MINI-REVIEW

Baeyer-Villiger oxidations: biotechnological approach

Marek Bučko1 & Peter Gemeiner1 & Andrea Schenkmayerová1 & Tomáš Krajčovič1 &

Florian Rudroff2 & Marko D. Mihovilovič2

Received: 23 March 2016 /Revised: 2 June 2016 /Accepted: 7 June 2016 /Published online: 21 June 2016# Springer-Verlag Berlin Heidelberg 2016

Abstract Baeyer-Villiger monooxygenases (BVMOs) are avery well-known and intensively studied class of flavin-dependent enzymes. Their substrate promiscuity, high che-mo-, regio-, and enantioselectivity are prerequisites for theuse in synthetic chemistry and should pave the way for suc-cessful industrial processes. Nonetheless, only a very limitednumber of industrial relevant transformations are known,mainly due to the lack of BVMOs stability and cofactor de-pendency. In this review, we focus on novel BVMO-mediatedtransformations, BVMOs in cascade type reactions, potentialindustrial applications, and how limitations have been tackledby the community. Special attention will be put on whole-cellimmobilization strategies. We emphasize to bridge recent de-velopments in fundamental research to industrial applications.

Keywords Baeyer-Villiger monooxygenases . Cascadereactions .Whole-cell immobilization . Polyelectrolytecomplex capsules . Asymmetric synthesis . Biocatalysis

Introduction

Enzymes from the group of Baeyer-Villiger monooxygenases(BVMOs, E.C. 1.14.13.xx) are bioequivalent to chemical cat-alysts of Baeyer-Villiger (BV) oxidations for biotechnologicalproduction of high added value chemicals. This class of en-zymes was systematically studied and described in severalrecent review articles (Balke et al. 2012; Leisch et al. 2011;Pazmino et al. 2010a). From the biotechnological point ofview, the main advantage of BVMOs compared to chemicalcatalysts includes high selectivity together with broad sub-strate scope, which can be illustrated by the ability of the moststudied enzyme—cyclohexanone monooxygenase fromAcinetobacter sp. (CHMOAcineto) (Donoghue et al. 1976) tocatalyze BV oxidation of several dozens of substrates (Finket al. 2012; Kayser 2009). BVMOs also provide access to anoverwhelming number of valuable compounds including chi-ral precursors for bioactive compounds (Mihovilovic 2006).This feature of BVMOs arised mainly from the large diversityof this enzyme class (Balke et al. 2012) which enables cata-lytic BV oxidations of many types of substrates includinglinear, cyclic, and aromatic ketones as well as steroids andterpenoids (Leisch et al. 2011). Moreover, BVMOs may cat-alyze BVoxidation towards products, which are not accessibleby synthetic route. Examples include production of “abnor-mal” lactone from bicyclo[3.2.0]bicyclo-2-en-6-one (Leischet al. 2011) or enantioselective BV oxidation of 8-oxabicyclo[3.2.1]oct-6-ene-3-one to the corresponding lac-tone which was considered as potential chiral building blockfor synthesis of bioactive compounds including antiviral C-nucleosides (Mihovilovic 2006). Currently, synthetic routestowards carba analogs of C-nucleoside derivatives using re-combinant, whole-cell BVMOs were investigated and pub-lished by Bianchi et al. (2013).

* Peter GemeinerPeter.Gemeiner@savba.sk

* Florian Rudroffflorian.rudroff@tuwien.ac.at

1 Department of Glycobiotechnology, Institute of Chemistry – Centerfor Glycomics, Slovak Academy of Sciences, Dúbravská cesta 9,SK-845 38 Bratislava, Slovakia

2 Institute of Applied Synthetic Chemistry, TU Wien, Getreidemarkt9/163, A-1060 Vienna, Austria

Appl Microbiol Biotechnol (2016) 100:6585–6599DOI 10.1007/s00253-016-7670-x

Despite extensive research of BVMOs, their application inindustrial BV protocols is missing (Leisch et al. 2011). In fact,the only found industrial process in kg scale was performed bySigma Aldrich (Doig et al. 2002). The process utilizes recom-binant E. coli cells overexpressing CHMOAcineto for the BVoxidation of bicyclo[3.2.0]hept-2-en-6-one to correspondinglactones in continuous stirred tank reactor (CSTR) of 55 Lwith space-time yield of 1.1 g L−1 h−1 (Doig et al. 2002).Later, a successful 200-L scale asymmetric Baeyer-Villigeroxidation catalyzed by CHMOAcineto was performed with theyield of 495 g of final lactone product from 900 g of substratebicyclo[3.2.0]hept-2-en-6-one (molar yield 48 %) (Baldwinet al. 2008). Probably, the main challenging tasks to solvefor applications of BVMOs in industry are their low stability,the high demand of aerial oxygen, and therefore, the identifi-cation of target lactones where the utilization of BVMOs issuperior to conventional chemical route (Leisch et al. 2011).Within the scope of this review, we will cover single to cas-cade transformations and suitable techniques for the utiliza-tion of BVMOs in large scale industrial applications.

Baeyer-Villiger monooxygenases

Classification of BVMOs

Flavin containingmonooxygenases are in general classified asexternal and internal monooxygenases (Pazmino et al. 2010b).The first group relies on the reducing power of NADPH orNADH for the regeneration of flavin itself whereas the latterone applies electrons from the substrate to reduce flavin. Ingeneral, they are divided in six subclasses (A-F) according tostructural/sequence-based relations (van Berkel et al. 2006).BVMOs are external monooxygenases containing non-covalently bound FAD or FMN and catalyze NAD(P)H de-pendent aerial oxygen insertions under the release of water(Balke et al. 2012; Ceccoli et al. 2014; Pazmino et al.2010a). Up to now, two different types of BVMOs are known.Type I monooxygenases are FAD and NADPH dependentbiocatalysts and belong to the class B flavoenzymes, whereasType II BVMOs require FMN and NADH for catalysis(Pazmino et al. 2010b). Recently, the first crystal structure ofa t ype I I BVMO, name ly 3 ,6 -d ike tocamphanemonooxygenase, was published (Isupov et al. 2015) whichcomplements the intensive research in this area in the pastyears (Kadow et al. 2012, 2014). Nevertheless, most widelyapplied in synthetic chemistry are Type I BVMOs and morethan 50 protein sequences are available for recombinant ex-pression today. In the past 5 years, several comprehensivereviews have been published covering the classification andcharacteristics of flavin-dependent monooxygenases; hence,we focus in the next sections on recent developments in ap-plying enzyme-mediated Baeyer-Villiger reactions in

synthetic chemistry (de Gonzalo et al. 2010; Leisch et al.2011; Mihovilovic et al. 2004).

Recent BVMO applications

Single enzyme transformations

Since nearly two decades, many BVMOs have been intensive-ly used in the synthesis of chiral lactones. Different structuralmotifs, like aliphatic, cyclic, and polycyclic have been trans-formed to the corresponding esters or lactones. Despite the useof aerial oxygen under ambient reaction conditions, BVMOsdisp lay a remarkably high reg io- , chemo- , andstereoselectivity (Fink et al. 2011; Mihovilovic et al. 2008;Snajdrova et al. 2006). One remarkable example for the useof BVMOs in industry is given by Codexis, which producedEsomeprazole by applying a CHMO variant (41 mutations,104-fold increased stability and activity) in 99 %enantioselectivity (Bong et al., 2011).

Extending recent discoveries for enantiocomplementaryenzymes of this class (Mihovilovic et al . 2005),cycloketone-converting BVMOs from Xanthobacter sp. ZL5(CHMOXantho) (van Beilen et al. 2003) and sp. NCIMB 9872(CPMOComa) (Iwaki et al. 2002), were used in the keydesymmetrization step towards both enantiomers of bicycliclactone (+) and (−)-(2). These key building blocks were ap-plied in the synthesis of several different kumausyne,goniofufurone, and C-nucleoside analogs. With this biocata-lytic approach, classical chemical synthetic routes could beshorten considerably and access to carba-analogs (X = CH2),a novel approach to so far not tested bioactive compound classwas achieved (Fig. 1a) (Bianchi et al. 2013; Rudroff et al.2015). The group of Bornscheuer exploited another set ofenantiodivergent BVMOs (CHMOs from Arthrobacter orBrachymonas sp., and a cyclododecanone monooxygenasefrom Rhodococcus ruber SC1, CDMORhodo) which catalyzedthe formation of optically pure β-amino acids (8) (e.g., (R)-β-leucine, 33% yield, >99 % ee) after enzymatic cleavage of theintermediate esters (7) (Fig. 1b) (Rehdorf et al. 2010). Despitethe high stereospecificity of BVMOs, the unique feature ofchemo- as well as regioselectivity was subject of several stud-ies in the last years.

