NOVEL ANTIBIOTIC COMPOUND

Abstract

The present invention provides a new antibiotic compound termed nidaromycin derived from a new biosynthetic gene cluster (BGC), and its uses. Also provided herein are novel genes and nucleic acid molecules encoding the biosynthetic machinery for the production of nidaromycin, and to constructs, vectors, and host cells for expressing the BGC and methods for producing the compound.

Claims

1. A compound of formula (I): ##STR00023## wherein R.sup.1 is SO.sub.2OH, SO.sub.2OR or SO.sub.2R and R.sup.2 is H, or wherein R.sup.2 is SO.sub.2OH, SO.sub.2OR or SO.sub.2R and R.sup.1 is H; wherein R is a C.sub.1-C.sub.20 hydrocarbyl group; wherein each R.sup.3 is independently selected from H or a C.sub.1-C.sub.20 hydrocarbyl group; or a pharmaceutically acceptable salt, solvate or hydrate thereof.

2. A compound as claimed in claim 1, wherein the compound has the structure: ##STR00024## wherein R.sup.1 is SO.sub.2OH, SO.sub.2OR or SO.sub.2R and R.sup.2 is H, or wherein R.sup.2 is SO.sub.2OH, SO.sub.2OR or SO.sub.2R and R.sup.1 is H; wherein R is a C.sub.1-C.sub.20 hydrocarbyl group; or a pharmaceutically acceptable salt, solvate or hydrate thereof.

3. A compound as claimed in claim 1, wherein R.sup.1 is SO.sub.2OH or SO.sub.2OR and R.sup.2 is H, or wherein R.sup.2 is SO.sub.2OH or SO.sub.2OR and R.sup.1 is H; wherein R is a C.sub.1-C.sub.20 hydrocarbyl group; preferably wherein R.sup.1 is SO.sub.2OH and R.sup.2 is H, or wherein R.sup.2 is SO.sub.2OH and R.sup.1 is H.

4. A compound as claimed in claim 1, wherein the compound has the structure: ##STR00025## or a pharmaceutically acceptable salt, solvate or hydrate thereof; or (B) ##STR00026## or a pharmaceutically acceptable salt, solvate, or hydrate thereof.

5. (canceled)

6. A nucleic acid molecule comprising: (a) a nucleotide sequence as shown in SEQ ID NO. 1; or (b) a nucleotide sequence which is the complement of SEQ ID NO. 1; or (c) a nucleotide sequence which is degenerate with SEQ ID NO. 1; or (d) a nucleotide sequence having at least 85% sequence identity with SEQ ID NO. 1; or (e) a part of any one of (a) to (d); wherein said nucleic acid molecule encodes or is complementary to a nucleic acid molecule encoding one or more polypeptides, or comprises or is complementary to a nucleic acid molecule comprising one or more genetic elements, having functional activity in the synthesis of an antibiotic compound.

7. The nucleic acid molecule as claimed in claim 6, wherein: (i) the compound is a compound of formula (I): ##STR00027## wherein R.sup.1 is SO.sub.2OH, SO.sub.2OR or SO.sub.2R and R.sup.2 is H, or wherein R.sup.2 is SO.sub.2OH, SO.sub.2OR or SO.sub.2R and R.sup.1 is H: wherein R is a C.sub.1-C.sub.20 hydrocarbyl group; wherein each R.sup.3 is independently selected from H or a C.sub.1-C.sub.20 hydrocarbyl group; or a pharmaceutically acceptable salt, solvate or hydrate thereof; (ii) encodes a biosynthetic system for the synthesis of said compound; (iii) comprises a nucleotide sequence as shown in any one or more of SEQ ID NOs 2-29, or a nucleotide sequence which is complementary or degenerate to any one or more of SEQ ID NOs 2-29, or which has at least 85% sequence identity with any one or more of SEQ ID NOs. 2-29; and/or (iv) comprises a nucleotide sequence which encodes an amino acid sequence as shown in any one or more of SEQ ID NOs 30-57, or an amino acid which has at least 85% sequence identity with any one or more of SEQ ID NOs. 30-57.

8-9. (canceled)

10. A polypeptide encoded by a nucleic acid molecule as defined in claim 6.

11. A recombinant construct comprising a nucleic acid molecule as defined in claim 6.

12. A vector comprising a nucleic acid molecule of claim 6 or a recombinant construct comprising the same.

13. A microbial host cell comprising: a nucleic acid molecule as defined in claim 6, or a recombinant construct or vector comprising the same.

14. (canceled)

15. The host cell as claimed in claim 13, wherein the host cell is: (i) a production host cell for production of an antibiotic compound and is an actinomycete; (ii) Streptomyces sp.; (iii) Streptomyces coelicolor; (iv) Streptomyces coelicolor strain M145 (ATCC BAA-471); (v) Streptomyces coelicolor strain M1152, being a derivative of (ii) comprising the modifications act red cpk cda rpoB(C.sub.1298T); or (vi) Streptomyces coelicolor strain M1152matAB, being a derivative of (iii) further comprising a deletion of the locus matAB.

16. A method of producing an antibiotic compound, said method comprising introducing into a microbial host cell a nucleic acid molecule as defined in claim 6 or a recombinant construct or vector comprising said nucleic acid molecule, and allowing the nucleic acid molecule to be expressed and the compound to be synthesised by the expressed biosynthetic system.

17. The method as claimed in claim 16, wherein: (i) the host cell is; (a) a production host cell of an antibiotic compound and is an actinomycete; (b) Streptomyces sp.; (c) Streptomyces coelicolor; (d) Streptomyces coelicolor strain M145 (ATCC BAA-471); (e) Streptomyces coelicolor strain M1152, being a derivative of (c) comprising the modifications act Ared Acpk Acda rpoB (C.sub.1298T); or (f) Streptomyces coelicolor strain M1152matAB, being a derivative of (d) further comprising a deletion of the locus matAB; (ii) the method further comprises recovering the compound; and/or (iii) the method further comprises purifying the compound.

18-19. (canceled)

20. A compound obtained or obtainable by the method as claimed in claim 16.

21. (canceled)

22. A method of treating a microbial infection in a subject or in a plant, said method comprising administering to the subject or to the plant an effective amount of a compound as claimed in claim 1.

23. The method as claimed in claim 22, wherein the microbial infection is caused by Gram-positive bacteria, preferably wherein the bacteria are Staphylococcus aureus or Enterococcus faecium, including antibiotic-resistant strains thereof.

24. (canceled)

25. A pharmaceutical composition comprising a compound as claimed in claim 1, further comprising at least one carrier, additive and/or excipient.

26. A method of controlling bacteria, comprising contacting said bacteria, or a surface or location carrying said bacteria, with a compound as claimed in claim 1.

