Developing the E. coli platform for efficient production of UMP-derived chemicals.

5-Methyluridine (5-MU) is a prominent intermediate for industrial synthesis of several antiviral-drugs, however, its availability over the past decades has overwhelmingly relied on chemical and enzymatic strategies. Here, we have realized efficient production of 5-MU in E. coli, for the first time, via a designer artificial pathway consisting of a two-enzyme cascade (UMP 5-methylase and phosphatase). More importantly, we have engineered the E. coli cell factory to boost 5-MU production by systematic evaluation of multiple strategies, and as a proof of concept, we have further developed an antibiotic-free fermentation strategy to realize 5-MU production (10.71 g/L) in E. coli MB229 (a ΔthyA strain). Remarkably, we have also established a versatile and robust platform with exploitation of the engineered E. coli for efficient production of diversified UMP-derived chemicals. This study paves the way for future engineering of E. coli as a synthetic biology platform for acceleratively accessing UMP-derived chemical diversities.

In vitro functional analysis and in silico structural modelling of pathogen-secreted polyglycine hydrolases.

Polyglycine hydrolases are fungal effectors composed of an N-domain with unique sequence and structure and a C-domain that resembles ß-lactamases, with serine protease activity. These secreted fungal proteins cleave Gly-Gly bonds within a polyglycine sequence in corn ChitA chitinase. The polyglycine hydrolase N-domain (PND) function is unknown. In this manuscript we provide evidence that the PND does not directly participate in ChitA cleavage. In vitro analysis of site-directed mutants in conserved residues of the PND of polyglycine hydrolase Es-cmp did not specifically impair protease activity. Furthermore, in silico structural models of three ChitA-bound polyglycine hydrolases created by High Ambiguity Driven protein-protein DOCKing (HADDOCK) did not predict significant interactions between the PND and ChitA. Together these results suggest that the PND has another function. To determine what types of PND-containing proteins exist in nature we performed a computational analysis of Foldseek-identified PND-containing proteins. The analysis showed that proteins with PNDs are present throughout biology as either single domain proteins or fused to accessory domains that are diverse but are usually proteases or kinases.

Genetic and chemical control of tuberculostearic acid production in Mycobacterium avium subspecies paratuberculosis.

Tuberculostearic acid (TBSA) is a fatty acid unique to mycobacteria and some corynebacteria and has been studied due to its diagnostic value, biofuel properties, and role in membrane dynamics. In this study, we demonstrate that TBSA production can be abrogated either by addition of pivalic acid to mycobacterial growth cultures or by a bfaA gene knockout encoding a flavin adenine dinucleotide (FAD)-binding oxidoreductase. Mycobacterium avium subspecies paratuberculosis (Map) growth and TBSA production were inhibited in 0.5-mg/mL pivalic acid-supplemented cultures, but higher concentrations were needed to have a similar effect in other mycobacteria, including Mycobacterium smegmatis. While Map C-type strains, isolated from cattle and other ruminants, will produce TBSA in the absence of pivalic acid, the S-type Map strains, typically isolated from sheep, do not produce TBSA in any condition. A SAM-dependent methyltransferase encoded by bfaB and FAD-binding oxidoreductase are both required in the two-step biosynthesis of TBSA. However, S-type strains contain a single-nucleotide polymorphism in the bfaA gene, rendering the oxidoreductase enzyme vestigial. This results in the production of an intermediate, termed 10-methylene stearate, which is detected only in S-type strains. Fatty acid methyl ester analysis of a C-type Map bfaA knockout revealed the loss of TBSA production, but the intermediate was present, similar to the S-type strains. Collectively, these results demonstrate the subtle biochemical differences between two primary genetic lineages of Map and other mycobacteria as well as explain the resulting phenotype at the genetic level. These data also suggest that TBSA should not be used as a diagnostic marker for Map.IMPORTANCEBranched-chain fatty acids are a predominant cell wall component among species belonging to the Mycobacterium genus. One of these is TBSA, which is a long-chain middle-branched fatty acid used as a diagnostic marker for Mycobacterium tuberculosis. This fatty acid is also an excellent biolubricant. Control of its production is important for industrial purposes as well as understanding the biology of mycobacteria. In this study, we discovered that a carboxylic acid compound termed pivalic acid inhibits TBSA production in mycobacteria. Furthermore, Map strains from two separate genetic lineages (C-type and S-type) showed differential production of TBSA. Cattle-type strains of Mycobacterium avium subspecies paratuberculosis produce TBSA, while the sheep-type strains do not. This important phenotypic difference is attributed to a single-nucleotide deletion in sheep-type strains of Map. This work sheds further light on the mechanism used by mycobacteria to produce tuberculostearic acid.

