A Comparative Biochemical Study of Oleate Hydratases Authors: Maxim van Delft, Alejandro Gran‐Scheuch, Ulf Hanefeld, Peter‐Leon Hagedoorn Published: 2026 ABSTRACT AbstractIn the search for more efficient chemical processes to valorize renewable feedstocks, oleate hydratases (Ohys) represent promising biocatalysts. These enzymes can convert cis‐Δ9 fatty acids like oleic acid into their hydroxy fatty acid counterparts using only water as co‐reagent and with unparallelled selectivity. A multitude of Ohys has been described, however, direct comparison is hard due to the use of different assays. For this work, we characterized and compared four Ohys across different homologous families (HFams): those from Elizabethkingia meningoseptica (EmOhy, HFam11), Stenotrophomonas nitritireducens (SnOhy, HFam11), Rhodococcus erythropolis (ReOhy, HFam3), and Rhodococcus pyridinivorans (RpOhy, HFam2). All enzymes, especially SnOhy, expressed well in Escherichia coli (28–99 mg/Lculture). FAD occupancy after purification was 51% or lower for all Ohys. ReOhy had lost the flavin altogether, making supplementation a necessity for its activity. Reduction of the cofactor to FADH2 clearly showed a favorable effect on the activity of all Ohys, with product yields increasing from modest to 12‐fold. Although SnOhy showed the highest initial reaction rates, EmOhy reached the highest 10‐hydroxystearic acid (10‐HSA) yields at higher oleic acid loading (10 mM) and prolonged reaction times. Thermostability was analyzed by thermal shift assays, with EmOhy appearing as most thermostable (T m app = 54 °C).Keywords: biocatalysis, enzymology, fatty acids, oleate hydratase, reducing conditions Introduction A greener chemical industry that is less reliant on fossil resources requires improved ways of valorizing renewable feedstocks like vegetable oils [1]. In this context, oleate hydratases (Ohys, EC 4.2.1.53) are promising biocatalysts. Whereas oleochemistry classically has a focus on transformation of the carboxyl group, these enzymes hydrate the double bond of several cis‐Δ9 fatty acids to form their hydroxy fatty acid counterpart, primarily oleic acid to (R)‐10‐hydroxystearic acid (10‐HSA, Scheme 1) [2, 6]. The complete regio‐ and near‐perfect stereoselectivity of the reaction remains unattainable with other chemical strategies [7]. Hydroxy fatty acids have useful properties for, among others, surfactant, emollient, and lubricant applications, and DSM‐Firmenich already employs oleate hydratases for the commercial production of 10‐HSA as a cosmetic ingredient [8]. The conversion of oleic acid to (R)‐10‐hydroxystearic acid by oleate hydratase. Due to their interesting reactivity and practical potential, the discovery, characterization, and engineering of oleate hydratases are increasingly the subject of research [9, 13]. To date, the crystal structure of five different Ohys has been reported [14, 18]. Interestingly, all Ohys require a flavin adenine dinucleotide (FAD) cofactor for activity, although its redox state is expected to remain unchanged during catalysis, as hydration is a redox‐neutral reaction. The exact role of FAD remains elusive. No indications were found for involvement in protein stabilization or direct participation in the reaction mechanism but an intricate role in repositioning amino acids into the catalytic conformation has been proposed [15, 18]. The FAD affinity of oleate hydratases is modest, and the cofactor is lost to a great extent during purification [15, 16, 19]. Interestingly, a clear activity‐enhancing effect was observed by chemical reduction of the flavin for multiple Ohys, and it is hypothesized to be reduced under physiological conditions in vivo [15, 20, 21]. Based on sequence identity, Ohys are classified into 11 different homologous families (HFams) as established in the Hydratase Engineering Database (HyED), with HFam2, 3, and 