Very recently two novel BVMOs with a distinct and so farunknown substrate specificity and especially regioselectivitywere reported by the group of Alphand (Reignier et al. 2014).They investigated the synthesis of chiral ene- (10) and enol-lactones (11) starting from α,β-unsaturated ketones ((9),Fig. 2). These compound classes ((10) and (11)) are frequentlyused motifs in organic synthesis for the production of bioac-tive compounds and natural products. The authors screened 60putative Type I BVMOs and found one BVMO fromOceanicola batsensis DSM 15984 (BVMOOcean) and a

6586 Appl Microbiol Biotechnol (2016) 100:6585–6599

BVMO from Parvibaculum lavamentivorans (BVMOParvi)DSM 13023 which were tested on various α,β-unsaturatedsubstrates (9) . Remarkably, transformations withBVMOOcean gave exclusively the normal lactone (10), where-as BVMOParvi resulted in the enol-lactone (11). Yields of up to93 % and stereoselectivities of 98 % ee were obtained.

Recently, another diastereoselective desymmetrizationwas published by the group of Reetz (Fig. 3): 4-methylenecyclohexanones (12) could be selectively convertedto E-configured 5-methylene-ε-caprolactones (13)-E (>96:4E:Z, four examples) using the wild-type CHMOAcineto

(Zhang et al. 2013). After creation of an efficient variant libraryof this catalyst over three rounds of mutation, complete inver-sion of diastereoselectivity towards the corresponding Z-lac-tones (13)-Z (4:96 d.r.) could be achieved. The group ofBornscheuer realized the first complete switch in regioselectivityby rational protein design of CHMO from Arthrobacter sp. onthe substrate (+)-trans-dihydrocarvolactone. After introducingdocking studies, a triple mutant resulted in a complete inversionof the regioselectivity compared to the wild-type enzyme (0:100vs 99:1 normal/abnormal lactone (Balke et al. 2016).

A surprising new class of BVMOs was found during thecharacterization of eight flavin monooxygenases (FMOs)from Rhodococcus jostii RHA1: three biocatalysts of the testset (Riebel et al. 2013) readily performed BVoxidations on arange of small-ring ketones, and they also have a heretoforeindifference towards the nicotinamide cofactor consequentlyutilizing NADH and NADPH equally well. This feature mayprove to be highly useful in enzymatic redox cascades,circumventing the need for complex cofactor shuttling.

The large and highly diverse information available onBVMOs, their substrate promiscuity, variable activity, andselectivity on structurally diverse ketones made it increasinglydifficult to set a newly discovered catalyst into context bycomparison of its performance to existing BVMOs. Hence, astatistical scoring algorithm for BVMOs was developed basedon the application-oriented properties activity and selectivity(Fink et al. 2012). Nine enzymes were ranked according totheir performance in various substrate classes and their overallversatility. The model is complemented by a clear graphical

Fig. 1 Desymmetrization ofbicycloketones (1) towards bothenantiomers of lactones (2) usingtwo BVMOs (a); regioselectivekinetic resolution ofβ-aminoketones (6) towardsβ-amino acids (8);CAL-B—Candida antarcticalipase B (b)

Fig. 2 Novel BVMO activities towards α,β-unsaturated ketones for thesynthesis of chiral enol- and ene-lactones

Fig. 3 Diastereoselective desymmetrization of 4-methylenecyclohexanonesby CHMOAcineto and evolved mutants

Appl Microbiol Biotechnol (2016) 100:6585–6599 6587

output for facile interpretation and operates with minimal bias.Consequently, the catalytic performance and potential synthet-ic applicability of novel BVMOs are quantifiable and, there-fore, an alleviated comparison to existing biocatalysts of thisenzyme class is possible.

BVMOs applied in cascade type reactions

The attention in the biocatalysis community increased signif-icantly towards cascade reactions in the past decade (Bayeret al. 2015; Kohler and Turner 2015; Muschiol et al. 2015).Synthesis of highly complex molecules based on multi-enzyme processes, inspired by Nature’s paradigm, becomemore and more important. Multistep cascades commonlyfunction without the separation of intermediates, display rath-er low intermediate concentrations, and result in processeswith high selectivity of products and without formation ofbyproducts. Additionally, limited productivity due to unfavor-able equilibrium can be enhanced by the cooperative effect ofmultiple catalysts. Especially, the shift of equilibrium by in-corporation of an irreversible reaction pulls the intermediatesto the end of the cascade and results in accumulation of thedesired product. Recently, several enzyme cascade examplesincluding redox enzymes have been published. Thereby, theirreversibility of BVMO reactions and the aspect of redoxbalancing was exploited to produce high value compounds.Several examples were published by the group of Park (Songet al. 2013). They focused on the use of renewable startingmaterials like fat or fatty acids (14) (γ-linolenic acid (Kimet al. 2015), ricinoleic acid (Jang et al. 2014), linoleic acid(Oh et al. 2015)) and applied artificial enzymatic cascadesin vivo for the production of flavors, antifungal agents, plas-tics, waxes, nylons, and secondary metabolites. In general,they combined a hydratase for the introduction of an alcoholgroup with an alcohol dehydrogenase (ADH) to produce thecorresponding ketone (16). Additionally, degradation of keto-acids was performed by a regioselective BVMO-mediatedoxygen insertion followed by a hydrolytic esterase transfor-mation ((17), Fig. 4). First, the selectivity of varioushydratases was exploited for the specific hydroxylation ofdiverging double bonds in unsaturated fatty acid precursors,secondly different BVMOs (BVMO from P. fluorescenceDSM 50106 (Kirschner et al. 2007)—normal lactone;P. putita KT2440 (Rehdorf et al. 2007)—abnormal lactone)were applied to control the direction of the oxygen insertion.The pulling effect of the irreversible BVMO transformationresulted in a highly efficient biosynthetic pathway for the pro-duction of hydroxy acids, diacids, carboxylic acids, and fattyalcohols (17). These pathways were expressed recombinantlyin E. coli and gave up to 97 % GC yield for γ-linolenic acid(14) as starting material after optimized expression (Kim et al.2015).

Recently, two different enzyme cascades have been pub-lished by the group of Kroutil (Sattler et al. 2014) andBornscheuer (Schmidt et al. 2015b) for the synthesis of valu-able bulk chemicals (Fig. 5), harnessing the concept of a redoxself-sufficient system. Kroutil and coworkers started fromcyclohexanol ((18), Fig. 5, top) and applied a NADP+ depen-dent ADH for the first oxidation step, followed by a NADPHdependent BVMO-mediated oxidation to the correspondingε-caprolactone (20). Afterwards, they tempted to open thelactone in the presence of a lactonase to yield the ε-hydroxyacid which could be transformed by an additional ADH(NAD+ dependent) and an ω-transaminase (in the presenceof an alanine dehydrogenase, AlaDH, NADH dependent) intothe desired 6-aminohexanoic acid (22). It turned out that oncethe lactone was opened, the cascade stopped and did not pro-ceed further. As a consequence, they developed a so-called“capping strategy”, trapping the free acid with methanol byhorse liver esterase to form the corresponding methyl ester(21), which could be converted to the desired 6-aminohexanoic acid (22) with good to reasonable yields (up to75 % starting from ε-caprolactone (20) and 24 % starting fromcyclohexanol (18)). The beauty of this cascade lies in its redox

Fig. 4 Designed biosynthetic pathway for the synthesis of differentvaluable compounds by the group of Park (Song et al. 2013), startingfrom renewable sources; nL normal lactone; abnL abnormal lactone

6588 Appl Microbiol Biotechnol (2016) 100:6585–6599

neutral nature, water as a reaction medium, and the supply ofstoichiometric ammonia and molecular oxygen solely.

The second example, recently published by the group ofGröger and Bornscheuer (Staudt et al. 2013), was based onthe conversion of cyclohexanol (18) to poly-ε-caprolactone(23) in presence of an ADH, BVMO, and a lipase (Fig. 5).The first two steps were similar as depicted before. Thoughthe authors used a different ADH for the first step, both groupspresented a productivity of the desired ε-caprolactone (20) inup to 99 % yield. Whereas the groups of Gröger andBornscheuer demonstrated a severe inhibition effect and en-zyme deactivation by ε-caprolactone (20) (at approximately80 mM cyclohexanol) (Staudt et al. 2013) on the BVMO, thegroup of Kroutil did not encounter any problems up to a con-centration of 200 mM cyclohexanol. Consequently, the groupof Bornscheuer investigated an in situ removal strategy to in-crease overall productivity. They envisaged a lipase mediatedpolymerization towards poly-ε-caprolactone (23) in water.Several lipases were tested and interestingly Cal-A was foundto have unique acyltransferase activity that enables the forma-tion of the desired ester product even in the presence of bulkwater. Finally, they were able to synthesize poly-ε-caprolactone(23) in 75 % yield (based on consumed starting material) in aone-pot process in water. Very recently, they could extent theirconcept towards chiral lactone and were able to produceenantiopure oligomers of 4-methylcaprolactone (Schmidtet al. 2015a).

