Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Towards an understanding of oleate hydratases and their application in industrial processes

Towards an understanding of oleate hydratases and their application in industrial processes Fatty acid hydratases are unique to microorganisms. Their native function is the oxidation of unsaturated C–C bonds to enable detoxification of environmental toxins. Within this enzyme family, the oleate hydratases (Ohys), which catalyze the hydroxylation of oleic acid to 10-(R)-hydroxy stearic acid (10-HSA) have recently gained particular industrial interest. 10-HSA is considered to be a replacement for 12-(R)-hydroxy stearic acid (12-HSA), which has a broad application in the chemical and pharmaceutical industry. As 12-HSA is obtained through an energy consuming synthesis process, the biotechnological route for sustainable 10-HSA production is of significant industrial interest. All Ohys identified to date have a non-redox active FAD bound in their active site. Ohys can be divided in several sub - families, that differ in their oligomerization state and the decoration with amino acids in their active sites. The latter observation indicates a different reaction mechanism across those subfamilies. Despite intensive biotechnological, biochemical and structural investigations, surprising little is known about substrate binding and the reaction mecha- nism of this enzyme family. This review, summarizes our current understanding of Ohys with a focus on sustainable biotransformation. Keywords: Oleate hydratase, Biocatalysis, Industrial biotechnology, Whole cell and enzymatic oleic acid transformation, Green chemistry, Protein engineering, Structure–function relation, Bioeconomy Introduction acids. This is prevented by the expression of enzymes Adaption to the outer environment is a crucial factor for called fatty acid hydratases [1–5], which are unique to survival of living organisms. Many microorganisms have microorganisms [6]. Moreover, long chain fatty acids found a way to survive toxins by producing detoxifying can cause prevention of protein and amino acid uptake, small molecules or proteins. One example is the detoxi- particularly in gram-positive bacteria due to the inher- fication of free long chain fatty acids by microorganisms, ent character of their cell membranes [2, 7–9]. Con- which in free form could potentially destroy outer mem- sequently, several microorganisms, which live in close branes causing lysis of protoplasts, subsequent leakage contact to free fatty acids, are reported to express fatty of proteins, cell-associated fatty acids as well as nucleic acid hydratases as an adaption and defence to their outer environment [10]. Two functions of oleate hydratases (Ohys) for micro- *Correspondence: daniel.garbe@tum.de; loll@chemie.fu-berlin.de organisms are currently discussed. Crude oils such as Werner Siemens-Chair of Synthetic Biotechnology, Dept. of Chemistry, oils from plants but also from the skin typically contain Technical University of Munich ( TUM), Lichtenbergstr. 4, 85748 Garching, Germany a certain percentage of free, unsaturated fatty acids [11, Institute for Chemistry and Biochemistry, Laboratory of Structural 12], which are toxic for microorganisms, and thus it is Biochemistry, Freie Universität Berlin, Takustr. 6, 14195 Berlin, Germany © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Prem et al. Microbial Cell Factories (2022) 21:58 Page 2 of 15 Scheme 1 Hydroxylation of oleic acid to 10-(R)-hydroxy stearic acid as performed by Ohys thought that they are being detoxified via Ohys. Staphy - level of defence and additionally, cytoplasmatic fatty acid lococcus aureus has been found to express a functional hydratases could complement the response mechanism. Ohy even though it does not synthesize unsaturated fatty Fatty acid hydratases are able to hydroxylate unsatu- acids. However, one of S. aureus’ natural habitats is the rated fatty acids. A plethora of fatty acid hydratases, human skin, where the high abundance of free, unsatu- which convert substrates with different acyl-chain length, rated fatty acids leads to an evolutionary pressure. ranging from C11:1 to C22:6, have been reported [20, An Ohy has been discovered in S. aureus (OhySa) 23–26]. Many fatty acid hydratases have low specificity, that conveys resistance against palmitoleic acid. The in respect to acyl-chain length, but demonstrate high hydroxylated form does not further convey toxicity and regio- and stereospecifity. For instance, Ohys are regio - is not incorporated into the phospholipid membrane specific for the cis-9 C–C double bond position and but is rather exported into the outer environment [13]. enantiospecific for the 10-(R) isomer (Scheme 1). Recently, it was shown that OhySa are able to convert Hydroxylated fatty acids have first been found in host cis-9 unsaturated fatty acids to their 10-hydroxy human steatorrhoeic faeces and since a standard diet derivatives in human serum and at the infection site in a does not contain such unusual fatty acids, it was assumed mouse neutropenic thigh model, suggesting that OhySa that microorganisms synthesize them in the gut [27]. could play a role in immune modulation in S. aureus This has subsequently been demonstrated, as a Pseu- pathogenesis [14]. Furthermore, fatty acid hydratases domonas sp. strain 3266 has been found to convert oleic have been reported to be involved in stress responses of acid to 10-(R)-hydroxy stearic acid (10-HSA; Scheme  1) microorganisms. In Bifidobacterium breve, the expres - [28]. Numerous other microorganisms, mostly discov- sion of a fatty acid hydratase increases stability against ered by investigating human or animal faeces, have been heat and solvents [15, 16]. shown to produce 10-HSA [29–31]. Notably, 47  years Ohys only convert free, unsaturated fatty acids, which passed by between the discovery of 10-HSA production is rather unique. Usually, bacteria can take up exogenous of Pseudomonas sp. strain 3266, later found to be Eliza- unsaturated fatty acids, but not all are incorporated into bethkingia meningoseptica, and the purification and char - their phospholipid layer [17]. Furthermore, it is not fully acterization of the responsible enzyme [24]. understood, where exactly Ohys act. They could either Prior to the discovery and characterization of the first function in the cytoplasm or in the outer environment. Ohy, the first patent has been filed regarding the indus - For S. aureus, it has been reported that Ohys were found trial use of an Ohy from Streptococcus pyogenes, includ- in vesicles, which were secreted from the cell in the pres- ing its direct homologues with more than 40% sequence ence of linoleic acid [18]. Furthermore, an Ohy from Lac- overlap [32]. In an industrial context, oleate hydratases tobacillus plantarum was found to be a protein, bound are of special interest, due to the high-value product to a membrane by electrostatic attachment and addi- 10-(R)-hydroxy stearic acid (10-HSA). tionally it was reported that the conversion of linoleate It was considered that 10-HSA can be a replacement to 10-hydroxy-cis-12-octadecenoic acid occurs at the for 12-(R)-hydroxy stearic acid (12-HSA), which is widely periphery of the cell [19]. Since a few microorganisms used in the chemical and pharmaceutical industry. As are known to contain several oleate hydratases, a com- surfactant, 12-HSA is added to soaps and body washes. plementary effect of defence might apply [20–22]. Mem - As molecule with emollient and thickening properties, brane-hydratases and secreted ones could serve as a first it is used in skin creams and lotions. Other common P rem et al. Microbial Cell Factories (2022) 21:58 Page 3 of 15 applications are as an additive in grease, lubricating-oils Architecture of Ohys and paints, in manufacturing PVC and as lubricants in To date, there is very limited structural information avail- synthetic or natural rubbers. Furthermore, it can be used able for Ohys (see Additional file  1: Table S1). Structural as an adhesive and as a fine chemical in the food and characterization of Ohys from only five different organ - pharmaceutical industry [33–35]. 10- and 12-HSA can isms was performed so far [26, 46–49]. For the sake of additionally be converted into valuable secondary prod- understanding, we will employ a uniform terminology for ucts using cascade reactions. Those include keto-fatty the description of the discussed Ohys. All Ohys will be acids, estolides and wax esters [36, 37]. Advantages com- referred to Ohy and the first two letters of the organism pared to similar products derived from petrochemicals name of origin. We will mainly discuss: the Ohy of Rho- are that 12-HSA can be manufactured from renewable dococcus erythropolis (OhyRe; Uniprot: T5I9M6), Ohy of recourses and it is considered as a low-risk compound Staphylococcus aureus (OhySa; Uniprot: A0A0D6GJV1), [35]. In large industrial scale, 12-HSA is produced by Ohy of Lactobacillus acidophilus (OhyLa; Uniprot: chemical hydrogenation of castor oil mainly consisting Q5FL96), Ohy of Stenotrophomonas sp. KCTC 12332 of ricinoleic acid [38, 39]. For the hydrogenation of cas- (OhySt; Uniprot: A0A126NKL7) as well as Ohy of Eliza- tor oil, either hydrogen and a metallic-catalyst such as bethkingia meningoseptica (OhyEm; Uniprot: OLHYD). Raney-Nickel is applied. Alternatively, a catalytic trans- Currently the structures cover three HFam families fer hydrogenation without hydrogen can be performed. of the in total 11 Ohy families [25]. A superposition of Often high pressure and temperatures are required to the available crystal structures clearly reveals a higher obtain sufficient yields of 12-HSA [40–43]. similarity of structures within one clade of HFam fami- Recently, there is a call for more sustainability in the lies compared to lower similarity between families (see chemical industry, and the use of a biocatalyst could Additional file  1: Tables  S2 and S3). Two structures of potentially support that demand. However, when produc- the HFam2 family, OhyLa as well as OhySa are avail- ing 10-HSA using biocatalysts, free oleic acid is needed, able, which superimpose with a root mean square devia- which can be produced from oil of different types of tion (rmsd) of 1.1  Å (see Additional file  1: Table  S3). In renewable sources. This could either be plant-based oils contrast, the superposition of the overall architecture such as high-oleic sunflower oil or when available in of Ohys belonging to different HFam families, is signifi - larger scales in immediate future, hydrolysed oil from cantly different with higher rmsd values (see Additional microorganisms such as Cutaneotrichosporon oleagi- file 1: Table S3). nosus [44] or free fatty acids produced from engineered The reported structures provide interesting insights bacteria [45]. As a result, there is less dependency on just into the binding of flavin adenine dinucleotide (FAD) and one type of oil. substrate binding sites as well as the different oligomeric More and more new types of hydratases have been elu- states of Ohys. Common for all HFam families are three cidated in recent years. This can be attributed to a grow - core domains (Fig.  1), but some subfamilies have addi- ing interest in the industrial production of 10-HSA using tional N- and C-terminal extensions (Figs.  1, 2). Based biocatalysts. For industry and academia, an understand- on the available structural and biochemical informa- ing of the precise mechanism of Ohys, including the role tion, Ohys can occur as monomers or dimers. OhyLa, of a potential cofactor as well as substrate recognition OhyEm, OhyLa, OhySa and OhySt, all members of the is a fundamental prerequisite for protein engineering HFam2 or HFam11 family arrange as dimers. In contrast, in respect to industrial application. Currently, the high requirements on the performance and process stability properties of these enzymes, which will be discussed in this review in detail, prevent their application in indus- trial processes. Successes in protein engineering are only achieved steadily and this can be attributed to many open questions regarding substrate and cofactor binding and the mechanism. Additionally, low substrate and product solubility hinder the appropriate capturing of enzymatic kinetic parameters. These struggles and ways to over - come them to establish well-functioning and stable Ohys Fig. 1 Schematic domain architecture of OhyEm and OhyRe. A will be the topic of this review. Moreover, we are going Domain architecture of OhyEm coored in grey-shading for domain I to discuss sequence specific differences within the Ohy to domain IV. In yellow marked the position of the Rossman signature families, potentially leading to differences in the catalytic motif. B Domain architecture of OhyRe with identical grey-shading for its domains as in OhyEm mechanism. Prem et al. Microbial Cell Factories (2022) 21:58 Page 4 of 15 Fig. 2 Overall structure of OhyEm and OhyRe. A Proteins are shown in cartoon representation. Dashed lines indicate un-modelled loop regions. Domain organization of OhyEm (PDB-ID: 4uir; [46]): Protomer I is shown in cartoon representation: Domain I in light blue, domain II in blue, domain III in deep teal and domain IV in marine. Protomer II is shown in a transparent surface representation and cartoon representation depicted in light gray. The FAD cofactor is shown as black stick representation. B Domain organization of OhyRe (PDB-ID: 5odo; [48]): Domain I in green, domain II in orange, domain III in deep teal and domain IV in red. The shown FAD cofactor is derived from the superposition with OhyEm. The FAD is depicted in black stick representation. C Superposition of protomer I of OhyEm and OhyRe in identical orientation as in panel B. One protomer of OhyEm is shown in light blue. The terminal extensions of OhyEm are clearly visible on the right site of the panel. Figures were prepared with PYMOL (Schrödinger Inc.) OhyRe is a monomeric enzyme belonging to HFam3. For sequence of monomeric OhyRe [48] (Figs.  1, 2). These dimeric Ohys common are N- and C-terminal amino noticeable differences in the overall structure as well as acid sequence extensions, which are not present in the in different oligomerization states of Ohys (Figs.  1, 2) P rem et al. Microbial Cell Factories (2022) 21:58 Page 5 of 15 could hint to variations in co-factor binding or substrate the FAD is non-covalently bound to the protein. There - recognition. In the following, differences and similarities fore, binding of FAD induces a conformational change in between various members of Ohy families with a focus domain I, which leads to closure of the FAD-binding site on the domain arrangement, substrate and FAD-binding and enfolding of the FAD [46, 47]. will be discussed. All known Ohys strictly require FAD for functioning even though the FAD is likely not to function as a con- Overall structure ventional redox cofactor known from other enzyme fami- In general, all Ohy structures are composed of three lies [54]. The function of the FAD cofactor in Ohys is still core domains (Figs.  1, 2), that are related to other FAD- under debate and could likely play a role in the polariza- dependent enzymes. In the structures of the dimeric tion of the substrate, involvement in substrate binding HFam2 and HFam11 family members, the proteins fold or the stabilization of reaction intermediates [55, 56]. in an α-helix N-terminal of domain  1, which is involved A merely structural role of FAD cannot be completely in stabilization of the dimers. In Ohys of the HFam3 fam- excluded and might contribute to stabilization of the ily the α-helix N-terminal of domain  1 is absent and the protein. Thus, the crucial cofactor binding for structural protein is monomeric. Domain  I is a mixed α/β domain integrity and function of Ohys remains an elusive ques- composed of a parallel five-stranded β-sheet packed tion until now and hampers industrial approaches so far. between two α-helices on one side and a three-stranded For most industrial processes, heterogeneous catalysis antiparallel β-sheet on the other side (Fig.  2). Domain  I is the most common and preferred method. For economic resembles a variant of the Rossmann fold. Domain II con- reasons, enzymes are often preferred to be immobilized sists of an antiparallel β-sheet (Fig.  2) flanked by three on solid supports [57–59]. However, each cycle of reuse α  helices defining the cofactor- and substrate-binding induces a new equilibrium between medium and enzyme site in conjunction with domain  I. Domain  III is exclu- and thus over time, part of the cofactors can be lost, par- sively α-helical (Fig.  2) and its fold is structurally related ticularly in those enzymes with low binding affinity. This to monoamine oxidases [50]. Together, domain II and III applies to Ohys, since they have weak binding affinity form a tunnel to guide the substrate into the active site. towards FAD [26, 48]. This leads to either partial or com - The C-terminal domain IV differs in size and if extended, plete loss of FAD and activity. OhyRe loses both cofactor contributes to the dimer interface (Fig.  2A, C). Domain and activity, and OhyEm has only 86% of cofactor load IV undergoes a large conformational change upon sub- [46]. OhyLa has been reported to lose FAD after exten- strate binding [51], suggesting a role of domain IV in sub- sive washing on an ion-exchange or affinity column and strate recognition in conjunction with domain II and III. after gel-filtration [26]. In former immobilization experi - Notably, the most significant structural differences are ments with OhyEm, a loss of activity after each round of found for domain IV of all known Ohys, which could be reaction has been observed. The loss of FAD might be a caused by the size of domain IV and/or its involvement possible explanation [37]. Thus, elucidating crucial amino in substrate recognition. Hence, cofactor recognition and acids for binding of FAD would aid in engineering the binding play a crucial role for Ohy activity in the differ - enzymes towards optimized variants, with a higher affin - ent families implementing different catalytic pathways. ity towards FAD. Therefore, the role of cofactor binding will be discussed In domain I, the FAD-binding pocket