11 containing the currently most prominent Ohys [22]. For instance, the oleate hydratase from Elizabethkingia meningoseptica (EmOhy, HFam11) was the first Ohy of which the coding gene was discovered and has been extensively characterized [15, 23]. A reaction mechanism was proposed with a crucial role for a glutamate (E122) and a tyrosine (Y241) residue: for activation of a water molecule and protonation of the double bond, respectively. With respect to oligomeric state, HFam11 Ohys have been described as homodimers in solution; a large part of the dimeric interface is made up of N‐ and C‐terminal extension loops. HFam2 is the largest family and includes Ohys from Lactobacillus acidophilus (LaOhy) and Rhodococcus pyridinivorans (RpOhy) [14, 24]. Previous analyses indicated the occurrence of both dimeric and monomeric forms of these enzymes. Lastly, HFam3 Ohys are clearly distinct from the enzymes in HFam2 and 11. The oleate hydratase from Rhodococcus erythropolis (ReOhy) can be considered a model enzyme for this family [16]. More recently, an Ohy from Pediococcus parvulus (PpOhy) was added [19]. Both occur only as a monomer in solution, presumably because of the lack of N‐ and C‐terminal extensions. Also, in terms of mechanism, HFam3 distinguishes itself from HFam2 and 11. Instead of the catalytic glutamate, HFam3 Ohys contain a methionine at the corresponding position (e.g., M77 in ReOhy), strongly suggesting a different catalytic mechanism. Additionally, the FAD affinity seems distinct; while Ohys from HFam2 and 11 show the presence of FAD after purification, HFam3 Ohys are apo‐enzymes upon isolation. Still, more characterization data remains desirable, and a direct comparison of different Ohys based on literature data is challenging due to different methods of analysis and conditions used. In particular, activity is difficult to compare, even when kinetic data are obtained with purified enzymes. Partially, this is due to the hydrophobic nature of fatty acids, which strongly affects substrate availability. Hence, activity is highly dependent on, for instance, substrate and cosolvent concentration, pH, and mixing speed. Thus, to obtain a more comprehensive understanding of Ohys and directly compare their properties, we selected four different oleate hydratases for characterization under identical conditions. The selection comprises EmOhy and ReOhy, the Ohy from Stenotrophomonas nitritireducens (SnOhy, HFam11), and RpOhy. SnOhy was added due to its exceptionally high reported turnover number (Table 1) [25] and RpOhy to include an Ohy from HFam2. Although not fully doing justice to the complete diversity of the homologous families, this selection of enzymes represents a diverse Ohy subset. The enzymes were heterologously expressed and characterized, in particular, aiming to better understand their relative activities and the role of FAD. TABLE 1Oleate hydratases selected for this study, with corresponding homologous family, reported activity, effect of FAD reduction on activity, and PDB entry. The amino acid sequence identity matrix of these enzymes is presented in Table S1.OrganismOhyHFamOligomeric state k cat , s−1 Effect of FAD reductionPDB entryReferences Elizabethkingia meningoseptica EmOhy11Dimer1.27‐fold increase4UIR[15, 23] Stenotrophomonas nitritireducens SnOhya 11Dimer78n.d.b 5Z70[17, 25, 26] Rhodococcus erythropolis ReOhyc 3Monomer0.57n.d.b 5ODO[16] Rhodococcus pyridinivorans RpOhy2Monomer and dimer20n.d.b n.a.