Discovery of a Phosphonoacetic Acid Derived Natural Product by

Jan 19, 2017 - Ulysses Amancio de Frias , Greicy Kelly Bonifacio Pereira , María-Eugenia Guazzaroni , Rafael Silva-Rocha. BioMed Research Internation...
1 downloads 0 Views 1MB Size
Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Letter

Discovery of a phosphonoacetic acid derived natural product by pathway refactoring Todd S. Freestone, Kou-San Ju, Bin Wang, and Huimin Zhao ACS Synth. Biol., Just Accepted Manuscript • DOI: 10.1021/acssynbio.6b00299 • Publication Date (Web): 19 Jan 2017 Downloaded from http://pubs.acs.org on January 21, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Synthetic Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

Submitted to ACS Synthetic Biology

Discovery of a phosphonoacetic acid derived natural product by pathway refactoring

Todd S. Freestone1,2, Kou-San Ju2,4, Bin Wang2, and Huimin Zhao1,2,3*

1

Department of Chemical and Biomolecular Engineering, 2Carl R. Woese Institute for Genomic

Biology, 3Departments of Biochemistry, Bioengineering, and Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801

4

Present address: Department of Microbiology and the Division of Medicinal Chemistry &

Pharmacognosy, The Ohio State University, Columbus, OH 43210

* To whom correspondence should be addressed. Phone: (217) 333-2631. Fax: (217) 333-5052. E-mail: [email protected].

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract The activation of silent natural product gene clusters is a synthetic biology problem of great interest. As the rate at which gene clusters are identified outpaces the discovery rate of new molecules, this unknown chemical space is rapidly growing, as too are the rewards for developing technologies to exploit it. One class of natural products that has been underrepresented is phosphonic acids, which have important medical and agricultural uses. Hundreds of phosphonic acid biosynthetic gene clusters have been identified encoding for unknown molecules. Although methods exist to elicit secondary metabolite gene clusters in native hosts, they require the strain to be amenable to genetic manipulation. One method to circumvent this is pathway refactoring, which we implemented in an effort to discover new phosphonic acids from a gene cluster from Streptomyces sp. NRRL F-525. By reengineering this cluster for expression in the production host Streptomyces lividans, utility of refactoring is demonstrated with the isolation of a novel phosphonic acid, O-phosphonoacetic acid serine, and the characterization of its biosynthesis. In addition, a new biosynthetic branch point is identified with a phosphonoacetaldehyde dehydrogenase, which was used to identify additional phosphonic acid gene clusters that share phosphonoacetic acid as an intermediate.

Keywords Pathway refactoring, Natural products, Phosphonic acids, Phosphonoacetic acid,

ACS Paragon Plus Environment

Page 2 of 24

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

Actinomycetes are a rich source of bioactive molecules and have been estimated to harbor thousands of unique natural products that are currently unknown to mankind (1). A major bottleneck to the discovery of these compounds is that most cannot be detected from laboratory cultures. Advances in synthetic biology are promising in revolutionizing natural product discovery by activating silent natural product gene clusters (NPGCs). There is great interest to use DNA sequences to explore nature’s chemical space, but with the ever-increasing rate bacterial genomic data is becoming available (2), our ability to determine chemistries from identified clusters is severely lacking. Although several strategies exist to express inactive NPGCs in native hosts (3, 4), they rely on the target strain to be genetically tractable, which drastically limits the number of exploitable clusters. Pathway refactoring is a method that decouples gene clusters from native regulatory contexts and can be applied to essentially all identified NPGCs. This strategy entails placing pathway genes downstream of characterized promoters in a production host and has been successful in discovering new molecules (5, 6).

One class of natural products that has notable bioactivity is phosphonic acids. This group of molecules carry carbon-phosphorous bonds and include compounds with beneficial antibacterial (fosfomycin) (7), antifungal (rhizocticin) (8), antiviral (foscarnet) (9), antimalarial (FR900098) (10), and herbicidal (phosphinothricin) (11) activities. The notable inhibitory capabilities of these molecules can be attributed to the configuration of their stable phosphonic acid moiety, which often mimics the phosphate esters and carboxylic acids of substrates in critical metabolic pathways. Examples of this include FR900098, a phosphonic analogue of 1-deoxy-D-xylulose 5phosphate

in

the

non-mevalonate

pathway

of

isoprenoid

ACS Paragon Plus Environment

biosynthesis

(10);

L-1-

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

aminoethylphosphonic acid, which inhibits alanine racemase that makes the essential bacterial peptidoglycan component D-alanine (12); and phosphinothricin, an inhibitor of glutamine synthase that prevents ammonia detoxification in bacteria, fungi, and plants (13). Similar to many other NPGCs, phosphonic acid clusters contain a signature gene that allows for straightforward identification. Nearly all phosphonic acid biosynthesis begins with the catalysis of phosphoenolpyruvate (PEP) to phosphonopyruvate (PnPy) by the enzyme PEP mutase (pepM) (14). This reaction is either followed by an acetylation to form 2-phosphonomethylmalic acid (2PMM) or a decarboxylation to make phosphonoacetaldehyde (PnAA). From there, pathways diverge to yield a diversity of different phosphonic acids with beneficially uses. In an effort to discover new phosphonic acids, we applied the synthetic biology tool of pathway refactoring and show the utility of this strategy by identifying a new compound, elucidating its biosynthetic route, and confirming a branch point for phosphonic acid synthesis that can guide the discovery of new molecules.