The tunicamycin derivative TunR2 exhibits potent antibiotic properties with low toxicity in an in vivo Mycobacterium marinum-zebrafish TB infection model.

Tunicamycins (TUN) are well-defined, Streptomyces-derived natural products that inhibit protein N-glycosylation in eukaryotes, and by a conserved mechanism also block bacterial cell wall biosynthesis. TUN inhibits the polyprenylphosphate-N-acetyl-hexosamine-1-phospho-transferases (PNPT), an essential family of enzymes found in both bacteria and eukaryotes. We have previously published the development of chemically modified TUN, called TunR1 and TunR2, that have considerably reduced activity on eukaryotes but that retain the potent antibacterial properties. A mechanism for this reduced toxicity has also been reported. TunR1 and TunR2 have been tested against mammalian cell lines in culture and against live insect cells but, until now, no in vivo evaluation has been undertaken for vertebrates. In the current work, TUN, TunR1, and TunR2 are investigated for their relative toxicity and antimycobacterial activity in zebrafish using a well-established Mycobacterium marinum (M. marinum) infection system, a model for studying human Mycobacterium tuberculosis infections. We also report the relative ability to activate the unfolded protein response (UPR), the known mechanism for the eukaryotic toxicity observed with TUN treatment. Importantly, TunR1 and TunR2 retained their antimicrobial properties, as evidenced by a reduction in M. marinum bacterial burden, compared to DMSO-treated zebrafish. In summary, findings from this study highlight the characteristics of recently developed TUN derivatives, mainly TunR2, and its potential for use as a novel anti-bacterial agent for veterinary and potential medical purposes.

Precursor-Directed Biosynthesis and Biological Testing of omega-Alicyclic- and neo-Branched Tunicamycin N-Acyl Variants.

Tunicamycins (TUNs) are Streptomyces-derived natural products, widely used to block protein N-glycosylation in eukaryotes or cell wall biosynthesis in bacteria. Modified or synthetic TUN analogues that uncouple these activities have considerable potential as novel mode-of-action antibacterial agents. Chemically modified TUNs reported previously with attenuated activity on yeast have pinpointed eukaryotic-specific chemophores in the uridyl group and the N-acyl chain length and terminal branching pattern. A small molecule screen of fatty acid biosynthetic primers identified several novel alicyclic- and neo-branched TUN N-acyl variants, with primer incorporation at the terminal omega-acyl position. TUNs with unique 5- and 6-carbon ω-cycloalkane and ω-cycloalkene acyl chains are produced under fermentation and in yields comparable with the native TUN. The purification, structural assignments, and the comparable antimicrobial properties of 15 of these compounds are reported, greatly extending the structural diversity of this class of compounds for potential medicinal and agricultural applications.

Crystal structure of a polyglycine hydrolase determined using a RoseTTAFold model.

Polyglycine hydrolases (PGHs) are secreted fungal proteases that cleave the polyglycine linker of Zea mays ChitA, a defensive chitinase, thus overcoming one mechanism of plant resistance to infection. Despite their importance in agriculture, there has been no previous structural characterization of this family of proteases. The objective of this research was to investigate the proteolytic mechanism and other characteristics by structural and biochemical means. Here, the first atomic structure of a polyglycine hydrolase was identified. It was solved by X-ray crystallography using a RoseTTAFold model, taking advantage of recent technical advances in structure prediction. PGHs are composed of two domains: the N- and C-domains. The N-domain is a novel tertiary fold with an as-yet unknown function that is found across all kingdoms of life. The C-domain shares structural similarities with class C ß-lactamases, including a common catalytic nucleophilic serine. In addition to insights into the PGH family and its relationship to ß-lactamases, the results demonstrate the power of complementing experimental structure determination with new computational techniques.

Efficient biosynthesis of nucleoside cytokinin angustmycin A containing an unusual sugar system.

