Hydroxamate formation in the lipopeptide containing siderophore taiwachelin Authors: Aaron J. Gringer, Kernen P. Agwaza, Isabel Da Fonseca, Pablo Sobrado Published: 2026 ABSTRACT Cupriavidus taiwanensis is a nitrogen-fixing bacterium found in root nodules of the invasive tropical weed Mimosa pudica. C. taiwanensis has been shown to grow in environments contaminated with heavy metals such as Pb, Cu, and Cd. Taiwachelin, a recently identified siderophore produced by C. taiwanensis, is an iron chelator that contains a hydroxamate functional group for metal binding. In its biosynthesis, a flavin-dependent N-monooxygenase (NMO) was predicted to hydroxylate l-ornithine, leading to formation of the metal binding hydroxamate moiety. Here, we report the cloning, expression, and biochemical characterization of this enzyme, herein referred to as CtNMO. CtNMO was isolated with bound FAD and catalyzed the hydroxylation of l-ornithine with a kcat of 0.72 ± 0.04 s−1 and a KM of 156 ± 26 μM. A ∼25-fold preference for NADPH over NADH was observed. Expression of CtNMO as a fusion to maltose-binding protein resulted in enhanced enzymatic activity. These results validate CtNMO as an efficient l-ornithine hydroxylase and establish its potential utility in future applications, including bioremediation. Introduction The nitrogen-fixing bacterium Cupriavidus taiwanensis is found in root nodules of the legume Mimosa pudica. M. pudica, commonly known as the “touch-me-not”, is characterized by its leaves closing upon mechanical stimulation and is a widespread sub-tropical invasive species [[1], [2], [3]]. C. taiwanensis converts atmospheric dinitrogen into ammonia, which is subsequently assimilated by M. pudica host in its nitrogen metabolism. In this symbiotic relationship, M. pudica provides essential nutrients to C. taiwanensis. A significant nutrient is iron, which is required for enzymes whose activity depends on complex metallocofactors, including iron-sulfur clusters and molybdenum-iron cofactor [4,5]. To meet the requirement for iron demand by several iron-dependent metabolic enzymes, C. taiwanensis synthetizes and secretes an iron chelating molecule known as siderophores, which scavenge iron from M. pudica nodules [6]. The structure of the C. taiwanensis siderophore, taiwachelin, has been determined. Taiwachelin is a lipopeptide consisting of a dodecanoate moiety linked to hydroxy-aspartate, threonine, serine, 3-hydroxybutanoyl, and N5-hydroxy-l-ornithine residues (Fig. 1) [7]. The gene cluster responsible for taiwachelin biosynthesis contain genes that code for nonribosomal peptide synthetases (NRPSs) and a flavin-dependent N-monooxygenase (NMO) enzyme predicted to catalyze the formation of N5-hydroxy-L-ornithine. NMOs are a specialized class of flavoenzymes that catalyze the regio- and chemoselective hydroxylation of aminecontaining-substrates using molecular oxygen and reduced flavin cofactors [8,9]. These enzymes typically generate a reactive C4a-hydroperoxyflavin intermediate that enables oxygen transfer to nucleophilic nitrogen atoms, most commonly converting primary amines such as ornithine or lysine into their corresponding N-hydroxylated- derivatives [10,11].Fig. 1(A) Reaction catalyzed by CtNMO. B) Structure of the siderophore taiwachelin highlighting the CtNMO generated hydroxylated ornithines in blue. (A) Reaction catalyzed by CtNMO. B) Structure of the siderophore taiwachelin highlighting the CtNMO generated hydroxylated ornithines in blue. C. taiwanensis NMO (CtNMO) is predicted to play an essential role in taiwachelin function as it forms the hydroxamate moiety responsible for iron binding (Fig. 1). Here we present the cloning, expression, and purification of CtNMO. The enzyme was