is defined by the in the next paragraph. Rossman-fold as well as a lid region, that undergoes a conformational change upon binding of FAD. Latter con- Functional role of FAD in Ohys formational change ultimately leads to a closure of the Ohys are lyases, which don’t necessarily require a redox- FAD-binding pocket with the lid segment in close prox- active cofactor. However, all known Ohys display a imity to the isoalloxazine ring, the diphosphate function strictly conserved Rossmann-fold or Rossmann-fold like as well as the ribose of FAD. Interestingly, the length of secondary structure motif [51], which are specific for the lid segment differs between HFam family members. binding of FAD or nicotinamide adenine dinucleotide The lid segment has a length of 17 amino acids in all Ohys phosphate NAD(P)H. In flavoproteins, FAD can either so far structurally characterized, with just one exception be bound covalently or non-covalently [52]. In the case for OhyRe (Fig.  3A). Here, the lid is significantly longer of non-covalent binding, van-der-Waals and ionic inter- with 27 amino acids (Fig.  3A). One might ask, whether actions play a crucial role. As a result, FAD is bound via this could be a structural feature of members of the an on–off mechanism, that depending on the strength HFam3 family. Hence, we aligned all available amino of binding can be more or less profound. Upon dilution, acid sequences of subfamily HFam3, available in the flavin molecules can be released from a protein even assembled “hydratase engineering database” [25]. Our when they have picomolar binding-affinity [53]. In Ohys, sequence analysis clearly revealed that all HFam3 family Prem et al. Microbial Cell Factories (2022) 21:58 Page 6 of 15 Fig. 3 Sequence conservation of the FAD lid and the activation loop and architecture of the active site of OhySa, OhyEm, and OhyRe. A Amino acid sequence alignment of OhyRe (Uniprot: T5I9M6), OhySa (Uniprot: A0A0D6GJV1), OhyLa (Uniprot: Q5FL96), OhyEm (Uniprot: OLHYD), OhySt (Uniprot: A0A126NKL7) restricted to the FAD-lid and the activation loop. Conserved residues in lid and activation loop are highlighted by yellow background. The catalytic residue in the loop is highlighted with a light blue or light orange box, respectively. Highly conserved residues are indicated with asterisk, moderate conservation with two points, low conservation with one point. Primary sequences of Ohys were aligned using Clustal Omega [60]. B Active site of OhyEm (PDB-ID 4uir; [46]), shown with catalytic important residues. The bound PEG molecule in close proximity of the active site is shown in orange. Structural elements shown in cartoon representation. C Active site of OhySa (PDB-ID: 7kaz; [47]) shown with important residues lining the active site. The ternary complex of OhySa with bound FAD and oleate was obtained with the OhySa variant E82A. For clarity, we have computationally re-introduced the wild-type situation. FAD, oleic acid and indicated residues shown in stick representation. D Active site of OhyRe (PDB-ID: 5odo; [48]) shown with important residues lining the active site in stick representation. The shown FAD cofactor and oleic acid were obtained by a superposition of the OhySa structure A and derived from the superposition with the structure of OhySa. E Superposition of the active site of OhyRe and OhySa members contain an extended lid with a length of 27 or binding pocket for the approaching substrate. In the 25 amino acid residues, indicating a distinct structural structure of OhySa, the activation loop comprises resi- feature of this family. Despite its length, in all structures, dues from 78 to 83 (Fig. 3A,). Notably, the catalytic E122 the lid segment contains the highly conserved signature of OhyEm as well as E82 of OhySa are located within lat- motif GGXXXG (X any amino acid; Fig.  3A). Notably, ter activation segment (Fig. 3B, C). in HFam3, the motif is altered to GXXXG. Concomi- Analysing the sequence conservation of the FAD-bind- tant with FAD-binding, a loop region, termed “activation ing pocket, clearly reveals a very high degree of sequence loop” by Radka et  al. [47], undergoes a large conforma- conservation of the surface shaping the pocket (Fig.  4). tional change (Fig. 3E). As consequence of FAD-binding, The observed differences in affinity towards FAD could the activation loop almost rotates by 180°, otherwise it be likely attributed to differences in the length and amino would lead to a steric clash. In the FAD-bound state, the acid sequence pattern of the FAD lid, which have con- activation loop is in proximity to the isoalloxazine func- sequences for the conformational flexibility of the lid tion of FAD and secondly, it pre-shapes the substrate region. Such conformational flexibility is also structurally P rem et al. Microbial Cell Factories (2022) 21:58 Page 7 of 15 Fig. 4 Conservation of FAD-binding pocket. A Surface representation of OhyRe (PDB-ID: 5odo; [48]) with conservation of residues from variable to conserved as indicated in the legend. The shown FAD cofactor is depicted as black stick representation and derived from a superposition with OhySa. In the structure of OhyRe, the FAD-lid is disordered and could not be modelled. B Surface representation of OhySa (PDB-ID: 7kaz; [47]) with conservation of residues from variable to conserved as indicated in the legend. The FAD-lid covers the bound FAD molecule. Conservation of Ohys was calculated with the Consurf server [61] reflected. For instance, in the structure of OhySa, a not yet formed. Analysis of the OhyRe structure reveals weaker electron density compared to the protein was that α-helices of domain III are in closer proximity to interpreted as a not fully occupied FAD and fragmented each other, narrowing the channel. In addition, a num- electron density was observed for the lid region, support- ber of amino acid side chains with hydrophobic char- ing an inherent flexibility of the lid region [47]. acter point into the putative channel. Interestingly, many of these residues are conserved or at least simi- lar to OhySa. In the product or substrate bound state Substrate binding of the variant OhySa E82A, the α-helices of domain III Recently, the crystal structure of an OhySa variant bound and side chain rotamers adopt a different conforma - to oleate and FAD was reported [47], giving insights into tion, opening a channel in direction towards the FAD the active site configuration. Previously, structures of molecule. Ohys from other organisms were reported with polyeth- Further, it remains unknown how the substrate per- ylene glycol (PEG) molecules, originating from the crys- suades along the approximately 30 Å long ligand chan- tallization experiment, bound in the cavity in proximity nel from the protein exterior towards the catalytic to FAD. It was proposed, that the PEG might resemble site. A hypothesis could be that the substrate diffuses the substrate [46, 47]. The structure of OhyLa was through the channel towards the active site. Yet another reported with a bound linoleic acid in domain IV, distinct possibility could be a partial opening of the protein with from the active site [26]. subsequent binding of the substrate. The initially bound Superposition of all bound ligands in the structure substrate could then further diffuse along the channel. of OhySa (Fig.  5) illustrates, that a substrate channel is However, passive diffusion of the substrate seems to be built from the distal part (linoleic acid) to the active site rather atypical for such long ligand channels, since also (10-HSA). Oleic acid occupies this tunnel in between the product needs to diffuse through this channel to the linoleic acid and 10-HSA (Fig.  5). Mainly domain III exterior of the enzyme and passive diffusion would also and domain IV build up the ligand channel, which is not be time efficient in the catalytic process. Moreover, lined by hydrophobic amino acids, allowing the mainly a pure diffusion mechanism might be unlikely since the hydrophobic substrate to diffuse into the active site chemical structure of oleic acid with its C9 cis double niche. The role of the flexible domain  IV in the cata bond makes the substrate rather rigid. In absence of lytic cycle of Ohys remains elusive. Interestingly upon substrate or product, water molecules should, at least binding of ligands a conformational shift of the domain partially fill the empty ligand channel in the apo state IV is observed in the structure of OhyLa [26]. Notably, of Ohys. Latter water molecules need to be expelled for the monomeric OhyRe, belonging to the HFam3 upon substrate binding or substrate diffusion along family, the domain IV is significantly reduced in size the cavity. Moreover, we detected a side opening in the compared to dimeric Ohys. Moreover, calculation of structure of OhySa (Fig.  5), through which water mol- potential ligand channels in the structure of OhyRe in ecules could be pushed out by the moving substrate on its apo state was not possible, indicating that the sub its trajectory towards its binding site. This shorter side strate channel in the structure of OhyRe is blocked or Prem et al. Microbial Cell Factories (2022) 21:58 Page 8 of 15 Fig. 5 Substrate, product tunnel and FAD-binding site in OhySa. A OhySa in light orange cartoon representation and the predicted substrate, product and FAD cavities in surface representation in yellow and green. The substrate/product channel from the exterior of the protein towards the FAD is shown in green and numbered “3”. The cavity with bound FAD is depicted in yellow and numbered with “2”. A side channel in vicinity of the FAD cavity is labelled with “3”. The tunnels were calculated with Caver 3.0 [62]. FAD is shown in black stick representation. Oleic acid (PDB ID: 7kaz; [47]) is shown in light green stick representation; 10-HSA bound to OhySa (PDB ID: 7kaz; [47]) is shown in dark blue stick representation as well as linoleic acid bound to OhyLa (PDB ID: 4ia6; [26]) is shown in light violet stick representation; Dashed box displays magnification area as shown in B. B Magnification of dashed box in A with the protein omitted. The ligands linoleic acid, oleic acid and polyethylene glycol are depicted as in A. The ligands guide the substrate channel from the distal end of the channel to the proximal catalytic cleft close to the FAD cavity is mainly lined by hydrophilic amino acids and with the carbon skeleton are possible and only the was described in the structure of OhyEm [46]. charged carboxyl moiety could be sensed by the protein Interestingly, linoleic acid binds to OhyLa with the environment. One could imagine that for a stereospecific carboxylate function facing outward and the hydropho- hydroxylation, the substrate would have to be held in a bic poly-carbon tail threaded into the channel formed very precise, defined position to avoid any side reactions. by domain IV ([26] Fig.  5B). It should be noted, that the electron density interpreted as linoleic acid is weak. A clear decision on the orientation of the carboxy- Reaction mechanism late is very difficult and modelling of the carboxylate is The overall architecture of the active site is well pre - based on the observation of Arg and Lys residues in the served within all structurally characterized Ohys. Nev- closer neighbourhood to the carboxylate. In contrast, in ertheless, substantial differences can be noted between the structure of OhySa, the substrate oleic acid and the the members of the different HFam subfamily members product 10-HSA are both bound with the carboxyl group and will be discussed here. Based on the crystal structure facing inward (Fig.  5B). Additionally, the binding mode and docking studies, a reaction mechanism for OhyEm of the ligands is not thoroughly understood yet. Several was proposed, where Y241 initially protonates the dou- amino acids lining the active site of Ohys need to rear- ble bond of the substrate. E122 activates a water molecule range during binding of the substrates. Especially R81 that can quench the carbocation [46, 56, 63]. Recently, of OhySa is a crucial residue in binding of the substrate a similar function was proposed for the catalytic E82 of (Fig.  3B), since it acts as a block before the entry of the OhySa (Fig.  3C) [47]. In contrast to the earlier proposed substrate and its site chain rotates about 180° upon FAD- function, Y201 is hydrogen bonded to the backbone car- binding. As a consequence, the guanidinium function of bonyl of V505. Consequently, Y201 cannot donate a pro- R81 points in the direction of the approaching substrate. ton to the double bond. Conversely, our modelling data Surprisingly, based on the structure of OhySa with bound indicates that it is involved in a hydrogen bonding net- oleic acid, the positively charged guanidinium function work including the hydroxylated oleate. A hydronium is not directly involved in recognition of the negatively ion is stabilized by an α-helical dipole and a cation of an charged carboxylate of oleic acid [47]. Given the chemical acidic proton of E122. Subsequently, the hydronium ion structure of oleic acid, only van der Waals interactions attacks the substrate. Upon release of FAD, the proton P rem et al. Microbial Cell Factories (2022) 21:58 Page 9 of 15 is released to the hydrated active site and in turn to the but oleic acid is not equally distributed, hence care has C10-hydroxylated product [47]. to be taken on using stock suspensions for enzymatically Instead of a glutamate in the active site at an equiva- determined kinetic measurements. This is critical for the lent position of E122 in OhyEm or E82 in OhySa, belong- measurement of kinetic parameters but also for the reac- ing to HFam11 and HFam2, respectively, a methionine tion itself. When oleic acid is not brought into suspen- residue is located in OhyRe, belonging to HFam3 family sion, droplets are formed, lowering the access towards (Fig. 3E). We were puzzled whether a methionine residue the catalyst and thus a decreasing yield, as observed by within the activation loop could be a conserved feature Jeon et. al., which could be resolved by more efficient within the HFam3. Amino acid sequence analysis of the mixing resulting in the formation of a suspension [67]. “hydratase engineering database” [25] revealed a strict Furthermore, the pH and the temperature have effect conservation of the methionine residue. Mutational stud- on the formation of certain fatty acid species. At a pH ies of OhyRe variants M77E showed a drastically reduced lower than 6, fatty acids usually form crystals and at a hydroxylation activity compared to the wild type enzyme pH between 7 and 9, they are in a 1:1 acid to soap ratio, [48]. A plausible role of the methionine could be a sta- which have extremely low solubility in water [68]. Ohys bilization of the emerging carbocation [64]. Given the can have diverse pH-optima, OhyEm has two pH-optima fundamental difference in chemistry of glutamate versus at pH 6 and 8 [24], OhyRe at 7 [48] and Ohy from Rhodo- methionine poses the question, whether Ohys belong- coccus pyridinivorans has an optimum at pH 5 [69]. It is ing to the HFam3 family employ a different reaction unclear what kind of effects those fatty acid species have mechanism. The family-specific patterns such as dimeric on the enzyme and the reaction. Additionally, 10-HSA is versus monomeric enzymes; differences in the length of a product that has no solubility in water and depending the FAD-lid as well as different decoration of the active on the experimental reaction environment, presence of site niche could hint at different reaction mechanisms certain surfactants or mixing speeds generates different and explain the differences in substrate recognition. In forms of white aggregates in a solution. However, tak- summary, the observed differences might indicate a con - ing samples from a solution can only provide an initial vergent evolution of Ohy families from different ances - overview due to misleading distributive effects and enzy - tors. Consequently, these observations path the way for matic kinetic measurements shall be performed in single a deeper understanding and implementation of Ohys in reactions. biotechnological pathways and will help to employ such Also, 10-HSA has low solubility in many organic sol- enzymes for the chemical industry. In the following chap- vents. That is why full extraction of product and substrate ter a more detailed overview on biotechnological and can’t be achieved under certain extraction conditions and industrial application of Ohys will be given. consequently their measurement is distorted. A common extraction solvent is ethyl acetate after acidification of Industrial application—up and downstream the reaction solution, but also chloroform/methanol is innovation used [48, 70]. However, the solubility of 10-HSA in those Enzymes are currently used in a wide variety of industrial extraction solvents for analytical purposes has not been processes. These traditionally include the food, feed, pol - reported in most studies. Furthermore, several methods ymer, leather and cosmetics sectors. Moreover, enzymes for purification of 10-HSA in a preparative scale have are also used as functional detergent additives and in been performed. In one study, fractionation using ace- organic synthesis of specialty chemicals [65]. However, tone and acetonitrile has been performed at low temper- not all enzymes make it into an industrial process for atures resulting in a purity of over 99% [71]. several reasons and overcoming those challenges is one Mostly, gas chromatography is the analysis method of of the major tasks of protein scientists. The performance choice and for that, derivatization of the fatty acids has to and the costly development of processes are the main take place to reduce adsorption effects. This is achieved hurdles for using enzymes in an industrial environment by methylation or silylation of the carboxy and hydroxy [66]. groups [72, 73]. Additionally, at present, non-derivatized 10-HSA cannot be commercially purchased and thus the Measuring kinetic parameters standard has to be prepared in-house. For that, however, For many Ohys, enzymatically determined kinetic an internal standard is crucial, to evaluate the derivati- parameters have been published, and