In an effort to discover new phosphonic acids, a large genome mining initiative was recently carried out where 10,000 actinomycete genomes were analyzed for pepM, revealing 278 strains positive for this gene (15). With over 200 of the pepM positive strains not producing any phosphonic acids on the tested production media, many new phosphonic acids remain to be discovered. One of the pepM positive strains that did not produce phosphonic acids from the screen of 10,000 strains is Streptomyces sp. strain NRRL F-525. Analysis of the gene cluster indicated the presence of two known enzymes, pepM and PnPy decarboxylase (PDC), but the absence of genes encoding for homologs of enzymes in biosynthetic pathways known to further transform their product, PnAA. This intermediate is a prominent node to four major precursors in

ACS Paragon Plus Environment

Page 4 of 24

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

phosphonic acid biosynthesis, including phosphonoacetic acid (PnA, Figure 1). PnA formation has been confirmed to be carried out in fosfazinomycin biosynthesis by an α-ketoglutarate dependent dioxygenase, which also performs another oxidation in its pathway (16). Interestingly, a NADP+-dependent aldehyde dehydrogenase has also been found to carry out this same step but in phosphonic acid degradation. This PnAA dehydrogenase is used by the soil bacterium Sinorhizobium meliloti 1021 as it metabolizes phosphonic acids as a source of phosphorous (17). Although a PnAA dehydrogenase has been predicted to be involved in phosphonolipid biosynthesis, it has yet to be verified (18).

The genome contig from NRRL F-525 encoding the pepM gene was analyzed to identify the cluster boundaries. At the time of the original analysis, no other sequenced phosphonic acid cluster fell into the same gene cluster family as the F-525 cluster (1, 15), preventing the use of synteny to determine which genes to refactor. To hypothesize cluster boundaries, each gene in the neighborhood of the pepM gene was aligned with possible homologous proteins using BLAST alignment (blast.ncbi.nlm.nih.gov). For the top 100 hits of each gene, the number of homologues with greater than 50% and 70% sequence identity were tabulated. When this data is charted for each open reading frame in the pepM neighborhood, thirteen genes are found to lie in a region of low homology (Supplementary Figure 1 and Supplementary Table 1). The presence of pepM and PDC suggested that PnAA would be an intermediate in the pathway. This gene cluster was of particular interest as genes encoding for known biosynthetic enzymes that use PnAA as a substrate were notably absent. That the cluster’s product is a bioactive molecule is suggested by the presence of two putative ligases, which are enzymes found in pathways for phosphonopeptides (19, 20). This subclass of phosphonic acids are taken up by cells via peptide

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

transporters, and once inside, the peptide is often cleaved to release a bioactive head group that can target essential enzymatic machinery (12, 21).

The thirteen genes from the F-525 phosphonic acid cluster were named fpnA-M (for F-525 phosphonic acid) and were fully refactored using promoters that have been previously characterized (22) with the final construct integrated into Streptomyces lividans to make strain HZ6787. 31P-NMR analysis of the concentrated extract of the engineered strain revealed a unique phosphonic acid peak at 19.3 ppm. This molecule was purified and its structure elucidated by NMR and mass spectrometry (Supplementary Methods and Supplementary Figures 2-12). A novel compound, O-phosphonoacetic acid serine (O-PnAS), was discovered and consisted of a PnA head group attached to the alcohol side chain of serine via a phosphate ester bond (Figure 2).