27-28. (canceled)

29. A method of treating a microbial infection in a subject or in a plant, said method comprising administering to the subject or plant an effective amount of a compound as claimed in claim 20.

30. A pharmaceutical composition comprising a compound as claimed in claim 20, further comprising at least one carrier, additive and/or excipient.

Description

BRIEF DESCRIPTION OF FIGURES

[0186] FIG. 1 shows LC-DAD-isoplots of extracts of M1152matAB (P08-G05_C16) (A) and the control M1152matAB (B) cultivated in well plate with 2.5MG-2.5w/NaCl and 0.1% inducer-caprolactame. The two compounds eluting between 13 and 15 min are only observed in the extract of the transconjugant, Nidaromycin eluting at 14.6 min and a derivative eluting at 13.2 min.

[0187] FIG. 2 shows MS spectrum of the compound produced by the transconjugant P08-g05_C16 (corresponding to the main UV peak observed in extracts of the transconjugant). The compound also forms Na+ adducts in the MS. The red bars show the theoretical isotopic distribution of the proposed molecular formula, whereas the black bars show the measured isotopic distribution of the compound;

[0188] FIG. 3 shows a LC-DAD isoplot of the HPLC purified compound produced by the transconjugant M1152matAB (P08-G05_C16). Nidaromycin is eluting as a single peak at approximately 16 min in this chromatogram, and no other UV absorbing compounds are observed in the chromatogram.

[0189] FIG. 4 shows toxicity data for the compound nidaromycin, as produced by the transconjugant M1152matAB (P08-G05_C16) as described in Example 2, against the cell lines HepG and LLC-PK1 at varying concentrations from 0.5 to 50 g/ml (A) viability after 48 hrs exposure expressed as % and (B) LDH leakage (%) after 48 hrs exposure;

[0190] FIG. 5 shows MS spectra of .sup.15N labelled nidaromycin (A) and .sup.13C and .sup.15N labelled nidaromycin (B) showing that the molecular weight increases with 2 Da and 63 Da respectively which demonstrate that the formula contains 2N and 61C.

[0191] FIG. 6 shows MSMS fragmentation pattern of purified nidaromycin. The precursor mass is M+H+1349.5674

[0192] FIG. 7 shows the predicted structure of the compound produced by the transconjugant M1152matAB (P08-G05_C16) as determined by NMR studies in Example 8, including atom numbering for atom-specific assignments in NMR spectra used in the Example.

EXAMPLES

Example 1

Preparation of Host Strain Streptomyces coelicolor M1152matAB

[0193] S. coelicolor strain M1152 as described in Gomez-Escribano 2010 (supra) was obtained from the John Innes Centre, Norwich, UK.

[0194] In-frame deletion mutants for SCO2963/SCO2962 in S. coelicolor M1152 were created as described earlier (van Dissel et al., 2015. Microbial Cell Factories, 14 (1), pp. 1-10). In brief, the upstream region of SCO2963 ranging from 1326 to +43 relative to the start codon and the downstream region of SCO2962 from +2190 to +3610 were amplified by PCR from the S. coelicolor genome using the primers listed in Table 2. The amplified flanks were cloned into the unstable shuttle vector pWHM3-onT (Wu et al., 2019. Angewandte Chemie, 131 (9), pp.2835-2840) using the EcoRI and HindIll restriction site. The Xbal site, featured in both amplified regions, was used for insertion of the apramycin resistance cassette aacC4 flanked by loxP sites between the flanking regions. The completed vector (pMAT1) was introduced into E. coli ET12567+PUZ8002, which allowed transfer of pMAT1 towards S. coelicolor M1152 by conjugation. Mutants where the matAB locus was replaced by the aacC4 cassette and where the pWHM3 vector was lost were selected by replicate plating for a Thio/Apra+phenotype. A marker free S. coelicolor M1152 matAB strain was obtained by introduction of the pUWLcre plasmid, expressing the Cre recombinase, which incised the loxP sites surrounding the apramycin resistance gene.

TABLE-US-00002 TABLE2 matA_1326 AGTCGAATTCCAGCCGGGCGGTGAGATTCC matA_+43 ACTGTCTAGACGAGCACTCGTCGGCCGAAC matB_+2190 AGTCTCTAGAAGGCCGGTCGGATGACCACC matB_+3610 AGTCAAGCTTCCCTGTTCACTCCCGCAACCG

Example 2

Identification of the Biosynthetic Gene Cluster (Cluster 16) from Marine Actinobacteria Isolate P08-G05
Origin of marine isolate strain P08-G05

[0195] The marine isolate P08-G05 was obtained from the SINTEF/NTNU marine Actinobacteria strain collection, which was built from water, sediment, and sponge samples taken from the Trondheim fjord. The strain was selected based on a comprehensive assessment of the draft genomes of 1200 isolates from the strain collection based on different criteria: phylogenetic novelty, gene cluster diversity, and previously observed bioactivity, as described below.

[0196] The frozen glycerol culture from the collection was streaked on TSA (Trypton soya broth agar) supplemented with 0.5 artificial sea water (Engelhardt et al. 2010, Applied and Environmental Microbiology 76 (15): 4969-4976.). A pure isolate was cultivated in TSB with artificial sea water to produce mycelia for a working cell bank.

Illumina Sequencing and De Novo Assembly of the Genome of Strain P08-G05

[0197] Biomass of strain P08-G05 for genome sequencing was produced in TSB medium supplemented with 50% artificial sea water at 30 C. The biomass was collected by centrifugation and sent for sequencing to BaseClear BV, where the extraction of DNA, sequencing, and post-sequencing data processing were carried out.

[0198] Paired-end sequence reads were generated using the Illumina HiSeq2500 system. FASTQ sequence files were generated using the Illumina Casava pipeline version 1.8.3. Initial quality assessment was based on data passing the Illumina Chastity filtering. Subsequently, reads containing adapters and/or Phix control signal were removed using BaseClear's in-house filtering protocol. The second quality assessment was based on the remaining reads using the FASTQC quality control tool version 0.10.0. The quality of the FASTQ sequences was enhanced by trimming off low-quality bases using the Trim sequences option of the CLC Genomics Workbench version 8.0.

[0199] For genome assembly and scaffolding, the quality-filtered sequence reads were assembled into contig sequences. The analysis was performed using the De novo assembly option of the CLC Genomics Workbench version 8.0. Mis-assemblies and nucleotide disagreement between the Illumina data and the contig sequences were corrected with Pilon version 1.11. The contigs were linked and placed into scaffolds or super-contigs. This resulted in an assembly of 7,315,765 bp and 980 scaffolds. The orientation, order, and distance between the contigs was estimated using the insert size between the paired-end and/or mate-pair reads. The analysis was performed using the SSPACE Premium scaffolder version 2.3. The gapped regions within the scaffolds were (partially) closed in an automated manner using GapFiller version 1.10, taking advantage of the insert size between the paired-end and/or mate-pair reads. The obtained draft genomes were subsequently used for phylogenetic analyses and genome annotations. The quality of the Illumina sequencing de novo genome assembly of P08-G05 was evaluated by the checkM software (version 1.07) showing the high completeness of 95.9% with the low contamination of 1.6%.