In 2015 Zuhse et al. investigated investigated a bi enzymat-ic cascade for the production of ε-caprolactone (Fig. 6a—rect-angle) (Zuhse et al. 2015). The authors introduced a concep-tually novel term for redox-neutral reactions, so-called con-vergent cascade which involves two substrates that result in asingle product without the formation of intermediates. First,the authors applied a BVMO (CHMOAcineto) mediated andNADPH dependent oxidation of cyclohexanone (24, n = 1)towards ε-caprolactone (28, n = 1). This reaction step requiredNADPH and molecular oxygen. In parallel, they used the oxida-tive power of different NADPH dependent alcohol dehydroge-nases (Burdette and Zeikus 1994) for the synthesis of ε-caprolactone (28) starting from 1,6-hexanediol (24). In this par-ticular reaction step, 1 mol 1,6-hexandiol (24) regenerated 2 molNADP+ and provided necessary amounts of NADPH for theBVMO oxidation. This redox neutral reaction sequence demon-strated the power of a smart reaction design and again the asset of

BVMOs by shifting the reaction equilibrium towards the productside.

The group of Oppermann reported a multi-enzyme biocat-alytic pathway in which non-activated cycloalkanes (25) wereconverted to lactones, an important class of building block forpolymers (Fig. 6b). This cascade consists of a cytochromeP450 monooxygenase (Cyp450 (Pennec et al. 2015)), an al-cohol dehydrogenase (TADH (Hollrigl et al. 2008), TeSADH(Burdette and Zeikus 1994)) and finally a BVMO (Oppermanand Reetz 2010) for the synthesis of the desired lactone (28).In the first attempt, the authors created a designer cell, byexpressing all enzymes in E. coli, but unfortunately, the ex-pression levels were low and the expression reproducibilityfor the ADH and the Cyp450 turned out to be unpredictable.Instead of optimizing the cellular host, they changed from anin vivo to an in vitro system, by expressing all enzymes indi-vidually in E. coli and use cell free extracts. This approachfacilitates the optimization of this process for example bybalancing biocatalyst concentrations and increases the fluxthrough the cascade. Since the second and the third step ofthis cascade is redox self-sufficient, whereas the hydroxyl-ation still requires NAD(P)H, the authors implemented a co-factor recycling system specific for NAD+/NADH and applieda Cyp450 variant that is solely NADH dependent. After sev-eral optimization efforts, production of up to 23 mMenantholactone (28, n = 2) in less than 12 h could be achieved.In summary, they developed a redox-balanced one-pot cas-cade composed of four oxidoreductases with total turnovernumbers (TTN) calculated on the Cyp450 monooxygenaseofmore than 4000 (maximumμmol total oxygenated productsformed within 24 h μmol−1 P450). A few remarkable exam-ples of the use of BVMOs in enzymatic cascade type reactionspublished recently substantiate the great potential of this en-zyme class in complex organic syntheses (Liu and Li 2013;Oberleitner et al. 2013, 2014; Song et al. 2013). Quite recent-ly, Oberleitner et al. complemented their studies on artificialenzyme cascades by valorization of orange peel as startingmaterial for the production of a chiral carvolactone (33), apromising building block for thermoplastic polymers(Fig. 6c). Four different concepts were applied to augmentlimonene availability either based on water extraction solely,addition of extraction enhancers, or biomass dissolution.Subsuming in a one-pot resting cell mixed culture approach,they combined the selective hydroxylation by CumDO

Fig. 5 Redox neutral enzymecascade synthesis of nylon-6monomer and poly-ε-caprolactone starting fromcyclohexanol

Appl Microbiol Biotechnol (2016) 100:6585–6599 6589

expressed in P. putida S12 with a previously established syn-thetic mini-pathway in E. coli BL21(DE3), where carveol canserve as starting material. Thereby, limonene extracted by wa-ter solely (29) could be directly transformed to chiralcarvolactone (33) via carveol (30), carvone (31), anddihydro-carvone (32). Carvolactones, interesting buildingblocks for syntheses of bioactive or natural products, can alsoserve as monomers for polymer production as they can besubjected to ring-opening polymerization and their olefinicside chains can be easily functionalized and crosslinked.This study demonstrates the potential of artificial enzyme cas-cades and future industrial applications (Oberleitner et al.,2016).

Potential BVMO-mediated industrial processes

Despite the application of BVMOs for the synthesis of highvaluable chemicals or chiral building blocks, several potentialindustrial processes for the synthesis of bulk chemicals havebeen published recently. Fink et al. presented the synthesis of3-hydroxypropionates (35) starting from levulinic acid deriv-atives (34), a major by-product from carbohydrate dehydra-t ion (F ig . 7 ) (F ink and Mihovi lov ic 2015) . 3 -Hydroxypropionates (35) readily hydrolyze to 3-hydroxypropionic acid which is a versatile C3 precursor forthe production of acrylates, malonates, and 1,3-propandiol.The authors screened a library of 13 different BVMOs andfound 9 potential biocatalysts for the synthesis of the required

3-hydroxypropionates. They were able to produce the desiredcompounds in gram scale at ambient temperature, in water, inthe presence of aerial oxygen instead of propellant-gradeH2O2 which is required for the chemical reaction. Anotherquite appealing approach was published by the group ofFraaije (van Beek et al. 2014). They managed to find severalBVMOs for the production of ethyl acetate (36) or methylpropionate (37) starting from 2-butanone (34). Oxidation of2-butanone (34) (Fig. 7) in the presence of CHMO fromRhodococcus HI-31 gave a 6:1 ratio of ethyl acetate (36)and methyl propionate (37). Since the normal lactone forma-tion was preferred, the authors tried to influence the regiose-lectivity by modulation of the microenvironment. Substantialincrease in the amount of the starting ketone shifted theproduct distribution towards a 1.5:1 ratio still in favor of ethylacetate (36).

Concluding this section, Fink et al. investigated a chemo-enzymatic synthesis towards an important aroma compound

Fig. 6 a A convergent synthesisof ε-caprolactone employing1,6-hexanediol as a “smartcosubstrate”. b Enzymaticsynthesis of lactones starting fromsimple cycloalkanes.c Valorization of limonenecontaining orange peel for theproduction of chiral carvolactone(33) via a four step enzymaticcascade

Fig. 7 Recently published application of BVMOs for the production ofbulk chemicals

6590 Appl Microbiol Biotechnol (2016) 100:6585–6599

from African orchids, the natural isomer of Aerangis lactone(41) (Fink et al. 2013). Thereby, a diastereoselective Rh-catalyzed continuous flow hydrogenation was combined withthe Baeyer-Villiger reaction in a single operation, yielding thedesired fragrance lactone (41) in excellent optical purity withhigh scalability potential (>99 % ee, >99 % de, 3.4 g L−1,<60 min; Fig. 8). Remarkably, the applied CDMORhodo

displayed a perfect cis-selectivity which resulted in the finalhigh diastereomeric ratio and enantioselectivity. Access to theepimeric lactone (43) was obtained by employing CPMOComa

in the presence of the trans-ketone (42), which was receivedafter acidic epimerization, with comparable, but not perfectdiastereomeric purity. Subsuming, a 7-step stereoselectivesynthesis could be shortened to a single operation process byapplying heterogeneous metal assisted and enzyme catalysis.

Although none of the presented potential applications didachieve readiness for industrial processes, these examplesdemonstrate a future trend for BVMOs.

BVMOs in biotechnology—overcoming problems

A number of drawbacks have been identified to be overcometo enable industrial applications of BVMO-catalyzed oxida-tions. Low enzyme stability, substrate, and product inhibition(Doig et al. 2003) as well as low product concentration, ade-quate oxygen transfer, and tolerance of organic cosolvents(Law et al. 2006) are among the most relevant obstacles.Several ways to overcome these limitations have been intro-duced during recent years.

Immobilization

Immobilization in biotechnology is defined by IUPAC as thetechnique used for the physical or chemical fixation of cells,organelles, enzymes, or other proteins (e.g., monoclonal anti-bodies) onto a solid support, into a solid matrix or retained bya membrane, in order to increase their stability and make pos-sible their repeated or continued use (http://goldbook.iupac.org/I02973.html). Consequently, several attempts either forimmobilization of BVMO expressing whole-cells or isolatedenzymes have been investigated to improve overall

biocatalyst stability. Such immobilization strategies may ren-der improvements regarding increase of harvested product percell mass, which is still challenging task at a larger scale BVoxidations (Lima-Ramos et al. 2014).

Another major challenge in the application of BVMOs istheir cofactor dependency (NAD(P)H) which in fact compli-cates their utilization in organic synthesis. Different strategieswere investigated to overcome this problem in small scaleeither by including a second enzyme loop for the cofactorregeneration using a cheap cosubstrate or by the applicationof a whole-cell system. Maintenance of the host metabolismprovides continuous enzyme production and cofactor supply(Doig et al. 2001). Recent progress in the area of immobiliza-tion of isolated enzymes and whole cells with BVMO activitywill be described in the following chapter.

Immobilization of isolated BVMOs

Since the application of BVMOs in industry is limited byenzyme stability and product inhibition, immobilization ofisolated BVMOs might be a good strategy to overcome theseproblems and facilitate recovery of BVMOs as well as toimprove cofactor regeneration of co-immobilization withrecycling enzymes (Balke et al. 2012). Table 1 shows exam-ples of immobilization techniques and characterization param-eters of immobilized biocatalysts applying isolated BVMOs.As shown in Table 1, covalent binding is the most commonimmobilization technique used for immobilization of isolatedBVMOs. It is also true that utilization of isolated BVMOsresulted in loss of their enzyme activity despite immobiliza-tion (Mallin et al. 2013; Cuetos et al. 2012; Atia 2005).Moreover, it is not possible to compare results from differentimmobilization techniques due to inconsistency in choice ofcharacterization parameters for immobilized isolated BVMOslisted in Table 1. Thus, the selection of proper immobilizationtechnique is still based purely on trial and error approach.

