobi M, Winkelman G, Kaiser D, Kempter C, Jung G, Berg G Bahl H (1996) Maltophilin: an new antifugal compound produced by Srenotrophomonas maltophilia R 3089. J Antibiot 49:1101–1104

  16. Jin ZH, Cheng X, Cen PL (2006a) Effect of glucose and phosphate on spinosad fermentation by Saccharopolyspora spinosa. Chin J Chem Eng 14:542–546

  17. Jin ZH, Wu JP, Zhang Y (2006b) Improvement of spinosad producing Saccharopolyspora spinosa by rational screening. Journal of Zhejiang University 7:366–370

  18. Kim HJ, Pongdee R, Wu Q, Hong L, Liu HW (2007) The biosynthesis of spinosyn in Saccharopolyspora spinosa: synthesis of the cross-bridging precursor and identification of the function of SpnJ. J Am Chem Soc 129:14582–14584

  19. Kirst HA, Michel KH, Mynderse JS, Chio EH, Yao RC, Nakatsukasa WM, Boeck L, Occolowitz JL, Paschal JW, Deeter JB, Thompson GD (1992) Discovery, isolation, and structure elucidation of a family of structurally unique fermentation-derived tetracyclic macrolides. In: Baker DR, Fenyes JG, Steffens JJ (eds) Synthesis and chemistry of agrochemicals, vol. 3. American Chemical Society, Washington, DC, pp 214–225

  20. Kirst HA, Michel KH, Mynderse JS, Chio EH, Yao RC, Nakatsukasa WM, Boeck L, Occolowitz JL, Paschal JW, Deeter JB, Thompson GD (1993) Discovery and Identification of a novel fermentation derived insecticide. In: Brown WC (ed) Development in industrial microbiology series: microbial metabolites, vol 32. Society for Industrial Microbiology, Washington, DC, pp 109–116

  21. Liang Y, Lu WY, Wen JP (2008) Improvement of Saccharopolyspora spinosa and the kinetic analysis for spinosad production. Appl Biochem Biotechnol (in press)

  22. Lewer P, Hahn DR, Karr LL, Graupener PR, Gilbert JR, Worden T, Yao R, Norton DW (2002) Pecticidal macrolides. US patent 6,455,504

  23. Maaheimo H, Fiaux J, Petek J, Bailey JE, Sauer U, Szyperski T (2001) Central carbon metabolism of Saccharomyces cerevisiae explored by biosynthetic fractional 13C labeling of common amino acids. Eur J Biochem 268:2464–2479

  24. MacNeil DJ (1995) Avermectin. Biotechnology 28:421–442

  25. Madduri K, Waldron C, Matsushima P, Broughton MC, Crawford K, Merlo DJ, Baltz RH (2001a) Genes for the biosynthesis of spinosyns: applications for yield improvement in Saccharopolyspora spinosa. J Ind Microbiol Biotechnol 27:399–402

  26. Madduri K, Waldron C, Merlo DJ (2001b) Rhamnose biosynthesis pathway supplies precursors for primary and secondary metabolism in Saccharopolyspora spinosa. J Bacteriol 183:5632–5638

  27. Martin CJ, Timoney MC, Sheridan RM, Kendrew SG, Wilkinson B, Staunton J, Leadlay PF (2003) Heterologous expression in Saccharopolyspora erythraea of a pentaketide synthase derived from the spinosyn polyketide synthase. Org Biomol Chem 1:4144–4147

  28. Mergott DJ, Frank SA, Roush WR (2004) Total synthesis of (−)-spinosyn A. Proc Natl Acad Sci USA 101:11955–11959

  29. Mertz FP, Yao RC (1990) Saccharopolyspora spinosa sp. Nov. isolated from soil collected in sugar mill rum still. Int J Syst Bacteriol 37:19–22

  30. Matsushima P, Baltz RH (1994) Transformation of Saccharopolyspora spinosa protoplasts with plasmid DNA modified in vitro to avoid host restriction. Microbiology 140:139–143