Because O-PnAS appeared to be derived from PnA, we hypothesized that the aldehyde dehydrogenase in the cluster was responsible for PnA formation. A four-gene cassette containing the F-525 pepM, PDC, and aldehyde dehydrogenase genes (fpnHIJG) was integrated into S. lividans (HZ6807), which produced PnA as expected (Figure 3). This is the first instance of an aldehyde dehydrogenase responsible for PnA formation in phosphonic acid biosynthesis. The F525 PnA dehydrogenase, fpnG, showed 31% identity to phnY, the PnAA dehydrogenase used in phosphonic acid degradation by Sinorhizobium meliloti 1021. When a BLAST search was performed on fpnG, of the top 50 hits, nine of the PnAA dehydrogenases were in close proximities to a pepM gene while 14 hits were near a PnA hydrolase, which breaks down PnA into acetate and phosphate. The remaining BLAST hits were associated with neither pepM nor

ACS Paragon Plus Environment

Page 6 of 24

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

PnA hydrolase. Homologues of phnY were more conserved and were nearly always associated with a PnA hydrolase. The genes fpnG and phnY also differed in the conserved protein domain families they were grouped into: fpnG with an α-ketoglutarate semialdehyde dehydrogenase and phnY with a phosphonoformaldehye dehydrogenase. These dehydrogenase domain families shared the same catalytic residues but differed slightly in their NAD(P)+ binding domains.

We predicted that for O-PnAS synthesis, PnA was conjugated onto a nucleotide by the cluster’s nucleotidyltransferase

and

then

attached

to

serine

by

the

putative

CDP-alcohol

phosphatidyltransferase, comparable to the attachment of serine to phospholipids by similar enzymes. Partial pathways were refactored where two five-gene cassettes were integrated into S. lividans, each containing genes for PnA biosynthesis and either the nucleotidyltransferase (HZ6808) or the phosphatidyltransferase (HZ6809). Neither of these strains produced O-PnAS when cultures were analyzed by

31

P-NMR (Spectra 1 and 2, Figure 4). However, a strain that

produced PnA and expressed both the nucleotidyltransferase and phosphatidyltransferase (HZ6810) did produce O-PnAS as hypothesized (Spectra 3 and 4, Figure 4). Thus, both the nucleotidyltransferase and phosphatidyltrasferase are needed for O-PnAS production. Although no new phosphonic acid signals were observed in the 31P-NMR spectrum of the strain integrated with the four-gene cassette containing the nucleotidyltransferase, it is possible that the nucleotide conjugated intermediate is short-lived. This is similar to the pathway of the phosphonic acid FR900098, where cleavage of the cytidine monophosphate moiety can occur in the absence of the pathway’s nucleotide hydrolase (23).

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

O-PnAS and PnA were tested for bioactivity in a preliminary disc diffusion assay using E. coli strain WM6242, engineered with IPTG inducible overexpression of broad specificity phosphonic acid transporters (24). Although no zones of inhibition were observed for O-PnAS, inhibited growth was seen for PnA (Supplementary Figure 12). Antibacterial activity was also seen for PnA against S. lividans. We attempted to cultivate S. lividans strain HZ6811, which overexpressed the F-525 nucleotidyltransferase (fpnM) and phosphotidyltransferase (fpnK) genes, on ISP2 supplemented with 5 mM PnA in an effort to see if the partial pathway would produce O-PnAS; however, the strain’s growth was severely inhibited. PnA is known as a DNA polymerase inhibitor and has been studied as an antiviral for Herpes infections (25, 26), but no antibacterial activity has previously been reported for the compound. Phosphonoformate, a similar antiviral, was also tested against the phosphonic acid susceptible E. coli strain but did not inhibit growth. To further evaluate PnA’s antibacterial capabilities, six additional bacterial strains used as benchmarks to measure minimum inhibitory concentrations (MIC) (27) were also tested against PnA, with only Staphylococcus aureus ATCC 29213 showing some growth inhibition with a MIC of 256 µg/mL (Supplementary Figure 13). The bioactivity of O-PnAS was not tested due to its failure to inhibit growth in the preliminary screen and because only a limited amount of the compound was available. Although the inhibitory properties of PnA are weak, additional modifications to it may increase potency. It is possible that PnA derived phosphonic acids have similar modes of action as PnA but have improved cellular penetration or enzyme inhibition properties.

By identifying gene clusters that encode both pepM and PnAA dehydrogenase, ten additional gene clusters were found that likely have products derived from PnA (Figure 5). Four of these

ACS Paragon Plus Environment

Page 8 of 24

Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

appear to belong in the same gene cluster family with F-525 (white arrows, Figure 5) and have been uploaded to the NCBI database over the course of a year, evidence of the increasing rate at which new pathways are becoming available for study. The presence of an asparagine synthetase homologue (fpnL) hints at the possible addition of another nitrogen on one of O-PnAS’s carboxylic acids, whereas further nitrogen tailoring may occur considering the presence of enzymes involved in nitrogen chemistry (fpnE and fpnF).

Of the other clusters with a PnA

intermediate, four appear to be related to each other in a gene cluster family that contains 22 homologous genes (gray arrows, Figure 5) that includes possible ATP-grasp enzymes that may be involved in phosphonopeptide biosynthesis. The remaining two clusters are unrelated to one another and the other identified clusters. Bioinformatic comparisons among these PnA clusters can aid in pathway elucidation, such as helping determine cluster boundaries, gene order, and possible common intermediates.