Utilization of immobilized isolated BVMOs requires anefficient cofactor recycling system. Two research groups usedcombination of CHMO and glucose-6-phosphate dehydroge-nase (G6PDH) for cofactor regeneration (Table 1). Althoughentrapment of CHMO in polyacrylamide gel resulted in highproduct yields, no data on operational stability were provided

Fig. 8 Kinetic resolution of cis-and trans-tetrahydrojasmones(39) towards natural Aerangislactone (41) and its epimer (43)using two BVMOs

Appl Microbiol Biotechnol (2016) 100:6585–6599 6591

(Abril et al. 1989). Stabilization effect of co-immobilization ofCHMO and G6PDH by c o v a l e n t b i n d i n g o npolyethyleneimine porous agarose beads was improved bygamma irradiation (Atia 2005). Regardless of this, relativeactivity of immobilized enzymes decreased to less than50 % of initial activity after 15 repeated cycles of (2-oxooxepan-3-yl)acetic acid production. Similarly, reuse ofCHMO and alcohol dehydrogenase (ADH) co-immobilizedon Eupergit C resulted in progressive decrease of conversionrates after the third cycle of repeated Baeyer-Villigerbiooxidations of bicyclo[3.2.0]hept-2-en-6-one (Zambianchiet al. 2002). One of the current new routes to enable enzymerecycling and stabil ization is represented by co-immobilization of thermostable polyol dehydrogenase(PDH) and CHMO by covalent binding on glutaraldehyde-activated support (Mallin et al. 2013). Immobilization enabledproduction of 600 mg l−1 of pure ε-caprolactone, and CHMOactivity towards higher substrate concentrations was im-proved. On the other side, immobilized CHMO lost its activityduring repeated biotransformation cycles. Decrease of relativeactivity of phenylacetone monooxygenase (PAMO) co-immobilized with G6PDH on polyphosphazene carrier during5 cycles of benzyl acetate production was ascribed to lowstability of PAMO (Cuetos et al. 2012). A unique concept ofpolymersome nanoreactors was used for encapsulation of fu-sion protein CRE2-PAMO containing thermostable PAMOand NADPH regenerating phosphite dehydrogenase

(Meeuwissen et al. 2011). The most effective production ofphenyl acetate was achieved by encapsulation of fused proteinin aqueous compartment of nanoreactors. No data on opera-tional stability of nanoreactors were provided.

Immobilization of whole-cells expressing BVMOs

Utilization of intact whole cells as biocatalysts of BV oxida-tions enable to avoid problems connected with regeneration ofexpensive cofactors and protein instability upon enzyme pu-rification (Mihovilovic 2006). Recombinant, whole-cellBVMOs are also more suitable for biotransformations if thenatural strain grows slowly or displays pathogenicity. Furtherimprovement of biocatalytic performance of BVMOs may beachieved by several strategies and immobilization of whole-cells is among them. The immobilization of cells is mainlyconsidered when there is a need of easy biocatalyst removalfrom reaction mixture, reuse of biocatalyst, continuous modeof reaction, and increase of biocatalyst concentration(Carballeira et al. 2009). Significant progress in biocatalyticperformance and characterization of recombinant whole-cellBVMOs has been achieved utilizing encapsulation of cellswithin polyelectrolyte complex (PEC) capsules and entrap-ment in polyvinyl alcohol particles. (Table 2).

The main benefits of BVMOs encapsulation in PEC cap-sules (Table 2) include (i) highly controlled encapsulationprocess (Scheme 1) with defined parameters, (ii)

Table 1 Immobilization materials and characterization parameters used for isolated BVMOs

Enzymes Immobilization technique/matrix Products Characterization of immobilizedbiocatalyst

Reference

CHMOAcineto +PDH

Covalent binding/Relizyme™HA403 glutaraldehyde activated

ε-Caprolactone -Stability towards substrate-Operational stability

(Mallin et al.2013)

CHMOAcineto +G6PDH

Entrapment/polyacrylamide gel 2-Oxabicyclo [3.2.1]octan-3-one2-Oxabicyclo[3.2.1]1,7,

7-trimethyloctan-3-one6-Isoprenyl-3-methyl-2-

oxacyclopentanone2 regioisomers of 1,2-fencholide

and 2,3-fencholide

-Substrate specificity,regioselectivity andenantioselectivity

-Reaction time-Yield

(Abril et al.1989)

CHMOAcineto +ADH

Covalent binding/Eupergit C R-methyl phenyl sulphoxide(1R,5S)-3-oxabicyclo-[3.3.0]

oct-6-en-3-one, (1S,5R)-2-oxabicyclo-[3.3.0]oct-6-en-3-one

-Operational stability-Effect of cofactor concentration

on conversion

(Zambianchiet al. 2002)

CHMOAcineto +G6PDH

Covalent binding/polyethyleneimine-glyoxyl-agarose beads

(2-oxooxepan-3-yl)acetic acid -Effect of gamma-irradiation onenzyme activity and leakage

-Effect of enzyme loading,temperature, support ratios,pH on activity, enzymekinetics, operational stability

(Atia 2005)

PAMO +G6PDH

Covalent binding/polyphosphazeneparticles and glutaraldehyde

Benzyl acetate -Operational stability-Stereoselectivity

(Cuetos et al.2012)

CRE2-PAMO Encapsulation/polymersomenanoreactors

Phenyl acetate -Time course of biooxidations (Meeuwissenet al. 2011)

6592 Appl Microbiol Biotechnol (2016) 100:6585–6599

biocompatible microenvironment for viable cells, (iii) adjust-ability of mechanical resistance and permeability of capsulemembranes, (iv) proper storage and operational stability, and(v) ability to scale-up production of capsules (Hucik et al.2010). PEC capsules preserved storage stability of E. coliwithCPMO, which was two-fold higher as compared to free cells.The time required to reach the end-point of Baeyer-Villigerbiooxidation of 8-oxabicyclo[3.2.1]oct-6-en-3-one to 4,9-dioxabicyclo[4.2.1]non-7-en-3-one was the same (48 h) forboth encapsulated and free cells (Hucik et al. 2010). In asubsequent study, a concentration-dependent inhibition actionof 4-methylcyclohexanone towards encapsulated E. coli withCPMO was observed. Parallely, negligible toxic effect of 4-methylcyclohexanone was detected by cell viability measure-ments via confocal laser scanning microscopy (CLSM)(Schenkmayerova et al. 2012). High operational stability dur-ing 14 repeated Baeyer-Vill iger biooxidations ofbicyclo[3.2.0]hept-2-en-6-one to corresponding lactones aswell as high storage stability during 91 days using E. coliwithCHMO encapsulated in PEC capsules was achievedemploying continuous packed-bed minireactor. The reactorwas connected with flow calorimeter and reservoir with reac-tion mixture saturated with oxygen by bubble free oxygena-tion (Bucko et al. 2011). Moreover, robust thermometric sig-nal, appropriate for measurement of reaction enthalpy ofBaeyer-Villiger biooxidation, was obtained by flow calorime-try (Scheme 1). The later technique was used for kinetic studyusing immobilized cells for determination of apparent kinetic

constants. Also, mild substrate inhibition above concentrationof 4.0 mM was proven. Utilization of PEC membrane andE. coli with CPMOComa for the construction of amperometricbiosensor for monitoring of the Baeyer-Villiger biooxidation(Schenkmayerova et al. 2013) proved high adaptability ofused polyelectrolyte complexation for biotechnological pur-poses. Interestingly, control over the selectivity of BVbiooxidation of cyclohexanol to ε-caprolactone can beachieved by immobilization ofGeotrichum candidum in poly-acrylamide or agar particles (Carballeira et al. 2004).