[24]a The crystal structure was reported for the oleate hydratase from Stenotrophomonas sp. KCTC 12332 (accession number AMJ57405.1), but its sequence is identical to that of SnOhy (AND01240.1).b n.d.: not determined.c While most work on ReOhy concerns ReOhyCCM2595, this study uses ReOhy (PR4) as in our previous report. Sequences differ by only three amino acids [16, 27, 29]. Oleate hydratases selected for this study, with corresponding homologous family, reported activity, effect of FAD reduction on activity, and PDB entry. The amino acid sequence identity matrix of these enzymes is presented in Table S1. The crystal structure was reported for the oleate hydratase from Stenotrophomonas sp. KCTC 12332 (accession number AMJ57405.1), but its sequence is identical to that of SnOhy (AND01240.1). n.d.: not determined. While most work on ReOhy concerns ReOhyCCM2595, this study uses ReOhy (PR4) as in our previous report. Sequences differ by only three amino acids [16, 27, 29]. Results and Discussion Expression and Oligomeric State Analysis The N‐terminally His‐tagged oleate hydratases were expressed in Escherichia coli (E. coli) and purified by immobilized metal affinity chromatography (IMAC) in good yield, generally tens of mg per liter of cell culture (Figure S1, Table S2). Especially for SnOhy, high expression yields were obtained (99 mg/Lculture). Surprisingly, upon codon optimization of the gene and use of a high‐level protein expression system (pET in BL21(DE3) instead of pBAD in TOP10), EmOhy yield dropped from 28 to 9 mg/Lculture. The oligomeric state of the Ohys was determined by size‐exclusion chromatography coupled to multi‐angle light scattering (SEC‐MALS). Consistent with earlier reports, the HFam11 enzymes EmOhy and SnOhy were observed as dimers, and ReOhy as a monomer (Figure 1A–C) [15, 16, 25]. RpOhy occurs both in the dimeric and monomeric form, in an approximate 1:1 ratio (Figure 1D), as also indicated before [24]. This co‐existence of both oligomeric states may be rationalized based on sequence and structural analysis. In HFam11 Ohys, a large part of the dimeric interface is made up of both a C‐terminal and an N‐terminal extension. While RpOhy does have a C‐terminal extension, it lacks the long N‐terminal extension (Figures 1E, S2), thus occurrence as both dimer and monomer in equilibrium is reasonable [30]. SEC‐MALS analysis of the oleate hydratases and structural explanation of the monomer‐dimer equilibrium of RpOhy. (A–D) show the chromatograms, in the order EmOhy, SnOhy, ReOhy and RpOhy. The calculated molecular weights are given with their corresponding peaks, along with the corresponding apparent oligomeric state (AOS). (E) shows an overlay of RpOhy (orange, AlphaFold2 model) and chain A of EmOhy (gray, PDB: 4UIR). The boxes indicate the N‐ and C‐terminal extensions of EmOhy, together making up a major part of the dimerization interface. Figure made with ChimeraX. FAD Occupancy Purified EmOhy, SnOhy, and RpOhy stocks appeared pale yellow while ReOhy was always colorless, even when FAD was supplemented during cell lysis prior to purification. Indeed, only the former three Ohys showed a UV–Vis spectrum characteristic of flavoenzymes (Figure 2). The FAD occupancy after purification was determined by denaturing the enzyme and determining the free FAD concentration based on the absorbance at 450 nm. SnOhy showed the highest FAD occupancy at 51%, followed by EmOhy (30%) and RpOhy (16%). The substoichiometric FAD occupancy does not seem to be a consequence of insufficient intracellular FAD concentration during expression, as the best‐expressed Ohy (SnOhy) also shows the highest FAD occupancy. As expected, no FAD‐typical absorbance peak was observed for ReOhy, confirming the loss of its cofactor during purification and thus a lower FAD affinity compared to the other Ohys (Table 2). UV–Vis absorption spectra of the oleate hydratases before and after denaturation: (A) EmOhy (17 µM), (B) SnOhy (104 µM), (C) ReOhy (72 µM) and (D) RpOhy (39 µM). TABLE 2FAD occupancies of the oleate hydratases after purification.OhyFAD occupancy, % EmOhy30 SnOhy51 ReOhyn.d.a RpOhy16a n.d.: not determined due to absence of A 450 peak. FAD occupancies of the oleate hydratases after purification. n.d.: not determined due to absence of A 450 