Although enzyme homology and synteny analyses of the F-525 clusters support that the correct set of genes were refactored for S. lividans expression, the final product was not isolated. One possible explanation for this is that the pathway is improperly balanced, which is evidenced by the buildup of side products. 2-aminoethylphosphonic acid (2-AEP) and two unknown phosphonic acids were produced by each S. lividans strain integrated with a partial or full pathway. These compounds are observed even when only pepM and PDC are expressed in S. lividans (Supplementary Figure 14). This limitation we have found with refactored phosphonic acid pathways is not seen when fosmids are used to introduce phosphonic acid clusters. For instance, when the pathway for the phosphonopeptide dehydrophos is integrated into the S. lividans genome, only the final product is detected in the phosphonic acid region of the 31P-NMR

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

spectrum (20). While expression profiles for native clusters have evolved for optimal pathway flux, the introduction of strong promoters can imbalance fluxes and lead to the accumulation of intermediates and side products. This may be more evident with phosphonic acids than other natural products because the small size of phosphonic acids and their resemblances to primary metabolites may make them more readily metabolized by endogenous enzymes.

By identifying a new phosphonic acid and establishing its biosynthetic route, we have shown that pathway refactoring can be a helpful tool in elucidating uncharacterized pathways. Confirming the activity of the early pathway enzyme PnAA dehydrogenase has also led to the classifying of additional clusters that likely share PnA as a pathway intermediate. This information will be helpful in guiding the investigation of these other pathways. The research presented here also highlights the difficulty of properly balancing gene expression with de novo pathway refactoring, especially with larger clusters. A complete refactoring, such as was done with the F-525 cluster, can easily disrupt coordinated expression that has been fine-tuned by evolution. This problem can be seen in the refactoring of the Klebsiella oxytoca nitrogen fixation gene cluster where, after optimizing expression of gene subsets and testing over a hundred variants of the full pathway, only 57% of wild-type activity is recovered in the native host (28). Given how nuanced native expression is and how difficult it can be to mimic, a partial refactoring approach may be more promising in discovering new molecules. For instance, by keeping operonic structures intact and introducing a small set of promoters, researchers were able to successfully activate the silent lazarimide pathways using a minimalistic approach (6). Despite its difficulties, the broad applicability of pathway refactoring will continue to make it a promising technology in achieving the paramount goal of activating NPGCs in a high-throughput manner.

ACS Paragon Plus Environment

Page 11 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

Methods Strains and reagents All cloning was performed in E. coli strain BW25141 (29), conjugation with E. coli strain WM6026 (30), and all phosphonic acid production and testing in Streptomyces lividans 66. Plasmids and strains used in this study are listed in Supplementary Tables 2 and 3, respectively. Media and reagents were from Sigma-Aldrich (St. Louis, MO), Fisher Scientific (Pittsburgh, PA), or Becton Dickinson (Franklin Lakes, NJ). PCR reactions were performed in FailSafe PCR PreMix G (Epicentre Biotechnologies, Madison, WI) with Q5 DNA polymerase (New England Biolabs, Ipswich, MA). Restriction enzyme Esp3I was from Thermo-Scientific (Pittsburgh, PA) and all other restriction enzymes, T4 ligase, pUC19 plasmid, and HiFi DNA Assembly Kit (used for Gibson Assembly) were from New England Biolabs. All primers ordered from Integrated DNA Technologies (Coralville, Iowa) and are listed in Supplementary Tables 4 through 8. Promoter backbones used as templates were previously constructed in our laboratory (22).

Assembly of full and partial Streptomyces sp. F-525 refactored pathways A multi-tier assembly was used to assemble full and partial pathways as outlined in Supplementary Figure 15. The vector backbone for promoter plasmids (pProBB) was constructed by Golden Gate assembly (31) with restriction enzyme Esp3I. DNA fragments were amplified by PCR using primers listed in Supplementary Table 4 with plasmid pAE4 used as template (24). Promoters gapdhp(EL), gapdhp(KR), gapdhp(TP), rpsLp(AC), rpsLp(SG), and rpsLp(XC) were amplified from templates previously constructed in our laboratory (22) using primers listed in Supplementary Table 5. Promoters were inserted into the vector backbone using Golden Gate

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

assembly with restriction enzyme BsaI. Some promoters had mutations introduced to make them Golden Gate compatible, and a xylE reporter assay (32) was used to verify the expression of all mutated promoters.