As described in Table 2, significant improvements of im-mobilization of encapsulated cells in PEC have been achievedapplying a lab-scale encapsulator, the main task regardingindustrial applications of PEC capsules is still the scale-upregarding production of PEC capsules. There are several highperformance encapsulators on the market which are based ondifferent principles with throughput from tens of grams toseveral kilograms of calcium alginate particles per hour(Prusse et al. 2008). Therefore, the main challenging task isto adapt and increase the throughput of more complex encap-sulation materials such as PEC capsules using high perfor-mance devices. Besides this, entrapment of whole-cellBVMOs in polyvinyl alcohol lenses LentiKats® (Table 2)(Rebros et al. 2014; Schenkmayerova et al. 2014) representspromising immobilization technique with potential industrialapplications, since LentiKats® were already utilized in indus-trial biotransformations (www.lentikats.eu). Moreover, keyphysical and chemical-engineering properties of PEC capsules

Table 2 Immobilization materials and characterization parameters used for whole-cell BVMOs

Enzyme in cells Immobilization technique/matrix Product Characterization of immobilizedbiocatalyst

Reference

CPMO/E. coli Entrapment/polyvinyl alcoholgel lens-shaped particlesLentiKats®

(1R,5S)-3-Oxabicyclo[3.3.0]oct-6-en-3-one

(1S,5R)-2-Oxabicyclo-[3.3.0]oct-6-en-3-one

-Enzyme expression-Oxygen demand-Operational stability

(Rebros et al. 2014)

CPMO/E. coli Entrapment/polyvinyl alcoholgel lens-shaped particlesLentiKats® and encapsulation/polyelectrolyte complex capsules

(1R,5S)-3-Oxabicyclo[3.3.0]oct-6-en-3-one

(1S,5R)-2-Oxabicyclo-[3.3.0]oct-6-en-3-one

-Particle size distribution-Inner morphology-Mechanical resistance-Size-exclusion properties-Effective diffusion coefficient

(Schenkmayerovaet al. 2014)

CPMO/E. coli Encapsulation/polyelectrolytecomplex capsules

5-Methyloxepane-2-one -Operational stability-Inhibition by substrate-Cell viability

(Schenkmayerovaet al. 2012)

CHMO/E. coli Encapsulation/polyelectrolytecomplex capsules

(1R,5S)-3-Oxabicyclo[3.3.0]oct-6-en-3-one

(1S,5R)-2-Oxabicyclo-[3.3.0]oct-6-en-3-one

-Thermometric signal by flowcalorimetry

-Operational stability-Storage stability

(Bucko et al. 2011)

CPMO/E. coli Encapsulation/polyelectrolytecomplex capsules

4,9-Dioxabicyclo[4.2.1]non-7-en-3-one

-Cell viability-Storage stability-Course of biotransformation

(Hucik et al. 2010)

Geotrichumcandidum

Entrapment/ polyacrylamide,calcium alginate, agar,κ-carrageenan, polyvinylalcohol, chitosan

ε-Caprolactone -Variations of polyacrylamideand biomass content,immobilization volume

(Carballeira et al. 2004)

Appl Microbiol Biotechnol (2016) 100:6585–6599 6593

and LentiKats® with immobilized E. coli with CPMOComa

(Table 2) were directly compared (Schenkmayerova et al.2014). Despite differences in inner structure, surface proper-ties and effective diffusion coefficients, both immobilizationmatrices rendered high operational stability during 18 cyclesof BV biooxidations of (±)-cis-bicyclo [3.2.0] hept-2-en-6-one to corresponding regioisomeric lactones (Table 2).

The efficient implementation of immobilization biocatalystsin industry requires their proper characterization. Besides char-acterization parameters described in Table 2, six types of char-acterization areas have been identified and proposed for stan-dard description of immobilization materials including charac-terization of polymers used, their permeability, mechanicalproperties, surface properties, biocompatibility, and storageconditions (de Vos et al. 2009). Although viability is the mainmarker describing the biocompatibility of the complex interac-tions of immobilization matrix and living cells, non-invasivetechniques such as CLSM (Schenkmayerova et al. 2012) andhigh resolution environmental scanning electron microscopy(Schenkmayerová et al. 2014) are also very promising for

characterization of cell viability and morphology ofimmobilized biocatalysts. Figure 9 shows CLSM image ofE. coli with CPMOComa in transmission mode (Fig. 9a), greenflavin autofluorescence within the cells (Fig. 9b) and red ne-crotic cells (Fig. 9c) in superimposed images.

Two-phase system strategies

Utilization of two-phase reaction media with water-immiscible organic solvents belongs to strategies utilized forBVMO-catalyzed oxidations to keep the substrate and productconcentrations under inhibitory levels. The organic phase actsas substrate reservoir and also as extraction solvent for productisolation. For example, inhibition of a PAMO mutant duringwhole-cell BV oxidation was overcome by utilization ofbuffer/dioctyl phthalate in a ratio 1:1 (Reetz et al. 2006).Yang et al. used recombinant E. coliwith cyclopentadecanonemonooxygenase for the improved BV oxidation ofcyclododecanone to lauryl lactone exploiting two-phasesemicontinuous reactor with hexadecane as organic solvent

Scheme 1 Encapsulation of recombinant cells E. coli withoverexpressed Baeyer-Villiger monooxygenases in polyelectrolytecomplex (PEC) capsules with semipermeable membrane and theirutilization as biocatalysts of Baeyer-Villiger oxidations. 1 air pressure, 2air flow, 3 flow calorimeter used for measurement of kinetic constants viathermometric signal, 4 amplifier, 5 computer, 6 continuous process, 7

batch process, PA polyanion aqueous solution composed of sodiumalginate, cellulose sulfate, and NaCl, PC polycation aqueous solutioncomposed of poly(methylene-co-guanidine), CaCl2, and NaCl. Schemewas compiled according to results fromHucik et al. 2010 and Bucko et al.2011

6594 Appl Microbiol Biotechnol (2016) 100:6585–6599

(Yang et al. 2009). Recently, a mathematical model for a threephase partitioning bioreactor for BV biooxidation of modelketone catalyzed by E. coli with CHMO was described(Melgarejo-Torres et al. 2015). Less cell deactivation and de-crease of oxygen transfer rate in the presence of ionic liquidwas observed using developed mathematical model. It wasalso found out that cell inactivation occurred at higher oxygentransfer rate.

Substrate-feeding product-removal technique (SFPR) in-troduced byHilker et al. (Hilker et al. 2004) is another strategyto solve recombinant whole-cell BVMOs sensitivity to highersubstrate and product concentrations. SFPR uses an adsorbingpolymer resin which acts as a reservoir for the substrate andtrap for the product allowing to maintain their concentrationsbelow critical level. A recent application of the technique im-pressively demonstrated the capacity of this approach by analmost 30-fold increase of applicable 3-phenyl-2-butanoneconcentration for the kinetic resolution catalyzed by recombi-nant whole-cell 4-hydroxyacetophenone monooxygenase(Geitner et al. 2010). Though such two-phase strategies mayeffectively solve the problems associated with the inhibitoryeffect of substrates and products of BVoxidations, appropriateconcentration of oxygen as the cosubstrate of reaction must betaken into account in bioreactor configuration.

Oxygen issues

An adequate oxygen transfer during BVMO-catalyzed BVoxidations is a difficult task mainly due to low solubility ofoxygen in aqueous solutions at atmospheric pressure. Sincethe oxygen is a substrate for BVoxidations, limitation in ox-ygen transfer rate may have significant influence on reactionrate. The latter was proved by Baldwin et al., where the in-crease in product formation rate was observed as a conse-quence of increased maximum oxygen transfer rate at a highcell concentration using recombinant E. coli overexpressingCHMO (Baldwin and Woodley 2006). Oxygen transfer ratewas strongly influenced by geometry of reaction flasks, wherethe baffled flasks significantly improved oxygen transfer rate

of a model BV biooxidation catalyzed by recombinant E. coliwith CHMO, immobilized in LentiKats® (Rebros et al. 2014).Another issue connected with oxygen transfer is the deleteri-ous effect of high oxygen concentrations on the CHMO titer.Therefore, the oxygen transfer rate should be adjusted withregard to maximum CHMO synthesis but minimum oxidativedamage of the enzyme (Doig et al. 2001). Regardless of this,the oxygen inactivation constant for whole-cell CHMO wasdetermined only recently, demonstrating that CHMO was twotimes more sensitive to inhibition by bicyclo[3.2.0]hept-2-en-6one than oxygen (Melgarejo-Torres et al. 2014).

Conclusion

Increased efforts in the field of Baeyer-Vil l igermonooxygenases during recent years are aimed at biotechno-logical production of value added chemicals. Discovery ofnovel catalytic abilities of known BVMOs, their variationsby mutations, characterization of missing BVMO in metabo-lism and discovery of new classes of BVMOs with novelfeatures regarding cofactor indifference are milestones whichunderscore the important role of BVMOs in nature and theirbiotechnological potential. The latter was also strengthened byutilization of BVMOs in artificial cascade reactions whichexploited the pulling effect of irreversible BVMO transforma-tion resulting in increased product accumulation. On the otherside, biotechnological potential of BVMOs for production ofvalue added chemicals and chiral building blocks have notbeen fully exploited in industrial processes, yet. Another ap-pealing approaches include utilization of BVMOs for synthe-sis of bulk chemicals and employment of chemo-enzymaticsynthesis. These approaches may significantly reduce thenumber of synthetic steps to minimum and decrease negativeenvironmental impact of the whole production process.

Important progress has been achieved also in techniqueswhich may overcome the main obstacles of BVMO-catalyzed oxidations. Immobilization of isolated and whole-cell BVMOs improved their operational and storage stability

Fig. 9 Example of CLSM image of E. coli with CPMOComa intransmission mode (a) compared to green flavin autofluorescencewithin the same cells (b). Cell necrosis was monitored by propidium

iodide fluorescence imaged in the red channel (c). This figure originallypublished in Schenkmayerova et al. (2012) Biotechnol Lett 34:309 wasreused with permission of Springer

Appl Microbiol Biotechnol (2016) 100:6585–6599 6595

and rendered the tool for continuous and repeated use ofBVMOs with increased harvested product per cell mass.Utilization of recombinant whole-cell BVMOs circumventscofactor regeneration and protein instability problems.Therefore, the use of immobilized whole-cell BVMOs shouldbe favorable as compared to immobilized isolated BVMOs.Moreover, utilization of high performance devices for produc-tion of immobilized whole-cell BVMOs may significantlyincrease the scalability potential. In this view, standardizationof characterization methods for immobilized BVMOs, ratio-nal use of substrate-feeding product-removal technique, rigor-ous control of oxygen transfer, and scale-up from laboratory toindustrial scale manufacture of immobilizates is desirable forselection and development of a proper BVMO biocatalyst forpotential industrial applications.