peak. Relative Activities and Attempted Reconstitution of FAD The activities of the Ohys were assessed with short (10‐minute) reactions with 1 mM oleic acid (OA) concentration. SnOhy reached the highest product concentrations, followed by EmOhy and then the Rhodococcus enzymes (Figure 3). Because the Ohys occurred partially in their apo‐form, FAD was supplemented during the biotransformations. However, for EmOhy, SnOhy, and RpOhy, 10‐HSA yields remained unaltered upon moderate addition (4:1 molar ratio to Ohy) of FAD compared to control reactions without exogenous FAD. Only for ReOhy, which was isolated as an apo‐enzyme after purification, FAD supplementation was shown to be essential for activity. Even higher concentrations and pre‐incubation with FAD did not increase activity (Figure S3). The lack of effect of FAD supplementation on Ohy activity was surprising, since the concentrations were well above the reported dissociation constant of EmOhy (K D = 1.8 µM) [15]. It is considered unlikely that flavin‐deficient Ohys are active, based on earlier studies with EmOhy and ReOhy [15, 29], and on the data presented here for ReOhy. However, our data indicate that reconstitution of the cofactor in already folded oleate hydratase is not trivial. Effect of FAD supplementation on product yields of the Ohys in biotransformations. Reaction conditions: 1.0 mM OA, 1.0 µM Ohy, 0 or 4.0 µM FAD, 5.0% v/v DMSO, in 100 mM KPi 7.0, total volume 500 µL; 25 °C, 1000 rpm orbital shaking, reaction time 10 min. Each bar represents the mean of triplicates, and the error bars represent their standard deviation. To better understand the role of FAD, and because earlier studies indicated a large effect of reduction of FAD to FADH2 on Ohy activity [15, 20, 21], similar reactions were conducted under reducing conditions. Interestingly, under these conditions, all enzymes, including the less active Rhodococcus Ohys, reached full conversion (Figure 4). For ReOhy, this corresponds to at least a 12‐fold increase in activity. The clear activity‐enhancing effect of FAD supplementation under reductive conditions, in contrast to the above‐described results, may indicate an important role for the conformation of the isoalloxazine ring. Upon reduction, the FAD undergoes a conformational shift from a planar, aromatic structure to a bent “butterfly” shape [31]. However, this remains speculative, and detailed biophysical analyses would be required to further investigate this hypothesis. Effect of reducing conditions on the product yields of the oleate hydratases in biotransformations at 1.0 mM oleic acid loading. Reaction conditions: 1.0 mM OA, 1.0 µM Ohy, 4.0 µM FAD, 5.0% v/v DMSO in 100 mM KPi, pH 7; 25 °C, 1000 rpm, 10 min reaction time. For reducing conditions, 2.0 mM dithionite was added, and the reaction was performed anaerobically. Each bar represents the mean of triplicates, and error bars are their standard deviation [23, 29]. To investigate differences in pH preference among the Ohys, reactions were conducted at varying pH values besides pH 7.0. Also at pH 6.0 and 8.0, the highest product formation was achieved in the order of SnOhy, EmOhy, RpOhy, and ReOhy (Figure 5). EmOhy and RpOhy show relatively consistent activity across the different pHs; for SnOhy and ReOhy, pH 6.0 is preferred. The observed rate of EmOhy at pH 6.0 (k obs = 1.0 s−1) is in the same range as its reported k cat of 1.2 s−1 under similar conditions [15]. Observed activities of SnOhy and RpOhy are not as high as perhaps would be expected based on their reported k cat values. For instance, the k obs values of SnOhy (1.5 s−1) and RpOhy (0.24 s−1) at pH 7.0 are much lower than their reported turnover numbers of 78 and 20 s−1, respectively [24, 25]. However, the k obs and k cat values should be compared with caution, as differences in assay conditions are to be considered. The observation