Mengjiao Xu,Zhuotao Tan,Chenjie Zhu*,Wei Zhuang,Hanjie Ying*,Pingkai Ouyang
College of Biotechnology and Pharmaceutical Engineering,Nanjing Tech University,Nanjing 211816,China
ABSTRACT Chemoenzymatic catalysis can give full play to the advantages of versatile reactivity of chemocatalysis and excellent chemo-,regio-,and stereoselectivities of biocatalysis.These chemoenzymatic methods can not only save resource,cost,and operating time but also reduce the number of reaction steps,and avoid separating unstable intermediates,leading to the generation of more products under greener circumstances and thereby playing an indispensable role in the fields of medicine,materials and fine chemicals.Although incompatible challenges between chemocatalyst and biocatalyst remain,strategies such as biphasic system,artificial metalloenzymes,immobilization or supramolecular host,and protein engineering have been designed to overcome these issues.In this review,chemoenzymatic catalysis according to different chemocatalysis types was classifiably described,and in particular,the classic dynamic kinetic resolutions(DKR)and cofactor regeneration were summarized.Finally,the bottlenecks and development of chemoenzymatic catalysis were summarized,and future development was prospected.
Keywords:Chemoenzymatic Dynamic kinetic resolution Cofactor regeneration Biocatalysis Chemocatalysis
Chemocatalysis has been an essential tool for human lives,and plays an important role in pharmaceutical,food,material,energy,etc.,owing to its versatile reactivity [1–3].However,high energy consumption,complex workup and serious pollutants often ensue,which makes it difficult to promote economic benefits in industry.Biocatalysis offers high selectivity,mild reaction conditions,and environmental friendliness[4,5],but biocatalysts are always sensitive to harsh reaction conditions or reagents.Recently,it has become a new trend to combine chemocatalyst with enzyme to achieve mutual promotion and synergistic catalysis.For example,in the1980s,the Bekkum group first combined Pd/C with glucose isomerase to catalyze hydrogenation and isomerization of Dglucose [6].The Williams group integrated Pd withPseudomonas fluorescenslipase (PFL) to catalyze the DKR of secondary alcohols and proposed the concept of enzyme and metal cascade catalysis[7,8].This chemoenzymatic DKR approach overcomes the maximum yield restriction (50%) of traditional kinetic resolution (KR)methods.Cofactor regeneration,as a classic chemoenzymatic method,was established to solve the problems of high cost and inhibition of natural cofactors during the bioprocess.Thus,combining biocatalysis with chemocatalysis in a reasonable manner represents an opportunity to break through certain limits and form an efficient and green route to sustainable development in terms of environment,energy and other aspects [9].The Kroutil [10],Zhu[11],and Zhao [12]groups have described the recent progress of chemoenzymatic catalysis in terms of catalyst type,combination mode of biocatalysis and chemocatalysis,etc.,and showed that such a combination could not only save resources,cost,and operating time but also reduce the number of reaction steps and avoid separating the unstable intermediates,thereby obtaining more products under greener circumstances.The vital functions of chemoenzymatic catalysis in the synthesis of medicine,chemical,material and others were also highlighted by these groups.
Despite these advantages,mutual inactivation between chemocatalysis and biocatalysis has been challenging,because biocatalysts other than lipases or serine proteases cannot maintain their stable catalytic activity in a nonaqueous solvent and at high temperature,and chemocatalyst such as metal complexes can be inhibited in aqueous medium or lose their activity due to the coordination with enzymes.Several approaches have been developed to solve these problems,such as biphasic systems,artificial metalloenzymes,immobilization or supramolecular host,and protein engineering [13].
In this review,biocatalysis combined with four types of chemocatalysis,namely metal catalysis,small organic molecule catalysis,photocatalysis,and electrocatalysis,was described in different sections.As a representative chemoenzymatic method,DKR and cofactor regeneration were reviewed in detail;furthermore,the main challenges and the future development of chemoenzymatic approaches were expounded.