expressed as a fusion to maltose binding protein (MBP) and the recombinant enzyme isolated with and without MBP. The data shows that CtNMO is an effective ornithine hydroxylase, in addition we show that activity of the protein as a fusion to MBP is higher compared to the cleaved enzyme. Methods Cloning of C. taiwanensis NMO in pVP56K The gene coding for CtNMO was synthetized by GenScript and cloned into pUC19. For protein expression, the gene was subcloned into pVP56K in frame for expression as a fusion to maltose binding protein (MBP) containing an N- terminus 8x-His tag as done previously [12]. The pVP56KHis8-MBP-CtNMO was sequenced to ensure that no unwanted mutations were incorporated during the cloning process. Expression and purification The pVP56Khis8-mbp-Ctnmo was transformed into E. coli BL21 and plated onto agar plates supplemented with Kanamycin (25 μg/mL) and incubated at 37 °C overnight. A single colony was used to inoculate a 60 mL Luria-Broth (LB) containing Kanamycin (25 μg/mL), which was incubated for ∼16 h at 37 °C. The next day, six 1 L flasks containing autoinduction media and Kanamycin (25 μg/mL), were each inoculated with 10 mL of the cell culture [13,14]. The flasks were incubated at 37 °C with agitation set at 250 rpm until the optical density at 600 nm reached a value ∼ 3, at which point the temperature was decreased to 18 °C and incubated overnight. Cells were harvested by centrifugation at 5,000xg for 20 min and the pellet stored at −80 °C. For purification, the cell pellet (∼60 gr) was resuspended in 300 mL of 50 mM NaPO4, 300 mM NaCl, pH 7.5 buffer (Buffer A), supplemented with 25 μg/mL of lysozyme, DNase, and RNase, and sonicated for 10 min (5 on 15 s off). The lysate was centrifuged at 20,000xg for 60 min. The resulting supernatant was loaded onto an immobilized metal affinity chromatography (IMAC) using an AKTA prime system. The column (10 mL) was washed with 50 mL of Buffer A and the bound protein eluted with a 50 mL of 50 mM sodium phosphate, 300 mM NaCl, 200 mM Imidazole, pH 7.5. The yellow fractions containing His8-MBP-CtNMO were pooled. To half of the sample, 5 mg of tobacco etch virus (His8-TEV) protease was added and incubated overtight to remove the His8-MBP [15]. The solution was dialyzed into Buffer A and then loaded unto an IMAC column (10 mL) equilibrated with Buffer A. In this step the His8-MBP and His8-TEV bind to the column, while the free CtNMO does not and is present in the flowthrough, which was collected and concentrated. Aliquots of 50 μL were frozen in liquid nitrogen and stored at −80 °C. The His-MBP–CtNMO (referred to as MBP-CtNMO) fusion protein was generated by bypassing the TEV protease cleavage step during purification. Bioinformatic analysis A Multiple Sequence Alignment (MSA) of six enzymes with high percent identity to CtFMO (37.89% to 59.77%) was created using the Clustal Omega web server of the European Bioinformatics Institute (EMBL-EBI). The MSA file was visualized using ESpript3 server. A 3-D model for CtFMO with bound FAD and NADPH was generated using the Aphafold3 web server with the l-ornithine N5-monooxygenase (SidA) from A. fumigatus serving as the template [10,16]. Prediction of the binding residues for CtFMO was done by superimposing the generated CtFMO model unto experimentally solved structure of SidA (PDB code: 4B63) using PYMOL software [17]. Oxygen consumption assay The activity of CtNMO was monitored by measuring the oxygen consumption during catalysis using a Hansatech Oxygraph (Norfolk, UK). Assays were done in 1 mL, 100 mM sodium phosphate, pH 7.5 under constant stirring at 25 °C. When NADPH or NADH were varied l-ornithine was kept constant at 5 mM. When l-ornithine was varied NAD(P)H was kept