Fig 1.
Tilivallin (1) biosynthetic gene cluster (BGC) composition in (A) Klebsiella oxytoca sp. and Xenorhabdus beddingii and (B) additional Xenorhabdus strains.
The TV BGC contains the aroX operon and encodes a bimodular NRPS (grey arrows; in Klebsiella sp. NpsAB and ThdA, in Xenorhabdus sp. XtvAB) with corresponding domain architecture annotation according to antiSMASH analysis and [28]). Note the differences in NRPS architecture in (A, freestanding) and (B, fused A-T didomain). Abbreviations: A (adenylation), T (thiolation), C (condensation), Re (reductase) domain. Colored arrows show genes of the aroX operon encoding the following enzymes: Orange (HmoX, 4-hydroxyphenyl acetate-3-monoxygenase), violet (AdsX, 2-amino-2-deoxy-isochorismate synthase), red (IcmX or PhzD homolog, respectively, isochorismatase), blue (DhbX, 2,3-dihydro-dehydrogenase), green (AroX, 2-keto-3-deoxy-D-arabinoheptulosonate phosphate synthase), white (2,3-dihydroxybenzoate-AMP ligase of chrysobactin BGC). Arrows with shaded colors and asterisk indicate that corresponding genes are not encoded in TV BGC but elsewhere in the genome.
Fig 2.
UPLC-HRMS/MS analysis of crude extracts of X. eapokensis grown in SF-900 medium.
(A) Extracted ion chromatograms (EIC) of natural produced TV (1), and derivatives thereof (2–5) with structures according to (C). (B) EICs of 1 and derivatives thereof (2, 6–8, 10–15) with structures according to (D) of precursor directed feeding experiment (details see text). Note: 1* shows TV signal intensity in an extract without supplementation of 3-hydroxy anthranilic acid.
Table 1.
Overview of TV, derivatives thereof and putative biosynthetic precursors described in this work.
Fig 3.
TV specific subnetwork analysis of a X. eapokensis cultivation in SF-900 insect medium.
Nodes (circles) are labeled with precursor ion masses [M+H]+ of TV (1) and derivatives thereof (2–5). Precursor mass differences are depicted on connecting edges (grey lines).
Fig 4.
UPLC-HRMS/MS analysis of a crude extract of X. eapokensis grown in SF-900 medium.
(A) MS/MS spectrum of 9-deoxy-TV (2). (B) MS/MS spectrum of dihydroxy-TV (3). Signals are annotated with predicted sum formulas, detection errors and putative molecule fragment structures. Red diamonds indicate precursor ions.
Fig 5.
UPLC-HRMS/MS analysis of a crude extract of induced E. coli DH10B entD::mtaA carrying pHW10 (containing xtvAB from X. indica) grown in LB medium supplemented with different ABAs according to S3 Table.
(A) Extracted ion chromatograms of TV (1), 9-deoxy TV (2) and TV analogs (6–10 and 12–14) with substitution patterns listed in (B). Identified ions and predicted sum formulas are listed in (Table 1), MS/MS data is shown in (Fig 4 and S3 Fig). Control C1 (non-induced E. coli DH10B entD::mtaA pHW10 with 1 mM 3-hydroxy anthranilic acid) and C2 (induced E. coli DH10B entD::mtaA pHW10 without supplementation).
Fig 6.
Proposed tilivallin (1) biosynthesis of X. eapokensis.
Chorismate is a main product of the shikimic acid pathway modified by various enzymes to precursors accepted by XtvA. NRPS enzymes XtvA and XtvB activate anthranilic acid or proline (and analogs thereof), respectively, due to ATP hydrolysis and bind resulting aminoacyl adenylates to corresponding thiolation domains (T). After condensation domain (C) mediated reaction of the T domain bound aminoacyls, resulting dipeptide undergoes a reductive cyclisation by action of the terminal reduction domain (Re) release mechanism resulting a pyrrolobenzodiazepine core structure (16, 17). Finally post NRPS, free indole (or derivatives thereof) is added non enzymatically, resulting in the production of 1. Compound abbreviations: ADIC = 2-amino-4-deoxy isochorismate, DHHA = 6-amino-5-hydroxycyclohexa-1,3-diene-1-carboxylate, 3-HAA = 3-hydroxyanthranilic acid, 5-HAA = 5-hydroxyanthranilic acid, 3,5-DHAA = 3,5-dihydroxyanthranilic acid, 4,5-DHAA = 4,5-dihydroxyanthranilic acid. Enzyme abbreviations: AdsX = ADIC synthase, PhzD = ADIC hydrolase, DhbX = DHHA dehydrogenase, TomO = salicylyl-coenzyme A 5-hydroxylase (anthranlilic acid oxygenase), TomEF = phenol-2-monooxygenase reductase/oxygenase, XtvAB = TV NRPS module 1 and 2.