The complete refactoring of the Streptomyces species NRRL F-525 phosphonic acid cluster was performed in three subsequent assemblies. First, pathway genes (amplified by PCR using primers listed in Supplementary Table 6) were individually cloned into the promoter helper plasmids using Golden Gate assembly with restriction enzyme Esp3I and then were verified by Sanger sequencing. Second, each promoter-gene cassette was amplified by primers given in Supplementary Table 7 and assembled into four partial pathway plasmids (pUC19t-pfnIJM, pUC19t-Apr-fpnABD, pUC19t-fpnLKH, and pUC19t-fpnEFC), each containing three genes. The backbone the second-tier plasmids except pUC19t-Apr-fpnABD were made by amplifying a fragment from pUC19 with primers “BbsI.Esp3I.pUC19 for”/”BbsI.EcoRVpUC19 rev” and circularizing it in a Golden Gate assembly reaction with enzyme BbsI. The backbone for pUC19t-Apr-fpnABD, which included an apramycin (Apr) marker, included an additional fragment amplified from the pAE4 Apr cassette using primers “BbsI.EcoRV.Apr for”/”BbsI.Apr rev” and assembled with the pUC19 backbone mentioned above in a Golden Gate assembly with BbsI.

Before the second-tier assembly, the backbones were digested with EcoRV and

dephosphorylated with rSAP. The third and final assembly was made by mixing the four partial pathway plasmids in a Golden Gate assembly reaction with enzyme Esp3I and the backbone pF525-BB. pF525-BB was constructed by Golden Gate assembly with the primers prefixed with “BB” and their respective templates as listed in Supplementary Table 8. The presence of each gene in the final construct was confirmed both by restriction analysis and PCR. Partial pathways

ACS Paragon Plus Environment

Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

constructs of two-, three-, four-, and five-gene were made by amplifying individual promotergene cassettes by PCR and using Golden Gate assembly with Esp3I and Final-BB. Conjugation of all plasmids into Streptomyces lividans was performed as previously described (33).

Culturing of strains and compound purification All engineered strains of Streptomyces lividans were grown on ISP2 (10 g/L malt extract, 4 g/L yeast extract, 4 g/L dextrose) with 25 mg/L Apr at 30 ºC unless otherwise noted. Strains containing pathways of interest were streaked out onto agar plates and grown for 2-3 days. Individual colonies were then inoculated into 2 mL liquid medium in culture tubes and grown for another two days. 1 mL of the starter culture was then used to inoculate 50 mL liquid medium in 250 mL baffled flasks with beads shaken at 250 rpm for three days to high density. Agar plates were then inoculated with 0.4 mL of final liquid culture and incubated for ten days. Three plates were cultured for preliminary production of the full pathway and for the partial pathway strains. Media extracts (50 mL) were harvest by freezing plates at -20 ºC overnight, thawing at room temperature, compressing the agar and clarifying the liquid through milk filters (KenAG, Ashland, OH). To produce enough compound for purification, S. lividans engineered with the full F-525 pathway was cultured for 10 days on 10 L of ISP2+10 mg/L Apr agar plates, with about 6 L of liquid collected after freezing, thawing, and compressing plates. Full details of compound purification and characterization are given in Supplementary Methods.

Nuclear magnetic resonance (NMR) and mass spectrometry analysis NMR spectra were collected using an Agilent 600 MHz spectrometer with OneNMR probe (Agilent, Santa Clara, CA), with 243 MHz used for 31P, 600 MHz for 1H, and 150 MHz for 13C.

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

All samples were dissolved in D2O (Sigma-Aldrich). Spectra were analyzed using MestReNova v. 10.0.2 software (Mestrelab Research, Santiago de Compostela, Spain). Mass of O-PnAS was determined for the purified sample using a Q-Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer (ThermoFisher Scientific, Waltham, MA; sheath gas flow rate of 45 psi, auxiliary gas flow rate of 10 psi, sweep gas flow rate of 2 psi, voltage of 2.5 kV, capillary temperature of 250 ºC and auxiliary gas heater temperature of 400 ºC) at the Roy J. Carver Biotechnology Center (University of Illinois).

Bioactivity assays 25 µL of E. coli WM6242 (24) grown in LB to OD600=0.6 was used to inoculate 5 mL of LB or M9-glucose with 0.7% agar which was spread over agar plates of same medium. To induce the expression of the phosphonic acid transporters, 1 mM IPTG was included in the top agar layers before pouring. Paper discs with 20 µL of either 10 mM O-PnAS, 10 mM PnA or 45 mM FR900098 (positive control) were placed in the center of the agar plates, which were then incubated at 37 ºC. Zones of inhibition were measured after 18 hours (LB) or 24 hours (M9glucose) from top plating. A similar test was carried out with 10 mM phosphonoformate but only on M9-glucose.