Acknowledgments This work was supported by the Slovak GrantAgency for Science VEGA 2/0090/16 and by the Slovak Research andDevelopment Agency under contract no. APVV-15-0227. This publica-tion is the result of the project implementation: applied research in thefield of industrial biocatalysis, ITMS code: 26240220079 supported bythe Research & Development Operational Programme funded by theERDF. Dr. Florian Rudroff and Prof. Marko D. Mihovilovic thank theFWF (grant no. I723-N17, P24483-B20), TU Wien (ABC-TOPAnschubfinanzierung), and the COST action systems biocatalysis WG2for financial support.

Compliance with ethical standards

Conflict of interest The authors declare that they have no competinginterests.

Research involving human participants and/or animals This articledoes not contain any studies with human participants or animals per-formed by any of the authors.

References

Abril O, Ryerson CC, Walsh C, Whitesides GM (1989) EnzymaticBaeyer-Villiger type oxidations of ketones catalyzed by cyclohexa-none oxygenase. Bioorg Chem 17:41–52. doi:10.1016/0045-2068(89)90006-0

Atia KS (2005) Co-immobilization of cyclohexanone monooxygenaseand glucose-6-phosphate dehydrogenase onto polyethylenimine-porous agarose polymeric composite using gamma irradiation touse in biotechnological processes. Radiat Phys Chem 73:91–99.doi:10.1016/j.radphyschem.2004.07.005

Baldwin CVF, Woodley JM (2006) On oxygen limitation in a whole cellbiocatalytic Baeyer-Villiger oxidation process. Biotechnol Bioeng95:362–369. doi:10.1002/bit.20869

Baldwin CVF, Wohlgemuth R, Woodley JM (2008) The first 200-L scaleasymmetric Baeyer-Villiger oxidation using a whole-cell biocata-lyst. Org Process Res Dev 12:660–665. doi:10.1021/op800046t

Balke K, Kadow M, Mallin H, Sass S, Bornscheuer UT (2012)Discovery, application and protein engineering of Baeyer-Villigermonooxygenases for organic synthesis. Org Biomol Chem 10:6249–6265. doi:10.1039/c2ob25704a

Balke K, Schmidt S, Genz M, Bornscheuer UT (2016) Switching theregioselectivity of a cyclohexanone monooxygenase towards (+)-trans-dihydrocarvone by rational design. ChemCatChem 7:3951–3955. doi:10.1021/acschembio.5b00723

Bayer T, Milker S, Wiesinger T, Rudroff F, Mihovilovic MD (2015)Designer microorganisms for optimized redox cascade reactions—challenges and future perspectives. Adv Synth Catal 357:1587–1618. doi:10.1002/adsc.201500202

Bianchi DA, Moran-Ramallal R, Iqbal N, Rudroff F, Mihovilovic MD(2013) Enantiocomplementary access to carba-analogs of C-nucleoside derivatives by recombinant Baeyer-Villigermonooxygenases. Bioorg Med Chem Lett 23:2718–2720.doi:10.1016/j.bmcl.2013.02.085

Bong YK, Clay MD, Collier SJ, Mijts B, Vogel M, Zhang X, Zhu J,Nazor J, Smith D, Song S (2011) Engineered cylohexanonemonooxygenases for synthesis of prazole compounds. USAWO2011071982 A2

Bucko M, Schenkmayerova A, Gemeiner P, Vikartovska A, MihovilovicMD, Lacik I (2011) Continuous testing system for Baeyer-Villigerbiooxidation using recombinant Escherichia coli expressing cyclo-hexanone monooxygenase encapsulated in polyelectrolyte complexcapsules. Enzyme Microb Technol 49:284–288. doi:10.1016/j.enzmictec.2011.05.013

Burdette D, Zeikus JG (1994) Purification of acetaldehyde dehydroge-nase and alcohol dehydrogenases from Thermoanaerobacterethanolicus 39e and characterization of the secondary-alcohol de-hydrogenase (2-degrees Adh) as a bifunctional alcohol-dehydrogenase acetyl-CoA reductive thioesterase. Biochem J302:163–170

Carballeira JD, Álvarez E, Sinisterra JV (2004) Biotransformation ofcyclohexanone using immobilized Geotrichum candidumNCYC49. J Mol Catal B Enzym 28:25–32. doi:10.1016/j.molcatb.2004.01.009

Carballeira JD, Quezada MA, Hoyos P, Simeo Y, Hernaiz MJ, AlcantaraAR, Sinisterra JV (2009) Microbial cells as catalysts forstereoselective red-ox reactions. Biotechnol Adv 27:686–714.doi:10.1016/j.biotechadv.2009.05.001

Ceccoli RD, Bianchi DA, Rial DV (2014) Flavoprotein monooxygenasesfor oxidative biocatalysis: recombinant expression in microbialhosts and applications. Front Microbiol 5. doi:10.3389/Fmicb.2014.00025

Cuetos A, Rioz-Martinez A, Valenzuela ML, Lavandera I, de Gonzalo G,Carriedo GA, Gotor V (2012) Immobilized redox enzymatic cata-lysts : Baeyer-Vil l iger monooxygenases supported onpolyphosphazenes. J Mol Catal B Enzym 74:178–183.doi:10.1016/j.molcatb.2011.10.002

de Gonzalo G, Mihovilovic MD, Fraaije MW (2010) Recent develop-ments in the application of Baeyer-Villiger monooxygenases asbiocatalysts. ChemBioChem 11:2208–2231. doi:10.1002/cbic.201000395

de Vos P, BuckoM, Gemeiner P, Navratil M, Svitel J, FaasM, Strand BL,Skjak-Braek G, Morch YA, Vikartovska A, Lacik I, Kollarikova G,Orive G, Poncelet D, Pedraz JL, Ansorge-Schumacher MB (2009)Multiscale requirements for bioencapsulation in medicine and bio-technology. Biomaterials 30:2559–2570. doi:10.1016/j.biomaterials.2009.01.014

Doig SD, O’Sullivan LM, Patel S, Ward JM, Woodley JM (2001) Largescale production of cyclohexanone monooxygenase fromEscherichia coli TOP10 pQR239. Enzyme Microb Technol 28:265–274. doi:10.1016/S0141-0229(00)00320-3

Doig SD, Avenell PJ, Bird PA, Gallati P, Lander KS, Lye GJ,Wohlgemuth R, Woodley JM (2002) Reactor operation and scale-up of whole cell Baeyer-Villiger catalyzed lactone synthesis.Biotechnol Prog 18:1039–1046. doi:10.1021/bp0200954

Doig SD, Simpson H, Alphand V, Furstoss R, Woodley JM (2003)Characterization of a recombinant Escherichia coli TOP10

6596 Appl Microbiol Biotechnol (2016) 100:6585–6599

[pQR239] whole-cell biocatalyst for stereoselective Baeyer-Villigeroxidations. Enzyme Microb Technol 32:347–355. doi:10.1016/S0141-0229(02)00317-4

Donoghue NA, Norris DB, Trudgill PW (1976) Purification and proper-ties of cyclohexanone oxygenase from Nocardia-globerula Cl1 andAcinetobacterNCIB-9871. Eur J Biochem 63:175–192. doi:10.1111/j.1432-1033.1976.tb10220.x

Fink MJ, Mihovilovic MD (2015) Non-hazardous Baeyer-Villiger oxida-tion of levulinic acid derivatives: alternative renewable access to 3-hydroxypropionates. Chem Commun 51:2874–2877. doi:10.1039/c4cc08734h

FinkMJ, Fischer TC, Rudroff F, Dudek H, Fraaije MW,MihovilovicMD(2011) Extensive substrate profiling of cyclopentadecanonemonooxygenase as Baeyer-Villiger biocatalyst reveals novelregiodivergent oxidations. J Mol Catal B Enzym 73:9–16.doi:10.1016/j.molcatb.2011.07.003

FinkMJ, Rial DV, Kapitanova P, Lengar A, Rehdorf J, Cheng Q, RudroffF, Mihovilovic MD (2012) Quantitative comparison of chiral cata-lysts selectivity and performance: a generic concept illustrated withcyclododecanone monooxygenase as Baeyer-Villiger biocatalyst.Adv Synth Catal 354:3491–3500. doi:10.1002/adsc.201200453

Fink MJ, Schoen M, Rudroff F, Schnuerch M, Mihovilovic MD (2013)Single operation stereoselective synthesis of Aerangis lactones:combining continuous flow hydrogenation and biocatalysts in achemoenzymatic sequence. ChemCatChem 5:724–727.doi:10.1002/cctc.201200753