  31. Matsushima P, Broughton MC, Turner JR, Baltz RH (1994) Conjugal transfer of cosmid DNA from Escherichia coli to Saccharopolyspora spinosa: effects of chromosomal insertions on macrolide A83543 production. Gene 146:39–45

  32. Millar NS, Denholm I (2007) Nicotinic acetylcholine receptors: targets for commercially important insecticides. Invert Neurosci 7:53–66

  33. Mironov VA, Sergienko OV, Nastasyak IN, Danilenko VN (2004) Biogenesis and regulation of biosynthesis of erythomycins in Saccharopolyspora erythraea. Appl Biochem Microbiol 40:531–541

  34. Ochi K (2007) From microbial differentiation to ribosome engineering. Biosci Biotechnol Biochem 71:1373–1386

  35. Parekh S, Vinci VA, Strobel RJ (2000) Improvement of microbial strains and fermentation processes. Appl Microbiol Biotechnol 54:287–301

  36. Salgado VL, Sparks TC (2005) The spinosyns: chemistry, biochemistry, mode of action, and resistance. In: Gilbert LI, Iatrou K, Gill SS (eds) Comprehensive insect molecular sciences, vol. 6. Elsevier, New York, pp 136–173

  37. Shaw SJ (2008) The structure activity relationship of discodermolide analogues. Mini Rev Med Chem 8:276–284

  38. Sheehan LS, Lill RE, Wilkinson B, Sheridan RM, Vousden WA, Kaja AL, Crouse GD, Gifford J, Graupner PR, Karr L, Lewer P, Sparks TC, Leadlay PF, Waldron C, Martin CJ (2006) Engineering of the spinosyn PKS: directing starter unit incorporation. J Nat Prod 69:1702–1710

  39. Sparks TC, Crouse GD, Dripps JE, Anzeveno P, Martynow J, Deamicis CV, Gifford J (2008) Neural network-based QSAR and insecticide discovery: spinetoram. J Comput Aided Mol Des 22:393–401

  40. Strobel RJ, Nakatsukasa WM (1993) Response surface methods for optimizing Saccharopolyspora spinosa, a novel macrolide producer. J Ind Microbiol 11:121–127

  41. Thompson GD, Michel KH, Yao RC, Mynderse JS, Mosburg CT (1997) The discovery of Saccharopolyspora spinosa and a new class of insect control products. Down Earth 52:1–5

  42. Vinci VA, Byng G (1999) Strain improvement by nonrecombinant methods. In: Demain AL, Davies JE (eds) Manual of industrial microbiology and biotechnology. American Chemical Society, Washington, DC, pp 103–113

  43. Waldron C, Madduri K, Crawford K, Merlo DJ, Treadway P, Broughton MC, Baltz RH (2000) A cluster of genes for the biosynthesis of spinosyns, novel macrolide insect control agents produced by Saccharopolyspora spinosa. Antonie Van Leeuwenhoek 78:385–390

  44. Waldron C, Matsushima P, Rosteck PR Jr, Broughton MC, Turner J, Madduri K, Crawford KP, Merlo DJ, Baltz RH (2001) Cloning and analysis of the spinosad biosynthetic gene cluster of Saccharopolyspora spinosa. Chem Biol 8:487–499

  45. Wiechert W (2001) 13C metabolic flux analysis. Metab Eng 3:195–206

  46. Zhao Z, Hong L, Liu HW (2005) Characterization of protein encoded by spnR from the spinosyn gene cluster of Saccharopolyspora spinosa: mechanistic implications for forosamine biosynthesis. J Am Chem Soc 127:7692–7693

Download references

Acknowledgements

This work was supported by grants from the National Natural Science Foundation of China (no. 30870027) to KH.

Author information

Correspondence to Ke-xue Huang.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Huang, K., Xia, L., Zhang, Y. et al. Recent advances in the biochemistry of spinosyns. Appl Microbiol Biotechnol 82, 13–23 (2009). https://doi.org/10.1007/s00253-008-1784-8

Download citation

Keywords

  • Spinosyn
  • Saccharopolyspora spinosa
  • Insecticide
  • Polyketide
  • Butenyl-spinosyn