those could be used zation efficiency and evaporation effects. Only then, the for industrial process simulations and cost evaluations. instrument can properly be calibrated. However, for Ohys they have to be considered with cau- To sum up, for enzymatic kinetic measurements, which tion, as the substrate, oleic acid, is not fully miscible in are important for industrial process simulations, sev- water. Suspensions can be prepared by vigorous mixing, eral considerations are necessary. Only under certain Prem et al. Microbial Cell Factories (2022) 21:58 Page 10 of 15 conditions, appropriate enzymatic kinetic parameters bulky substrates such as Ohys [25]. One study aimed to can be determined. Those apply, when oleic acid is added overcome this effect by decreasing the route between purely to the reaction, when complete extraction of sub- catalyst and substrate. For that, the enzyme was tar- strate and product is performed and the standards are geted into the periplasm using a signal peptide. The prepared carefully. Furthermore, enzyme kinetics should whole-cell reaction using the periplasmatic enzyme be performed in single reactions and it is important resulted in a tenfold higher hydration rate compared to to keep in mind that they are not comparable between the cytoplasmatic reaction. It is known, that the redox- enzymes and studies due to varying reaction conditions. environment plays a role on the activity of Ohys but this has not been discussed [76]. Currently the highest reported STYs for producing Performance of whole‑cell catalysts −1 −1 10-HSA with a whole-cell catalyst is 8–12 g  L  h . For The performance of a catalyst is crucial for every indus - comparison, the production of acrylamide with nitrile trial process, in heterogenous catalysis the space–time −1 −1 hydratase, which is one of the most efficient whole- yield (STY) can be around 1 to 10  kg  L  h . However, cell biocatalytic processes in the industry, gives STYs when looking at biocatalysts, STYs can be decreased by −1 −1 −1 −1 between 53 and 93  g  L  h [66, 77]. To reach such a up to 1000 times to around 0.001–0.3  kg  L  h com- level for the production of hydroxylated fatty acids, sig- pared to conventional processes [66]. While this certainly nificant process optimization is required. However, using can be tolerated by the pharmaceutical industry with a whole-cell catalysts also brings disadvantages particu- need for enantiomeric purity and high-quality products, larly for this certain application. Free, unsaturated fatty expensive processes for a final product, that is mostly acids might convey toxic effects on the whole-cell cata - used as an additive such as 12-HSA, will most likely not lysts upon a certain concentration since the detoxifying sustain. fatty acid hydratases usually are expressed in the cytosol. An enzymatic industrial process can be installed in dif- Additionally, as already mentioned the mass transport ferent modes. First, either wild-type or genetically engi- of substrate and product is hindered by the membrane. neered whole-cells can be used to convert oleic acid. Therefore, it is important to keep the fatty acid content The advantage here is, that no further purification of the under a critical toxic concentration and for the latter enzyme is needed, only the extraction of fatty acids and issue, organic solvents or surfactants such as Tween80 the purification of 10-HSA. Usually, whole-cell conver - can be added. Those additives, however, can influence the sions apply, when large gene clusters and cascades are energy metabolism within the cell, increase the costs of a involved in the formation of a product or if enzymes are process and might complicate the purification. So other not soluble or active when being isolated. strategies such as genetic engineering are investigated Recombinant Escherichia coli expressing an Ohy [78, 79]. from Stenotrophomonas maltophilia has been used as Additionally, in the aforementioned studies, samples whole-cell catalysts in a 1 mL scale leading to a STY of −1 −1 of the reaction medium were taken and extracted using 12.3 g  L  h , however, when they scaled-up to 1 L, the −1 −1 organic solvents such as ethyl acetate. Consequently, STY decreased to 8.2 g  L  h , presumably due to the endogenous fatty acids and hydrophobic molecules changed reaction conditions omitting the buffer and from the cells are extracted and appear as impurities in oxygen-depletion [67, 74]. Furthermore, the authors the final product. Whereas this might be no problem for mention that genetically modified E. coli has a three - industrial products, formulations for the pharmaceuti- fold higher formation rate than the wild-type strain. In cal or cosmetic industry certainly have higher standards other studies, the original organisms have been used, regarding the purity and more laborious downstream however the STYs were quite low compared to E. coli, processing is required to further purify the product. except for one study with Stenotrophomonas nitritire- −1 −1 Other possibilities are the filtration but this might come ducens, where 7.9  g  L  h was achieved [74]. How- with a significant product loss since substrate and prod - ever, the prolonged growth of S. nitritireducens and uct might adsorb to the cell exteriors. In one study about the maximum achievable cell concentration compared whole-cell biocatalysis, 30% of loss was observed after to E. coli were not considered. With S. maltophilia for downstream processing [67]. The high product losses instance, only 10  g/L of maximal cell concentration due to using whole-cell biocatalysts can also be attrib- can be achieved compared to 100  g/L in E. coli in fed- uted to the faster saturation of extraction solvents due to batch cultures [75]. Thus, the authors concluded that E. hydrophobic molecules from the cell. Lastly, it is much coli as whole-cell catalyst is more advantageous com- harder to recycle whole-cell catalysts, particularly when pared to wild-type strains. Mass transport limitations the product is solid and centrifugal forces do not lead to a are hurdles during reactions with whole-cell catalysts separation of product and catalyst. and this is particularly the case for enzymes converting P rem et al. Microbial Cell Factories (2022) 21:58 Page 11 of 15 Performance of Ohys in lysates and pure formes longer and thus the stability after each cycle might not be That is why in some cases, lysates or purified enzymes the same as shown in that study. Furthermore, it has not might be more desirable. Contrarily to some other fatty been investigated how the activity of immobilized Ohy acid converting enzymes or other hydratases [80, 81], changes during extended storage for days [37]. Ohys achieve high expression rates and solubility and Since isolated enzymes are less protected when they don’t rely on stoichiometric amounts of FAD [56], which are not part of a whole-cell catalyst, their stability and makes them excellent candidates for use in pure form or maintenance of activity over a long time plays a crucial in lysates. Many screenings of activities of Ohy have been role for an efficient process. In a few studies, low stabil - performed using lysates and lyophilized lysates [25], and ity of Ohys has been observed. Some lose their activity a patent described the large-scale production of 10-HSA already after a short period of time [25, 37]. In a com- with lysate of R. erythropolis and S. maltophilia [82]. Fur- parative study on enzyme stability, five different Ohys thermore, a pilot scale with cell-free extract has been were analysed. It was found that all of them started to performed using a variant of Paracoccus aminophilus denature already after one day within lysates, leading to −1 −1 with a STY of 22.5 g  L  h [83]. an exposure of their hydrophobic sites. As a result, nei- Since lysates are difficult to recycle and have weak ther substrate nor product was measurable anymore, stability; pure, immobilized enzymes are in some cases most likely since they interact with the hydrophobic sites the method of choice. Additionally, immobilization can of the denatured protein bulk. Buffer optimization led result in higher stability, increased activity and improved to certain improvements regarding the protein stability stereoselectivity and efficient recycling lowers the costs [25]. In another study it was reported that OhyEm loses [84]. However, not all enzymes can be immobilized and 60% of its activity already after 7 days at 4 °C. Todea et al. recycled for several rounds in native form and not all suspected OhyEm to inactivate as a result of the disso- products have good biocompatibility with the solid sup- ciation of subunits [37]. In general, however, not much is ports. Ways to overcome these challenges are on the way published about the stability of Ohys over a longer period by developing novel supports and materials for immobi- of time since most studies have no industrial but rather a lization and by using state-of the art technologies in the medicinal background. Todea et al. have used additives in field of protein engineering [84]. order to overcome the stability problems. This has been One of the main issues for immobilization is the men- investigated by storing the protein for 7 days at 4 °C and tioned insolubility of 10-HSA in water and thus the testing its residual activity. However, no experiments catalyst cannot easily be separated by centrifugation. have been conducted what effects the additives have on Furthermore, oleate is a hydrophobic molecule and thus the process stability with several re-usage cycles and at attaches to certain materials used as solid supports. At elevated temperatures. Additionally, additives can com- present, only one study exists, where an Ohy has been plicate the process since they might have to be removed immobilized. Several issues occurred while testing dif- before the reaction starts and they increase the price of ferent kinds of support. The recovery of the product a process [37]. In general, the main reasons for a loss without harming the enzyme in form of cross-linked of protein activity is either the distortion of the tertiary enzyme aggregates was not possible. To ease the separa- structure, the dissociation of cofactors, chemical inac- tion of catalyst and product, magnetic beads were used, tivation when a reactive chemical is part of the reaction however that resulted in adsorption of substrate and or—as the first step of inactivation for multimeric pro - product to the support. The magnetic beads were coated teins—the dissociation of subunits [85]. Consequently, with a layer of chitosan to avoid the adsorption. Still, in multimeric and FAD-bound enzymes are more affected all immobilization techniques, not more than 24% of the and less advantageous in industrial processes. First of all, residual activity has been recovered. The least residual a monomeric enzyme can overcome the issues of subunit activity was observed for the entrapment of the enzyme dissociation and an enzyme working without FAD can’t since organic solvents were used in that method known be subject to cofactor loss. However, currently OhyRe to inactivate the enzyme. In general, for entrapment— is the only known monomeric Ohy and it loses FAD even by other means where stability is maintained—the during purification resulting in a loss of function [48]. biggest issue is the transport of oleic acid in aqueous Consequently, this particular enzyme still requires opti- solutions towards the active sites. The chitosan-coated mizations in order to be used in isolated form since the magnetic beads as best candidates were finally chosen addition of FAD renders it too costly. for testing rounds of recycling and after 5 cycles, still 70% of initial activity was left. Each reaction of a cycle Protein engineering of Ohys was performed for 2  h, however, usually, reactions with Protein engineering is one of the main methods to over- Ohys with high concentrations of oleic acid take much come the several drawbacks of Ohys. Substrate spectrum Prem et al. Microbial Cell Factories (2022) 21:58 Page 12 of 15 and selectivity, cofactor binding, stability and turnover using directed evolution. In that context, a coupled assay number are attributes, which are desirable to improve. has been used to screen for optimized variants. 10-HSA Directed evolution and site-directed mutagenesis are the was converted into 10-oxostearic acid by an alcohol two main methods for improving proteins. dehydrogenase, the occurring coenzyme NADH was col- The crystal structure of OhyEm has been reported orimetrically analysed and the variant with the highest with bound FAD, but not with substrate or product [46]. colorimetric output was further analysed. For this type However, an electron density in the proposed substrate of assay usually performed in lysates the specificity of the binding cavity has been interpreted as a PEG molecule alcohol dehydrogenase is crucial [83]. thought to be a substrate mimic. To manipulate the sub- strate spectrum, structure-guided protein engineering Conclusions using site-directed mutagenesis has been performed. Ohys belong to the fatty acid hydratases enzyme fam- Hence, amino acids belonging to the pocket of the fatty ily, which is unique for its co-factor free modification of acid head group were altered. Some of those variants free, unsaturated fatty acids. It is suggested, that Ohys could convert derivatives of oleic acid such as ethyl- and primarily evolved to protect microorganisms from toxic n-propyl oleic acid, stearyl alcohol or stearyl amine at effects by incorporation of free, unsaturated fatty acids higher rates [54]. In another study, the substrate spec- into the cell membrane. At the beginning of the 1960s trum was altered towards alkenes with a terminal or also Ohy’s products could be isolated from animal and internal double bond by the addition of a dummy carboxy human faeces, which were assumed to originate from acid to artificially expand the size of the substrate and bacteria colonizing the gut. Recently, Ohy evoked indus- by decreasing the size of the substrate binding pocket trial interest for the conversion of oleic acid to sustain- by mutagenesis. Since the location of PEG was not suffi - able 10-HSA, which can replace 12-HSA in oleochemical cient, a structure of OhyEm with a docked oleic acid was and cosmetics applications, that is currently generated used [86]. by hydrogenation of castor oil. Although 12-HSA has a Another attempt to alter the substrate spectrum of high application spectrum as additive ranging from oils Ohys by rational-mutagenesis has been demonstrated and paints via manufacturing of rubbers to use in food by Eser et  al. In their study, residues of the active side and pharmaceutical industry, the educt castor oil is lim- of Ohys with 76% homology originating both from Lac- ited and fluctuates in quality. Furthermore, the hydration tobacillus acidophilus have been compared and their process needs high pressure and temperature conditions functionalities have been estimated by using the crys- to obtain economically sound yields. Therefore, indus - tal structure of OhyEm. One of the enzymes FA-HY2 try demands for a more sustainable and quality sta- is unique since it is able to convert substrates up to the ble replacement, which can be provided in theory by length of 22, whereas the other one serves as a rather 10-HSA. typical Ohy (FA-HY1) converting a substrate length of As a consequence, scientists became interested in 16–18. In conclusion the substrate preference and regi- Ohys, which is documented by an increase in articles oselectivity of FA-HY1 could be changed by swapping characterising Ohys from different microorganisms. critical residues from FA-HY2 [63]. While numerous new Ohys have been described recently, These findings suggest that site-directed mutagenesis there is a limited understanding concerning structure– has great potential when the crystal structure is fully function relationships in this structurally diverse enzyme unravelled and the location of substrates and products is family. Specifically, more insights on detailed reaction clear or can be cleared by docking experiments. The crys - mechanisms are required. In that context, it is unclear tal structure of OhyRe is neither resolved with FAD nor how the highly elongated substrate reaches the active substrate and docking of neither cofactor nor substrate centre and how the hydroxylated products are released has been successful so far. This can be due to many rea - after commencement of the reaction. Additionally, a big sons but a high-quality structure after docking can only controversy exists regarding the role of the FAD mol- be achieved when the underlying biomolecule isn’t sub- ecule bound in the structure. ject to large conformational changes upon binding of the Moreover, with the exception of the monomeric OhyRe docked molecule [87]. That is why directed evolution is recently described, all other Ohys deciphered today are sometimes a much more powerful tool specifically for dimers. Hence, OhyRe is amenable for efficient immo - improvements that are difficult to address by structure- bilisation, which makes it attractive for industrial appli- based methods, such as melting temperature or affinity cations, which require extended residence times of the towards certain small molecules. Furthermore, results biocatalyst in target reactions to reduce costs. However, can be achieved at much higher pace. An Ohy from Par- current literature studies revealed that OhyRe has a sig- acoccus aminophilus has been successfully enhanced by nificantly different domain architecture to other Ohys, P rem et al. Microbial Cell Factories (2022) 21:58 Page 13 of 15 3. Stenz L, Francois P, Fischer A, Huyghe A, Tangomo M, Hernandez D, Cassat which suggests that this enzyme may follow a different J, Linder P, Schrenzel J. Impact of oleic acid (cis-9-octadecenoic acid) on reaction mechanism. bacterial viability and biofilm production in Staphylococcus aureus. FEMS Currently, it is the yet unknown OhyRe reaction mech- Microbiol Lett. 2008;287:149–55. 