Angustmycin A has anti-mycobacterial and cytokinin activities, and contains an intriguing structure in which an unusual sugar with C5'-C6' dehydration is linked to adenine via an N-glycosidic bond. However, the logic underlying the biosynthesis of this molecule has long remained obscure. Here, we address angustmycin A biosynthesis by the full deciphering of its pathway. We demonstrate that AgmD, C, A, E, and B function as D-allulose 6-phosphate 3-epimerase, D-allulose 6-phosphate pyrophosphokinase, adenine phosphoallulosyltransferase, phosphoribohydrolase, and phosphatase, respectively, and that these collaboratively catalyze the relay reactions to biosynthesize angustmycin C. Additionally, we provide evidence that AgmF is a noncanonical dehydratase for the final step to angustmycin A via a self-sufficient strategy for cofactor recycling. Finally, we have reconstituted the entire six-enzyme pathway in vitro and in E. coli leading to angustmycin A production. These results expand the enzymatic repertoire regarding natural product biosynthesis, and also open the way for rational and rapid discovery of other angustmycin related antibiotics.

Branched Chain Lipid Metabolism As a Determinant of the N-Acyl Variation of Streptomyces Natural Products.

Branched-chain fatty acids (BCFA) are encountered in Gram-positive bacteria, but less so in other organisms. The bacterial BCFA in membranes are typically saturated, with both odd- and even-numbered carbon chain lengths, and with methyl branches at either the ω-1 (iso) or ω-2 (anteiso) positions. The acylation with BCFA also contributes to the structural diversity of microbial natural products and potentially modulates biological activity. For the tunicamycin (TUN) family of natural products, the toxicity toward eukaryotes is highly dependent upon N-acylation with trans-2,3-unsaturated BCFA. The loss of the 2,3-unsaturation gives modified TUN with reduced eukaryotic toxicity but crucially with retention of the synergistic enhancement of the ß-lactam group of antibiotics. Here, we infer from genomics, mass spectrometry, and deuterium labeling that the trans-2,3-unsaturated TUN variants and the saturated cellular lipids found in TUN-producing Streptomyces are derived from the same pool of BCFA metabolites. Moreover, non-natural primers of BCFA metabolism are selectively incorporated into the cellular lipids of TUN-producing Streptomyces and concomitantly produce structurally novel neo-branched TUN N-acyl variants.

Harnessing synthetic biology-based strategies for engineered biosynthesis of nucleoside natural products in actinobacteria.

Antibiotic resistance poses an increasing threat to global health, and it is urgent to reverse the present trend by accelerating development of new natural product derived drugs. Nucleoside antibiotics, a valuable family of promising natural products with remarkable structural features and diverse biological activities, have played significant roles in healthcare and for plant protection. Understanding the biosynthesis of these intricate molecules has provided a foundation for bioengineering the microbial cell factory towards yield enhancement and structural diversification. In this review, we summarize the recent progresses in employing synthetic biology-based strategies to improve the production of target nucleoside antibiotics. Moreover, we delineate the advances on rationally accessing the chemical diversities of natural nucleoside antibiotics.

Exploring the Active Site of the Antibacterial Target MraY by Modified Tunicamycins.

The alarming growth of antibiotic resistance that is currently ongoing is a serious threat to human health. One of the most promising novel antibiotic targets is MraY (phospho-MurNAc-pentapeptide-transferase), an essential enzyme in bacterial cell wall synthesis. Through recent advances in biochemical research, there is now structural information available for MraY, and for its human homologue GPT (GlcNAc-1-P-transferase), that opens up exciting possibilities for structure-based drug design. The antibiotic compound tunicamycin is a natural product inhibitor of MraY that is also toxic to eukaryotes through its binding to GPT. In this work, we have used tunicamycin and modified versions of tunicamycin as tool compounds to explore the active site of MraY and to gain further insight into what determines inhibitor potency. We have investigated tunicamycin variants where the following motifs have been modified: the length and branching of the tunicamycin fatty acyl chain, the saturation of the fatty acyl chain, the 6″-hydroxyl group of the GlcNAc ring, and the ring structure of the uracil motif. The compounds are analyzed in terms of how potently they bind to MraY, inhibit the activity of the enzyme, and affect the protein thermal stability. Finally, we rationalize these results in the context of the protein structures of MraY and GPT.