To determine the minimum inhibitory concentration of PnA, the test strains were grown overnight at 37 ºC at 220 rpm in 10 mL of either brain heart infusion medium (from Becton Dickinson, used to culture Staphylococcus aureus ATCC 29213 and Enterococcus faecalis ATCC 19433), nutrient broth (from Becton Dickinson, used to culture Klebsiella pneumoniae ATCC 27736 and Escherichia coli ATCC 25922), or tyrpticase soy broth (15 g/L tryptone, 5 g/L

ACS Paragon Plus Environment

Page 15 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

soytone, 5 g/L NaCl; used to culture Pseudomonas aeruginosa PAO1 and Acinetobacter baumannii ATCC 19606). 100 µL of each overnight culture was used to inoculate 10 mL of Mueller-Hinton (MH) medium (Sigma-Aldrich) and cultured for 1 hour at similar conditions. In a 96-well plate, 10 µL of 5x106 cfu/mL of each strain was then added to 90 µL of MH medium with PnA concentrations ranging from 2 to 1024 µg/mL. OD600 of each well was measured and relative turbidity was calculated as (F-B)/(N-B), where F is the optical density (OD) of culture fed the molecule of interest, N is the OD of culture with no feeding, and B is the OD of the medium only.

Supporting Information Supplementary methods, tables, and figures are available free of charge via the Internet at http://pubs.acs.org.

Author Information T.S.F., K.S.J. and H.Z. designed the experiments and wrote the manuscript. T.S.F. and B. Wang performed experiments.

Acknowledgments This work was funded by the National Institutes of Health (GM077596).

Conflict of Interest Disclosure The authors declare no conflict of interest.

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

References 1.

2.

3.

4.

5.

6.

7.

8. 9.

10.

11. 12.

13.

14.

Doroghazi, J. R., Albright, J. C., Goering, A. W., Ju, K. S., Haines, R. R., Tchalukov, K. A., Labeda, D. P., Kelleher, N. L., and Metcalf, W. W. (2014) A roadmap for natural product discovery based on large-scale genomics and metabolomics, Nat Chem Biol 10, 963-968. Land, M., Hauser, L., Jun, S. R., Nookaew, I., Leuze, M. R., Ahn, T. H., Karpinets, T., Lund, O., Kora, G., Wassenaar, T., Poudel, S., and Ussery, D. W. (2015) Insights from 20 years of bacterial genome sequencing, Funct Integr Genomics 15, 141-161. Bergmann, S., Schumann, J., Scherlach, K., Lange, C., Brakhage, A. A., and Hertweck, C. (2007) Genomics-driven discovery of PKS-NRPS hybrid metabolites from Aspergillus nidulans, Nat Chem Biol 3, 213-217. Guo, F., Xiang, S. H., Li, L. Y., Wang, B., Rajasarkka, J., Grondahl-Yli-Hannuksela, K., Ai, G. M., Metsa-Ketela, M., and Yang, K. G. (2015) Targeted activation of silent natural product biosynthesis pathways by reporter-guided mutant selection, Metab Eng 28, 134142. Luo, Y., Huang, H., Liang, J., Wang, M., Lu, L., Shao, Z., Cobb, R. E., and Zhao, H. (2013) Activation and characterization of a cryptic polycyclic tetramate macrolactam biosynthetic gene cluster, Nat Commun 4, 2894. Montiel, D., Kang, H. S., Chang, F. Y., Charlop-Powers, Z., and Brady, S. F. (2015) Yeast homologous recombination-based promoter engineering for the activation of silent natural product biosynthetic gene clusters, Proc Natl Acad Sci USA 112, 8953-8958. Hendlin, D., Stapley, E. O., Jackson, M., Wallick, H., Miller, A. K., Wolf, F. J., Miller, T. W., Chaiet, L., Kahan, F. M., Foltz, E. L., and Woodruff, H. B. (1969) Phosphonomycin, a new antibiotic produced by strains of Streptomyces, Science 166, 122-123. Rapp, C., Jung, G., Kugler, M., and Loeffler, W. (1988) Rhizocticins - new phosphonooligopeptides with antifungal activity, Liebigs Annalen Der Chemie, 655-661. Helgstrand, E., Eriksson, B., Johansson, N. G., Lannero, B., Larsson, A., Misiorny, A., Noren, J. O., Sjoberg, B., Stenberg, K., Stening, G., Stridh, S., and Oberg, B. (1978) Trisodium phosphonoformate, a new antiviral compound, Science 201, 819-821. Jomaa, H., Wiesner, J., Sanderbrand, S., Altincicek, B., Weidemeyer, C., Hintz, M., Turbachova, I., Eberl, M., Zeidler, J., Lichtenthaler, H. K., Soldati, D., and Beck, E. (1999) Inhibitors of the nonmevalonate pathway of isoprenoid biosynthesis as antimalarial drugs, Science 285, 1573-1576. Hoerlein, G. (1994) Glufosinate (phosphinothricin), a natural amino acid with unexpected herbicidal properties, Rev Environ Contam Toxicol 138, 73-145. Atherton, F. R., Hall, M. J., Hassall, C. H., Lambert, R. W., Lloyd, W. J., and Ringrose, P. S. (1979) Phosphonopeptides as antibacterial agents - mechanism of action of alaphosphin, Antimicrob Agents Chemother 15, 696-705. Abell, L. M., and Villafranca, J. J. (1991) Investigation of the mechanism of phosphinothricin inactivation of Escherichia coli glutamine synthetase using rapid quench kinetic techniques, Biochemistry 30, 6135-6141. Seidel, H. M., Freeman, S., Seto, H., and Knowles, J. R. (1988) Phosphonate biosynthesis: isolation of the enzyme responsible for the formation of a carbonphosphorus bond, Nature 335, 457-458.