4. Maia MR, Chaudhary LC, Bestwick CS, Richardson AJ, McKain N, Larson anism and structure–function relationship of this enig- TR, Graham IA, Wallace RJ. Toxicity of unsaturated fatty acids to the bio- matic monomeric enzyme as well as the complicated hydrogenating ruminal bacterium Butyrivibrio fibrisolvens. BMC Microbiol. purification process for the target product 10-HSA, 2010;10:52. 5. Zheng CJ, Yoo JS, Lee TG, Cho HY, Kim YH, Kim WG. Fatty acid synthesis which hamper further industrial scaling of the reaction. is a target for antibacterial activity of unsaturated fatty acids. FEBS Lett. Therefore, a broader knowledge-base is needed to enable 2005;579:5157–62. industrial adaptation of biotechnological 10-HSA pro- 6. Kim KR, Oh DK. Production of hydroxy fatty acids by microbial fatty acid- hydroxylation enzymes. Biotechnol Adv. 2013;31:1473–85. duction to replace thermocatalytic 12-HSA synthesis. 7. Galbraith H, Miller TB. Eec ff t of metal cations and pH on the antibac- terial activity and uptake of long chain fatty acids. J Appl Bacteriol. 1973;36:635–46. Abbreviations 8. Parsons JB, Yao J, Frank MW, Jackson P, Rock CO. Membrane disruption FAD: Binding of flavin adenine dinucleotide; Ohy: Oleate hydratase; PEG: Poly- by antimicrobial fatty acids releases low-molecular-weight proteins from ethylene glycol; Rmsd: Root mean square deviation; STY: Space–time yield. Staphylococcus aureus. J Bacteriol. 2012;194:5294–304. 9. Galbraith H, Miller TB, Paton AM, Thompson JK. Antibacterial activity of long chain fatty acids and the reversal with calcium, magnesium, ergocal- Supplementary Information ciferol and cholesterol. J Appl Bacteriol. 1971;34:803–13. The online version contains supplementary material available at https:// doi. 10. McKain N, Shingfield KJ, Wallace RJ. Metabolism of conjugated linoleic org/ 10. 1186/ s12934- 022- 01777-6. acids and 18: 1 fatty acids by ruminal bacteria: products and mechanisms. Microbiology. 2010;156:579–88. Additional file 1. Additional Tables. 11. Pappas A. Epidermal surface lipids. Dermatoendocrinol. 2009;1:72–6. 12. Man YBC, Moh MH, van de Voort FR. Determination of free fatty acids in crude palm oil and refined-bleached-deodorized palm olein using Fou- Acknowledgements rier transform infrared spectroscopy. J Am Oil Chem Soc. 1999;76:485–90. T. B. gratefully acknowledges funding by the Werner Siemens foundation for 13. Subramanian C, Frank MW, Batte JL, Whaley SG, Rock CO. Oleate establishing the field of Synthetic Biotechnology at the Technical University of hydratase from Staphylococcus aureus protects against palmitoleic acid, Munich ( TUM). the major antimicrobial fatty acid produced by mammalian skin. J Biol Chem. 2019;294:9285–94. Author contributions 14. Radka CD, Batte JL, Frank MW, Rosch JW, Rock CO. Oleate hydratase BL and DG conceived the review. SP, CPOH, ND, SH, and TB contributed to the (OhyA) is a virulence determinant in Staphylococcus aureus. Microbiol manuscript. CPOH: generated the figures with advice of ND. All authors read Spectr. 2021;9:e0154621. and approved the final manuscript. 15. Rosberg-Cody E, Liavonchanka A, Gobel C, Ross RP, O’Sullivan O, Fitzger- ald GF, Feussner I, Stanton C. Myosin-cross-reactive antigen (MCRA) pro- Funding tein from Bifidobacterium breve is a FAD-dependent fatty acid hydratase Open Access funding enabled and organized by Projekt DEAL. N.D. and B.L. which has a function in stress protection. BMC Biochem. 2011;12:9. are supported by the BMBF Grant 031B0853. S.P. was supported by BMBF 16. O’Connell KJ, Motherway MO, Hennessey AA, Brodhun F, Ross RP, Grant 03SF0577A. Feussner I, Stanton C, Fitzgerald GF, van Sinderen D. Identification and characterization of an oleate hydratase-encoding gene from Bifidobacte - Availability of data and materials rium breve. Bioengineered. 2013;4:313–21. Not applicable. 17. Yao JW, Rock CO. How bacterial pathogens eat host lipids: implications for the development of fatty acid synthesis therapeutics. J Biol Chem. 2015;290:5940–6. Declarations 18. Kengmo Tchoupa A, Peschel A. Staphylococcus aureus releases proinflam- matory membrane vesicles to resist antimicrobial fatty acids. mSphere. Ethics approval and consent to participate 2020;5:e00804. Not applicable. 19. Ortega-Anaya J, Hernandez-Santoyo A. Functional characterization of a fatty acid double-bond hydratase from Lactobacillus plantarum and Consent for publication its interaction with biosynthetic membranes. Biochim Biophys Acta. Not applicable. 2015;1848:3166–74. 20. Hirata A, Kishino S, Park SB, Takeuchi M, Kitamura N, Ogawa J. A novel Competing interests unsaturated fatty acid hydratase toward C16 to C22 fatty acids from The author declares that there are no competing interests associated with this Lactobacillus acidophilus. J Lipid Res. 2015;56:1340–50. manuscript. 21. Kang WR, Seo MJ, Shin KC, Park JB, Oh DK. Comparison of biochemical properties of the original and newly identified oleate hydratases from Received: 9 February 2022 Accepted: 18 March 2022 Stenotrophomonas maltophilia. Appl Environ Microbiol. 2017;83:e03351. 22. Kim KR, Oh HJ, Park CS, Hong SH, Park JY, Oh DK. Unveiling of novel regio- selective fatty acid double bond hydratases from Lactobacillus acidophilus involved in the selective oxyfunctionalization of mono- and di-hydroxy fatty acids. Biotechnol Bioeng. 2015;112:2206–13. References 23. Kishino S, Takeuchi M, Park SB, Hirata A, Kitamura N, Kunisawa J, Kiyono H, 1. Dilika F, Bremner PD, Meyer JJ. Antibacterial activity of linoleic and oleic Iwamoto R, Isobe Y, Arita M, et al. Polyunsaturated fatty acid saturation by acids isolated from Helichrysum pedunculatum: a plant used during gut lactic acid bacteria affecting host lipid composition. Proc Natl Acad circumcision rites. Fitoterapia. 2000;71:450–2. Sci USA. 2013;110:17808–13. 2. Greenway DL, Dyke KG. Mechanism of the inhibitory action of lin- 24. Bevers LE, Pinkse MW, Verhaert PD, Hagen WR. Oleate hydratase catalyzes oleic acid on the growth of Staphylococcus aureus. J Gen Microbiol. the hydration of a nonactivated carbon-carbon bond. J Bacteriol. 1979;115:233–45. 2009;191:5010–2. Prem et al. Microbial Cell Factories (2022) 21:58 Page 14 of 15 25. Schmid J, Steiner L, Fademrecht S, Pleiss J, Otte KB, Hauer B. Bio- conformational plasticity surrounding the FAD binding site. Biochem catalytic study of novel oleate hydratases. J Mol Catal B Enzymatic. Biophys Res Commun. 2018;499:772–6. 2017;133:S243–9. 50. De Colibus L, Li M, Binda C, Lustig A, Edmondson DE, Mattevi A. Three- 26. Volkov A, Khoshnevis S, Neumann P, Herrfurth C, Wohlwend D, Ficner dimensional structure of human monoamine oxidase A (MAO A): relation R, Feussner I. Crystal structure analysis of a fatty acid double-bond to the structures of rat MAO A and human MAO B. Proc Natl Acad Sci U S hydratase from Lactobacillus acidophilus. Acta Crystallogr D Biol Crystal- A. 2005;102:12684–9. logr. 2013;69:648–57. 51. Rao ST, Rossmann MG. Comparison of super-secondary structures in 27. James AT, Webb JP, Kellock TD. The occurrence of unusual fatty acids in proteins. J Mol Biol. 1973;76:241–56. faecal lipids from human beings with normal and abnormal fat absorp- 52. Dym O, Eisenberg D. Sequence-structure analysis of FAD-containing tion. Biochem J. 1961;78:333–9. proteins. Protein Sci. 2001;10:1712–28. 28. Wallen LL, Benedict RG, Jackson RW. The microbiological produc- 53. Bollen YJ, Westphal AH, Lindhoud S, van Berkel WJ, van Mierlo CP. Distant tion of 10-hydroxystearic acid from oleic acid. Arch Biochem Biophys. residues mediate picomolar binding affinity of a protein cofactor. Nat 1962;99:249–53. Commun. 2012;3:1010. 29. Thomas PJ. Identification of some enteric bacteria which convert oleic 54. Engleder M, Strohmeier GA, Weber H, Steinkellner G, Leitner E, Muller acid to hydroxystearic acid in vitro. Gastroenterology. 1972;62:430–5. M, Mink D, Schurmann M, Gruber K, Pichler H. Evolving the promiscuity 30. Koritala S, Hosie L, Hou CT, Hesseltine CW, Bagby MO. Microbial conver- of Elizabethkingia meningoseptica oleate hydratase for the regio- and sion of oleic-acid to 10-hydroxystearic acid. Appl Microbiol Biotechnol. stereoselective hydration of oleic acid derivatives. Angew Chem Int Ed 1989;32:299–304. Engl. 2019;58:7480–4. 31. Hudson JA, MacKenzie CAM, Joblin KN. Conversion of oleic acid to 55. Hiseni A, Arends IWCE, Otten LG. New cofactor-independent hydration 10-hydroxystearic acid by two species of ruminal bacteria. Appl Microbiol biocatalysts: structural, biochemical, and biocatalytic characteristics of Biotechnol. 1995;44:1–6. carotenoid and oleate hydratases. ChemCatChem. 2015;7:29–37. 32. Feussner I, Hornung E, Liavonchanka A: Production of hydroxy fatty acids 56. Engleder M, Pichler H. On the current role of hydratases in biocatalysis. using fatty acid hydratase from Streptococcus pyogenes. (GÖTTINGEN Appl Microbiol Biotechnol. 2018;102:5841–58. G-A-U ed). 2008. 57. Basso A, Serban S. Industrial applications of immobilized enzymes—a 33. Fameau A-L, Rogers MA. The curious case of 12-hydroxystearic acid—the review. Mol Catal. 2019;479:110607. Dr. Jekyll & Mr. Hyde of molecular gelators. Curr Opin Colloid Interface Sci. 58. Rodrigues RC, Virgen-Ortiz JJ, Dos Santos JCS, Berenguer-Murcia A, 2020;45:68–82. Alcantara AR, Barbosa O, Ortiz C, Fernandez-Lafuente R. Immobilization 34. Borg P, Lê G, Lebrun S, Pées B. Example of industrial valorisation of deriva- of lipases on hydrophobic supports: immobilization mechanism, advan- tive products of Castor oil. Oléagineux Corps gras Lipides. 2009;16:211–4. tages, problems, and solutions. Biotechnol Adv. 2019;37:746–70. 35. Johnson W. Amended final report on the safety assessment of hydroxy- 59. Cole-Hamilton DJ. TRP: Catalyst separation recovery and recycling. Cataly- stearic acid. Int J Toxicol. 1999;18:1–10. sis by metal complexes. In: Cole-Hamilton DJ, editor. Homogeneous 36. Lee B, Lee S, Kim H, Jeong K, Park J, Lee E, Lee J. Biotransformation of catalysis—advantages and problems, vol. 30. Dordrecht: Springer; 2006. oleic acid into 10-ketostearic acid by recombinant Corynebacterium p. 1–8. glutamicum-based biocatalyst. Biotechnol Lett. 2015;37:1101–6. 60. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam 37. Todea A, Hiseni A, Otten LG, Arends IWCE, Peter F, Boeriu CG. Increase of H, Remmert M, Soding J, et al. Fast, scalable generation of high-quality stability of oleate hydratase by appropriate immobilization technique protein multiple sequence alignments using Clustal Omega. Mol Syst and conditions. J Mol Catal B Enzym. 2015;119:40–7. Biol. 2011;7:539. 38. Pari L, Suardi A, Stefanoni W, Latterini F, Palmieri N. Environmental and 61. Landau M, Mayrose I, Rosenberg Y, Glaser F, Martz E, Pupko T, Ben-Tal economic assessment of castor oil supply chain: a case study. Sustainabil- N. ConSurf 2005: the projection of evolutionary conservation scores of ity. 2020;12:6339. residues on protein structures. Nucleic Acids Res. 2005;33:W299-302. 39. Ogunniyi DS. Castor oil: a vital industrial raw material. Bioresour Technol. 62. Chovancova E, Pavelka A, Benes P, Strnad O, Brezovsky J, Kozlikova B, Gora 2006;97:1086–91. A, Sustr V, Klvana M, Medek P, et al. CAVER 3.0: a tool for the analysis of 40. de Cassia de Souza Schneider R, Lara LRS, Ceolin MM, Kaercher JA, Sch- transport pathways in dynamic protein structures. PLoS Comput Biol. neider M. Environmental impact of castor oil catalytic transfer hydrogena- 2012;8:e1002708. tion. Clean Technol Environ Policy. 2013;15:977–85. 63. Eser BE, Poborsky M, Dai R, Kishino S, Ljubic A, Takeuchi M, Jacobsen C, 41. Trivedi RK, Vasishtha AK. Low-pressure hydrogenation of castor-oil. J Am Ogawa J, Kristensen P, Guo Z. Rational engineering of hydratase from Oil Chem Soc. 1988;65:1467–9. Lactobacillus acidophilus reveals critical residues directing substrate 42. Sreenivasan B, Kamath NR, Kane JG. Studies on castor oil. II. Hydrogena- specificity and regioselectivity. ChemBioChem. 2020;21:550–63. tion of castor oil. J Am Oil Chem Soc. 1957;34:302–7. 64. Dixit M, Weitman M, Gao J, Major DT. Chemical control in the battle 43. Naughton FC. Production, chemistry, and commercial applications of against fidelity in promiscuous natural product biosynthesis: the case of various chemicals from castor-oil. J Am Oil Chem Soc. 1974;51:65–71. trichodiene synthase. ACS Catal. 2017;7:812–8. 44. Masri MA, Garbe D, Mehlmer N, Brück TB. A sustainable, high-perfor- 65. Singh R, Kumar M, Mittal A, Mehta PK. Microbial enzymes: industrial mance process for the economic production of waste-free microbial progress in 21st century. 3 Biotech. 2016;6:1–5. oils that can replace plant-based equivalents. Energy Environ Sci. 66. Hauer B. Embracing nature’s catalysts: a viewpoint on the future of bioca- 2019;12:2717–32. talysis. ACS Catal. 2020;10:8418–27. 45. Zheng Y, Li L, Liu Q, Qin W, Yang J, Cao Y, Jiang X, Zhao G, Xian M. Boosting 67. Jeon E-Y, Lee J-H, Yang K-M, Joo Y-C, Oh D-K, Park J-B. Bioprocess the free fatty acid synthesis of Escherichia coli by expression of a cytosolic engineering to produce 10-hydroxystearic acid from oleic acid by Acinetobacter baylyi thioesterase. Biotechnol Biofuels. 2012;5:76. recombinant Escherichia coli expressing the oleate hydratase gene of 46. Engleder M, Pavkov-Keller T, Emmerstorfer A, Hromic A, Schrempf S, Stenotrophomonas maltophilia. Process Biochem. 2012;47:941–7. Steinkellner G, Wriessnegger T, Leitner E, Strohmeier GA, Kaluzna I, et al. 68. Small DM: physical properties of fatty acids and their extracellular and Structure-based mechanism of oleate hydratase from Elizabethkingia intracellular distribution. In: Polyunsaturated fatty acids in human nutri- meningoseptica. ChemBioChem. 2015;16:1730–4. tion. Volume 28. Edited by Bracco U, Deckelbaum RJ. New York: Nestec 47. Radka CD, Batte JL, Frank MW, Young BM, Rock CO. Structure and mecha- Ltd., Vevey/Raven Press. Ltd.; 1992: 25–39.[Nestlé Nutrition Workshop nism of Staphylococcus aureus oleate hydratase (OhyA). J Biol Chem. Séries]. 2021;296:100252. 69. Busch H, Tonin F, Alvarenga N, van den Broek M, Lu S, Daran JM, Hanefeld 48. Lorenzen J, Driller R, Waldow A, Qoura F, Loll B, Brück T. Rhodococcus eryth- U, Hagedoorn PL. Exploring the abundance of oleate hydratases in the ropolis oleate hydratase: a new member in the oleate hydratase family genus Rhodococcus-discovery of novel enzymes with complementary tree-biochemical and structural studies. ChemCatChem. 2018;10:407–14. substrate scope. Appl Microbiol Biotechnol. 2020;104:5801–12. 49. Park AK, Lee GH, Kim DW, Jang EH, Kwon HT, Chi YM. Crystal structure 70. Takeuchi M, Kishino S, Park SB, Hirata A, Kitamura N, Saika A, Ogawa J. Effi- of oleate hydratase from Stenotrophomonas sp. KCTC 12332 reveals cient enzymatic production of hydroxy fatty acids by linoleic acid Delta9 P rem et al. Microbial Cell Factories (2022) 21:58 Page 15 of 15 hydratase from Lactobacillus plantarum AKU 1009a. J Appl Microbiol. 2016;120:1282–8. 71. An JU, Joo YC, Oh DK. New biotransformation process for production of the fragrant compound gamma-dodecalactone from 10-hydroxystea- rate by permeabilized Waltomyces lipofer cells. Appl Environ Microbiol. 2013;79:2636–41. 72. Griffiths MJ, van Hille RP, Harrison ST. Selection of direct transesterification as the preferred method for assay of fatty acid content of microalgae. Lipids. 2010;45:1053–60. 73. Topolewska A, Czarnowska K, Halinski LP, Stepnowski P. Evaluation of four derivatization methods for the analysis of fatty acids from green leafy vegetables by gas chromatography. J Chromatogr B Anal Technol Biomed Life Sci. 2015;990:150–7. 74. Kim BN, Yeom SJ, Oh DK. Conversion of oleic acid to 10-hydroxystearic acid by whole cells of Stenotrophomonas nitritireducens. Biotech Lett. 2011;33:993–7. 75. Joo YC, Seo ES, Kim YS, Kim KR, Park JB, Oh DK. Production of 10-hydroxy- stearic acid from oleic acid by whole cells of recombinant Escherichia coli containing oleate hydratase from Stenotrophomonas maltophilia. J Biotechnol. 2012;158:17–23. 76. Jung SM, Seo JH, Lee JH, Park JB, Seo JH. Fatty acid hydration activ- ity of a recombinant Escherichia coli-based biocatalyst is improved through targeting the oleate hydratase into the periplasm. Biotechnol J. 2015;10:1887–93. 77. Lee CY, Chang HN. Continuous production of acrylamide using immobi- lized Brevibacterium sp. CH2 in a two-stage packed bed reactor. Biotech- nol Lett. 1990;12:23–8. 78. Lin BX, Tao Y. Whole-cell biocatalysts by design. Microb Cell Fact. 2017;16:106. 79. Julsing MK, Cornelissen S, Buhler B, Schmid A. Heme-iron oxygenases: powerful industrial biocatalysts? Curr Opin Chem Biol. 2008;12:177–86. 80. Li Y, Dietrich M, Schmid RD, He B, Ouyang P, Urlacher VB. Identification and functional expression of a Δ9-fatty acid desaturase from Psychrobac- ter urativorans in Escherichia coli. Lipids. 2008;43:207–13. 81. Busch H, Alvarenga N, Abdelraheem E, Hoek M, Hagedoorn PL, Hanefeld U. Re-investigation of hydration potential of Rhodococcus whole-cell biocatalysts towards michael acceptors. ChemCatChem. 2019;12:193–8. 82. Brück T, Lorenzen J: A process for the cell-free enzymatic production of 10-hydroxystearic acid (10-hsa) from bio-based oils for lubricant formula- tion. (München TU ed). 2019. 83. Sun QF, Zheng YC, Chen Q, Xu JH, Pan J. Engineering of an oleate hydratase for efficient C10-Functionalization of oleic acid. Biochem Biophys Res Commun. 2021;537:64–70. 84. Federsel HJ, Moody TS, Taylor SJC. Recent trends in enzyme immobi- lization-concepts for expanding the biocatalysis toolbox. Molecules. 2021;26:2822. 85. Betancor L, Hidalgo A, Fernandez-Lorente G, Mateo C, Rodriguez V, Fuentes M, Lopez-Gallego F, Fernandez-Lafuente R, Guisan JM. Use of physicochemical tools to determine the choice of optimal enzyme: Stabilization of D-amino acid oxidase. Biotechnol Prog. 2003;19:784–8. 86. Demming RM, Hammer SC, Nestl BM, Gergel S, Fademrecht S, Pleiss J, Hauer B. Asymmetric enzymatic hydration of unactivated aliphatic alkenes. Angewandte Chemie-International Edition. 2019;58:173–7. 87. de Vries SJ, van Dijk M, Bonvin AM. The HADDOCK web server for data- driven biomolecular docking. Nat Protoc. 2010;5:883–97. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- fast, convenient online submission lished maps and institutional affiliations. thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Microbial Cell Factories Springer Journals

Towards an understanding of oleate hydratases and their application in industrial processes

Download PDF
15 pages

Loading next page...