Kilbournase, a protease-associated domain subtilase secreted by the fungal corn pathogen Stenocarpella maydis.

Subtilases are a large family of serine proteases that occur throughout biology. A small subset contain protease-associated (PA) domains that are structurally separate from but encoded within the active site. In bacteria, subtilase PA domains function to recruit specific protein substrates. Here we demonstrate that a protease secreted by the fungal corn pathogen Stenocarpella maydis, which truncates corn ChitA chitinase, is a PA domain subtilase. Protease was purified from S. maydis cultures and tryptic peptides were analyzed by LC-MS/MS. Ions were mapped to two predicted PA domain subtilases. Yeast strains were engineered to express each protease. One failed to produce recombinant protein while the other secreted protease that truncated ChitA. This protease, that we named kilbournase, was purified and characterized. It cleaved multiple peptide bonds in the amino-terminal chitin binding domain of ChitA while leaving the catalytic domain intact. Kilbournase was more active on the ChitA-B73 alloform compared to ChitA-LH82 and did not cleave AtChitIV3, a homolog from Arabidopsis thaliana, indicating a high level of specificity. Truncation of corn ChitA by kilbournase resembles truncation of human C5a by Streptococcus pyogenes ScpA, arguing that PA domain proteases in bacteria and fungi may commonly target specific host proteins.

Quantitative NIR determination of isoflavone and saponin content of ground soybeans.

Over 3200 discrete soybean samples were obtained from production locations around the United States during the years 2012-2016. Ground samples were scanned on near infrared spectrometers (NIRS) and analyzed by HPLC for total isoflavone and total saponin composition, as well as total carbohydrate composition. Multiple Linear Regression (MLR) analysis of preprocessed spectral data was used to develop optimized models to predict isoflavone content. The selection of a suitable calibration model was based on a high regression coefficient (R2), and lower standard error of calibration (SEC) values. Robust validated predictions were obtained for isoflavones, however less than robust calibrations were obtained for the total saponins. The correlations were not as robust for predicting the carbohydrate composition. NIRS is a suitable, rapid, nondestructive method to determine isoflavone composition in ground soybeans. Useful isoflavone composition predictions for large numbers of soybean samples can be obtained from quickly obtained NIRS scans.

Inhibition of Erwinia amylovora by Bacillus nakamurai.

A variety of potential inhibitors were tested for the first time for the suppression of Erwinia amylovora, the causal agent of fire blight in apples and pears. Strain variability was evident in susceptibility to inhibitors among five independently isolated virulent strains of E. amylovora. However, most strains were susceptible to culture supernatants from strains of Bacillus spp., and particularly to the recently described species B. nakamurai. Minimal inhibitory concentrations (MICs) were 5-20% (vol/vol) of culture supernatant from B. nakamurai against all five strains of E. amylovora. Although Bacillus species have been previously reported to produce lipopeptide inhibitors of E. amylovora, matrix-assisted laser desorption time of flight mass spectrometry (MALDI-TOF MS) and column chromatography indicated that the inhibitor from B. nakamurai was not a lipopeptide, but rather a novel inhibitor.

Comparative Investigation into Formycin A and Pyrazofurin A Biosynthesis Reveals Branch Pathways for the Construction of C-Nucleoside Scaffolds.

Formycin A (FOR-A) and pyrazofurin A (PRF-A) are purine-related C-nucleoside antibiotics in which ribose and a pyrazole-derived base are linked by a C-glycosidic bond. However, the logic underlying the biosynthesis of these molecules has remained largely unexplored. Here, we report the discovery of the pathways for FOR-A and PRF-A biosynthesis from diverse actinobacteria and propose that their biosynthesis is likely initiated by a lysine N6-monooxygenase. Moreover, we show that forT and prfT (involved in FOR-A and PRF-A biosynthesis, respectively) mutants are correspondingly capable of accumulating the unexpected pyrazole-related intermediates 4-amino-3,5-dicarboxypyrazole and 3,5-dicarboxy-4-oxo-4,5-dihydropyrazole. We also decipher the enzymatic mechanism of ForT/PrfT for C-glycosidic bond formation in FOR-A/PRF-A biosynthesis. To our knowledge, ForT/PrfT represents an example of ß-RFA-P (ß-ribofuranosyl-aminobenzene 5'-phosphate) synthase-like enzymes governing C-nucleoside scaffold construction in natural product biosynthesis. These data establish a foundation for combinatorial biosynthesis of related purine nucleoside antibiotics and also open the way for target-directed genome mining of PRF-A/FOR-A-related antibiotics.IMPORTANCE FOR-A and PRF-A are C-nucleoside antibiotics known for their unusual chemical structures and remarkable biological activities. Deciphering the enzymatic mechanism for the construction of a C-nucleoside scaffold during FOR-A/PRF-A biosynthesis will not only expand the biochemical repertoire for novel enzymatic reactions but also permit target-oriented genome mining of FOR-A/PRF-A-related C-nucleoside antibiotics. Moreover, the availability of FOR-A/PRF-A biosynthetic gene clusters will pave the way for the rational generation of designer FOR-A/PRF-A derivatives with enhanced/selective bioactivity via synthetic biology strategies.