Geitner K, Rehdorf J, Snajdrova R, Bornscheuer UT (2010) Scale-up ofBaeyer-Villiger monooxygenase-catalyzed synthesis of enantiopurecompounds. Appl Microbiol Biotechnol 88:1087–1093.doi:10.1007/s00253-010-2724-y

Hilker I, Alphand V, Wohlgemuth R, Furstoss R (2004) Microbial trans-formations, 56. Preparative scale asymmetric Baeyer-Villiger oxida-tion using a highly productive “two-in-one” resin-based in situSFPR concept. Adv Synth Catal 346:203–214. doi:10.1002/adsc.200303183

Hollrigl V, Hollmann F, Kleeb AC, Buehler K, Schmid A (2008) TADH,the thermostable alcohol dehydrogenase from Thermus sp ATN1: aversatile new biocatalyst for organic synthesis. Appl MicrobiolBiotechnol 81:263–273. doi:10.1007/s00253-008-1606-z

Hucik M, Bucko M, Gemeiner P, Stefuca V, Vikartovska A, MihovilovicMD, Rudroff F, Iqbal N, Chorvat D, Lacik I (2010) Encapsulation ofrecombinant E. coli expressing cyclopentanone monooxygenase inpolyelectrolyte complex capsules for Baeyer-Villiger biooxidationof 8-oxabicyclo[3.2.1]oct-6-en-3-one. Biotechnol Lett 32:675–680.doi:10.1007/s10529-010-0203-2

Isupov MN, Schröder E, Gibson RP, Beecher J, Donadio G, Saneei V,Dcunha SA, McGhie EJ, Sayer C, Davenport CF, Lau PC,Hasegawa Y, Iwaki H, Kadow M, Balke K, Bornscheuer UT,Bourenkov G, Jennifer A, Littlechild JA (2015) The oxygenatingconstituent of 3,6-diketocamphane monooxygenase from the CAMplasmid of Pseudomonas putida. Acta Cystal Sect D Biol Cystal 71:2344–2353. doi:10.1107/S1399004715017939

Iwaki H, Hasegawa Y, Wang SZ, Kayser MM, Lau PCK (2002) Cloningand characterization of a gene cluster involved in cyclopentanolmetabolism in sp strain NCIMB 9872 and biotransformationseffected by Escherichia coli-expressed cyclopentanone 1,2-monooxygenase. Appl Environ Microbiol 68:5671–5684.doi:10.1128/Aem.68.11.5671-5684.2002

Jang HY, Jeon EY, Baek AH, Lee SM, Park JB (2014) Production ofomega-hydroxyundec-9-enoic acid and n-heptanoic acid fromricinoleic acid by recombinant Escherichia coli-based biocatalyst.Process Biochem 49:617–622. doi:10.1016/j.procbio.2014.01.025

Kadow M, Loschinski K, Saß S, Schmidt M, Bornscheuer UT (2012)Completing the series of BMVOs involved in camphor metabolismof Pseudomonas putida NCIMB 10007 by identification of the twomissing genes, their functional, expression in E. coli and

biochemical characterization. Appl Microbiol Biotechnol 96:419–429. doi:10.1007/s00253-011-3859-1

Kadow M, Balke K, Willetts A, Bonscheuer UT, Bäckwall JE (2014)Functional assembly of camphor converting two-componentBaeyer-Villiger monooxygenase with a flavin reductase fromE.coli. Appl Microbiol Biotechnol 98:3975–3986. doi:10.1007/s00253-013-5338-3

Kayser MM (2009) ‘Designer reagents’ recombinant microorganisms:new and powerful tools for organic synthesis. Tetrahedron 65:947–974. doi:10.1016/j.tet.2008.10.039

Kim SU, Kim KR, Kim JW, Kim S, Kwon YU, Oh DK, Park JB (2015)Microbial synthesis of plant oxylipins from gamma-linolenic acidthrough designed biotransformation pathways. J Agric Food Chem63:2773–2781. doi:10.1021/jf5058843

Kirschner A, Altenbuchner J, Bornscheuer UT (2007) Cloning, expres-sion, and characterization of a Baeyer-Villiger monooxygenase fromPseudomonas fluorescens DSM 50106 in E. coli. Appl MicrobiolBiotechnol 73:1065–1072. doi:10.1007/s00253-006-0556-6

Kohler V, Turner NJ (2015) Artificial concurrent catalytic processes in-volving enzymes. Chem Commun 51:450–464. doi:10.1039/c4cc07277d

Law HEM, Baldwin CVF, Chen BH, Woodley JM (2006) Process limi-tations in a whole-cell catalysed oxidation: sensitivity analysis.Chem Eng Sci 61:6646–6652. doi:10.1016/j.ces.2006.06.007

Leisch H, Morley K, Lau PCK (2011) Baeyer-Villiger monooxygenases:more than just green chemistry. Chem Rev 111:4165–4222.doi:10.1021/cr1003437

Lima-Ramos J, Neto W, Woodley JM (2014) Engineering of biocatalystsand biocatalytic processes. Top Catal 57:301–320. doi:10.1007/s11244-013-0185-0

Liu J, Li Z (2013) Cascade biotransformations via enantioselective reduc-tion, oxidation, and hydrolysis: preparation of (R)-delta-lactonesfrom 2-Alkylidenecyclopentanones. ACS Catal 3:908–911.doi:10.1021/Cs400101v

Mallin H, Wulf H, Bornscheuer UT (2013) A self-sufficient Baeyer-Villiger biocatalysis system for the synthesis of epsilon-caprolactone from cyclohexanol. Enzyme Microb Technol 53:283–287. doi:10.1016/j.enzmictec.2013.01.007

Meeuwissen SA, Rioz-Martinez A, de Gonzalo G, Fraaije MW, Gotor V,van Hest JCM (2011) Cofactor regeneration in polymersomenanoreactors: enzymatically catalysed Baeyer-Villiger reactions. JMater Chem 21:18923–18926. doi:10.1039/c1jm12407b

Melgarejo-Torres R, Castillo-Araiza CO, López-Ordaz P, Torres-Martínez D, Gutierrez-Rojas M, Lye GJ, Huerta-Ochoa S (2014)Kinetic mathematical model for ketone bioconversion usingEscherichia coli TOP10 pQR239. Chem Eng J 240:1–9.doi:10.1016/j.cej.2013.11.047

Melgarejo-Torres R, Castillo-Araiza CO, Dutta A, Beny G, Torres-Martinez D, Gutierrez-Rojas M, Lye GJ, Huerta-Ochoa S (2015)Mathematical model of a three phase partitioning bioreactor forconversion of ketones using whole cells. Chem Eng J 260:765–775. doi:10.1016/j.cej.2014.08.097

Mihovilovic MD (2006) Enzyme mediated Baeyer-Villiger oxidations.Curr Org Chem 10:1265–1287. doi:10.2174/138527206777698002

Mihovilovic MD, Rudroff F, Grotzl B (2004) Enantioselective Baeyer-Villiger oxidations. Curr Org Chem 8:1057–1069. doi:10.2174/1385272043370159

Mihovilovic MD, Rudroff F, Grötzl B, Kapitan P, Snajdrova R, Rydz J,Mach R (2005) Family clustering of Baeyer–Vill igermonooxygenases based on protein sequence and stereopreference.Angew Chem Int Ed 44:3609–3613. doi:10.1002/anie.200462964

Mihovilovic MD, Kapitan P, Kapitanova P (2008) RegiodivergentBaeyer-Villiger oxidation of fused ketones by recombinant whole-cell biocatalysts. ChemSusChem 1:143–148. doi:10.1002/cssc.200700069

Appl Microbiol Biotechnol (2016) 100:6585–6599 6597

Muschiol J, Peters C, Oberleitner N, Mihovilovic MD, Bornscheuer UT,Rudroff F (2015) Cascade catalysis—strategies and challenges enroute to preparative synthetic biology. Chem Commun 51:5798–5811. doi:10.1039/c4cc08752f

Oberleitner N, Peters C,Muschiol J, KadowM, Sass S, Bayer T, Schaaf P,Iqbal N, Rudroff F, Mihovilovic MD, Bornscheuer UT (2013) Anenzymatic toolbox for cascade reactions: a showcase for an in vivoredox sequence in asymmetric synthesis. ChemCatChem 5:3524–3528. doi:10.1002/cctc.201300604

Oberleitner N, Peters C, Rudroff F, Bornscheuer UT, Mihovilovic MD(2014) In vitro characterization of an enzymatic redox cascade com-posed of an alcohol dehydrogenase, an enoate reductases and aBaeyer-Villiger monooxygenase. J Biotechnol 192:393–399.doi:10.1016/jpiotec.20111.0/1.008

Oberleitner N, Ressmann AK, Bica K, Gärtner P, Fraaije MW,Bornscheuer UT, Rudroff F, Mihovilovic MD (2016) From wasteto value—direct utilization of limonene from orange peel in biocat-alytic cascade reaction towards chiral carvolactone. Green Chem,accepted

Oh HJ, Kim SU, Song JW, Lee JH, Kang WR, Jo YS, Kim KR,Bornscheuer UT, Oh DK, Park JB (2015) Biotransformation oflinoleic acid into hydroxy fatty acids and carboxylic acids using alinoleate double bond hydratase as key enzyme. Adv Synth Catal357:408–416. doi:10.1002/adsc.201400893