 
/lp/springer-journals/towards-an-understanding-of-oleate-hydratases-and-their-application-in-Ly9E0tkuNe

References (95)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
1475-2859
DOI
10.1186/s12934-022-01777-6
Publisher site
See Article on Publisher Site

Abstract

Fatty acid hydratases are unique to microorganisms. Their native function is the oxidation of unsaturated C–C bonds to enable detoxification of environmental toxins. Within this enzyme family, the oleate hydratases (Ohys), which catalyze the hydroxylation of oleic acid to 10-(R)-hydroxy stearic acid (10-HSA) have recently gained particular industrial interest. 10-HSA is considered to be a replacement for 12-(R)-hydroxy stearic acid (12-HSA), which has a broad application in the chemical and pharmaceutical industry. As 12-HSA is obtained through an energy consuming synthesis process, the biotechnological route for sustainable 10-HSA production is of significant industrial interest. All Ohys identified to date have a non-redox active FAD bound in their active site. Ohys can be divided in several sub - families, that differ in their oligomerization state and the decoration with amino acids in their active sites. The latter observation indicates a different reaction mechanism across those subfamilies. Despite intensive biotechnological, biochemical and structural investigations, surprising little is known about substrate binding and the reaction mecha- nism of this enzyme family. This review, summarizes our current understanding of Ohys with a focus on sustainable biotransformation. Keywords: Oleate hydratase, Biocatalysis, Industrial biotechnology, Whole cell and enzymatic oleic acid transformation, Green chemistry, Protein engineering, Structure–function relation, Bioeconomy Introduction acids. This is prevented by the expression of enzymes Adaption to the outer environment is a crucial factor for called fatty acid hydratases [1–5], which are unique to survival of living organisms. Many microorganisms have microorganisms [6]. Moreover, long chain fatty acids found a way to survive toxins by producing detoxifying can cause prevention of protein and amino acid uptake, small molecules or proteins. One example is the detoxi- particularly in gram-positive bacteria due to the inher- fication of free long chain fatty acids by microorganisms, ent character of their cell membranes [2, 7–9]. Con- which in free form could potentially destroy outer mem- sequently, several microorganisms, which live in close branes causing lysis of protoplasts, subsequent leakage contact to free fatty acids, are reported to express fatty of proteins, cell-associated fatty acids as well as nucleic acid hydratases as an adaption and defence to their outer environment [10]. Two functions of oleate hydratases (Ohys) for micro- *Correspondence: daniel.garbe@tum.de; loll@chemie.fu-berlin.de organisms are currently discussed. Crude oils such as Werner Siemens-Chair of Synthetic Biotechnology, Dept. of Chemistry, oils from plants but also from the skin typically contain Technical University of Munich ( TUM), Lichtenbergstr. 4, 85748 Garching, Germany a certain percentage of free, unsaturated fatty acids [11, Institute for Chemistry and Biochemistry, Laboratory of Structural 12], which are toxic for microorganisms, and thus it is Biochemistry, Freie Universität Berlin, Takustr. 6, 14195 Berlin, Germany © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Prem et al. Microbial Cell Factories (2022) 21:58 Page 2 of 15 Scheme 1 Hydroxylation of oleic acid to 10-(R)-hydroxy stearic acid as performed by Ohys thought that they are being detoxified via Ohys. Staphy - level of defence and additionally, cytoplasmatic fatty acid lococcus aureus has been found to express a functional hydratases could complement the response mechanism. Ohy even though it does not synthesize unsaturated fatty Fatty acid hydratases are able to hydroxylate unsatu- acids. However, one of S. aureus’ natural habitats is the rated fatty acids. A plethora of fatty acid hydratases, human skin, where the high abundance of free, unsatu- which convert substrates with different acyl-chain length, rated fatty acids leads to an evolutionary pressure. ranging from C11:1 to C22:6, have been reported [20, An Ohy has been discovered in S. aureus (OhySa) 23–26]. Many fatty acid hydratases have low specificity, that conveys resistance against palmitoleic acid. The in respect to acyl-chain length, but demonstrate high hydroxylated form does not further convey toxicity and regio- and stereospecifity. For instance, Ohys are regio - is not incorporated into the phospholipid membrane specific for the cis-9 C–C double bond position and but is rather exported into the outer environment [13]. enantiospecific for the 10-(R) isomer (Scheme 1). Recently, it was shown that OhySa are able to convert Hydroxylated fatty acids have first been found in host cis-9 unsaturated fatty acids to their 10-hydroxy human steatorrhoeic faeces and since a standard diet derivatives in human serum and at the infection site in a does not contain such unusual fatty acids, it was assumed mouse neutropenic thigh model, suggesting that OhySa that microorganisms synthesize them in the gut [27]. could play a role in immune modulation in S. aureus This has subsequently been demonstrated, as a Pseu- pathogenesis [14]. Furthermore, fatty acid hydratases domonas sp. strain 3266 has been found to convert oleic have been reported to be involved in stress responses of acid to 10-(R)-hydroxy stearic acid (10-HSA; Scheme  1) microorganisms. In Bifidobacterium breve, the expres - [28]. Numerous other microorganisms, mostly discov- sion of a fatty acid hydratase increases stability against ered by investigating human or animal faeces, have been heat and solvents [15, 16]. shown to produce 10-HSA [29–31]. Notably, 47  years Ohys only convert free, unsaturated fatty acids, which passed by between the discovery of 10-HSA production is rather unique. Usually, bacteria can take up exogenous of Pseudomonas sp. strain 3266, later found to be Eliza- unsaturated fatty acids, but not all are incorporated into bethkingia meningoseptica, and the purification and char - their phospholipid layer [17]. Furthermore, it is not fully acterization of the responsible enzyme [24]. understood, where exactly Ohys act. They could either Prior to the discovery and characterization of the first function in the cytoplasm or in the outer environment. Ohy, the first patent has been filed regarding the indus - For S. aureus, it has been reported that Ohys were found trial use of an Ohy from Streptococcus pyogenes, includ- in vesicles, which were secreted from the cell in the pres- ing its direct homologues with more than 40% sequence ence of linoleic acid [18]. Furthermore, an Ohy from Lac- overlap [32]. In an industrial context, oleate hydratases tobacillus plantarum was found to be a protein, bound are of special interest, due to the high-value product to a membrane by electrostatic attachment and addi- 10-(R)-hydroxy stearic acid (10-HSA). tionally it was reported that the conversion of linoleate It was considered that 10-HSA can be a replacement to 10-hydroxy-cis-12-octadecenoic acid occurs at the for 12-(R)-hydroxy stearic acid (12-HSA), which is widely periphery of the cell [19]. Since a few microorganisms used in the chemical and pharmaceutical industry. As are known to contain several oleate hydratases, a com- surfactant, 12-HSA is added to soaps and body washes. plementary effect of defence might apply [20–22]. Mem - As molecule with emollient and thickening properties, brane-hydratases and secreted ones could serve as a first it is used in skin creams and lotions. Other common P rem et al. Microbial Cell Factories (2022) 21:58 Page 3 of 15 applications are as an additive in grease, lubricating-oils Architecture of Ohys and paints, in manufacturing PVC and as lubricants in To date, there is very limited structural information avail- synthetic or natural rubbers. Furthermore, it can be used able for Ohys (see Additional file  1: Table S1). Structural as an adhesive and as a fine chemical in the food and characterization of Ohys from only five different organ - pharmaceutical industry [33–35]. 10- and 12-HSA can isms was performed so far [26, 46–49]. For the sake of additionally be converted into valuable secondary prod- understanding, we will employ a uniform terminology for ucts using cascade reactions. Those include keto-fatty the description of the discussed Ohys. All Ohys will be acids, estolides and wax esters [36, 37]. Advantages com- referred to Ohy and the first two letters of the organism pared to similar products derived from petrochemicals name of origin. We will mainly discuss: the Ohy of Rho- are that 12-HSA can be manufactured from renewable dococcus erythropolis (OhyRe; Uniprot: T5I9M6), Ohy of recourses and it is considered as a low-risk compound Staphylococcus aureus (OhySa; Uniprot: A0A0D6GJV1), [35]. In large industrial scale, 12-HSA is produced by Ohy of Lactobacillus acidophilus (OhyLa; Uniprot: chemical hydrogenation of castor oil mainly consisting Q5FL96), Ohy of Stenotrophomonas sp. KCTC 12332 of ricinoleic acid [38, 39]. For the hydrogenation of cas- (OhySt; Uniprot: A0A126NKL7) as well as Ohy of Eliza- tor oil, either hydrogen and a metallic-catalyst such as bethkingia meningoseptica (OhyEm; Uniprot: OLHYD). Raney-Nickel is applied. Alternatively, a catalytic trans- Currently the structures cover three HFam families fer hydrogenation without hydrogen can be performed. of the in total 11 Ohy families [25]. A superposition of Often high pressure and temperatures are required to the available crystal structures clearly reveals a higher obtain sufficient yields of 12-HSA [40–43]. similarity of structures within one clade of HFam fami- Recently, there is a call for more sustainability in the lies compared to lower similarity between families (see chemical industry, and the use of a biocatalyst could Additional file  1: Tables  S2 and S3). Two structures of potentially support that demand. However, when produc- the HFam2 family, OhyLa as well as OhySa are avail- ing 10-HSA using biocatalysts, free oleic acid is needed, able, which superimpose with a root mean square devia- which can be produced from oil of different types of tion (rmsd) of 1.1  Å (see Additional file  1: Table  S3). In renewable sources. This could either be plant-based oils contrast, the superposition of the overall architecture such as high-oleic sunflower oil or when available in of Ohys belonging to different HFam families, is signifi - larger scales in immediate future, hydrolysed oil from cantly different with higher rmsd values (see Additional microorganisms such as Cutaneotrichosporon oleagi- file 1: Table S3). nosus [44] or free fatty acids produced from engineered The reported structures provide interesting insights bacteria [45]. As a result, there is less dependency on just into the binding of flavin adenine dinucleotide (FAD) and one type of oil. substrate binding sites as well as the different oligomeric More and more new types of hydratases have been elu- states of Ohys. Common for all HFam families are three cidated in recent years. This can be attributed to a grow - core domains (Fig.  1), but some subfamilies have addi- ing interest in the industrial production of 10-HSA using tional N- and C-terminal extensions (Figs.  1, 2). Based biocatalysts. For industry and academia, an understand- on the available structural and biochemical informa- ing of the precise mechanism of Ohys, including the role tion, Ohys can occur as monomers or dimers. OhyLa, of a potential cofactor as well as substrate recognition OhyEm, OhyLa, OhySa and OhySt, all members of the is a fundamental prerequisite for protein engineering HFam2 or HFam11 family arrange as dimers. In contrast, in respect to industrial application. Currently, the high requirements on the performance and process stability properties of these enzymes, which will be discussed in this review in detail, prevent their application in indus- trial processes. Successes in protein engineering are only achieved steadily and this can be attributed to many open questions regarding substrate and cofactor binding and the mechanism. Additionally, low substrate and product solubility hinder the appropriate capturing of enzymatic kinetic parameters. These struggles and ways to over - come them to establish well-functioning and stable Ohys Fig. 1 Schematic domain architecture of OhyEm and OhyRe. A will be the topic of this review. Moreover, we are going Domain architecture of OhyEm coored in grey-shading for domain I to discuss sequence specific differences within the Ohy to domain IV. In yellow marked the position of the Rossman signature families, potentially leading to differences in the catalytic motif. B Domain architecture of OhyRe with identical grey-shading for its domains as in OhyEm mechanism. Prem et al. Microbial Cell Factories (2022) 21:58 Page 4 of 15 Fig. 2 Overall structure of OhyEm and OhyRe. A Proteins are shown in cartoon representation. Dashed lines indicate un-modelled loop regions. Domain organization of OhyEm (PDB-ID: 4uir; [46]): Protomer I is shown in cartoon representation: Domain I in light blue, domain II in blue, domain III in deep teal and domain IV in marine. Protomer II is shown in a transparent surface representation and cartoon representation depicted in light gray. The FAD cofactor is shown as black stick representation. B Domain organization of OhyRe (PDB-ID: 5odo; [48]): Domain I in green, domain II in orange, domain III in deep teal and domain IV in red. The shown FAD cofactor is derived from the superposition with OhyEm. The FAD is depicted in black stick representation. C Superposition of protomer I of OhyEm and OhyRe in identical orientation as in panel B. One protomer of OhyEm is shown in light blue. The terminal extensions of OhyEm are clearly visible on the right site of the panel. Figures were prepared with PYMOL (Schrödinger Inc.) OhyRe is a monomeric enzyme belonging to HFam3. For sequence of monomeric OhyRe [48] (Figs.  1, 2). These dimeric Ohys common are N- and C-terminal amino noticeable differences in the overall structure as well as acid sequence extensions, which are not present in the in different oligomerization states of Ohys (Figs.  1, 2) P rem et al. Microbial Cell Factories (2022) 21:58 Page 5 of 15 could hint to variations in co-factor binding or substrate the FAD is non-covalently bound to the protein. There - recognition. In the following, differences and similarities fore, binding of FAD induces a conformational change in between various members of Ohy families with a focus domain I, which leads to closure of the FAD-binding site on the domain arrangement, substrate and FAD-binding and enfolding of the FAD [46, 47]. will be discussed. All known Ohys strictly require FAD for functioning even though the FAD is likely not to function as a con- Overall structure ventional redox cofactor known from other enzyme fami- In general, all Ohy structures are composed of three lies [54]. The function of the FAD cofactor in Ohys is still core domains (Figs.  1, 2), that are related to other FAD- under debate and could likely play a role in the polariza- dependent enzymes. In the structures of the dimeric tion of the substrate, involvement in substrate binding HFam2 and HFam11 family members, the proteins fold or the stabilization of reaction intermediates [55, 56]. in an α-helix N-terminal of domain  1, which is involved A merely structural role of FAD cannot be completely in stabilization of the dimers. In Ohys of the HFam3 fam- excluded and might contribute to stabilization of the ily the α-helix N-terminal of domain  1 is absent and the protein. Thus, the crucial cofactor binding for structural protein is monomeric. Domain  I is a mixed α/β domain integrity and function of Ohys remains an elusive ques- composed of a parallel five-stranded β-sheet packed tion until now and hampers industrial approaches so far. between two α-helices on one side and a three-stranded For most industrial processes, heterogeneous catalysis antiparallel β-sheet on the other side (Fig.  2). Domain  I is the most common and preferred method. For economic resembles a variant of the Rossmann fold. Domain II con- reasons, enzymes are often preferred to be immobilized sists of an antiparallel β-sheet (Fig.  2) flanked by three on solid supports [57–59]. However, each cycle of reuse α  helices defining the cofactor- and substrate-binding induces a new equilibrium between medium and enzyme site in conjunction with domain  I. Domain  III is exclu- and thus over time, part of the cofactors can be lost, par- sively α-helical (Fig.  2) and its fold is structurally related ticularly in those enzymes with low binding affinity. This to monoamine oxidases [50]. Together, domain II and III applies to Ohys, since they have weak binding affinity form a tunnel to guide the substrate into the active site. towards FAD [26, 48]. This leads to either partial or com - The C-terminal domain IV differs in size and if extended, plete loss of FAD and activity. OhyRe loses both cofactor contributes to the dimer interface (Fig.  2A, C). Domain and activity, and OhyEm has only 86% of cofactor load IV undergoes a large conformational change upon sub- [46]. OhyLa has been reported to lose FAD after exten- strate binding [51], suggesting a role of domain IV in sub- sive washing on an ion-exchange or affinity column and strate recognition in conjunction with domain II and III. after gel-filtration [26]. In former immobilization experi - Notably, the most significant structural differences are ments with OhyEm, a loss of activity after each round of found for domain IV of all known Ohys, which could be reaction has been observed. The loss of FAD might be a caused by the size of domain IV and/or its involvement possible explanation [37]. Thus, elucidating crucial amino in substrate recognition. Hence, cofactor recognition and acids for binding of FAD would aid in engineering the binding play a crucial role for Ohy activity in the differ - enzymes towards optimized variants, with a higher affin - ent families implementing different catalytic pathways. ity towards FAD. Therefore, the role of cofactor binding will be discussed In domain I, the FAD-binding pocket is