Metals or their complexes are universal and versatile chemocatalysts and have been successfully applied to many types of chemoenzymatic reactions.In the following section,the chemoenzymatic process catalyzed by metals or their complexes combined with enzymes,including DKR,cofactor regeneration,and other typical reactions will be discussed.
Chiral alcohols and amines are important building blocks of various organic compounds [14,15].Many stereoselective routes to access these valuable molecules have been developed including the construction of carbon-heteroatom bonds [16],hydrogenation of ketones or imines [17],nucleophilic addition of carbonyl compounds [18],and KR [19,20].Due to its excellent properties,enzymatic KR is a common method for optically pure alcohols and amines in industry[21].However,the major limitation of KR is that the maximum yield is theoretically 50%,while DKR consisting of racemization and continuous asymmetric transformation can break through this restriction,and allow the yield to reach 100%,thereby remarkably increasing the efficiency of enantiomeric resolution (Fig.1) [22–25].
Noble metals consist of Pd,Ru,Rh,Ir,etc.,and are often applied for DKR.In addition to the abovementioned Pd/lipase system,the Williams group first combined a Rh complex with PFL to catalyze the DKR of racemic phenylethanol [12,13].Although (R)-1-phenylethanol was obtained with 60% conversion and 98%ee,phenanthroline and acetophenone had to be added for efficient racemization (Fig.2).The Ostaszewski group also reported DKR catalyzed by Rh2(OAc)4andCandida antarcticalipase B (CALB).Novozym-435 was also assessed as a chiral resolution biocatalyst,and optically pure ester was obtained from racemic carboxylic acid with high yield andee[26].
Fig.1.Selected example of dynamic kinetic resolution using racemic alcohol or amine as substrates.
Ru,another classic metal catalyst,was also developed for chemoenzymatic DKR.Shov catalyst {[Ph4(η5-C4CO)]2H]}Ru2(-CO)4(μ-H)1and Novozym-435 were coimmobilized in acrylic resin to catalyze the DKR of aliphatic or aromatic alcohols,exhibiting high yields andeevalues(Fig.3) [27–31].However,one drawback of this system was that Shov catalyst1needs thermal activation at 70 °C to effectively split into two single species that function as racemic catalysts (Fig.3,1a and 1b) [32].Therefore,only thermostable lipase could be combined with Shov catalyst1.Another drawback was that the acyl donor of lipase was confined to aryl esters since side ketone products were produced during the reaction when simple alkenyl acetates were added.Later,the Park group combined (η5-indenyl)RuCl(PPh3)22with lipase to catalyze the DKR of racemic alcohols at room temperature[33].Few ketone side products were generated,but this system suffered from severe problems,as Ru complex2needed KOH for activation,which could lead to hydrolysis of the product and enzyme deactivation.Although it was found that KOH could be replaced by O2and Et3N,a higher temperature (60 °C) was essential for the efficient racemization of Ru catalyst2.Subsequently,Pseudomonas cepacialipase was combined with this metal complex for the DKR of some simple secondary alcohols(Fig.4)[34].From the above discussion,we found that enzymes combined with metals can catalyze various types of reactions,providing access to several kinds of high-value products.Actually,different metal complexes could also result in diverse chemoenzymatic reaction activities.For instance,the Park group investigated the influence of Ru complexes with different ligands on the racemization of secondary alcohols,finding that the activity of Ru with η5-cyclopentadienyl ligand was much lower than that with ninhydrin ligand;however,theeevalue with η5-cyclopentadienyl ligand was higher than that with ninhydrin ligand [33].The Somanathan group constructed a series of aromatic monosulfonamide ligands with different electronwithdrawing and electron-donating groups and combined them with Ru metal.Then,this group explored the effect of these Ru complexes with different ligands on the reduction reaction of acetophenone.The result showed that the yield of Ru complexes with electron-withdrawing ligands was higher than that with electrondonating ligands.While the enantioselectivity of the benzene ligand-Ru complexes was lower than that with cymene ligand-Ru complex,the yield was higher in contrast.This result may be due to the large steric hindrance of the isopropyl group of cymene[35].The Martín-Matute group reported the DKR of α-hydroxy ketones [36].[Ru(p-cymene)Cl2]2andPseudomonas stutzerilipase(lipase TL) were combined for enantioselective synthesis of hydroxy ketone esters,and the reaction could smoothly proceed at room temperature with a yield of 99% andeeof 99% (Fig.5).