Rhodium-catalyzed reductive modification of pyrimidine nucleosides, nucleotide phosphates, and sugar nucleotides.

Nucleosides and nucleotides are a group of small molecule effectors and substrates which include sugar nucleotides, purine and pyrimidine-based nucleotide phosphates, and diverse nucleotide antibiotics. We previously reported that hydrogenation of the nucleotide antibiotic tunicamycin leads to products with reduced toxicity on eukaryotic cells. We now report the hydrogenation of diverse sugar nucleosides, nucleotide phosphates, and pyrimidine nucleotides. UDP-sugars and other uridyl and thymidinyl nucleosides are quantitatively reduced to the corresponding 5,6-dihydro-nucleosides. Cytidyl pyrimidines are reduced, but the major products are the corresponding 5,6-dihydrouridyl nucleosides resulting from a deamination of the cytosine ring.

Divergent Biosynthesis of C-Nucleoside Minimycin and Indigoidine in Bacteria.

Minimycin (MIN) is a C-nucleoside antibiotic structurally related to pseudouridine, and indigoidine is a naturally occurring blue pigment produced by diverse bacteria. Although MIN and indigoidine have been known for decades, the logic underlying the divergent biosynthesis of these interesting molecules has been obscure. Here, we report the identification of a minimal 5-gene cluster (min) essential for MIN biosynthesis. We demonstrated that a non-ribosomal peptide synthetase (MinA) governs "the switch" for the divergent biosynthesis of MIN and the cryptic indigoidine. We also demonstrated that MinCN (the N-terminal phosphatase domain of MinC), MinD (uracil phosphoribosyltransferase), and MinT (transporter) function together as the safeguard enzymes, which collaboratively constitute an unusual self-resistance system. Finally, we provided evidence that MinD, utilizing an unprecedented substrate-competition strategy for self-resistance of the producer cell, maintains competition advantage over the active molecule MIN-5'-monophosphate by increasing the UMP pool in vivo. These findings greatly expand our knowledge regarding natural product biosynthesis.

Synergistic enhancement of beta-lactam antibiotics by modified tunicamycin analogs TunR1 and TunR2.

The ß-lactams are the most widely used group of antibiotics in human health and agriculture, but this is under threat due to the persistent rise of pathogenic resistance. Several compounds, including tunicamycin (TUN), can enhance the antibacterial activity of the ß-lactams to the extent of overcoming resistance, but the mammalian toxicity of TUN has precluded its use in this role. Selective hydrogenation of TUN produces modified compounds (TunR1 and TunR2), which retain the enhancement of ß-lactams while having much lower mammalian toxicity. Here we show that TunR1 and TunR2 enhance the antibacterial activity of multiple ß-lactam family members, including penems, cephems, and third-generation penicillins, to a similar extent as does the native TUN. Eleven of the ß-lactams tested were enhanced from 2 to >256-fold against Bacillus subtilis, with comparable results against a penicillin G-resistant strain. The most significant enhancements were obtained with third-generation aminothiazolidyl cephems, including cefotaxime, ceftazidime, and cefquinome. These results support the potential of low toxicity tunicamycin analogs (TunR1 and TunR2) as clinically valid, synergistic enhancers for a broad group of ß-lactam antibiotics.

Coordinated Biosynthesis of the Purine Nucleoside Antibiotics Aristeromycin and Coformycin in Actinomycetes.