Opperman DJ, Reetz MT (2010) Towards practical Baeyer-Villiger-monooxygenases: design of cyclohexanone monooxygenase mu-tants with enhanced oxidative stability. ChemBioChem 11:2589–2596. doi:10.1002/cbic.201000464

Pazmino DET, Dudek HM, Fraaije MW (2010a) Baeyer-Villigermonooxygenases: recent advances and future challenges. CurrOpin Chem Biol 14:138–144. doi:10.1016/j.cbpa.2009.11.017

Pazmino DET, Winkler M, Glieder A, Fraaije MW (2010b)Monooxygenases as biocatalysts: classification, mechanistic aspectsand biotechnological applications. J Biotechnol 146:9–24.doi:10.1016/j.jbiotec.2010.01.021

Pennec A, Jacobs CL, Opperman DJ, Smit MS (2015) Revisiting cyto-chrome P450-mediated oxyfunctionalization of linear and cyclicalkanes. Adv Synth Catal 357:118–130. doi :10.1002/adsc.201400410

Prusse U, Bilancetti L, Bucko M, Bugarski B, Bukowski J, Gemeiner P,Lewinska D, Manojlovic V, Massart B, Nastruzzi C, Nedovic V,Poncelet D, Siebenhaar S, Tobler L, Tosi A, Vikartovska A,Vorlop KD (2008) Comparison of different technologies for alginatebeads production. Chem Pap 62:364–374. doi:10.2478/s11696-008-0035-x

Rebros M, Liptak L, Rosenberg M, Bucko M, Gemeiner P (2014)Biocatalysis with Escherichia coli-overexpressing cyclopentanonemonooxygenase immobilized in polyvinyl alcohol gel. Lett ApplMicrobiol 58:556–563. doi:10.1111/lam.12227

Reetz MT, Wang LW, Bocola M (2006) Directed evolution ofenantioselective enzymes: iterative cycles of CASTing for probingprotein-sequence space. Angew Chem Int Ed 45:1236–1241.doi:10.1002/anie.200502746

Rehdorf J, Kirschner A, Bornscheuer UT (2007) Cloning, expression andcharacterization of a Baeyer-Villiger monooxygenase fromPseudomonas putida KT2440. Biotechnol Lett 29:1393–1398.doi:10.1007/s10529-007-9401-y

Rehdorf J, Mihovilovic MD, Fraaije MW, Bornscheuer UT (2010)Enzymatic synthesis of enantiomerically pure beta-amino ketones,beta-amino esters, and beta-amino alcohols with Baeyer-Villigermonooxygenases. Chem Eur J 16:9525–9535. doi:10.1002/chem.201001480

Reignier T, de Berardinis V, Petit JL, Mariage A, Hamze K, Duquesne K,Alphand V (2014) Broadening the scope of Baeyer-Villigermonooxygenase activities toward alpha,beta-unsaturated ketones: a

promising route to chiral enol-lactones and ene-lactones. ChemCommun 50:7793–7796. doi:10.1039/c4cc02541e

Riebel A, de Gonzalo G, Fraaije MW (2013) Expanding the biocatalytictoolbox of flavoprotein monooxygenases from Rhodococcus jostiiRHA1. J Mol Catal B Enzym 88:20–25. doi:10.1016/j.molcatb.2012.11.009

Rudroff F, Bianchi DA, Moran-Ramallal R, Iqbal N, Dreier D,Mihovilovic MD (2015) Synthesis of tetrahydrofuran-based naturalproducts and their carba analogs via stereoselective enzyme mediat-ed Baeyer-Villiger oxidation. Tetrahedron. doi:10.1016/j.tet.2015.11.048

Sattler JH, Fuchs M, Mutti FG, Grischek B, Engel P, Pfeffer J, WoodleyJM, Kroutil W (2014) Introducing an in situ capping strategy insystems biocatalysis to access 6-aminohexanoic acid. AngewChem Int Ed 53:14153–14157. doi:10.1002/anie.201409227

Schenkmayerova A, Bucko M, Gemeiner P, Chorvat D, Lacik I (2012)Viability of free and encapsulated Escherichia coli overexpressingcyclopentanone monooxygenase monitored during model Baeyer-Villiger biooxidation by confocal laser scanning microscopy.Biotechnol Lett 34:309–314. doi:10.1007/s10529-011-0765-7

Schenkmayerova A, Bucko M, Gemeiner P, Katrlik J (2013) Microbialmonooxygenase amperometric biosensor for monitoring of Baeyer-Villiger biotransformation. Biosens Bioelectron 50:235–238.doi:10.1016/j.bios.2013.06.061

Schenkmayerova A, Bucko M, Gemeiner P, Trel’ova D, Lacik I, ChorvatD, Acai P, Polakovic M, Liptak L, RebrosM, Rosenberg M, StefucaV, Nedela V, Tihlarikova E (2014) Physical and bioengineeringproperties of polyvinyl alcohol lens-shaped particles versus spheri-cal polyelectrolyte complex microcapsules as immobilisation matri-ces for a whole-cell Baeyer-Villiger monooxygenase. Appl BiochemBiotechnol 174:1834–1849. doi:10.1007/s12010-014-1174-x

Schmidt S, Bűchsenschűtz HC, Scherkus C, Liese A, Gröger H,Bornscheuer UT (2015a) Biocatalytic access chiral polyesters byan artificial enzyme cascade syntheses. ChemCatChem 7:3951–3955. doi:10.1002/cctc.201500823

Schmidt S, Scherkus C, Muschiol J, Menyes U, Winkler T, Hummel W,Groger H, Liese A, Herz HG, Bornscheuer UT (2015b) An enzymecascade synthesis of epsilon-caprolactone and its oligomers. AngewChem Int Ed 54:2784–2787. doi:10.1002/anie.201410633

Snajdrova R, Grogan G, Mihovilovic MD (2006) Resolution of fusedbicyclic ketones by a recombinant biocatalyst expressing theBaeyer-Vi l l iger monooxygenase gene Rv3049c fromMycobacterium tuberculosis H37Rv. Bioorg Med Chem Lett 16:4813–4817. doi:10.1016/j.bmcl.2006.06.072

Song JW, Jeon EY, SongDH, JangHY, Bornscheuer UT, OhDK, Park JB(2013) Multistep enzymatic synthesis of long-chain alpha,omega-dicarboxylic and omega-hydroxycarboxylic acids from renewablefatty acids and plant oils. Angew Chem Int Ed 52:2534–2537.doi:10.1002/anie.201209187

Staudt S, Bornscheuer UT, Menyes U, Hummel W, Groger H (2013)Direct biocatalytic one-pot-transformation of cyclohexanol withmolecular oxygen into epsilon-caprolactone. Enzyme MicrobTechnol 53:288–292. doi:10.1016/j.enzmictec.2013.03.011

van Beek HL, Winter RT, Eastham GR, Fraaije MW (2014) Synthesis ofmethyl propanoate by Baeyer-Villiger monooxygenases. ChemCommun 50:13034–13036. doi:10.1039/c4cc06489e

van Beilen JB, Mourlane F, Seeger MA, Kovac J, Li Z, Smits THM,Fritsche U, Witholt B (2003) Cloning of Baeyer-Villigermonooxygenases from Comamonas , Xanthobacter andRhodococcus using polymerase chain reaction with highly degener-ate primers. Environ Microbiol 5:174–182. doi:10.1046/j.1462-2920.2003.00401.x

van Berkel WJH, Kamerbeek NM, Fraaije MW (2006) Flavoproteinmonooxygenases, a diverse class of oxidative biocatalysts. JBiotechnol 124:670–689. doi:10.1016/j.jbiotec.2006.03.044

6598 Appl Microbiol Biotechnol (2016) 100:6585–6599

Yang JZ, Wang SZ, Lorrain MJ, Rho D, Abokitse K, Lau PCK (2009)Bioproduction of lauryl lactone and 4-vinyl guaiacol as value-addedchemicals in two-phase biotransformation systems. Appl MicrobiolBiotechnol 84:867–876. doi:10.1007/s00253-009-2026-4

Zambianchi F, Pasta P, Carrea G, Colonna S, Gaggero N, Woodley JM(2002) Use of isolated cyclohexanone monooxygenase from recom-binant Escherichia coli as a biocatalyst for Baeyer-Villiger and sul-fide oxidations. Biotechnol Bioeng 78:489–496. doi:10.1002/Bit.10207

Zhang Z-G, Roiban G-D, Acevedo JP, Polyak I, Reetz MT (2013) A newtype of stereoselectivity in Baeyer-Villiger reactions: access to E-and Z-olefins. Adv Synth Catal 355:99–106. doi:10.1002/adsc.201200759

Zuhse R, Leggewie C, Hollmann F, Kara S (2015) Scaling-up of “smartcosubstrate” 1,4-butanediol promoted asymmetric reduction of eth-yl-4,4,4-trifluoroacetoacetate in organic media. Org Process ResDev 19:369–372. doi:10.1021/op500374x

Appl Microbiol Biotechnol (2016) 100:6585–6599 6599