constant at 1 mM. The reaction was initiated by addition of enzyme (0.5 μM) and monitored for 60 s [18]. Product formation assay Assays were performed in 96 well plates and the final volume of each reaction was 104 μL in 100 mM sodium phosphate, pH 7.5. The reaction was started by addition of 2 μM CtNMO, and shaken for 10 min. The reaction was stopped by addition of 52.5 μL of 2 N perchloric acid. A volume of 47.5 μL was removed from each well and transfer to a clean 96 well plate. An additional 47.5 μL each of 10% sodium acetate and 1% sulfanilic acid are added. 19 μL of potassium iodine (1.3 w/v, in glacial acetic acid) was added and incubated on a shake plate for 15 min at 25 °C. Excess iodine was removed by addition of 19 μL of 0.1 N sodium thiosulfate. To create a chromophophore, 19 μL of 0.6% ⍺-naphthylamine was added and the absorbance measured at 562 nm in a SpectraMax M5e plate reader (Molecular Devices, San Jose, CA). A hydroxylamine hydrochloride standard curve was used to calculate the amount of hydroxylated product produced [[19], [20], [21], [22], [23]]. Results Protein expression and purification CtNMO was expressed as a fusion to His8-MBP in E. coli BL21T1 cells using autoinduction media. Generally, from 6 L of growth media ∼60 g of cells where obtained. Purification was done using an IMAC, following procedures previously published for related enzymes [19,[24], [25], [26]]. After elution, the samples were analyzed by SDS-PAGE and those fractions containing purified MBP-CtNMO were pooled. Half of the pooled sample was dialyzed, concentrated, and stored at −80 °C. The remaining half was combined with 5 mg of TEV to free CtNMO from MBP. Following a second IMAC step, isolated pure CtNMO was collected, concentrated, and stored. The UV-vis spectrum of purified CtNMO had peaks at 375 nm and 450 nm characteristic of flavin cofactors (Fig. S1). The protein concentration was calculated by measuring the absorbance at 450 nm using the extinction coefficient (ε) = 11.30 mM−1 cm−1. A yield of CtNMO of 5 mg/L and MBP-CtNMO of 12 mg/L was obtained. Kinetic characterization The steady-state kinetic parameters were calculated for CtNMO as a function of increasing concentration of l-ornithine at constant concentration of NAD(P)H and vice versa. Enzymatic activity was measured by monitoring both oxygen consumption and the formation of N5-hydroxy-l-ornithine. By comparing the kcat values obtained from these assays, it is possible to calculate the percent coupling of the reaction; that is, the amount of product formed per molecule of oxygen consumed. The results from the oxygen consumption assays are shown in Fig. 2 and the data on Table 1. The kcat value is less than 1 s−1, which is consistent with values reported for related enzymes [12,19,27]. Similarly, the KM value for l-ornithine and the enzyme's preference for NADPH over NADH are comparable to those reported for other NMOs [24,[27], [28], [29]].Fig. 2(A) Oxygen consumption of CtNMO as a function of l-ornithine concentration, keeping NADH (Black) or NADPH (Blue) at saturating concentrations (1 mM) t, and MBP-CtNMO (red) at saturating concentration of NADPH (1 mM). (B) Oxygen consumption of CtNMO as a function of NADPH (red) and NADH concentrations (black), keeping l-ornithine constant.Table 1Steady-state kinetic parameters determined by following the rate of oxygen consumption.a.Table 1Substratekcat (s−1)Km (μΜ)kcat/Km (M−1s−1)Ornithineb0.725 ± 0.036156 ± 264650 ± 810Ornithine-MBP-CtNMOb1.55 ± 0.0568 ± 9.022700 ± 3000Ornithinec0.410 ± 0.00854 ± 7.07550 ± 960NADPHd0.793 ± 0.01330 ± 2.026800 ± 2200NADHd1.01 ± 0.023865 ± 581170 ± 82aConditions: 100 mM sodium phosphate at pH 7.5 and 25 °C.bAssay was done with 1 mM NADPH.cAssay was done