defined by the in the next paragraph. Rossman-fold as well as a lid region, that undergoes a conformational change upon binding of FAD. Latter con- Functional role of FAD in Ohys formational change ultimately leads to a closure of the Ohys are lyases, which don’t necessarily require a redox- FAD-binding pocket with the lid segment in close prox- active cofactor. However, all known Ohys display a imity to the isoalloxazine ring, the diphosphate function strictly conserved Rossmann-fold or Rossmann-fold like as well as the ribose of FAD. Interestingly, the length of secondary structure motif [51], which are specific for the lid segment differs between HFam family members. binding of FAD or nicotinamide adenine dinucleotide The lid segment has a length of 17 amino acids in all Ohys phosphate NAD(P)H. In flavoproteins, FAD can either so far structurally characterized, with just one exception be bound covalently or non-covalently [52]. In the case for OhyRe (Fig.  3A). Here, the lid is significantly longer of non-covalent binding, van-der-Waals and ionic inter- with 27 amino acids (Fig.  3A). One might ask, whether actions play a crucial role. As a result, FAD is bound via this could be a structural feature of members of the an on–off mechanism, that depending on the strength HFam3 family. Hence, we aligned all available amino of binding can be more or less profound. Upon dilution, acid sequences of subfamily HFam3, available in the flavin molecules can be released from a protein even assembled “hydratase engineering database” [25]. Our when they have picomolar binding-affinity [53]. In Ohys, sequence analysis clearly revealed that all HFam3 family Prem et al. Microbial Cell Factories (2022) 21:58 Page 6 of 15 Fig. 3 Sequence conservation of the FAD lid and the activation loop and architecture of the active site of OhySa, OhyEm, and OhyRe. A Amino acid sequence alignment of OhyRe (Uniprot: T5I9M6), OhySa (Uniprot: A0A0D6GJV1), OhyLa (Uniprot: Q5FL96), OhyEm (Uniprot: OLHYD), OhySt (Uniprot: A0A126NKL7) restricted to the FAD-lid and the activation loop. Conserved residues in lid and activation loop are highlighted by yellow background. The catalytic residue in the loop is highlighted with a light blue or light orange box, respectively. Highly conserved residues are indicated with asterisk, moderate conservation with two points, low conservation with one point. Primary sequences of Ohys were aligned using Clustal Omega [60]. B Active site of OhyEm (PDB-ID 4uir; [46]), shown with catalytic important residues. The bound PEG molecule in close proximity of the active site is shown in orange. Structural elements shown in cartoon representation. C Active site of OhySa (PDB-ID: 7kaz; [47]) shown with important residues lining the active site. The ternary complex of OhySa with bound FAD and oleate was obtained with the OhySa variant E82A. For clarity, we have computationally re-introduced the wild-type situation. FAD, oleic acid and indicated residues shown in stick representation. D Active site of OhyRe (PDB-ID: 5odo; [48]) shown with important residues lining the active site in stick representation. The shown FAD cofactor and oleic acid were obtained by a superposition of the OhySa structure A and derived from the superposition with the structure of OhySa. E Superposition of the active site of OhyRe and OhySa members contain an extended lid with a length of 27 or binding pocket for the approaching substrate. In the 25 amino acid residues, indicating a distinct structural structure of OhySa, the activation loop comprises resi- feature of this family. Despite its length, in all structures, dues from 78 to 83 (Fig. 3A,). Notably, the catalytic E122 the lid segment contains the highly conserved signature of OhyEm as well as E82 of OhySa are located within lat- motif GGXXXG (X any amino acid; Fig.  3A). Notably, ter activation segment (Fig. 3B, C). in HFam3, the motif is altered to GXXXG. Concomi- Analysing the sequence conservation of the FAD-bind- tant with FAD-binding, a loop region, termed “activation ing pocket, clearly reveals a very high degree of sequence loop” by Radka et  al. [47], undergoes a large conforma- conservation of the surface shaping the pocket (Fig.  4). tional change (Fig. 3E). As consequence of FAD-binding, The observed differences in affinity towards FAD could the activation loop almost rotates by 180°, otherwise it be likely attributed to differences in the length and amino would lead to a steric clash. In the FAD-bound state, the acid sequence pattern of the FAD lid, which have con- activation loop is in proximity to the isoalloxazine func- sequences for the conformational flexibility of the lid tion of FAD and secondly, it pre-shapes the substrate region. Such conformational flexibility is also structurally P rem et al. Microbial Cell Factories (2022) 21:58 Page 7 of 15 Fig. 4 Conservation of FAD-binding pocket. A Surface representation of OhyRe (PDB-ID: 5odo; [48]) with conservation of residues from variable to conserved as indicated in the legend. The shown FAD cofactor is depicted as black stick representation and derived from a superposition with OhySa. In the structure of OhyRe, the FAD-lid is disordered and could not be modelled. B Surface representation of OhySa (PDB-ID: 7kaz; [47]) with conservation of residues from variable to conserved as indicated in the legend. The FAD-lid covers the bound FAD molecule. Conservation of Ohys was calculated with the Consurf server [61] reflected. For instance, in the structure of OhySa, a not yet formed. Analysis of the OhyRe structure reveals weaker electron density compared to the protein was that α-helices of domain III are in closer proximity to interpreted as a not fully occupied FAD and fragmented each other, narrowing the channel. In addition, a num- electron density was observed for the lid region, support- ber of amino acid side chains with hydrophobic char- ing an inherent flexibility of the lid region [47]. acter point into the putative channel. Interestingly, many of these residues are conserved or at least simi- lar to OhySa. In the product or substrate bound state Substrate binding of the variant OhySa E82A, the α-helices of domain III Recently, the crystal structure of an OhySa variant bound and side chain rotamers adopt a different conforma - to oleate and FAD was reported [47], giving insights into tion, opening a channel in direction towards the FAD the active site configuration. Previously, structures of molecule. Ohys from other organisms were reported with polyeth- Further, it remains unknown how the substrate per- ylene glycol (PEG) molecules, originating from the crys- suades along the approximately 30 Å long ligand chan- tallization experiment, bound in the cavity in proximity nel from the protein exterior towards the catalytic to FAD. It was proposed, that the PEG might resemble site. A hypothesis could be that the substrate diffuses the substrate [46, 47]. The structure of OhyLa was through the channel towards the active site. Yet another reported with a bound linoleic acid in domain IV, distinct possibility could be a partial opening of the protein with from the active site [26]. subsequent binding of the substrate. The initially bound Superposition of all bound ligands in the structure substrate could then further diffuse along the channel. of OhySa (Fig.  5) illustrates, that a substrate channel is However, passive diffusion of the substrate seems to be built from the distal part (linoleic acid) to the active site rather atypical for such long ligand channels, since also (10-HSA). Oleic acid occupies this tunnel in between the product needs to diffuse through this channel to the linoleic acid and 10-HSA (Fig.  5). Mainly domain III exterior of the enzyme and passive diffusion would also and domain IV build up the ligand channel, which is not be time efficient in the catalytic process. Moreover, lined by hydrophobic amino acids, allowing the mainly a pure diffusion mechanism might be unlikely since the hydrophobic substrate to diffuse into the active site chemical structure of oleic acid with its C9 cis double niche. The role of the flexible domain  IV in the cata bond makes the substrate rather rigid. In absence of lytic cycle of Ohys remains elusive. Interestingly upon substrate or product, water molecules should, at least binding of ligands a conformational shift of the domain partially fill the empty ligand channel in the apo state IV is observed in the structure of OhyLa [26]. Notably, of Ohys. Latter water molecules need to be expelled for the monomeric OhyRe, belonging to the HFam3 upon substrate binding or substrate diffusion along family, the domain IV is significantly reduced in size the cavity. Moreover, we detected a side opening in the compared to dimeric Ohys. Moreover, calculation of structure of OhySa (Fig.  5), through which water mol- potential ligand channels in the structure of OhyRe in ecules could be pushed out by the moving substrate on its apo state was not possible, indicating that the sub its trajectory towards its binding site. This shorter side strate channel in the structure of OhyRe is blocked or Prem et al. Microbial Cell Factories (2022) 21:58 Page 8 of 15 Fig. 5 Substrate, product tunnel and FAD-binding site in OhySa. A OhySa in light orange cartoon representation and the predicted substrate, product and FAD cavities in surface representation in yellow and green. The substrate/product channel from the exterior of the protein towards the FAD is shown in green and numbered “3”. The cavity with bound FAD is depicted in yellow and numbered with “2”. A side channel in vicinity of the FAD cavity is labelled with “3”. The tunnels were calculated with Caver 3.0 [62]. FAD is shown in black stick representation. Oleic acid (PDB ID: 7kaz; [47]) is shown in light green stick representation; 10-HSA bound to OhySa (PDB ID: 7kaz; [47]) is shown in dark blue stick representation as well as linoleic acid bound to OhyLa (PDB ID: 4ia6; [26]) is shown in light violet stick representation; Dashed box displays magnification area as shown in B. B Magnification of dashed box in A with the protein omitted. The ligands linoleic acid, oleic acid and polyethylene glycol are depicted as in A. The ligands guide the substrate channel from the distal end of the channel to the proximal catalytic cleft close to the FAD cavity is mainly lined by hydrophilic amino acids and with the carbon skeleton are possible and only the was described in the structure of OhyEm [46]. charged carboxyl moiety could be sensed by the protein Interestingly, linoleic acid binds to OhyLa with the environment. One could imagine that for a stereospecific carboxylate function facing outward and the hydropho- hydroxylation, the substrate would have to be held in a bic poly-carbon tail threaded into the channel formed very precise, defined position to avoid any side reactions. by domain IV ([26] Fig.  5B). It should be noted, that the electron density interpreted as linoleic acid is weak. A clear decision on the orientation of the carboxy- Reaction mechanism late is very difficult and modelling of the carboxylate is The overall architecture of the active site is well pre - based on the observation of Arg and Lys residues in the served within all structurally characterized Ohys. Nev- closer neighbourhood to the carboxylate. In contrast, in ertheless, substantial differences can be noted between the structure of OhySa, the substrate oleic acid and the the members of the different HFam subfamily members product 10-HSA are both bound with the carboxyl group and will be discussed here. Based on the crystal structure facing inward (Fig.  5B). Additionally, the binding mode and docking studies, a reaction mechanism for OhyEm of the ligands is not thoroughly understood yet. Several was proposed, where Y241 initially protonates the dou- amino acids lining the active site of Ohys need to rear- ble bond of the substrate. E122 activates a water molecule range during binding of the substrates. Especially R81 that can quench the carbocation [46, 56, 63]. Recently, of OhySa is a crucial residue in binding of the substrate a similar function was proposed for the catalytic E82 of (Fig.  3B), since it acts as a block before the entry of the OhySa (Fig.  3C) [47]. In contrast to the earlier proposed substrate and its site chain rotates about 180° upon FAD- function, Y201 is hydrogen bonded to the backbone car- binding. As a consequence, the guanidinium function of bonyl of V505. Consequently, Y201 cannot donate a pro- R81 points in the direction of the approaching substrate. ton to the double bond. Conversely, our modelling data Surprisingly, based on the structure of OhySa with bound indicates that it is involved in a hydrogen bonding net- oleic acid, the positively charged guanidinium function work including the hydroxylated oleate. A hydronium is not directly involved in recognition of the negatively ion is stabilized by an α-helical dipole and a cation of an charged carboxylate of oleic acid [47]. Given the chemical acidic proton of E122. Subsequently, the hydronium ion structure of oleic acid, only van der Waals interactions attacks the substrate. Upon release of FAD, the proton P rem et al. Microbial Cell Factories (2022) 21:58 Page 9 of 15 is released to the hydrated active site and in turn to the but oleic acid is not equally distributed, hence care has C10-hydroxylated product [47]. to be taken on using stock suspensions for enzymatically Instead of a glutamate in the active site at an equiva- determined kinetic measurements. This is critical for the lent position of E122 in OhyEm or E82 in OhySa, belong- measurement of kinetic parameters but also for the reac- ing to HFam11 and HFam2, respectively, a methionine tion itself. When oleic acid is not brought into suspen- residue is located in OhyRe, belonging to HFam3 family sion, droplets are formed, lowering the access towards (Fig. 3E). We were puzzled whether a methionine residue the catalyst and thus a decreasing yield, as observed by within the activation loop could be a conserved feature Jeon et. al., which could be resolved by more efficient within the HFam3. Amino acid sequence analysis of the mixing resulting in the formation of a suspension [67]. “hydratase engineering database” [25] revealed a strict Furthermore, the pH and the temperature have effect conservation of the methionine residue. Mutational stud- on the formation of certain fatty acid species. At a pH ies of OhyRe variants M77E showed a drastically reduced lower than 6, fatty acids usually form crystals and at a hydroxylation activity compared to the wild type enzyme pH between 7 and 9, they are in a 1:1 acid to soap ratio, [48]. A plausible role of the methionine could be a sta- which have extremely low solubility in water [68]. Ohys bilization of the emerging carbocation [64]. Given the can have diverse pH-optima, OhyEm has two pH-optima fundamental difference in chemistry of glutamate versus at pH 6 and 8 [24], OhyRe at 7 [48] and Ohy from Rhodo- methionine poses the question, whether Ohys belong- coccus pyridinivorans has an optimum at pH 5 [69]. It is ing to the HFam3 family employ a different reaction unclear what kind of effects those fatty acid species have mechanism. The family-specific patterns such as dimeric on the enzyme and the reaction. Additionally, 10-HSA is versus monomeric enzymes; differences in the length of a product that has no solubility in water and depending the FAD-lid as well as different decoration of the active on the experimental reaction environment, presence of site niche could hint at different reaction mechanisms certain surfactants or mixing speeds generates different and explain the differences in substrate recognition. In forms of white aggregates in a solution. However, tak- summary, the observed differences might indicate a con - ing samples from a solution can only provide an initial vergent evolution of Ohy families from different ances - overview due to misleading distributive effects and enzy - tors. Consequently, these observations path the way for matic kinetic measurements shall be performed in single a deeper understanding and implementation of Ohys in reactions. biotechnological pathways and will help to employ such Also, 10-HSA has low solubility in many organic sol- enzymes for the chemical industry. In the following chap- vents. That is why full extraction of product and substrate ter a more detailed overview on biotechnological and can’t be achieved under certain extraction conditions and industrial application of Ohys will be given. consequently their measurement is distorted. A common extraction solvent is ethyl acetate after acidification of Industrial application—up and downstream the reaction solution, but also chloroform/methanol is innovation used [48, 70]. However, the solubility of 10-HSA in those Enzymes are currently used in a wide variety of industrial extraction solvents for analytical purposes has not been processes. These traditionally include the food, feed, pol - reported in most studies. Furthermore, several methods ymer, leather and cosmetics sectors. Moreover, enzymes for purification of 10-HSA in a preparative scale have are also used as functional detergent additives and in been performed. In one study, fractionation using ace- organic synthesis of specialty chemicals [65]. However, tone and acetonitrile has been performed at low temper- not all enzymes make it into an industrial process for atures resulting in a purity of over 99% [71]. several reasons and overcoming those challenges is one Mostly, gas chromatography is the analysis method of of the major tasks of protein scientists. The performance choice and for that, derivatization of the fatty acids has to and the costly development of processes are the main take place to reduce adsorption effects. This is achieved hurdles for using enzymes in an industrial environment by methylation or silylation of the carboxy and hydroxy [66]. groups [72, 73]. Additionally, at present, non-derivatized 10-HSA cannot be commercially purchased and thus the Measuring kinetic parameters standard has to be prepared in-house. For that, however, For many Ohys, enzymatically determined kinetic an internal standard is crucial, to evaluate the derivati- parameters have been published, and those could be used zation efficiency and