PacBio Sequencing and Hybrid De Novo Genome Assembly

[0200] Cell mass for PacBio sequencing and direct cloning was produced in 500 ml shake flasks containing 3 g of 3 mm glass bead and 120 ml TSB medium supplemented with 0.5 artificial sea water at 30 C., 200 rpm with 2.5 orbital movement until OD.sub.600=5.6. The cell mass was harvested by centrifugation, kept on 40 C. until shipping, and shipped to BaseClear BV, The Netherlands, on dry ice. Long read PacBio sequencing was carried out at BaseClear using the PacBio Sequel instrument, and the obtained data were processed and filtered using the SMRT Link software suite with subreads shorter than 50 bp being discarded. This resulted in a number of 622,557 reads with the yield of 2,886,923, 195 bp.

[0201] The quality of the Illumina HiSeq reads was improved by trimming off low-quality bases using BBDuk, which is a part of the BBMap suite version 36.77. High-quality reads were assembled into contigs using ABySS version 2.0.2. The long reads were mapped to the draft assembly using BLASR version 1.3.1. Based on these alignments, the contigs were linked together and placed into scaffolds. The orientation, order, and distance between the contigs were estimated using SSPACE-LongRead version 1.0. Using Illumina reads, gapped regions within scaffolds were (partially) closed using GapFiller version 1.10. Finally, assembly errors and the nucleotide disagreements between the Illumina reads and scaffold sequences were corrected using Pilon version 1.21. This resulted in an assembly of 7,840,734 bp with 22 scaffolds.

Phylogenetic Positioning of Strain P08-G05

[0202] The phylogenetic position of isolate P08-G05 was determined as part of a comprehensive phylogenetic analysis performed on the Illumina HiSeq2500 sequenced genomes of 1200 selected strains of the SINTEF/NTNU marine Actinobacteria strain collection, similarly generated to the one of P08-G05 (as described above), and 576 Actinobacteria type strains retrieved from public databases. Analysis was carried out using IQTREE software (IQ-TREE MPI multicore version 1.6.7.1) with 92 house-keeping genes as reference to identify the phylogenetic novelty of the strain P08-G05. P08-G05 strain was placed among other strains in the Actinobacteria strain collection, not with other type strains, indicating that the strain likely represents a new Actinobacteria species.

Identification of the Biosynthetic Gene Cluster P08-G05_c16

[0203] An in-house Python script was used to evaluate the abundance and diversity of different biosynthetic gene cluster (BGC) classes of the 1200 strains from the Actinobacteria strain collection based on a collection of BGC's profile Hidden Markov models (pHMMs). The obtained matrixes containing the counts of pHMM hits from the corresponding strains were used to cluster the strains into different populations with the usage of an implementation in the programming language R of the t-distributed stochastic neighbor embedding (t-SNE) algorithm by AFG. P08-G05 strain was clustered in the cluster 34 (out of 40 t-SNE clusters) along with other strains. The strain was selected together with other strains from different t-SNE clusters for a shortlist of 86 strains for further characterization, including long-read PacBio sequencing.

[0204] The novel cluster, which encodes (at least) the (core) metabolic machinery to synthesize the nidaromycin compound was identified through manual curation based on the antiSMASH result of the PacBio sequenced genome of P08-G05. Analysis of resistance genes on the gene cluster was performed using an in-house script. No resistance genes were identified in the cluster P08-G06_c16.

Example 3

Cloning and Expression of Gene Cluster P08-G06_c16

Cloning and Conjugation of Gene Cluster P08-G06_c16

[0205] Based on antiSMASH results, gene cluster P08-G06_c16 was hypothesized to code for a moenomycin-like novel compound. Moenomycin has molecular formula C.sub.68H.sub.106N.sub.5O.sub.34P and mass 1567.645683 g/mol.

[0206] Cloning of cluster P08-G05_c16 in an inducible Bacterial Artificial Chromosome (BAC) vector (pDualP, proprietary to Varigen Biosciences (Madison, WI, USA)) was purchased from Varigen Biosciences based on chromosomal DNA of strain P08-G05. The construct was received from Varigen Bioscience in an E. coli strain suitable for the propagation of large constructs (10Beta). The cluster was transferred to S. coelicolor M1152matAB, prepared according to Example 1, by tri-parental conjugation by a procedure that is similar to the previously described methods (Jones et al 2013, PLOS ONE 8 (7): e69319.doe: 10.1371). In short: The BGC containing construct together with the driver plasmid pR9406 were transferred to E. coli ET12567 by triparental mating. For this each strain was first cultivated overnight on LB agar without selection. For all three strains, using an inoculation loop, a couple of colonies were scooped and were streaked together in a patch on LB agar containing apramycin, chloramphenicol and ampicillin. As control, each of the strains were also patched individually on the sample selection. Single colonies of ET12567+pR9406+DualP-BGC were growth in 13 ml culture tubes containing 5 ml LB+ampicillin, chloramphenicol and apramycin until an OD of 0.6, after which the culture was pelleted and washed twice with cold LB media. In parallel the S. coelicolor spores were pre-germinated by heat shock for 10 min at 50 C. and incubation at 30 C. for at 2-3h. E. coli and S. coelicolor spores were mixed and plated on soy flour mannitol (SFM) agar plates and incubated between 18 and 24 h at 30 C., before being overlaid with apramycin+nalidixic acid to select for transconjugant Streptomyces colonies. Single colonies were subsequently patched on selective SFM plates and expanded to confluent plates for spore harvest and storages through standard procedures. The new transconjugant strain carrying the nidaromycin gene cluster was given the short name M1152matAB (P08-G05_C.sub.16).

Cultivations of the Transconjugant M1152matAB (P08-G05_C.SUB.16.) and Expression of Gene Cluster P08-G05_c16

[0207] Well plate cultivations of M1152matAB (P08-G05_C.sub.16) and M1152matAB (control) were performed as follows: Seed cultures were produced in 250 ml shake flask with 50 ml of 0.5 Trypton Soya Broth (TSB) and 1.5 g of 3 mm glass bead (without antibiotics) for two days at 30 C. and 225 rpm until OD600=5-7. Production in 24 well plates (AXYGP-DW10ML24C) were performed in both 5254SW medium (Krlov et al 2021, Frontiers in Microbiology 12:2131) or MG-2.5 medium (Doull and Vining 1990, Applied Microbiology and Biotechnology, 32, 449-454; Martnez-Castro et al. 2013, Applied Microbiology and Biotechnology, 97, 2139-2152) supplemented with 1 g/L NaCl. The wells were filled with 2.5 ml medium and 43 mm glass beads and inoculated with 1.3% from the seed culture. The plates were incubated at 30 C. in a New Brunswick incubator at 800 rpm and 85% humidity for six days. The broth was freeze dried and extracted with one broth volume of DMSO for one hour.