that HFam11 Ohys exhibit a relatively high activity is consistent with earlier reports [12, 22]. Effect of pH on the product yields of the Ohys in biotransformations. Reaction conditions: 1.0 mM OA, 1.0 µM Ohy, 4.0 µM FAD, 5.0% v/v DMSO, in 100 mM KPi pH 6.0, 7.0 or 8.0, total volume 500 µL; 25 °C, 1000 rpm orbital shaking, reaction time 10 min. Each bar represents the mean of triplicates, and error bars represent their standard deviation. Enzyme Stability As an indicator of enzyme thermostability, the apparent melting temperatures of the Ohys were determined by thermal shift assays using both the ThermoFluor and ThermoFAD method [32, 33]. These assays follow protein unfolding by gradually increasing the temperature, with the melting temperature representing the inflection point of the transition from folded to fully unfolded protein. However, the ThermoFluor and ThermoFAD assays apply different indicators. The ThermoFluor assay utilizes a dye that binds to the hydrophobic parts of proteins, which become exposed during denaturation. Upon binding to these hydrophobic regions, the dye emits a fluorescent signal. ThermoFAD is a similar approach that works only for flavoproteins by monitoring the increase in fluorescence resulting from FAD release during denaturation. Although the methodologies slightly differ, the measured values were averaged, as they determine the same property (apparent melting temperature) under very similar conditions. These ranged from 36.0 °C for ReOhy to 53.8 °C for EmOhy (Table 3, Figure S4). For EmOhy, the value is well in line with an earlier report; for ReOhy, it is substantially lower than the earlier reported 45 °C [19]. Differences in buffer type and concentrations of protein and dye are likely to contribute to this. The inability to obtain a melting curve of ReOhy by ThermoFAD is explained by the absence of the cofactor. As suggested before, the higher stability of EmOhy versus monomeric ReOhy could possibly be attributed to stabilization conferred by its protein–protein interface. However, a general stabilizing effect of dimerization is not evident from this study: dimeric SnOhy has a lower apparent melting temperature (41.5 °C) than partially monomeric RpOhy (47.9 °C). TABLE 3Apparent melting temperatures of the oleate hydratases.Ohy T m app, °CThermoFluorThermoFADAverage EmOhy53.7 ± 0.153.9 ± 0.353.8 ± 0.3 SnOhy41.2 ± 0.641.7 ± 0.841.5 ± 1.0 ReOhy36.0 ± 0.3n.d.a 36.0 ± 0.3 RpOhy43.4 ± 0.352.4 ± 0.647.9 ± 0.6a n.d.: not detected. Apparent melting temperatures of the oleate hydratases. n.d.: not detected. Activity at Higher Substrate Loading To compare the performance of the different Ohys under more process‐like conditions, reactions were conducted at increased OA loading (10 mM). Reactions were conducted for 10 min, 2, and 24 h (Figure 6). Under these conditions, substrate saturation is achieved for all Ohys: the concentration is well above the K M values of EmOhy (110 µM), SnOhy (22 µM), ReOhy (490 µM), and RpOhy (720 µM) [3]. EmOhy showed the highest product formation, also compared to SnOhy, being the only enzyme to reach full conversion in 24 h. This implies that EmOhy is active for longer due to higher intrinsic stability, which is consistent with its relatively high apparent melting temperature. Plausibly, enzyme deactivation occurs within 24 h, at least for SnOhy, ReOhy, and RpOhy. Although substrate inhibition has not been observed for Ohys [3], the high concentration of OA may cause enzyme deactivation, given its highly apolar character. Time‐dependent product yields of the Ohys in biotransformations at higher oleic acid loading. Reaction conditions: 10 mM OA, 1.0 µM Ohy, 4.0 µM FAD, 5.0% v/v DMSO, in 50 mM PIPES, pH 7.0, total volume 500 µL; 25 °C, 1000 rpm orbital shaking, reaction time 10 min, 2, or 24 h. Each bar represents the mean of triplicates, and error bars represent their