with 1 mM NADH.dOrnithine was kept constant at 1 mM. (A) Oxygen consumption of CtNMO as a function of l-ornithine concentration, keeping NADH (Black) or NADPH (Blue) at saturating concentrations (1 mM) t, and MBP-CtNMO (red) at saturating concentration of NADPH (1 mM). (B) Oxygen consumption of CtNMO as a function of NADPH (red) and NADH concentrations (black), keeping l-ornithine constant. Table 1Steady-state kinetic parameters determined by following the rate of oxygen consumption.a.Table 1Substratekcat (s−1)Km (μΜ)kcat/Km (M−1s−1)Ornithineb0.725 ± 0.036156 ± 264650 ± 810Ornithine-MBP-CtNMOb1.55 ± 0.0568 ± 9.022700 ± 3000Ornithinec0.410 ± 0.00854 ± 7.07550 ± 960NADPHd0.793 ± 0.01330 ± 2.026800 ± 2200NADHd1.01 ± 0.023865 ± 581170 ± 82aConditions: 100 mM sodium phosphate at pH 7.5 and 25 °C.bAssay was done with 1 mM NADPH.cAssay was done with 1 mM NADH.dOrnithine was kept constant at 1 mM. Steady-state kinetic parameters determined by following the rate of oxygen consumption.a. Conditions: 100 mM sodium phosphate at pH 7.5 and 25 °C. Assay was done with 1 mM NADPH. Assay was done with 1 mM NADH. Ornithine was kept constant at 1 mM. Surprisingly, the MBP–CtNMO construct exhibited, as a function of l-ornithine concentration, an increased kcat and a decreased KM. Together, these changes result in an approximately threefold increase in catalytic efficiency (kcat/KM) relative to the untagged enzyme (Table 1 and Fig. 2). Enzymatic activity monitored by the formation of hydroxylated l-ornithine is shown in Fig. 3 and summarized in Table 2. The activity measured as a function of l-ornithine concentration is approximately 14% lower than that obtained from the oxygen consumption assay. This difference indicates that the reaction is ∼86% coupled. The apparent affinity for l-ornithine determined from the product-formation assay is higher than that measured using the oxygen consumption assay. This discrepancy is likely an assay-dependent artifact and has been reported previously for similar systems [11,12,30].Fig. 3(A) Product formation assays of CtNMO as a function of l-ornithine with NADPH (MBP) (black), or NADH (red) constant, and MBP-CtNMO (blue) with NADPH constant. (B) NADPH (blue) and NADH (black) substrate hydroxylation of CtNMO hydroxylation with l-ornithine constant.Table 2Steady-state kinetic parameters determined by following the rate of product formation.a.Table 2Substratekcat (s−1)Km (μΜ)kcat/Km (M−1s−1)Ki (mM)Ornithineb0.623 ± 0.0331000 ± 190623 ± 12392.0 ± 1.80Ornithine-(MBP-CtNMO)b1.20 ± 0.100976 ± 2321230 ± 31033.5 ± 9.30Ornithinec0.160 ± 0.005404 ± 63.9395 ± 63.684.0 ± 9.68NADPHd0.765 ± 0.275711 ± 3831080 ± 7002.09 ± 1.45NADHd0.579 ± 0.1492890 ± 104201 ± 87.02.45 ± 0.93aConditions: 100 mM sodium phosphate at pH 7.5 and 25 °C w/2μΜ substrate.bAssay was done using 1 mM NADPH.cAssay was done using 1 mM NADH.dOrnithine was kept constant at 1 mM. (A) Product formation assays of CtNMO as a function of l-ornithine with NADPH (MBP) (black), or NADH (red) constant, and MBP-CtNMO (blue) with NADPH constant. (B) NADPH (blue) and NADH (black) substrate hydroxylation of CtNMO hydroxylation with l-ornithine constant. Table 2Steady-state kinetic parameters determined by following the rate of product formation.a.Table 2Substratekcat (s−1)Km (μΜ)kcat/Km (M−1s−1)Ki (mM)Ornithineb0.623 ± 0.0331000 ± 190623 ± 12392.0 ± 1.80Ornithine-(MBP-CtNMO)b1.20 ± 0.100976 ± 2321230 ± 31033.5 ± 9.30Ornithinec0.160 ± 0.005404 ± 63.9395 ± 63.684.0 ± 9.68NADPHd0.765 ± 0.275711 ± 3831080 ± 7002.09 ± 1.45NADHd0.579 ± 0.1492890 ± 104201 ± 87.02.45 ± 0.93aConditions: 100 mM sodium phosphate at pH 7.5 and 25 °C w/2μΜ substrate.bAssay was done using 1 mM NADPH.cAssay was done using 1 mM NADH.dOrnithine was kept