[0208] Cell free extracts were analyzed by an Agilent LC-DAD-QTOF equipped with a Zorbax Bonus RP 2.150 mm, 3.5 L. 50 mM ammonium acetate [A] and an acetonitrile [B] were used as mobile phases. The gradient was 5% acetonitrile from 0-2 min, then increasing to 95% for the next 25 min. The QTOF was operated in positive and negative ionization mode with capillary voltage: 3.5 kV, Fragmentor voltage: 150 V, Skimmer: 65V, gas temperature 325 C., drying gass: 10 l/min, Nebulizer: 50. Data was processed using the Mass Hunter and Mass Profiler Professional software from Agilent.

[0209] The LC-DAD-isoplots showed that two peaks were observed in the transconjugant extract but not in the control (FIG. 1). The abundance of these two peaks were higher in the MG-2.5 w/0.5 seawater than in the 5254SW medium. The MS data showed that a cluster of masses was observed at the retention time that corresponded to the main UV peak. The three dominating masses were M+H=1349.5668 and it's adduct M+Na=1371.5496 (FIG. 2). These masses were not found in extracts of the control M1152matAB. The compound was concluded to be associated with the heterologous expression of the introduced P08-G05_c16 and named nidaromycin.

Example 4

Up-Scaled Production and Purification of the Heterologous Expressed Compound

[0210] Up-scaled production of active compound was performed in 500 ml shake flasks with 125 ml of MG-2.5 w/NaCl. The medium was inoculated with 3% from seed culture and incubated at 30 C. for six days at 200 rpm with 2.5 cm orbital movement.

[0211] The broth was freeze dried and homogenized with mortar. The material was extracted with DMSO acidified with trifluoroacetic acid (TFA) to 0.1% final concentration. The amount of organic solvent was 0.4 original broth volume. The DMSO extract was fractionated using an Agilent preparative HPLC equipped with a Zorbax Bonus RP, 9.4250 mm, 7 m column (Agilent), diode array detector (DAD) and a fraction collector. Mobile phases were water with 20 mM ammonium acetate [A] and acetonitrile [B]. The gradient was 5% [B] during injection, then a gradient increase from 55% to 75% [B] over 10 min. The column was washed for 1 min with 95% [B] before column equilibration with 5% [B]. The acetonitrile in the HPLC fractions was removed by rotational evaporator, and the aqueous phase was further purified and concentrated using 500 mg HLB solid phase extraction columns (Waters). The compound was eluted from the SPE column with methanol. The methanol was removed by evaporation using a Speedvac (ThermoFisher) at 50 C. The sample was added water, frozen at 80 C. and freeze dried.

[0212] The DAD plot in FIG. 3 confirms that a purified compound was obtained. The material obtained was used for inhibition and toxicity assays as described in Example 5, determining the molecular formula of nidaromycin as described in Example 6, as well as structure elucidation by NMR and determining the position of the sulphate group as described in Examples 7 and 8.

Example 5

Assay of Activity of Crude Extract and Purified Compound

Bioassay of Crude Extract

[0213] From cultures of strain M1152matAB (P08-G05_C.sub.16) (prepared as described in Example 3), cell free extract was prepared and tested in bioassay against a panel of strains, i.e., Enterococcus faecium CCUG 37832, M. luteus TO-09 ATCC9341, Pseudomonas aeruginosa ATCC 15692 and C. albicans CCUG. The extract showed activity against E. faecium CCUG37832.

[0214] In particular, extracts of the transconjugant inhibited growth of E. faecium CCUG 37832 at 16 dilution (MG-2.5 w/NaCl) and 4 dilution (5254SW), whereas extracts of the control did not inhibit any of the strains.

In Vitro MIC Bioassay

[0215] Minimal inhibitory concentrations against a selection of Gram-positive indicator organisms (MICs) were determined according to Clinical and Laboratory Standards Institute protocols by microdilution tests using in a 384 well format. The indicator strains were incubated in TSB medium over night until OD600-0.4, then diluted to OD600=0.1 in TSB medium and further 45 in assay medium (Mueller-Hinton broth, Difco). Inoculated medium was distributed into assay plates. Inoculated wells were added a 2 dilution series of either isolated compound or vancomycin (reference) diluted in DMSO, giving a final DMSO concentration in each well of 2.7%, and final concentrations of active compound between 0 and 540 g/ml (23 different concentrations). Four parallels were assayed for each compound and concentration. 0.5 mg of the compound produced by strain M1152matAB (P08-G05_C.sub.16) was purified on preparative HPLC, and the pure compound was tested in bioassay against a panel of strains. The indicator organisms were M. luteus ATCC 9341, Staphylococcus aureus ATCC 29213, Staphylococcus aureus ATCC 43300 (MRSA), Enterococcus faecium CCUG 37832, Enterococcus faecium CTC 492. The results are shown in Table 3 below.

TABLE-US-00003 TABLE 3 MIC70 (g/ml) Nidaromycin Vancomycin MIC.sub.70 S. aureus ATCC 29213: 0.53 1.06 MIC.sub.70 S. aureus ATCC 43300 0.53 1.06 MIC.sub.70: E. faecium CTC 492 8.45 1.06 MIC.sub.70: E. faecium CCUG 37832 2.11 270

In Vitro Toxicity Assay

[0216] Toxicity of nidaromycin against human cell lines was evaluated with the human cell lines HepG2, LLC-PK1 and L929 cultivated in RPMI 1640 supplemented with 10% Fetal Bovine Serum (FBS), 2 mM L-glutamine and 100 U/ml Pen-Strep (HepG2), Medium 199 supplemented with 3% FBS, 2 mM L-Glutamine, 100 U/ml Pen-Strep (LLC-PK1) and Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% FBS, 2 mM L-glutamine, 1 mM Sodium pyruvate and 100 U/ml Pen-Strep (L929). The cells were sub-cultured according to standard protocols, and the cell suspensions was transferred from a stirred reservoir and seeded into 384-well plates (Corning Assay Plate, 3712) using Tecan EVO robotic workstation with MCA384 pipetting unit using disposable tips (Tecan MCA 125 ui, Cat No. 300-5-1-808). The reservoir (flat base, 300 mL, Thermo Scientific, 10723363) was equipped with sterile magnetic stirring bars (154.5 mm VWR 442-4522) stirring at 350 rpm. The number of cells in each well was 50.000 (HepG2), 25.000 (LLC-PK1) and 10.000 (L929). The microplates with cell suspension were shaken at 1600 rpm with 2.5 mm amplitude (Bioshake) for 20 seconds after seeding. The microplates with the cells were incubated at 37 C. with 5% CO2 atmosphere. At the day of the exposure of the cells, serial dilutions were made in DMSO. The serial dilutions with the compounds were further diluted in cell culture medium and transferred to the assay wells, giving a total DMSO concentration in the assay wells of 0.6%. After exposure, the plate was further incubated at 37 C. with 5% CO2 atmosphere for 48 hours. The viability of the cells after incubation for 24 hours and 48 hours was measured using the Promega CellTiter-GLO 2.0 viability assay. The highest concentration tested was 50 g/ml of Nidaromycin, and no toxic effect was detected for this concentration or lower. The data are shown in FIG. 4.