standard deviation. Like for the 1 mM substrate reactions, reducing conditions improved the activity of all Ohys in reactions with 10 mM OA (Figure 7). However, the effect is much smaller, and product yield plateaued around 80%. Again, this may indicate enzyme inactivation over time, caused or amplified by the presence of the reducing agent dithionite. Alternatively, under these conditions, it may not be the concentration of active Ohy that is the rate‐limiting factor, but instead the availability of free OA molecules to the enzyme. Above the solubility limit, OA occurs as micelles, vesicles, or other structures that may render it less accessible for the reaction [34]. Effect of reducing conditions on the product yields of the oleate hydratases in biotransformations at 10 mM oleic acid loading. Reaction conditions: 10 mM OA, 1.0 µM Ohy, 4.0 µM FAD, 5.0% DMSO in 100 mM KPi, pH 7; 25 °C, 1000 rpm, 2 h reaction time. For reducing conditions, 2.0 mM dithionite was added, and the reaction was performed anaerobically. Each bar represents the mean of triplicates, and error bars represent their standard deviation. Conclusion This work presented a comparative biochemical study of four oleate hydratases across different HFams: those from Elizabethkingia meningoseptica (EmOhy, HFam11), Stenotrophomonas nitritireducens (SnOhy, HFam11), Rhodococcus erythropolis (ReOhy, HFam3), and Rhodococcus pyridinivorans (RpOhy, HFam2). All enzymes were expressed in E. coli and obtained in good yield, especially SnOhy (99 mg/Lculture). Oligomeric states were in line with the literature, including the occurrence of both monomeric and dimeric forms of RpOhy. This was explained based on the presence of the C‐terminal extension but the absence of the N‐terminal extension of RpOhy, with both making up a major part of the interface of dimeric Ohys. All Ohys had substoichiometric FAD content after purification, up to 51% of the protein concentration. For ReOhy, the FAD seems to have dissociated completely, indicating particularly low affinity. Indeed, ReOhy required FAD supplementation for product formation. However, for the other Ohys, FAD supplementation did not improve conversions, despite also containing apo‐enzyme. Reductive conditions did clearly improve product yields, for ReOhy at least 12‐fold, emphasizing the important role of the redox state of the cofactor. Although SnOhy was the most active enzyme at low oleic acid loadings, EmOhy achieved higher product yields at increased oleic acid loading (10 mM). Presumably, this is due to its higher stability, as demonstrated by thermal shift assays: EmOhy showed the highest apparent melting temperature of 54 °C. At higher substrate loading and longer reaction times, either deactivation of the Ohys or limitations in substrate availability occur. In conclusion, in this side‐by‐side biochemical study, EmOhy was identified as a preferred oleate hydratase for further research compared to SnOhy, ReOhy, and RpOhy: it shows relatively good expression and FAD affinity, and relatively high stability and activity at increased substrate loading and reaction times. The results provide a good basis for diverse avenues, like protein engineering, studying the phase behavior of the fatty acids, and immobilization, and experiments in these directions are currently underway in our laboratories. Experimental Section Enzyme Expression and Purification For expression of EmOhy and RpOhy (pBAD/HisA vector) in E. coli TOP10 cells, a preculture of 10 mL LB medium with 100 µg/mL ampicillin was inoculated with a colony from an LB‐agar plate. The preculture was grown overnight at 37 °C and 180 rpm. 5 mL of preculture was added to 500 mL of sterilized TB medium with 100 µg/mL ampicillin and grown again at 37 °C and 180 rpm until an OD600 of 0.6–0.8 was reached. Expression was induced by the addition of 0.2% (13.3 mM) L‐arabinose, and