Another precious metal Ir was also used for DKR.In 2008,the Janssen group designed a DKR process for the synthesis of optically pure epoxy compounds using an Ir complex as the racemic catalyst[37],and a halohydrin dehalogenase(HheC)mutant was combined to catalyze epoxidation of β-chlorohydrine.(R)-Products were obtained with high yields andeevalues(Fig.6a).The Kroutil group also reported the DKR of halohydrin,in which ADH and Ir complex4were combined (Fig.6b) [38].
Due to the scarcity of precious metal complexes and their high cost,researchers also designed DKR processes using nonprecious metal complexes.The Berkessel group reported the combination of AlMe3/2,2′-biphenol with Novozym-435 to catalyze the DKR of aliphatic or benzyl alcohols(Fig.7).Using cyclohexane as the substrate,98% yield andee>99% were obtained in 9 hours at 60 °C[39].The Moberg group prepared a Ti complex containing a lewis acid and base for cyanation of α-ketonitrile and aldehyde;then,CALB was combined to synthesize chiral cyanohydrin (Fig.8)[40].The V complex [OV(OSiPh3)3)]was also combined with CALB to catalyze the resolution of racemic allyl alcohol through 1,3-transfer of the hydroxyl group(Fig.9)[41].Subsequently,heterogeneous vanadium complexes containing oxygen were prepared using polymer and mesoporous silica gel as carriers.This catalyst could be recycled 6 times without loss of activity[42].The Jacobs group also reported that the V complex combined with lipase catalyzed chemoenzymatic DKR of other secondary alcohols at 80 °C [43].
Fig.2.DKR of 1-phenylethanol catalyzed by PFL and Rh2(OAc)4.
Fig.3.DKR of 1-phenylethanol catalyzed by Shvo’s dimeric Ru Complex 1 and Novozyme-435.
Fig.5.DKR of α-hydroxy ketones catalyzed by lipase TL and Ru complex.
Fig.6.Chemoenzymatic reactions catalyzed by Ir complex.
Fig.7.DKR of 1-cyclohexylethanol catalyzed by aluminum complex and Novozym 435.
Due to the lack of effective amine racemic catalysts,the DKR of amines has been less effective than that of alcohols.The main reason was that amine substrates can act as strong ligands,which may inhibit or even completely inactivate the metal catalyst.Although high temperature could alleviate this issue,thermostable enzymes should be available.Another issue of amine racemization was that the produced imine intermediate was highly active resulting in the occurrence of some side reactions.Regardless,several satisfactory results have been achieved by researchers.In 1996,the Reetz group first integrated Pd/C with Novozym-435 to catalyze the DKR of 1-phenylethylamine (Fig.10) [44].Despite the favorable yield andee,the reaction took 8 days,and thus improving catalytic efficiency was necessary.The Jacobs group replaced the support charcoal of Pd/C with alkaline earth including BaSO4,CaCO3,BaCO3or SrCO3.Among these catalysts,Pd/BaSO4exhibited the best activity [45,46].The Pedrozo group also demonstrated that the Pd/BaSO4combination was suitable for the DKR of benzyl primary amines,but for aliphatic primary amines,the activity was relatively low [47].
Ru Shov catalyst was also combined with Novozym-435 for the DKR of benzylamine[48].In this system,2,4-dimethyl-3-pentanol was used as hydrogen donor and successfully hindered the generation of byproducts.The Hollmann group combined monoamine oxidase (MAO-N) with the artificial transfer hydrogenase [Cp*Ir(Biot-p-L)Cl]·Sav S112A for the DKR of amine [49].Generally,free iridium complex and MAO-N could mutually inactivate each other,but the researchers ingeniously incorporated the metal complex within streptavidin to form an artificial transfer hydrogenase based on supramolecular self-assembly (Fig.11).Applying with this excellent artificial metal enzyme,various amines were kinetically separated,exhibiting superior activity.