Example 6

Determining the Molecular Formula of Nidaromycin

MS1 Analyses of Isotope Labelled Broth

[0217] To determine the molecular formula of nidaromycin, strain M1152matAB (P08-G05_C.sub.16) was cultivated in isotope labelled medium where all carbon sources were .sup.13C labelled and all nitrogen sources were .sup.15N labelled as follows: 3 ml of seed culture, produced in 0.5 TSB medium, was washed once with ml of sterile 0.9% NaCl, resuspended in 0.9% NaCl to OD=5 and used to inoculate production cultures (0.5% in 20 ml medium). Productions were performed in 250 ml shake flasks with 1.5 g of 3 ml glass beads and 20 ml of the following media from Silante: 1 g/100 ml .sup.13C Silex Media Powder for E. coli (115204100), E. coli OD2N (110301402) or E. coli OD2CN (110601402). The culture was harvested after six days.

[0218] Freeze dried isotope labelled broths were extracted one hour with 2 ml DMSO added trifluoroacetic acid to 0.1% per 20 ml original broth volume The volumetric yield of nidaromycin in these media was very low, but high enough to detect the masses of isotope labelled nidaromycin. The masses of .sup.13C, .sup.15N, and .sup.13C and .sup.15N labelled nidaromycin were 1410.7769 Da, 1351.5562 Da and 1412.7585 Da, respectively, showing that the molecular formula of nidaromycin contains 61 carbons and 2 nitrogens (FIG. 5). Based on MS1 data (FIG. 2), the most likely molecular formula was C.sub.61H.sub.92N.sub.2O.sub.29S giving a theoretical monoisotopic mass of 1348.5507 Da.

MS2 Analysis of Purified Nidaromycin

[0219] Purified nidaromycin was analysed with LC-MSMS with fragmentation of the precursor mass m/z=1349.567. The LC-conditions were the same as described in Example 4, and the MSMS data was generated using a Bruker Impact II QTOF in positive mode. The MS conditions were: Spectra rate: 12 Hz, Capillary voltage: 4500 V, Endplate offset: 500V, drying gas: 10 L/min, Nebulizer gas: 220, Data acquisition control: dynamic MSMS, collision energy 5V with multiCE 20, 50 and 100.

[0220] The molecular mass and the MSMS fragmentation pattern (FIG. 6) suggested that the molecular formula was C.sub.61H.sub.92N.sub.2O.sub.29S and 29 of the MSMS fragments could be explained by this formula (Data not shown).

Example 7

Structure Determination by NMR

[0221] The compound produced by the transconjugant M1152matAB (P08-G05_C16) was subjected to structure determination by 1D and 2D NMR spectroscopy by Red Glead Discovery AB.

[0222] The determined structure is shown in FIG. 7, which also shows the atom numbering for the atom-specific assignments which have been made. The structure consists of four substituted sugar moieties A-D, a linking 2,3-dihydroxypropionic acid (E) and a hydrocarbon moiety with the formula C30H45 (F). There are several alternative positions for the proposed sulphate group, which is shown at position 4 on the uronic acid unit D, that may be valid. The alternative positions are position 3 in unit D and position 4 in unit A, as shown below:

##STR00006##

[0223] The NMR studies made are detailed below.

Sample Information

[0224] The studied sample was provided to ReadGlead as solid material. The material was stored at 20 C. upon reception. Prepared NMR samples were stored dark at 4-8 C. in between measurements. The following sample information was provided: [0225] Sample ID: P08-G05_c16 [0226] Monoisotopic mass: 1348.5604 [0227] Formula sum: C61H92N2O29S [0228] Obtained Amount: 5.13 mg [0229] Solubility: 10 mg/mL in DMSO upon ultrasonication

Material & Methods

NMR samples

PN102-62-01

[0230] The NMR sample was prepared by weighing up 2.899 mg of sample P08-G05_c16 in a screw cap vial and adding 540 L DMSO-d.sub.6. The sample dissolved slowly and was heated to 40 C. for 1-2 minutes and before being ultrasonicated for 310 seconds. The sample still showed traces of finely dispersed undissolved particles, as controlled by visual inspection, but was transferred to a 5 mm NMR tube.

PN102-62-01B

[0231] The NMR sample was prepared by adding 20 L D.sub.2O to the NMR tube of sample PN102-62-01 above.

PN102-62-01C

[0232] The NMR sample was prepared by adding 2 L TFA-d to the NMR tube of sample PN102-62-01B above.

PN102-62-02

[0233] The NMR sample was prepared by adding 540 L CD3OD directly to the Eppendorf tube containing the remains of P08-G05_c16 (approx. 2.2 mg). The sample dissolved slowly and was heated to 40 C. for 1-2 minutes and before being ultrasonicated for 310 seconds. The sample still contained significant amounts of undissolved material, as controlled by visual inspection, but the supernatant was transferred to a 5 mm NMR tube.

Chemicals and Materials

Equipment:

[0234] Mettler Toledo MT5 balance [0235] Bandelin Sonorex ultrasonic bath, model no. RK 31 [0236] Agilent 2 mL clear screw neck vials, Part No. 5190-9062 [0237] Agilent Technologies screw caps, 9 mm with PTFE/Silicone septa, Part No. 5190-9068 [0238] Hilgenberg Standard NMR tubes, 5 mm diameter, Item No. 2001745 [0239] Hilgenberg closing caps for NMR tubes, 5 mm diameter, Item No. 9400312

TABLE-US-00004 NMR solvents and chemicals Name CAS# Lot# Supplier Cat# DMSO-d.sub.6 (99.9 2206- 11578 ARMAR 015200.0009 atom % D) 27-1 Chemicals D.sub.2O (99.9 7789- B 19908 Deutero 00506-25ml atom % D) 20-0 CD.sub.3OD (99.8 811- B 15576 Deutero 01105-25ml atom % D) 98-3 TFA-d (99.5 599- MKCH3593 Aldrich 152005- atom % D) 00-8 10X0.5ML

NMR Spectroscopy

[0240] A 500 MHz Bruker Avance Neo spectrometer equipped with a 5 mm iProbe BBF/H/D probe and a 500 MHz Varian Inova spectrometer equipped with a 5 mm 1H/.sup.13C/.sup.15N triple resonance probe were used for the performed NMR experiments. Data were recorded at 25 C. or 40 C. The recorded spectra are listed in Table 4 below.