evaporation effects. Only then, the for industrial process simulations and cost evaluations. instrument can properly be calibrated. However, for Ohys they have to be considered with cau- To sum up, for enzymatic kinetic measurements, which tion, as the substrate, oleic acid, is not fully miscible in are important for industrial process simulations, sev- water. Suspensions can be prepared by vigorous mixing, eral considerations are necessary. Only under certain Prem et al. Microbial Cell Factories (2022) 21:58 Page 10 of 15 conditions, appropriate enzymatic kinetic parameters bulky substrates such as Ohys [25]. One study aimed to can be determined. Those apply, when oleic acid is added overcome this effect by decreasing the route between purely to the reaction, when complete extraction of sub- catalyst and substrate. For that, the enzyme was tar- strate and product is performed and the standards are geted into the periplasm using a signal peptide. The prepared carefully. Furthermore, enzyme kinetics should whole-cell reaction using the periplasmatic enzyme be performed in single reactions and it is important resulted in a tenfold higher hydration rate compared to to keep in mind that they are not comparable between the cytoplasmatic reaction. It is known, that the redox- enzymes and studies due to varying reaction conditions. environment plays a role on the activity of Ohys but this has not been discussed [76]. Currently the highest reported STYs for producing Performance of whole‑cell catalysts −1 −1 10-HSA with a whole-cell catalyst is 8–12 g  L  h . For The performance of a catalyst is crucial for every indus - comparison, the production of acrylamide with nitrile trial process, in heterogenous catalysis the space–time −1 −1 hydratase, which is one of the most efficient whole- yield (STY) can be around 1 to 10  kg  L  h . However, cell biocatalytic processes in the industry, gives STYs when looking at biocatalysts, STYs can be decreased by −1 −1 −1 −1 between 53 and 93  g  L  h [66, 77]. To reach such a up to 1000 times to around 0.001–0.3  kg  L  h com- level for the production of hydroxylated fatty acids, sig- pared to conventional processes [66]. While this certainly nificant process optimization is required. However, using can be tolerated by the pharmaceutical industry with a whole-cell catalysts also brings disadvantages particu- need for enantiomeric purity and high-quality products, larly for this certain application. Free, unsaturated fatty expensive processes for a final product, that is mostly acids might convey toxic effects on the whole-cell cata - used as an additive such as 12-HSA, will most likely not lysts upon a certain concentration since the detoxifying sustain. fatty acid hydratases usually are expressed in the cytosol. An enzymatic industrial process can be installed in dif- Additionally, as already mentioned the mass transport ferent modes. First, either wild-type or genetically engi- of substrate and product is hindered by the membrane. neered whole-cells can be used to convert oleic acid. Therefore, it is important to keep the fatty acid content The advantage here is, that no further purification of the under a critical toxic concentration and for the latter enzyme is needed, only the extraction of fatty acids and issue, organic solvents or surfactants such as Tween80 the purification of 10-HSA. Usually, whole-cell conver - can be added. Those additives, however, can influence the sions apply, when large gene clusters and cascades are energy metabolism within the cell, increase the costs of a involved in the formation of a product or if enzymes are process and might complicate the purification. So other not soluble or active when being isolated. strategies such as genetic engineering are investigated Recombinant Escherichia coli expressing an Ohy [78, 79]. from Stenotrophomonas maltophilia has been used as Additionally, in the aforementioned studies, samples whole-cell catalysts in a 1 mL scale leading to a STY of −1 −1 of the reaction medium were taken and extracted using 12.3 g  L  h , however, when they scaled-up to 1 L, the −1 −1 organic solvents such as ethyl acetate. Consequently, STY decreased to 8.2 g  L  h , presumably due to the endogenous fatty acids and hydrophobic molecules changed reaction conditions omitting the buffer and from the cells are extracted and appear as impurities in oxygen-depletion [67, 74]. Furthermore, the authors the final product. Whereas this might be no problem for mention that genetically modified E. coli has a three - industrial products, formulations for the pharmaceuti- fold higher formation rate than the wild-type strain. In cal or cosmetic industry certainly have higher standards other studies, the original organisms have been used, regarding the purity and more laborious downstream however the STYs were quite low compared to E. coli, processing is required to further purify the product. except for one study with Stenotrophomonas nitritire- −1 −1 Other possibilities are the filtration but this might come ducens, where 7.9  g  L  h was achieved [74]. How- with a significant product loss since substrate and prod - ever, the prolonged growth of S. nitritireducens and uct might adsorb to the cell exteriors. In one study about the maximum achievable cell concentration compared whole-cell biocatalysis, 30% of loss was observed after to E. coli were not considered. With S. maltophilia for downstream processing [67]. The high product losses instance, only 10  g/L of maximal cell concentration due to using whole-cell biocatalysts can also be attrib- can be achieved compared to 100  g/L in E. coli in fed- uted to the faster saturation of extraction solvents due to batch cultures [75]. Thus, the authors concluded that E. hydrophobic molecules from the cell. Lastly, it is much coli as whole-cell catalyst is more advantageous com- harder to recycle whole-cell catalysts, particularly when pared to wild-type strains. Mass transport limitations the product is solid and centrifugal forces do not lead to a are hurdles during reactions with whole-cell catalysts separation of product and catalyst. and this is particularly the case for enzymes converting P rem et al. Microbial Cell Factories (2022) 21:58 Page 11 of 15 Performance of Ohys in lysates and pure formes longer and thus the stability after each cycle might not be That is why in some cases, lysates or purified enzymes the same as shown in that study. Furthermore, it has not might be more desirable. Contrarily to some other fatty been investigated how the activity of immobilized Ohy acid converting enzymes or other hydratases [80, 81], changes during extended storage for days [37]. Ohys achieve high expression rates and solubility and Since isolated enzymes are less protected when they don’t rely on stoichiometric amounts of FAD [56], which are not part of a whole-cell catalyst, their stability and makes them excellent candidates for use in pure form or maintenance of activity over a long time plays a crucial in lysates. Many screenings of activities of Ohy have been role for an efficient process. In a few studies, low stabil - performed using lysates and lyophilized lysates [25], and ity of Ohys has been observed. Some lose their activity a patent described the large-scale production of 10-HSA already after a short period of time [25, 37]. In a com- with lysate of R. erythropolis and S. maltophilia [82]. Fur- parative study on enzyme stability, five different Ohys thermore, a pilot scale with cell-free extract has been were analysed. It was found that all of them started to performed using a variant of Paracoccus aminophilus denature already after one day within lysates, leading to −1 −1 with a STY of 22.5 g  L  h [83]. an exposure of their hydrophobic sites. As a result, nei- Since lysates are difficult to recycle and have weak ther substrate nor product was measurable anymore, stability; pure, immobilized enzymes are in some cases most likely since they interact with the hydrophobic sites the method of choice. Additionally, immobilization can of the denatured protein bulk. Buffer optimization led result in higher stability, increased activity and improved to certain improvements regarding the protein stability stereoselectivity and efficient recycling lowers the costs [25]. In another study it was reported that OhyEm loses [84]. However, not all enzymes can be immobilized and 60% of its activity already after 7 days at 4 °C. Todea et al. recycled for several rounds in native form and not all suspected OhyEm to inactivate as a result of the disso- products have good biocompatibility with the solid sup- ciation of subunits [37]. In general, however, not much is ports. Ways to overcome these challenges are on the way published about the stability of Ohys over a longer period by developing novel supports and materials for immobi- of time since most studies have no industrial but rather a lization and by using state-of the art technologies in the medicinal background. Todea et al. have used additives in field of protein engineering [84]. order to overcome the stability problems. This has been One of the main issues for immobilization is the men- investigated by storing the protein for 7 days at 4 °C and tioned insolubility of 10-HSA in water and thus the testing its residual activity. However, no experiments catalyst cannot easily be separated by centrifugation. have been conducted what effects the additives have on Furthermore, oleate is a hydrophobic molecule and thus the process stability with several re-usage cycles and at attaches to certain materials used as solid supports. At elevated temperatures. Additionally, additives can com- present, only one study exists, where an Ohy has been plicate the process since they might have to be removed immobilized. Several issues occurred while testing dif- before the reaction starts and they increase the price of ferent kinds of support. The recovery of the product a process [37]. In general, the main reasons for a loss without harming the enzyme in form of cross-linked of protein activity is either the distortion of the tertiary enzyme aggregates was not possible. To ease the separa- structure, the dissociation of cofactors, chemical inac- tion of catalyst and product, magnetic beads were used, tivation when a reactive chemical is part of the reaction however that resulted in adsorption of substrate and or—as the first step of inactivation for multimeric pro - product to the support. The magnetic beads were coated teins—the dissociation of subunits [85]. Consequently, with a layer of chitosan to avoid the adsorption. Still, in multimeric and FAD-bound enzymes are more affected all immobilization techniques, not more than 24% of the and less advantageous in industrial processes. First of all, residual activity has been recovered. The least residual a monomeric enzyme can overcome the issues of subunit activity was observed for the entrapment of the enzyme dissociation and an enzyme working without FAD can’t since organic solvents were used in that method known be subject to cofactor loss. However, currently OhyRe to inactivate the enzyme. In general, for entrapment— is the only known monomeric Ohy and it loses FAD even by other means where stability is maintained—the during purification resulting in a loss of function [48]. biggest issue is the transport of oleic acid in aqueous Consequently, this particular enzyme still requires opti- solutions towards the active sites. The chitosan-coated mizations in order to be used in isolated form since the magnetic beads as best candidates were finally chosen addition of FAD renders it too costly. for testing rounds of recycling and after 5 cycles, still 70% of initial activity was left. Each reaction of a cycle Protein engineering of Ohys was performed for 2  h, however, usually, reactions with Protein engineering is one of the main methods to over- Ohys with high concentrations of oleic acid take much come the several drawbacks of Ohys. Substrate spectrum Prem et al. Microbial Cell Factories (2022) 21:58 Page 12 of 15 and selectivity, cofactor binding, stability and turnover using directed evolution. In that context, a coupled assay number are attributes, which are desirable to improve. has been used to screen for optimized variants. 10-HSA Directed evolution and site-directed mutagenesis are the was converted into 10-oxostearic acid by an alcohol two main methods for improving proteins. dehydrogenase, the occurring coenzyme NADH was col- The crystal structure of OhyEm has been reported orimetrically analysed and the variant with the highest with bound FAD, but not with substrate or product [46]. colorimetric output was further analysed. For this type However, an electron density in the proposed substrate of assay usually performed in lysates the specificity of the binding cavity has been interpreted as a PEG molecule alcohol dehydrogenase is crucial [83]. thought to be a substrate mimic. To manipulate the sub- strate spectrum, structure-guided protein engineering Conclusions using site-directed mutagenesis has been performed. Ohys belong to the fatty acid hydratases enzyme fam- Hence, amino acids belonging to the pocket of the fatty ily, which is unique for its co-factor free modification of acid head group were altered. Some of those variants free, unsaturated fatty acids. It is suggested, that Ohys could convert derivatives of oleic acid such as ethyl- and primarily evolved to protect microorganisms from toxic n-propyl oleic acid, stearyl alcohol or stearyl amine at effects by incorporation of free, unsaturated fatty acids higher rates [54]. In another study, the substrate spec- into the cell membrane. At the beginning of the 1960s trum was altered towards alkenes with a terminal or also Ohy’s products could be isolated from animal and internal double bond by the addition of a dummy carboxy human faeces, which were assumed to originate from acid to artificially expand the size of the substrate and bacteria colonizing the gut. Recently, Ohy evoked indus- by decreasing the size of the substrate binding pocket trial interest for the conversion of oleic acid to sustain- by mutagenesis. Since the location of PEG was not suffi - able 10-HSA, which can replace 12-HSA in oleochemical cient, a structure of OhyEm with a docked oleic acid was and cosmetics applications, that is currently generated used [86]. by hydrogenation of castor oil. Although 12-HSA has a Another attempt to alter the substrate spectrum of high application spectrum as additive ranging from oils Ohys by rational-mutagenesis has been demonstrated and paints via manufacturing of rubbers to use in food by Eser et  al. In their study, residues of the active side and pharmaceutical industry, the educt castor oil is lim- of Ohys with 76% homology originating both from Lac- ited and fluctuates in quality. Furthermore, the hydration tobacillus acidophilus have been compared and their process needs high pressure and temperature conditions functionalities have been estimated by using the crys- to obtain economically sound yields. Therefore, indus - tal structure of OhyEm. One of the enzymes FA-HY2 try demands for a more sustainable and quality sta- is unique since it is able to convert substrates up to the ble replacement, which can be provided in theory by length of 22, whereas the other one serves as a rather 10-HSA. typical Ohy (FA-HY1) converting a substrate length of As a consequence, scientists became interested in 16–18. In conclusion the substrate preference and regi- Ohys, which is documented by an increase in articles oselectivity of FA-HY1 could be changed by swapping characterising Ohys from different microorganisms. critical residues from FA-HY2 [63]. While numerous new Ohys have been described recently, These findings suggest that site-directed mutagenesis there is a limited understanding concerning structure– has great potential when the crystal structure is fully function relationships in this structurally diverse enzyme unravelled and the location of substrates and products is family. Specifically, more insights on detailed reaction clear or can be cleared by docking experiments. The crys - mechanisms are required. In that context, it is unclear tal structure of OhyRe is neither resolved with FAD nor how the highly elongated substrate reaches the active substrate and docking of neither cofactor nor substrate centre and how the hydroxylated products are released has been successful so far. This can be due to many rea - after commencement of the reaction. Additionally, a big sons but a high-quality structure after docking can only controversy exists regarding the role of the FAD mol- be achieved when the underlying biomolecule isn’t sub- ecule bound in the structure. ject to large conformational changes upon binding of the Moreover, with the exception of the monomeric OhyRe docked molecule [87]. That is why directed evolution is recently described, all other Ohys deciphered today are sometimes a much more powerful tool specifically for dimers. Hence, OhyRe is amenable for efficient immo - improvements that are difficult to address by structure- bilisation, which makes it attractive for industrial appli- based methods, such as melting temperature or affinity cations, which require extended residence times of the towards certain small molecules. Furthermore, results biocatalyst in target reactions to reduce costs. However, can be achieved at much higher pace. An Ohy from Par- current literature studies revealed that OhyRe has a sig- acoccus aminophilus has been successfully enhanced by nificantly different domain architecture to other Ohys, P rem et al. Microbial Cell Factories (2022) 21:58 Page 13 of 15 3. Stenz L, Francois P, Fischer A, Huyghe A, Tangomo M, Hernandez D, Cassat which suggests that this enzyme may follow a different J, Linder P, Schrenzel J. Impact of oleic acid (cis-9-octadecenoic acid) on reaction mechanism. bacterial viability and biofilm production in Staphylococcus aureus. FEMS Currently, it is the yet unknown OhyRe reaction mech- Microbiol Lett. 2008;287:149–55. 4. Maia MR, Chaudhary LC, Bestwick CS, Richardson AJ, McKain N, Larson anism and structure–function relationship of this enig- TR, Graham IA, Wallace RJ. Toxicity of unsaturated fatty acids to the bio- matic monomeric enzyme as well as the complicated hydrogenating ruminal bacterium Butyrivibrio fibrisolvens. BMC Microbiol. purification process for the target product 10-HSA, 2010;10:52. 