ACS Paragon Plus Environment

Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

15.

16. 17.

18.

19.

20.

21.

22.

23.

24.

25. 26.

27. 28.

29.

Ju, K. S., Gao, J., Doroghazi, J. R., Wang, K. K., Thibodeaux, C. J., Li, S., Metzger, E., Fudala, J., Su, J., Zhang, J. K., Lee, J., Cioni, J. P., Evans, B. S., Hirota, R., Labeda, D. P., van der Donk, W. A., and Metcalf, W. W. (2015) Discovery of phosphonic acid natural products by mining the genomes of 10,000 actinomycetes, Proc Natl Acad Sci USA 112, 12175-12180. Huang, Z. D., Wang, K. K. A., Lee, J., and van der Donk, W. A. (2015) Biosynthesis of fosfazinomycin is a convergent process, Chem Sci 6, 1282-1287. Agarwal, V., Peck, S. C., Chen, J. H., Borisova, S. A., Chekan, J. R., van der Donk, W. A., and Nair, S. K. (2014) Structure and function of phosphonoacetaldehyde dehydrogenase: The missing link in phosphonoacetate formation, Chem Biol 21, 125-135. Yu, X., Doroghazi, J. R., Janga, S. C., Zhang, J. K., Circello, B., Griffin, B. M., Labeda, D. P., and Metcalf, W. W. (2013) Diversity and abundance of phosphonate biosynthetic genes in nature, Proc Natl Acad Sci USA 110, 20759-20764. Blodgett, J. A. V., Zhang, J. K., and Metcalf, W. W. (2005) Molecular cloning, sequence analysis, and heterologous expression of the phosphinothricin tripeptide biosynthetic gene cluster from Streptomyces viridochromogenes DSM 40736, Antimicrob Agents Chemother 49, 230-240. Circello, B. T., Eliot, A. C., Lee, J. H., van der Donk, W. A., and Metcalf, W. W. (2010) Molecular cloning and heterologous expression of the dehydrophos biosynthetic gene cluster, Chem Biol 17, 402-411. Kugler, M., Loeffler, W., Rapp, C., Kern, A., and Jung, G. (1990) Rhizocticin-a, an antifungal phosphono-oligopeptide of Bacillus subtilis ATCC-6633: biological properties, Arch Microbiol 153, 276-281. Shao, Z., Rao, G., Li, C., Abil, Z., Luo, Y., and Zhao, H. (2013) Refactoring the silent spectinabilin gene cluster using a plug-and-play scaffold, ACS Synthetic Biology 2, 662669. Johannes, T. W., DeSieno, M. A., Griffin, B. M., Thomas, P. M., Kelleher, N. L., Metcalf, W. W., and Zhao, H. (2010) Deciphering the late biosynthetic steps of antimalarial compound FR-900098, Chem Biol 17, 57-64. Eliot, A. C., Griffin, B. M., Thomas, P. M., Johannes, T. W., Kelleher, N. L., Zhao, H., and Metcalf, W. W. (2008) Cloning, expression, and biochemical characterization of Streptomyces rubellomurinus genes required for biosynthesis of antimalarial compound FR900098, Chem Biol 15, 765-770. Purifoy, D. J., and Powell, K. L. (1977) Herpes simplex virus DNA polymerase as the site of phosphonoacetate sensitivity: temperature-sensitive mutants, J Virol 24, 470-477. Overby, L. R., Robishaw, E. E., Schleicher, J. B., Rueter, A., Shipkowitz, N. L., and Mao, J. C. (1974) Inhibition of herpes simplex virus replication by phosphonoacetic acid, Antimicrob Agents Chemother 6, 360-365. Andrews, J. M. (2001) Determination of minimum inhibitory concentrations, J Antimicrob Chemother 48 Suppl 1, 5-16. Smanski, M. J., Bhatia, S., Zhao, D., Park, Y., L, B. A. W., Giannoukos, G., Ciulla, D., Busby, M., Calderon, J., Nicol, R., Gordon, D. B., Densmore, D., and Voigt, C. A. (2014) Functional optimization of gene clusters by combinatorial design and assembly, Nat Biotechnol 32, 1241-1249. Datsenko, K. A., and Wanner, B. L. (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products, Proc Natl Acad Sci USA 97, 6640-6645.