constant at 1 mM. Steady-state kinetic parameters determined by following the rate of product formation.a. Conditions: 100 mM sodium phosphate at pH 7.5 and 25 °C w/2μΜ substrate. Assay was done using 1 mM NADPH. Assay was done using 1 mM NADH. Ornithine was kept constant at 1 mM. Substrate inhibition is observed when the concentration of NAD(P)H is increased, presumably due to excess NADPH binding to the already reduced enzyme. We also observed substrate inhibition at high l-ornithine concentrations; however, this effect is relatively modest compared to that seen with NAD(P)H. Structural analysis Multiple sequence alignment shows that CtNMO is closely related to other flavin-dependent monooxygenases from Class B enzymes. In particular, CtNMO is homologous to several characterized NMOs. For example, CtNMO shares 37% sequence identity with the ornithine hydroxylase SidA from Aspergillus fumigatus (SidA), 59% identity with the enzyme from Pseudomonas aeruginosa (PvdA), and 38% identity with the enzyme from Amycolatopsis alba (AMO) (Fig. S2) [19,24,31,32]. A three-dimensional structural model of CtNMO was generated using AlphaFold3 [16]. The predicted structure reveals a high degree of conservation in both the overall fold and the active-site residues when compared to other characterized NMOs (Fig. 4).Fig. 4A) Overlay of the crystal structure of SidA (green) with the AlphaFold model of CtNMO (orange). B) Comparison of active-site residues from SidA (white carbons) and CtNMO (orange carbons). Ornithine (green), flavin (yellow), and NADP+ (blue) correspond to the ligands bound to SidA in the PDB structure 4B63. A) Overlay of the crystal structure of SidA (green) with the AlphaFold model of CtNMO (orange). B) Comparison of active-site residues from SidA (white carbons) and CtNMO (orange carbons). Ornithine (green), flavin (yellow), and NADP+ (blue) correspond to the ligands bound to SidA in the PDB structure 4B63. Discussion In this work, we report the cloning, expression, purification, and functional characterization of the flavin-dependent NMO from C. taiwanensis (CtNMO). CtNMO was predicted to catalyze the formation of N5-hydroxy-L-ornithine during taiwachelin siderophore biosynthesis. Our combined biochemical and kinetic analyses demonstrate that CtNMO is an efficient ornithine hydroxylase whose catalytic properties closely resemble those of previously characterized NMOs involved in siderophore assembly, thereby confirming its proposed role in taiwachelin biosynthesis. The recombinant enzyme was successfully expressed in E. coli as a fusion to maltose-binding protein, yielding sufficient quantities of soluble protein for biochemical characterization. Amino acid analysis show that CtNMO is structurally related to Class B NMOs, which all use FAD for catalysis. Consistent with this, the UV–visible absorption spectrum of purified CtNMO displays peaks at 375 and 450 nm, characteristic of bound FAD cofactor. The overall yields obtained for both the tagged and untagged enzyme are comparable to those reported for related NMOs, indicating that the heterologous expression and purification strategy is well suited for mechanistic and structural studies. Steady-state kinetic analysis revealed that CtNMO operates with a turnover number below 1 s−1, a feature commonly observed for NMO that utilize molecular oxygen to generate a reactive C4a-hydroperoxyflavin intermediate. Similarly, KM values determined for l-ornithine and the strong preference for NADPH over NADH are in good agreement with values reported for other ornithine and lysine hydroxylases, including SidA and PvdA [12,33]. By comparing oxygen consumption with direct product formation, we determined that the CtNMO catalyzed reaction is