Purine nucleoside antibiotic pairs, concomitantly produced by a single strain, are an important group of microbial natural products. Here, we report a target-directed genome mining approach to elucidate the biosynthesis of the purine nucleoside antibiotic pair aristeromycin (ARM) and coformycin (COF) in Micromonospora haikouensis DSM 45626 (a new producer for ARM and COF) and Streptomyces citricolor NBRC 13005 (a new COF producer). We also provide biochemical data that MacI and MacT function as unusual phosphorylases, catalyzing an irreversible reaction for the tailoring assembly of neplanocin A (NEP-A) and ARM. Moreover, we demonstrate that MacQ is shown to be an adenosine-specific deaminase, likely relieving the potential "excess adenosine" for producing cells. Finally, we report that MacR, an annotated IMP dehydrogenase, is actually an NADPH-dependent GMP reductase, which potentially plays a salvage role for the efficient supply of the precursor pool. Hence, these findings illustrate a fine-tuned pathway for the biosynthesis of ARM and also open the way for the rational search for purine antibiotic pairs.IMPORTANCE ARM and COF are well known for their prominent biological activities and unusual chemical structures; however, the logic of their biosynthesis has long been poorly understood. Actually, the new insights into the ARM and COF pathway will not only enrich the biochemical repertoire for interesting enzymatic reactions but may also lay a solid foundation for the combinatorial biosynthesis of this group of antibiotics via a target-directed genome mining strategy.

Thiazolidine Peracetates: Carbohydrate Derivatives that Readily Assign cis-, trans-2,3-Monosaccharides by Gas Chromatography-Mass Spectrometry Analysis.

A novel group of carbohydrate derivatives is described that uniquely assign cis/ trans-2,3-aldose stereoisomers at low nanomolar concentrations. Aldopentoses, aldohexoses, or component aldoses from hydrolysis of polysaccharides or oligosaccharides react with cysteamine in pyridine to give quantitative formation of thiazolidines, which are subsequently peracetylated in a one-pot reaction. The nonpolar thiazolidines peracetate (TPA) derivatives are analyzed by gas chromatography and electron impact mass spectrometry (GC/EI-MS), each aldose giving rise to two TPA geometric isomers. The quantitative ratio of these diastereomers is dependent upon whether the parent monosaccharide is cis-2,3-(Rib, Lyx, Man, All, Gul, and Tal), or trans-2,3-aldose (Xyl, Ara, Glc, Gal, Ido, and Alt). TPAs generate observed EI-MS fragment ions characteristic of C1-C2 and C3-C4 bond cleavage of the parent sugars. This has been used to estimate the extent of metabolic labeling of microbial cell-wall carbohydrates, especially into the defining anomeric carbons and during aldolase / ketolase -catalyzed rearrangements.

An ATP-Dependent Ligase with Substrate Flexibility Involved in Assembly of the Peptidyl Nucleoside Antibiotic Polyoxin.

Polyoxin (POL) is an unusual peptidyl nucleoside antibiotic, in which the peptidyl moiety and nucleoside skeleton are linked by an amide bond. However, their biosynthesis remains poorly understood. Here, we report the deciphering of PolG as an ATP-dependent ligase responsible for the assembly of POL. A polG mutant is capable of accumulating multiple intermediates, including the peptidyl moiety (carbamoylpolyoxamic acid [CPOAA]) and the nucleoside skeletons (POL-C and the previously overlooked thymine POL-C). We further demonstrate that PolG employs an ATP-dependent mechanism for amide bond formation and that the generation of the hybrid nucleoside antibiotic POL-N is also governed by PolG. Finally, we determined that the deduced ATP-binding sites are functionally essential for PolG and that they are highly conserved in a number of related ATP-dependent ligases. These insights have allowed us to propose a catalytic mechanism for the assembly of peptidyl nucleoside antibiotic via an acyl-phosphate intermediate and have opened the way for the combinatorial biosynthesis/pathway engineering of this group of nucleoside antibiotics.IMPORTANCE POL is well known for its remarkable antifungal bioactivities and unusual structural features. Actually, elucidation of the POL assembly logic not only provides the enzymatic basis for further biosynthetic understanding of related peptidyl nucleoside antibiotics but also contributes to the rational generation of more hybrid nucleoside antibiotics via synthetic biology strategy.