TABLE-US-00005 TABLE 4 NMR experiment PN102-62-01 PN102-62-01B PN102-62-01C PN102-62-02 1D .sup.1H X X X X (standard sequence zg30 in TopSpin 4.0.8 with a 30 pulse or standard sequence s2pul in VnmrJ 2.3, 45 pulse; or quantitative measurements with 90 pulse and d1 = 57 s) 1D .sup.13C X X (standard sequence zgpg30 in TopSpin 4.0.8) 1D .sup.31P X X (standard sequence zgpg30 in TopSpin 4.0.8) 2D .sup.1H COSY X X X X (standard version cosygpppqf in TopSpin 4.0.8 or standard version gCOSY in VnmrJ 2.3) 2D .sup.1H TOCSY X X X X (standard version mlevphpp in TopSpin 4.0.8 or standard version zTOCSY in VnmrJ 2.3, mixing time set to 60 or 80 ms) 2D .sup.1H NOESY X X X X (standard version noesygpphpp in TopSpin 4.0.8 or standard version NOESY in VnmrJ 2.3, mixing time set to 200 or 300 ms) 2D .sup.1H ROESY X (standard version roesyphpp.2 in TopSpin 4.0.8, mixing time set to 200 ms) 2D .sup.1H-.sup.13C HSQC X X X X (multiplicity edited: standard version hsqcedetgpsisp2.3 in TopSpin 4.0.8, or in VnmrJ 2.3: gradient version kindly provided by Lewis Kay's NMR laboratory, University of Toronto, Canada; implemented by Yamazaki 1993) 2D .sup.1H-.sup.13C HMBC X X X X (standard version hmbcetgpl3nd in TopSpin 4.0.8 or standard version gHMBC in VnmrJ 2.3, optimized for J = 5, 8 or 10 Hz) 2D .sup.1H-.sup.15N HSQC X (standard version hsqcetgpsi2 in TopSpin 4.0.8) 2D .sup.1H-.sup.15N HMBC X (standard version hmbcgpndqf in TopSpin 4.0.8)

[0241] The solvent residual signals of DMSO-d.sub.6 (2.50/39.52 ppm) and CD.sub.3OD (3.31/49.00 ppm) were used as the reference for .sup.1H and .sup.13C chemical shifts.

[0242] NMR data were processed and analysed using MestreNova 12.0.1 (Mestrelab Research S. L.). Chemical shift predictions were performed with the plug-in NMRPredict in MestReNova 12.0.1, using the predictor Mnova Best with solvent set to DMSO-d.sub.6.

Results and Conclusions

NMR Experiments and Conditions

[0243] The NMR experiments were performed on the obtained material dissolved in DMSO-d.sub.6 or CD.sub.3OD. NMR data have been recorded on a 500 MHz Bruker Avance NMR spectrometer and a 500 MHz Varian Inova spectrometer.

[0244] 1D and 2D .sup.1H/.sup.13C/.sup.15N NMR spectral data of moderate quality were acquired for the provided sample material in DMSO-de. A significant number of low-intensity signals were observed, possibly referring to structurally related impurities and/or minor conformers of the main species in solution. The spectral region of the sugar moieties was complicated by signal overlapping and signal broadening, to the extent that .sup.1H-.sup.13C HSQC cross-peaks in a few cases were not readily observable. Heating the DMSO-de sample to 40 C. resulted in no or only very minute sharpening of the signals and thus all data used in the structure elucidation were recorded at 25 C. Acidification of the DMSO-de sample with TFA-d, however, resulted in significant sharpening of some signals, and also chemical shift changes of the sugars and sugar derivatives (while the chemical shifts of the hydrocarbon tail remained essentially unaffected).

[0245] The overall appearance of the CD.sub.3OD spectra were slightly clearer and sharper than the DMSO-d.sub.6 spectra. However, due to limitations in solubility, the signal-to-noise ratio intensity in methanol was too poor to achieve useful 2D long-range and through-space NMR data, critical for the structure identification process. The CD.sub.3OD data set could still provide useful information in a few cases where DMSO-de data were not unambiguous.

[0246] In all, four different NMR data sets were used for the structure elucidation. For completeness, the chemical shift assignment of the suggested structure is reported for both the DMSO-de sample (PN102-62-01) and the DMSO-de/TFA-d sample (PN102-62-01C), except for the hydrocarbon tail F where chemical shifts are very alike in the two samples. Besides the main compound, the sample also appears to contain significant amount of an unassigned small molecule.

Elucidation of Chemical Structure and Atom Specific Assignment

[0247] The suggested structure of P08-G05_c16 (FIG. 7) shares several structural features with related moenomycin compounds in that they all contain a substituted tetrasaccharide linked to a hydrocarbon tail. However, as opposed to the moenomycins, P08-G05_c16 lacks the linking phosphodiester and the hydrocarbon tail contains 30 instead of 25 carbon atoms. Structural evidence is strong for P08-G05_c16 as essentially all 1H/.sup.13C/.sup.15N atoms are observed and assigned, except for the amide nitrogen of sugar unit C and the carbonyl carbon of the 2,3-dihydroxypropionic acid unit E, and the atom connectivity is fully consistent with the obtained 2D data. Further, a quite good degree of agreement is observed between the experimental and predicted chemical shift values.

[0248] The sugar units denoted A and D are assigned as uronic acids and B and C as N-acetyl-glucosamines. Stereospecific assignments are not included in the present study. The connectivity of the sugar moieties was determined by the correlations between the anomeric proton signals and the corresponding carbon signals (C-4 of sugars B and C and C-2 of sugar D) through glycoside bonds in the HMBC spectra and/or NOE correlations between the anomeric proton signal and the corresponding proton in the next sugar moiety. The connectivity of sugar D and the linking 2,3-dihydroxypropionic acid E is confirmed by NOE correlations between the anomeric proton of D and the methylene protons of E (only observed for the acidified sample PN102-62-01C). The connectivity between E and the hydrocarbon tail F is also established through observed NOEs, between both the CH and CH2 protons of E and the two closest CH protons of F (atom no. 47 and 48 in Table 6).

[0249] The assigned chemical shifts for P08-G05_c16 are presented in Tables 5 and 6 below, along with the corresponding chemical shifts predicted from the proposed chemical structure. The overall picture is that predicted chemical shifts values fully support the suggested molecular structure, and that minor deviations from the predicted values are only observed for the D moiety.