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

30.

31. 32.

33.

Page 18 of 24

Blodgett, J. A. V., Thomas, P. M., Li, G. Y., Velasquez, J. E., van der Donk, W. A., Kelleher, N. L., and Metcalf, W. W. (2007) Unusual transformations in the biosynthesis of the antibiotic phosphinothricin tripeptide, Nat Chem Biol 3, 480-485. Engler, C., Kandzia, R., and Marillonnet, S. (2008) A one pot, one step, precision cloning method with high throughput capability, PLoS One 3, e3647. Ingram, C., Brawner, M., Youngman, P., and Westpheling, J. (1989) Xyle functions as an efficient reporter gene in Streptomyces spp - use for the study of Galp1, a catabolitecontrolled promoter, J Bacteriol 171, 6617-6624. Shao, Z. Y., Luo, Y. Z., and Zhao, H. M. (2011) Rapid characterization and engineering of natural product biosynthetic pathways via DNA assembler, Mol Biosyst 7, 1056-1059.

ACS Paragon Plus Environment

Page 19 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

Figure 1. Early steps of phosphonic acid biosynthesis with PnAA as a major branch point. Whereas an α-ketoglutarate dependent dioxygenase has been confirmed in PnA formation, an aldehyde dehydrogenase has only been predicted to carry out this step before this study. PEP, phosphoenolpyruvate; PnPy, phosphonopyruvate; 2-PMM, 2-phosphonomethylmalate; PnAA, phosphonoacetylaldehyde; 2-AEP, 2-aminoethylphosphonic acid; 2-HEP, 2hydroxyethylphosphonic acid; KHPTA, 2-keto-4-hydroy-5-phophonpentanoic acid; α-KG, αketoglutarate; PnA, phosphonoacetic acid.

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. Chemical structure of O-phosphonoacetic acid serine (O-PnAS).

ACS Paragon Plus Environment

Page 20 of 24

Page 21 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

PnA

Figure 3. 31P-NMR spectra confirming aldehyde dehydrogenase from F-525 phosphonic acid cluster is responsible for PnA formation. Spectra 1: Supernatant from S. lividans engineered strain HZ6806 integrated with F-525 pepM and PDC genes (fpnHIJ); Spectra 2: Supernatant from S. lividans engineered strain HZ6807 integrated with F-525 pepM, PDC, and aldehyde dehydrogenase genes (fpnHIJG); Spectra 3: HZ6807 supernatant spiked with PnA standard enlarging peak at 14.3 ppm. The major signal at 16.6 ppm is 2-aminoethylphosphonic acid.

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 24

a

O-PnAS b

Figure 4. (a) Proposed biosynthetic pathway for O-PnAS based on enzymes present in cluster. NTP, nucleoside triphosphate; NMP, nucleoside monophosphate. (b) 31P-NMR spectra validating proposed pathway where O-PnAS is present only when both the nucleotidyltransferase and phosphotidyltransferase are expressed. Spectra 1: Supernatant from engineered S. lividans strain HZ6808 integrated with F-525 pepM, PDC, PnAA dehydrogenase, and nucleotidyltransferase genes (fpnHIJGM) which lacks O-PnAS signal at 19.35 ppm; Spectra 2: Supernatant from engineered S. lividans strain HZ6809 integrated with F-525 pepM, PDC, PnAA dehydrogenase, and phosphatidyltransferase genes (fpnHIJGK) which also lacks O-PnAS signal; Spectra 3: Supernatant from engineered S. lividans strain HZ6810 integrated with F-525 pepM, PDC, PnAA dehydrogenase, nucleotidyltransferase, and phosphatidyltransferase genes (fpnHIJGMK); Spectra 4: HZ6810 supernatant spiked with purified O-PnAS increases the signal at 19.35 ppm, verifying that only when the nucleotidyltransferase and phosphatidyltransferase genes are expressed is O-PnAS formed. Minor signals at 18.0 and 18.5 ppm are side products observed when pepM and PDC genes are expressed.

ACS Paragon Plus Environment

Page 23 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

Figure 5. Identified phosphonic acid gene clusters with predicted PnA intermediate. Genes conserved with the Streptomyces sp. F-525 phosphonic acid cluster are highlighted in white; gray genes are conserved between S. kanasensis ZX01, Streptomyces sp. MMG1121, Streptomyces sp. 31A4, and Kitaspora sp. Root187.

ACS Paragon Plus Environment

ACS Synthetic Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Table of Contents Use Only

ACS Paragon Plus Environment

Page 24 of 24