5. Zheng CJ, Yoo JS, Lee TG, Cho HY, Kim YH, Kim WG. Fatty acid synthesis which hamper further industrial scaling of the reaction. is a target for antibacterial activity of unsaturated fatty acids. FEBS Lett. Therefore, a broader knowledge-base is needed to enable 2005;579:5157–62. industrial adaptation of biotechnological 10-HSA pro- 6. Kim KR, Oh DK. Production of hydroxy fatty acids by microbial fatty acid- hydroxylation enzymes. Biotechnol Adv. 2013;31:1473–85. duction to replace thermocatalytic 12-HSA synthesis. 7. Galbraith H, Miller TB. Eec ff t of metal cations and pH on the antibac- terial activity and uptake of long chain fatty acids. J Appl Bacteriol. 1973;36:635–46. Abbreviations 8. Parsons JB, Yao J, Frank MW, Jackson P, Rock CO. Membrane disruption FAD: Binding of flavin adenine dinucleotide; Ohy: Oleate hydratase; PEG: Poly- by antimicrobial fatty acids releases low-molecular-weight proteins from ethylene glycol; Rmsd: Root mean square deviation; STY: Space–time yield. Staphylococcus aureus. J Bacteriol. 2012;194:5294–304. 9. Galbraith H, Miller TB, Paton AM, Thompson JK. Antibacterial activity of long chain fatty acids and the reversal with calcium, magnesium, ergocal- Supplementary Information ciferol and cholesterol. J Appl Bacteriol. 1971;34:803–13. The online version contains supplementary material available at https:// doi. 10. McKain N, Shingfield KJ, Wallace RJ. Metabolism of conjugated linoleic org/ 10. 1186/ s12934- 022- 01777-6. acids and 18: 1 fatty acids by ruminal bacteria: products and mechanisms. Microbiology. 2010;156:579–88. Additional file 1. Additional Tables. 11. Pappas A. Epidermal surface lipids. Dermatoendocrinol. 2009;1:72–6. 12. Man YBC, Moh MH, van de Voort FR. Determination of free fatty acids in crude palm oil and refined-bleached-deodorized palm olein using Fou- Acknowledgements rier transform infrared spectroscopy. J Am Oil Chem Soc. 1999;76:485–90. T. B. gratefully acknowledges funding by the Werner Siemens foundation for 13. Subramanian C, Frank MW, Batte JL, Whaley SG, Rock CO. Oleate establishing the field of Synthetic Biotechnology at the Technical University of hydratase from Staphylococcus aureus protects against palmitoleic acid, Munich ( TUM). the major antimicrobial fatty acid produced by mammalian skin. J Biol Chem. 2019;294:9285–94. Author contributions 14. Radka CD, Batte JL, Frank MW, Rosch JW, Rock CO. Oleate hydratase BL and DG conceived the review. SP, CPOH, ND, SH, and TB contributed to the (OhyA) is a virulence determinant in Staphylococcus aureus. Microbiol manuscript. CPOH: generated the figures with advice of ND. All authors read Spectr. 2021;9:e0154621. and approved the final manuscript. 15. Rosberg-Cody E, Liavonchanka A, Gobel C, Ross RP, O’Sullivan O, Fitzger- ald GF, Feussner I, Stanton C. Myosin-cross-reactive antigen (MCRA) pro- Funding tein from Bifidobacterium breve is a FAD-dependent fatty acid hydratase Open Access funding enabled and organized by Projekt DEAL. N.D. and B.L. which has a function in stress protection. BMC Biochem. 2011;12:9. are supported by the BMBF Grant 031B0853. S.P. was supported by BMBF 16. O’Connell KJ, Motherway MO, Hennessey AA, Brodhun F, Ross RP, Grant 03SF0577A. Feussner I, Stanton C, Fitzgerald GF, van Sinderen D. Identification and characterization of an oleate hydratase-encoding gene from Bifidobacte - Availability of data and materials rium breve. Bioengineered. 2013;4:313–21. Not applicable. 17. Yao JW, Rock CO. How bacterial pathogens eat host lipids: implications for the development of fatty acid synthesis therapeutics. J Biol Chem. 2015;290:5940–6. Declarations 18. Kengmo Tchoupa A, Peschel A. Staphylococcus aureus releases proinflam- matory membrane vesicles to resist antimicrobial fatty acids. mSphere. Ethics approval and consent to participate 2020;5:e00804. Not applicable. 19. Ortega-Anaya J, Hernandez-Santoyo A. Functional characterization of a fatty acid double-bond hydratase from Lactobacillus plantarum and Consent for publication its interaction with biosynthetic membranes. Biochim Biophys Acta. Not applicable. 2015;1848:3166–74. 20. Hirata A, Kishino S, Park SB, Takeuchi M, Kitamura N, Ogawa J. A novel Competing interests unsaturated fatty acid hydratase toward C16 to C22 fatty acids from The author declares that there are no competing interests associated with this Lactobacillus acidophilus. J Lipid Res. 2015;56:1340–50. manuscript. 21. Kang WR, Seo MJ, Shin KC, Park JB, Oh DK. Comparison of biochemical properties of the original and newly identified oleate hydratases from Received: 9 February 2022 Accepted: 18 March 2022 Stenotrophomonas maltophilia. Appl Environ Microbiol. 2017;83:e03351. 22. Kim KR, Oh HJ, Park CS, Hong SH, Park JY, Oh DK. Unveiling of novel regio- selective fatty acid double bond hydratases from Lactobacillus acidophilus involved in the selective oxyfunctionalization of mono- and di-hydroxy fatty acids. Biotechnol Bioeng. 2015;112:2206–13. References 23. Kishino S, Takeuchi M, Park SB, Hirata A, Kitamura N, Kunisawa J, Kiyono H, 1. Dilika F, Bremner PD, Meyer JJ. Antibacterial activity of linoleic and oleic Iwamoto R, Isobe Y, Arita M, et al. Polyunsaturated fatty acid saturation by acids isolated from Helichrysum pedunculatum: a plant used during gut lactic acid bacteria affecting host lipid composition. Proc Natl Acad circumcision rites. Fitoterapia. 2000;71:450–2. Sci USA. 2013;110:17808–13. 2. Greenway DL, Dyke KG. Mechanism of the inhibitory action of lin- 24. Bevers LE, Pinkse MW, Verhaert PD, Hagen WR. Oleate hydratase catalyzes oleic acid on the growth of Staphylococcus aureus. J Gen Microbiol. the hydration of a nonactivated carbon-carbon bond. J Bacteriol. 1979;115:233–45. 2009;191:5010–2. Prem et al. Microbial Cell Factories (2022) 21:58 Page 14 of 15 25. Schmid J, Steiner L, Fademrecht S, Pleiss J, Otte KB, Hauer B. Bio- conformational plasticity surrounding the FAD binding site. Biochem catalytic study of novel oleate hydratases. J Mol Catal B Enzymatic. Biophys Res Commun. 2018;499:772–6. 2017;133:S243–9. 50. De Colibus L, Li M, Binda C, Lustig A, Edmondson DE, Mattevi A. Three- 26. Volkov A, Khoshnevis S, Neumann P, Herrfurth C, Wohlwend D, Ficner dimensional structure of human monoamine oxidase A (MAO A): relation R, Feussner I. Crystal structure analysis of a fatty acid double-bond to the structures of rat MAO A and human MAO B. Proc Natl Acad Sci U S hydratase from Lactobacillus acidophilus. Acta Crystallogr D Biol Crystal- A. 2005;102:12684–9. logr. 2013;69:648–57. 51. Rao ST, Rossmann MG. Comparison of super-secondary structures in 27. James AT, Webb JP, Kellock TD. The occurrence of unusual fatty acids in proteins. J Mol Biol. 1973;76:241–56. faecal lipids from human beings with normal and abnormal fat absorp- 52. Dym O, Eisenberg D. Sequence-structure analysis of FAD-containing tion. Biochem J. 1961;78:333–9. proteins. Protein Sci. 2001;10:1712–28. 28. Wallen LL, Benedict RG, Jackson RW. The microbiological produc- 53. Bollen YJ, Westphal AH, Lindhoud S, van Berkel WJ, van Mierlo CP. Distant tion of 10-hydroxystearic acid from oleic acid. Arch Biochem Biophys. residues mediate picomolar binding affinity of a protein cofactor. Nat 1962;99:249–53. Commun. 2012;3:1010. 29. Thomas PJ. Identification of some enteric bacteria which convert oleic 54. Engleder M, Strohmeier GA, Weber H, Steinkellner G, Leitner E, Muller acid to hydroxystearic acid in vitro. Gastroenterology. 1972;62:430–5. M, Mink D, Schurmann M, Gruber K, Pichler H. Evolving the promiscuity 30. Koritala S, Hosie L, Hou CT, Hesseltine CW, Bagby MO. Microbial conver- of Elizabethkingia meningoseptica oleate hydratase for the regio- and sion of oleic-acid to 10-hydroxystearic acid. Appl Microbiol Biotechnol. stereoselective hydration of oleic acid derivatives. Angew Chem Int Ed 1989;32:299–304. Engl. 2019;58:7480–4. 31. Hudson JA, MacKenzie CAM, Joblin KN. Conversion of oleic acid to 55. Hiseni A, Arends IWCE, Otten LG. New cofactor-independent hydration 10-hydroxystearic acid by two species of ruminal bacteria. Appl Microbiol biocatalysts: structural, biochemical, and biocatalytic characteristics of Biotechnol. 1995;44:1–6. carotenoid and oleate hydratases. ChemCatChem. 2015;7:29–37. 32. Feussner I, Hornung E, Liavonchanka A: Production of hydroxy fatty acids 56. Engleder M, Pichler H. On the current role of hydratases in biocatalysis. using fatty acid hydratase from Streptococcus pyogenes. (GÖTTINGEN Appl Microbiol Biotechnol. 2018;102:5841–58. G-A-U ed). 2008. 57. Basso A, Serban S. Industrial applications of immobilized enzymes—a 33. Fameau A-L, Rogers MA. The curious case of 12-hydroxystearic acid—the review. Mol Catal. 2019;479:110607. Dr. Jekyll & Mr. Hyde of molecular gelators. Curr Opin Colloid Interface Sci. 58. Rodrigues RC, Virgen-Ortiz JJ, Dos Santos JCS, Berenguer-Murcia A, 2020;45:68–82. Alcantara AR, Barbosa O, Ortiz C, Fernandez-Lafuente R. Immobilization 34. Borg P, Lê G, Lebrun S, Pées B. Example of industrial valorisation of deriva- of lipases on hydrophobic supports: immobilization mechanism, advan- tive products of Castor oil. Oléagineux Corps gras Lipides. 2009;16:211–4. tages, problems, and solutions. Biotechnol Adv. 2019;37:746–70. 35. Johnson W. Amended final report on the safety assessment of hydroxy- 59. Cole-Hamilton DJ. TRP: Catalyst separation recovery and recycling. Cataly- stearic acid. Int J Toxicol. 1999;18:1–10. sis by metal complexes. In: Cole-Hamilton DJ, editor. Homogeneous 36. Lee B, Lee S, Kim H, Jeong K, Park J, Lee E, Lee J. Biotransformation of catalysis—advantages and problems, vol. 30. Dordrecht: Springer; 2006. oleic acid into 10-ketostearic acid by recombinant Corynebacterium p. 1–8. glutamicum-based biocatalyst. Biotechnol Lett. 2015;37:1101–6. 60. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam 37. Todea A, Hiseni A, Otten LG, Arends IWCE, Peter F, Boeriu CG. Increase of H, Remmert M, Soding J, et al. Fast, scalable generation of high-quality stability of oleate hydratase by appropriate immobilization technique protein multiple sequence alignments using Clustal Omega. Mol Syst and conditions. J Mol Catal B Enzym. 2015;119:40–7. Biol. 2011;7:539. 38. Pari L, Suardi A, Stefanoni W, Latterini F, Palmieri N. Environmental and 61. Landau M, Mayrose I, Rosenberg Y, Glaser F, Martz E, Pupko T, Ben-Tal economic assessment of castor oil supply chain: a case study. Sustainabil- N. ConSurf 2005: the projection of evolutionary conservation scores of ity. 2020;12:6339. residues on protein structures. Nucleic Acids Res. 2005;33:W299-302. 39. Ogunniyi DS. Castor oil: a vital industrial raw material. Bioresour Technol. 62. Chovancova E, Pavelka A, Benes P, Strnad O, Brezovsky J, Kozlikova B, Gora 2006;97:1086–91. A, Sustr V, Klvana M, Medek P, et al. CAVER 3.0: a tool for the analysis of 40. de Cassia de Souza Schneider R, Lara LRS, Ceolin MM, Kaercher JA, Sch- transport pathways in dynamic protein structures. PLoS Comput Biol. neider M. Environmental impact of castor oil catalytic transfer hydrogena- 2012;8:e1002708. tion. Clean Technol Environ Policy. 2013;15:977–85. 63. Eser BE, Poborsky M, Dai R, Kishino S, Ljubic A, Takeuchi M, Jacobsen C, 41. Trivedi RK, Vasishtha AK. Low-pressure hydrogenation of castor-oil. J Am Ogawa J, Kristensen P, Guo Z. Rational engineering of hydratase from Oil Chem Soc. 1988;65:1467–9. Lactobacillus acidophilus reveals critical residues directing substrate 42. Sreenivasan B, Kamath NR, Kane JG. Studies on castor oil. II. Hydrogena- specificity and regioselectivity. ChemBioChem. 2020;21:550–63. tion of castor oil. J Am Oil Chem Soc. 1957;34:302–7. 64. Dixit M, Weitman M, Gao J, Major DT. Chemical control in the battle 43. Naughton FC. Production, chemistry, and commercial applications of against fidelity in promiscuous natural product biosynthesis: the case of various chemicals from castor-oil. J Am Oil Chem Soc. 1974;51:65–71. trichodiene synthase. ACS Catal. 2017;7:812–8. 44. Masri MA, Garbe D, Mehlmer N, Brück TB. A sustainable, high-perfor- 65. Singh R, Kumar M, Mittal A, Mehta PK. Microbial enzymes: industrial mance process for the economic production of waste-free microbial progress in 21st century. 3 Biotech. 2016;6:1–5. oils that can replace plant-based equivalents. Energy Environ Sci. 66. Hauer B. Embracing nature’s catalysts: a viewpoint on the future of bioca- 2019;12:2717–32. talysis. ACS Catal. 2020;10:8418–27. 45. Zheng Y, Li L, Liu Q, Qin W, Yang J, Cao Y, Jiang X, Zhao G, Xian M. Boosting 67. Jeon E-Y, Lee J-H, Yang K-M, Joo Y-C, Oh D-K, Park J-B. Bioprocess the free fatty acid synthesis of Escherichia coli by expression of a cytosolic engineering to produce 10-hydroxystearic acid from oleic acid by Acinetobacter baylyi thioesterase. Biotechnol Biofuels. 2012;5:76. recombinant Escherichia coli expressing the oleate hydratase gene of 46. Engleder M, Pavkov-Keller T, Emmerstorfer A, Hromic A, Schrempf S, Stenotrophomonas maltophilia. Process Biochem. 2012;47:941–7. Steinkellner G, Wriessnegger T, Leitner E, Strohmeier GA, Kaluzna I, et al. 68. Small DM: physical properties of fatty acids and their extracellular and Structure-based mechanism of oleate hydratase from Elizabethkingia intracellular distribution. In: Polyunsaturated fatty acids in human nutri- meningoseptica. ChemBioChem. 2015;16:1730–4. tion. Volume 28. Edited by Bracco U, Deckelbaum RJ. New York: Nestec 47. Radka CD, Batte JL, Frank MW, Young BM, Rock CO. Structure and mecha- Ltd., Vevey/Raven Press. Ltd.; 1992: 25–39.[Nestlé Nutrition Workshop nism of Staphylococcus aureus oleate hydratase (OhyA). J Biol Chem. Séries]. 2021;296:100252. 69. Busch H, Tonin F, Alvarenga N, van den Broek M, Lu S, Daran JM, Hanefeld 48. Lorenzen J, Driller R, Waldow A, Qoura F, Loll B, Brück T. Rhodococcus eryth- U, Hagedoorn PL. Exploring the abundance of oleate hydratases in the ropolis oleate hydratase: a new member in the oleate hydratase family genus Rhodococcus-discovery of novel enzymes with complementary tree-biochemical and structural studies. ChemCatChem. 2018;10:407–14. substrate scope. Appl Microbiol Biotechnol. 2020;104:5801–12. 49. Park AK, Lee GH, Kim DW, Jang EH, Kwon HT, Chi YM. Crystal structure 70. Takeuchi M, Kishino S, Park SB, Hirata A, Kitamura N, Saika A, Ogawa J. Effi- of oleate hydratase from Stenotrophomonas sp. KCTC 12332 reveals cient enzymatic production of hydroxy fatty acids by linoleic acid Delta9 P rem et al. Microbial Cell Factories (2022) 21:58 Page 15 of 15 hydratase from Lactobacillus plantarum AKU 1009a. J Appl Microbiol. 2016;120:1282–8. 71. An JU, Joo YC, Oh DK. New biotransformation process for production of the fragrant compound gamma-dodecalactone from 10-hydroxystea- rate by permeabilized Waltomyces lipofer cells. Appl Environ Microbiol. 2013;79:2636–41. 72. Griffiths MJ, van Hille RP, Harrison ST. Selection of direct transesterification as the preferred method for assay of fatty acid content of microalgae. Lipids. 2010;45:1053–60. 73. Topolewska A, Czarnowska K, Halinski LP, Stepnowski P. Evaluation of four derivatization methods for the analysis of fatty acids from green leafy vegetables by gas chromatography. J Chromatogr B Anal Technol Biomed Life Sci. 2015;990:150–7. 74. Kim BN, Yeom SJ, Oh DK. Conversion of oleic acid to 10-hydroxystearic acid by whole cells of Stenotrophomonas nitritireducens. Biotech Lett. 2011;33:993–7. 75. Joo YC, Seo ES, Kim YS, Kim KR, Park JB, Oh DK. Production of 10-hydroxy- stearic acid from oleic acid by whole cells of recombinant Escherichia coli containing oleate hydratase from Stenotrophomonas maltophilia. J Biotechnol. 2012;158:17–23. 76. Jung SM, Seo JH, Lee JH, Park JB, Seo JH. Fatty acid hydration activ- ity of a recombinant Escherichia coli-based biocatalyst is improved through targeting the oleate hydratase into the periplasm. Biotechnol J. 2015;10:1887–93. 77. Lee CY, Chang HN. Continuous production of acrylamide using immobi- lized Brevibacterium sp. CH2 in a two-stage packed bed reactor. Biotech- nol Lett. 1990;12:23–8. 78. Lin BX, Tao Y. Whole-cell biocatalysts by design. Microb Cell Fact. 2017;16:106. 79. Julsing MK, Cornelissen S, Buhler B, Schmid A. Heme-iron oxygenases: powerful industrial biocatalysts? Curr Opin Chem Biol. 2008;12:177–86. 80. Li Y, Dietrich M, Schmid RD, He B, Ouyang P, Urlacher VB. Identification and functional expression of a Δ9-fatty acid desaturase from Psychrobac- ter urativorans in Escherichia coli. Lipids. 2008;43:207–13. 81. Busch H, Alvarenga N, Abdelraheem E, Hoek M, Hagedoorn PL, Hanefeld U. Re-investigation of hydration potential of Rhodococcus whole-cell biocatalysts towards michael acceptors. ChemCatChem. 2019;12:193–8. 82. Brück T, Lorenzen J: A process for the cell-free enzymatic production of 10-hydroxystearic acid (10-hsa) from bio-based oils for lubricant formula- tion. (München TU ed). 2019. 83. Sun QF, Zheng YC, Chen Q, Xu JH, Pan J. Engineering of an oleate hydratase for efficient C10-Functionalization of oleic acid. Biochem Biophys Res Commun. 2021;537:64–70. 84. Federsel HJ, Moody TS, Taylor SJC. Recent trends in enzyme immobi- lization-concepts for expanding the biocatalysis toolbox. Molecules. 2021;26:2822. 85. Betancor L, Hidalgo A, Fernandez-Lorente G, Mateo C, Rodriguez V, Fuentes M, Lopez-Gallego F, Fernandez-Lafuente R, Guisan JM. Use of physicochemical tools to determine the choice of optimal enzyme: Stabilization of D-amino acid oxidase. Biotechnol Prog. 2003;19:784–8. 86. Demming RM, Hammer SC, Nestl BM, Gergel S, Fademrecht S, Pleiss J, Hauer B. Asymmetric enzymatic hydration of unactivated aliphatic alkenes. Angewandte Chemie-International Edition. 2019;58:173–7. 87. de Vries SJ, van Dijk M, Bonvin AM. The HADDOCK web server for data- driven biomolecular docking. Nat Protoc. 2010;5:883–97. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- fast, convenient online submission lished maps and institutional affiliations. thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Microbial Cell Factories – Springer Journals

Published: Apr 9, 2022

Keywords: Oleate hydratase; Biocatalysis; Industrial biotechnology; Whole cell and enzymatic oleic acid transformation; Green chemistry; Protein engineering; Structure–function relation; Bioeconomy

There are no references for this article.