[0250] Based on the suggested formula sum, a sulphate substituent on one of the sugar oxygens has been postulated. There are several available positions for this, i.e. sugar positions where the OH proton is not detected, and no other substituent/linkage is determined. As the .sup.13C chemical shifts for C-3 and C-6 are very similar in moiety B as compared to moiety C, it is deemed unlikely that any of those positions should bear a sulphate. Thus, the positions C-4 of unit A, C-3 of unit D and C-4 of unit D are left as plausible candidates. The position of the sulphate cannot be defined solely based on the NMR data. However, as ring system D is more sensitive to changes in pH and also appears to have a higher structural complexity than system A due to deviations from predicted chemical shift values, it is considered the more plausible option. O-sulphation can be expected to cause slightly downfield shifts for both the O-sulphated carbon and the proton bound to it, and for this reason position C-4 of unit D has been tentatively assigned.

Table 5

[0251] Chemical shift (o) values for P08-G05_c16, part A-D in FIG. 8, predicted with Mnova Predict and experimentally determined in DMSO-d.sub.6 (NMR sample PN102-62-01) and DMSO-d.sub.6 after addition of D.sub.2O and TFA-d (NMR sample PN102-62-01C). Experimental shifts are reported relative the solvent residual signal for 1H/.sup.13C (2.50/39.52 ppm) while indirect referencing is applied for .sup.15N. Data are recorded at C. NO=not observed.

TABLE-US-00006 Atom Predicted DMSO-d.sub.6, DMSO-d.sub.6/TFA, no. .sup.1H/.sup.13C [ppm] .sup.1H/.sup.13C [ppm] .sup.1H/.sup.13C [ppm] 2 3.84/75.6 3.18/75.1 3.22/75.0 3 3.94/79.0 3.27/81.3 3.22/81.5 4 4.11/72.3 3.52/72.2 3.50/72.2 5 3.74/55.4 3.45/55.1 3.50/54.7 6 5.27/101.3 4.51/101.7 4.53/101.8 8 4.02/79.3 3.75/NO 3.84/72.8 9 5.04/101.9 5.67/NO 5.77/95.4 10 3.97/75.9 4.39/NO 4.58/67.1 11 5.22/75.8 3.86/NO 3.92/70.5 12 4.09/73.1 3.71/NO 3.81/68.7 15 .sup.7.57/87.8 (.sup.15N) 7.94/NO NO/NO 16 172.3 169.7 169.3 17 2.02/23.2 1.82/23.0 1.82/22.8 19 3.77, 3.91/61.6.sup. 3.43, 3.59/60.0.sup. 3.43, 3.60/60.2.sup. 21 4.19 4.67 NO 22 5.24/101.6 4.37/101.9 4.36/101.8 23 4.11/72.4 3.49/72.3 3.46/71.8 24 3.98/79.8 3.30/80.3 3.35/79.9 25 3.86/75.9 3.31/75.2 3.35/74.9 27 3.75/55.2 3.51/54.9 3.48/54.8 28 4.57 NO NO 29 .sup.7.57/87.9 (.sup.15N) .sup.7.88/118.4 (.sup.15N) NO/NO 30 172.4 168.9 169.8 31 2.02/23.3 1.82/23.0 1.81/22.9 33 5.18/103.2 4.23/102.8 4.37/102.9 35 3.77/76.0 3.33/73.4 3.72/75.0 36 3.61/72.6 3.12/72.0 3.28/71.2 37 3.85/76.0 3.14/76.4 3.18/75.7 38 3.81/74.6 2.99/73.1 3.03/72.8 39 4.40 5.17 NO 40 4.72 4.98 NO 41 5.20 NO NO 42 173.4 NO 170.0 43 4.13 NO NO 46 3.82, 3.89/61.5.sup. 3.53, 3.81/60.2.sup. 3.52, 3.79/59.8.sup. 77 172.1 NO 170.3

[0252] Amide nitrogen no. 15 in sugar C is not observed from the recorded 1H-.sup.15N HSQC data. This is an expected result due to the observed broadening of the associated amide proton signal in the 1D 1H spectral data. Nonetheless, the structural assignment of the acetamido sugar C is very likely, due to the characteristic chemical shifts of both the amide proton and the adjacent C-2 carbon, along the with the overall conformity between moieties B and C and the anticipated formula sum. Similarly, the carbonyl carbon no. 85 in moiety E is not observed from the recorded .sup.1H-.sup.13C HMBC data, as would be expected due to the broadening observed for the adjacent methine no. 75. The suggested substructure of E is still probable, due to the close agreement between predicted and experimental .sup.13C chemical shift values for methine no. 75, the known presence of this structural motif in related moenomycins and the overall formula sum. The high number of quaternary carbon atoms and the splitting of the methylene proton signals observed for fragment F accounts for the cyclic subunits, that also agrees with the total number of cycles and double bonds expected for the suggested formula sum. The structural unit F significantly deviates from the moenomycin structure(s) and has not been evaluated from a biosynthetic point of view.

[0253] The structural moieties D and E display significantly broadened 1H signals in DMSO-d.sub.6, with crosspeaks in .sup.1H-.sup.13C HSQC data broadened beyond recognition. The signals are sharpened upon addition of TFA-d to the sample, allowing for .sup.13C chemical shift assignments, but data then also reveals a tendency for doubling of these signalsthe sharpening effect in this region are consistent with the presence of the carboxylic acid moieties being protonated upon acidification. The origin behind these observations have not been explored within the present study and only the major signals have been evaluated in the structure elucidation.

Table 6

[0254] Chemical shift () values for P08-G05_c16, part E-F in FIG. 8, predicted with Mnova Predict and experimentally determined in DMSO-de (NMR sample PN102-62-01). * Shifts reported in DMSO-de after addition of D.sub.20 and TFA-d (NMR sample PN102-62-01C). Experimental shifts are reported relative the solvent residual signal for 1H/.sup.13C (2.50/39.52 ppm) while indirect referencing is applied for .sup.15N. Data are recorded at 25 C. NO=not observed.