approximately 86% coupled. This relatively high degree of coupling indicates efficient channeling of activated oxygen toward productive hydroxylation rather than unproductive peroxide formation. Partial uncoupling is nevertheless common among flavin-dependent- monooxygenases and may reflect inherent instability of the C4a-hydroperoxyflavin intermediate. One of the most striking observations in this study is the enhanced catalytic efficiency of CtNMO when fused to MBP. The MBP-CtNMO construct displays both an increased kcat and a reduced apparent KM for L-ornithine, resulting in an approximately threefold-increase in kcat/KM relative to the cleaved enzyme. While affinity tags are often assumed to be catalytically neutral, increasing evidence suggests that fusion partners such as MBP can influence enzyme dynamics, stability, or oligomeric state. In the case of CtNMO, the MBP moiety may stabilize a conformation that is more favorable for substrate binding. This is supported by a decrease of ∼ 2-fold in KM value for l-ornithine. However, the observed increase in kcat suggests that product release may be less rate-limiting in MBP-CtNMO. Collectively, these findings indicate that MBP-CtNMO is a strong candidate for biosynthetic applications, as the purification protocol is simplified and the resulting enzyme exhibits enhanced catalytic activity. Substrate inhibition was observed at elevated concentrations of both NAD(P)H and l-ornithine, although the effect is substantially more pronounced for the reducing cofactor. Inhibition by NADPH is commonly attributed to formation of a deadend complex in which excess NADPH binds to the reduced enzyme. The weaker substrate inhibition observed at high -Lornithine concentrations may reflect nonproductive binding in the active site or perturbation of enzyme dynamics required-for efficient flavin turnover. Sequence analysis and structural modeling further support the functional assignment of CtNMO. The enzyme shares significant sequence identity with several characterized NMOs, and the AlphaFold predicted structure reveals conservation of the overall fold and key active-site- residues implicated in flavin binding, NADPH recognition, and substrate positioning. This structural conservation strongly suggests that CtNMO employs the same general catalytic strategy established for other Class B flavin dependent-monooxygenase, in which reductive activation of the flavin by NADPH precedes oxygen activation and substrate hydroxylation [34,35]. Taken together, these results establish CtNMO as an ornithine hydroxylase required for the biosynthesis of taiwachelin in C. taiwanensis. Given the central role of siderophores in iron acquisition during symbiosis and nitrogen fixation, the biochemical characterization of CtNMO provides an important foundation for understanding how iron homeostasis is coordinated with nitrogenase function in this organism. Furthermore, because C. taiwanensis has been shown to grow in the presence of harmful metals in the environment including Pb, Cu, and Cd, it is possible that taiwachelin may be involved in this process and can be employed in bioremediation programs [36]. Authors contributions PS designed the research and edited the manuscript. IDF cloned the CtOMO DNA into the pVP56K vector and collaborated with AJG in expression, purification, and size-exclusion chromatography. AJG performed data collection/analysis, and conducted the oxygen and iodine oxidation assays. AJG wrote the initial draft of the manuscript. KPA performed the bioinformatic analyses. Declaration of competing interest The authors Aaron J. Gringer, Kernen P. Agwaza, Isabel Da Fonseca, and Pablo Sobrado declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.