growth was resumed at 25 °C and 180 rpm for 15–20 h. To harvest the cells, the culture was centrifuged (7000 × g, 10 min, 4 °C) and the cell pellet was washed with 50 mM Tris‐HCl with 500 mM NaCl, pH 8.0, and centrifuged again (7000 × g, 10 min, 4 °C) before storage at −70 °C. Expression of SnOhy and ReOhy (on pET28a(+) vector) was performed identically, except that the E. coli strain was BL21(DE3) instead of TOP10, the antibiotic was kanamycin (50 µg/mL) instead of ampicillin, and the inducer was IPTG (0.5 mM final concentration) instead of arabinose. For cell lysis and purification, pellets were resuspended in 50 mM Tris‐HCl, 500 mM NaCl, pH 8.0, containing 1 mM PMSF, 10 mM MgCl2, 2 mM CaCl2, 0.02 mg/mL DNase, and 0.2 mg/mL lysozyme and shaken on a horizontal shaker for 30 min at room temperature. Cell lysis was done by sonication (Branson Sonifier 450) for 5 min, 40% duty cycle, and output 3.5. Insoluble debris was separated by centrifugation (20,000 rpm, 40 min, 4 °C), and the supernatant was 0.45 µm filtered. For purification by gravity column, the supernatant was loaded onto a Fast Flow nickel Sepharose column pre‐equilibrated with Buffer A (50 mM Tris‐HCl pH 8.0, 500 mM NaCl, 40 mM imidazole). The His‐tagged protein was eluted with elution buffer (50 mM Tris‐HCl pH 8.0, 500 mM NaCl, 400 mM imidazole). Lastly, eluted fractions were desalted using an EconoPac 10‐DG (Bio‐Rad) pre‐equilibrated with 50 mM KPi, pH 8.0. The purity of each protein sample was analyzed and verified by SDS‐PAGE and quantified by bichinchonic acid (BCA) assay. Purified protein aliquots were flash‐frozen and stored at −70 °C. SEC‐MALS Analysis Purified Ohy samples (concentrations: 1.2 mg/mL (EmOhy), 7.8 mg/mL (SnOhy), 4.8 mg/mL (ReOhy), and 3.0 mg/mL (RpOhy)) were centrifuged to remove aggregates (14000xg, 4 °C, 10 min), and 50 µL was transferred to a chromatography vial. The proteins were analyzed by SEC using an HPLC unit (1260 Infinity II, Agilent) with a Superdex200 Increase 10/300 GL column (Cytiva), running in series with an online ultraviolet (UV) detector (1260 Infinity II VWD, Agilent), an eight‐angle static light scattering detector (DAWN HELEOS 8+, Wyatt Technology) and a refractometer (Optilab T‐rEX, Wyatt Technology). The running buffer was 50 mM KPi, pH 8.0 with 100 mM NaCl. The molecular weight was calculated based on the measured Rayleigh scattering at different angles and the established differential refractive index increment of value of 0.185 mL/g for proteins in solution with respect to the change in protein concentration (dn/dc) using ASTRA software (Wyatt Technology, version 7.3.1). UV–Vis Spectroscopy Purified Ohy samples were centrifuged to remove aggregates (14000 × g, 10 min, 4 °C) and their UV–Vis spectra were recorded on a spectrophotometer (Agilent Cary 60). Denaturation was performed by adding sodium dodecyl sulfate (SDS) to a concentration of 0.2% from a 10% stock and heating at 95 °C for 10 min. Samples were centrifuged again (14000 × g, 10 min, 4 °C), and the spectrum of the supernatant with released FAD was recorded. For determination of the FAD occupancy, the FAD concentration was calculated using its extinction coefficient at 450 nm (ε450 = 11,300 M−1 cm−1) [35]. Thermal Shift Assays For the ThermoFluor assay, purified Ohy stocks were diluted to 1.1 mg/mL in 50 mM KPi, pH 7.6. 