TABLE-US-00007 Atom Predicted DMSO-d.sub.6, no. .sup.1H/.sup.13C [ppm] .sup.1H/.sup.13C [ppm] 47 6.58/147.7 6.51/148.6 48 5.19/118.4 5.82/105.3 49 135.4 130.2 50 5.44/127.7 4.95/121.5 51 1.80/14.2 1.68/20.8 52 2.14, 2.20/36.0.sup. 2.13, 2.21/35.3.sup. 53 43.0 41.0 54 150.2 149.1 55 1.81/37.4 1.71/34.5 56 2.08, 2.12/33.0.sup. 1.84, 2.42/31.2.sup. 57 5.63/129.1 5.66/123.5 58 139.5 135.8 59 0.95/20.3 0.93/21.9 60 0.91/15.9 0.74/16.0 61 5.30/113.5 4.89, 5.03/109.7 62 6.23/127.5 5.89/125.7 63 5.80/139.1 5.78/138.7 64 50.4 49.4 65 43.6 45.3 66 2.07/45.9 1.70/47.7 67 0.87/22.3 0.70/18.5 68 0.92/22.3 0.77/22.5 69 1.75, 2.25/28.5.sup. 1.74, 2.01/29.2.sup. 70 5.02/124.8 5.09/124.3 71 133.3 130.3 72 1.62/23.4 1.64/25.6 73 1.62/20.2 1.56/17.7 75 4.55/74.2 4.38/NO 75* 4.55/74.2 4.68/77.4* 76 3.97, 3.99/66.2.sup. 3.94, 4.04/NO.sup. 76* 3.97, 3.99/66.2.sup. 4.16, 4.27/65.8* 85 172.0 NO

Example 8

Determining the Position of the Sulphate Group in Nidaromycin

BACKGROUND

[0255] ReadGlead has determined the structure of Nidaromycin (the active compound produced by P08-G05_c16). However, there were some uncertainties regarding where the sulphate group (SO4-group) is. Here we used MSMS fragmentation followed by in silico fragmentation with the aim of determine the position of the SO4-group in Nidaromycin.

[0256] Based on the suggested formula sum, a sulphate substituent on one of the sugar oxygens has been postulated. There are several available positions for this, i.e. sugar positions where the OH proton is not detected, and no other substituent/linkage is determined. As the .sup.13C chemical shifts for C-3 and C-6 are very similar in moiety B as compared to moiety C, it is deemed unlikely that any of those positions should bear a sulphate. Thus, the positions C-4 of unit A, C-3 of unit D and C-4 of unit D are left as plausible candidates. The position of the sulphate cannot be defined solely based on the NMR data. However, as ring system D is more sensitive to changes in pH and also appears to have a higher structural complexity than system A due to deviations from predicted chemical shift values, it is considered the more plausible option. O-sulphation can be expected to cause slightly downfield shifts for both the O-sulphated carbon and the proton bound to it, and for this reason position C-4 of unit D has been tentatively assigned.

[0257] The three possible structures are given with SMILES:

TABLE-US-00008 NidaromycinA4 C/C(C)C\CC1CCC(C)(/CC/C2CCC(C)C(C)(C\CC(\C)/ CC/OC(COC3OC(C(O)O)C(O)C(O)C3OC3OC(CO)C(OC4OC (CO)C(OC5OC(C(O)O)C(OS(O)(O)O)C(O)C5O)C(O)C4/ NC(\C)O)C(O)C3\NC(/C)O)C(O)O)C2C)C1(C)C NidaromycinD3. CC(CCC1/CC/C2(C)C(C)(C)C(CCC(C)C)CC2)C(C)(C/ CC(\C)/CC/OC(COC(C(C(C2O)OS(O)(O)O)OC(C(C3O) NC(C)O)OC(CO)C3OC(C(C3O)NC(C)O)OC(CO)C3OC(C(C (C3O)O)O)OC3C(O)O)OC2C(O)O)C(O)O)C1C NidaromycinD4: CC1CCC(\CC\C2(C)CCC(CCC(C)C)C2(C)C)C(C)C1 (C)C\CC(/C)\CC\OC(COC1OC(C(OS(O)(O)O)C(O) C1OC1OC(CO)C(OC2OC(CO)C(OC3OC(C(O)C(O)C3O) C(O)O)C(O)C2NC(C)O)C(O)C1NC(C)O)C(O)O)C(O)O

Materials and Methods:

[0258] LC-MS-method: Cell free extracts were analyzed using an Agilent LC-DAD system connected to a Bruker Impact II QTOF. The LC was run with 10 mM ammonium acetate buffer [mobile phase A] and 90:10 acetonitrile:water with 10 mM ammonium acetate [mobile phase B]. The gradient was 5% B for 2 min, then 5-100% B for 2-25 min. The MS was performed at electrospray ionization at positive mode with the following MS parameters: Mass range 100-1800, Spectra rate: 12 Hz, Absolute threshold: 25 counts, threshold for fragmentation: 100 counts, capillary voltage: 4500V, endplate offset: 500V, drying gas: 10 L/min, Nebulizer: 31.9 psi, Drying temperature: 220 C., Precursor ion list: 1000-1500, Data acquisition control: Dynamic MSMS or Fixed MSMS, collision energy: 5V, CID: acqCtr+MultiCe, MultiCe20. In silico fragmentation. In silico fragmentation was performed using MetFrag (Schymanski et al., 2015 Anal Bioanal Chem 407 (21): 6237-6255).

Results:

[0259] Investigating the fragments: The in silico fragmentation strongly suggest that the SO4-group is positioned at either D3 or D4 and not A4. As shown in Table 7, there are several fragments that should not be formed if the SO4-group is positioned in A4. Further, it is difficult to distinguish between D3 and D4 since these structures in general will give the same fragments. In addition, the SO4-group is often lost in the fragmentation, and only a few low abundant fragments contain the SO4-group. However, we observe one fragment (M+H=398.1976) that can be explained by D3, but not by D4. This is an indication that the SO4-group is positioned in D3, but this is supported by only one matching fragment.

[0260] To support the QTOF-data, data from FT-ICR MSMS fragmentation at negative ionization was inspected. Not even the FT-ICR fragmentation pattern could distinguish between the position D3 and D4. However, several fragments could only be explained by sulphate in either position D3 or D4 and not by position A4 (Table 5).

Table 7

[0261] Fragments obtained with MSMS fragmentation using Bruker Impact II QTOF were compared with in silico fragmentation of the three suggested structures. Several of the fragments could not be explained by the structure with the SO4-group in the A4 position.

TABLE-US-00009 Mass and formula A4 D3 D4 380.1185Da [C14H22NO11]+ [00007] [00008] [00009] 398.198Da [C20H28O8+H]+H+ ND [00010] ND 429.1454Da [C15H24NO13+2H]+H+ ND [00011] [00012] 759.2290 [C28H42N2O22]+H+ ND [00013] [00014] [00015]

Table 8

[0262] Fragments obtained with MSMS fragmentation using Bruker FT-ICR were compared with in silico fragmentation of the three suggested structures. The fragments shown here could not be explained by the structure with the SO4-group in the A4 position.

TABLE-US-00010 Mass and formula A4 D3 D4 [C19H25O12S]H 476.09936 ND [00016] [00017] [C40H57NO14S+2H] 809.3678 ND [00018] [00019] [C42H61NO18S+2H] 901.37914 ND [00020] [00021] [00022]