2 µL of SYPRO Orange (250x) was added to 18 µL of pure Ohy solution (final Ohy concentration 1 mg/mL, SYPRO Orange 25×) in qPCR tubes with optically clear lids. Samples were transferred to a real‐time PCR machine (Analytik Jena, qTower3) with a temperature program 25–85 °C at a heating rate of 1 °C/min and analyzed with the JOE filter (excitation 515 nm, emission 545 nm) at gain 3/10. For the ThermoFAD assay, purified Ohy stocks (20 µL, 3.5 mg/mL) were transferred to qPCR tubes with optically clear lids. Samples were transferred to a real‐time PCR machine (Analytik Jena, qTower3) with a temperature program 25–85 °C at a heating rate of 1 °C/min and analyzed with the FAM filter (excitation 470 nm, emission 520 nm) at gain 10/10. For data analysis of both assays, raw data were smoothed using a second‐order polynomial with four neighbors (Savitzky–Golay filter, GraphPad Prism). Biotransformations For 1 mM reactions, biotransformations were performed with 100 mM KPi, pH 6.0, 7.0, or 8.0 containing 1 µM purified Ohy (concentration with respect to Ohy monomer, i.e., number of active sites) and 0 or 4 µM FAD in a total volume of 475 µL, in a 1.5 mL Eppendorf tube. Reactions were initiated by addition of 25 µL of 20 mM OA in DMSO for a total reaction volume of 500 µL with final concentrations of 1 mM OA and 5% v/v DMSO. Reaction samples were incubated at 25 °C with 1000 rpm orbital shaking for 10 min. For reactions under reductive conditions, these conditions were achieved by dissolving 2 mM dithionite (in its sodium salt form) in the buffer and performing the reactions in an anaerobic glove box (MBRAUN). Reactions were terminated by addition of 40 µL 1 M HCl. For 10 mM reactions, biotransformations were performed with 50 mM PIPES, pH 7.0, containing 1 µM purified Ohy (concentration with respect to Ohy monomer, i.e., number of active sites) and 4 µM FAD in a total volume of 475 µL, in a 1.5 mL Eppendorf tube. Reactions were initiated by addition of 25 µL of 200 mM OA in DMSO for a total reaction volume of 500 µL with final concentrations of 10 mM OA and 5% v/v DMSO. Reaction samples were incubated at 25 °C with 1000 rpm orbital shaking for 10 min, 2, or 24 h. For reactions under reductive conditions, these conditions were achieved by dissolving 2 mM dithionite (in its sodium salt form) in the buffer and performing the reactions in a glove box. Reactions were terminated by the addition of 40 µL 1 M HCl. GC‐FID Analysis Analysis of 1 mM reactions was done as follows: to the acidified reaction samples was added 500 µL of EtOAc containing 0.5 mM n‐dodecane as an internal standard, and samples were vortexed for 30 s. Samples were centrifuged briefly, and the extract was dried over MgSO4. Subsequently, 25 µL was added to 75 µL of derivatization reagent (BSTFA with 1% TMCS: pyridine, 1: 1) in glass chromatography vials. These were incubated at 60 °C for 1.5 h for silylation before their contents were transferred into plastic 0.3 mL chromatography vials. GC measurements were performed on a Shimadzu type GC‐2014 with a CP‐Sil5 CB column (25 m × 0.25 mm × 1.2 µm) with N2 (flow rate 30 mL min−1) as carrier gas. Injector and detector (FID) temperatures were 340 and 360 °C, respectively. 1 µL was injected at a split ratio of 100:1. The oven program started at 165 °C (3 min hold) with a 15 °C/min ramp until 345 °C (3 min hold). Analysis of 10 mM reactions was done as follows: to the acidified reaction samples was added 500 µL of EtOAc containing 5 mM decanoic acid methyl ester (DAME) as an internal standard, and samples were vortexed for 2 min. The extract was centrifuged briefly, dried over MgSO4, and transferred to a 0.3 mL chromatography vial. GC measurements were performed on a Shimadzu type GC‐2010 Pro with a MET‐Biodiesel capillary GC column (13 m × 0.32 mm, df 0.10 µm – the 2 m × 0.53 mm I.D. guard was removed) from Merck, and H2 as carrier gas (Parker 40H‐MD Carrier generator, flow rate: 70 cm s−1). Injector and detector temperatures were 270 and 350 °C, respectively. One microliter was injected at a split ratio of 60:1. The oven program started at 100 °C (1 min hold) with a 30 °C/min ramp until 340 °C (1 min hold). Funding This work was supported by the Dutch Research Council (NWO) domain Applied and Engineering Sciences under the Open Technology Programme grant 20781 (project name EnzIm2Flow). Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supplementary Material