Fig.8.DKR of chiral α-acetoxyphenylacetonitrile catalyzed by titanium complex and CALB.
Fig.9.DKR of racemic allyl alcohol catalyzed by vanadium complexes and CALB.
Fig.10.DKR of racemic 1-phenylethylamine catalyzed by Pd/C and Novozym-435.
Similarly,inexpensive metals were employed for the DKR of amine.In 2008,the Jacobs group reported the DKR of primary amines catalyzed by Raney Ni or Raney Co combined with Novozyme-435[50].However,the conversion andeevalues of these reactions were generally low,even at 70–80°C for 2–5 days.In 2014,the Wu group combined Raney Ni with CALB for the DKR of 1-phenylethylamine,2-heptylamine and 2-octylamine [51].Moderate yields andeevalues of most of the products could be achieved at 70 °C after 48 hours (Fig.12).The Rueping group reported a chemoenzymatic reaction for the synthesis of chiral acetate from prochiral ketones.Lipase was used as the resolution biocatalyst,and an iron complex was used as the hydrogenationracemization catalyst.Finally,the enantiopure chiral acetate was isolated in high yield.In this system,molecular hydrogen and ethyl acetate were applied as hydrogen donor and acyl donor,respectively,which resulted in high atom economy and environmental friendliness (Fig.13) [52].
In addition to DKR,another typical example of chemoenzymatic catalysis was cofactor regeneration.As indispensable molecules in most biocatalysis processes,cofactors can not only stabilize the conformation of enzymes but also transfer protons,energy and functional groups [53].However,the price of natural cofactors is so high that they cannot be added excessively in industry,and in some cases,a large amount of cofactor might inhibit the reactions.Thus,several cofactor regeneration strategies including chemical,electrochemical,photochemical,and enzymatic strategies have been developed[54].As a feasible method of cofactor regeneration,metal catalysis has been investigated.
Fig.11.DKR of racemic 1-methyl-1,2,3,4-tetrahydroisoquinoline catalyzed by artificial transfer hydrogenase and MAO-N.
Fig.12.DKR of racemic 1-phenylethylamine catalyzed by Raney nickel and CALB.
Fig.13.Combined iron with lipase for the transformation of acetophenone to the corresponding chiral acetate.
The classic metal complex applied for cofactor regeneration was pentamethylcyclopentadienyl rhodium bipyridine ([Cp*Rh(bpy)(H2O)]2+).It was first prepared in 1987 by the K?lle group and exhibited excellent proton transfer capability[55].Then,the Steckhan group further reported that it could catalyze the regeneration of nicotinamide cofactor NAD(P)H (Fig.14) [56].This Rh complex could function under a wide range of pH values and temperatures,and catalyze the conversion of NAD(P)+to bioactive 1,4-NAD(P)H,but it had disadvantages in that hydrogen gas was easily formed as a byproduct under the condition of strong acid conditions.The complex could also be inactivated by OH-of the base through nucleophilic attack.The Schmid group reported the detailed redox properties of [Cp*Rh(bpy)(H2O)]2+and demonstrated its potential for cofactor regeneration for a broad range of oxidoreductases[57].The Süss-Fink group designed five other kinds of watersoluble 1,10-phenanthroline complexes of Ru,Rh,and Ir and combined them with ADH to reduce ketones.All of these complexes could catalyze the conversion of NAD+to NADH in aqueous solution.The highest TOF value (2000 h-1) was achieved by[(η5-C5Me5)Rh(phen)Cl]+under optimized conditions [58].
The Jux group constructed a series of water-soluble artificial analogs of iron porphyrin [59].Based on these results,the Gr?ger group simplified these complexes,and used them for the regeneration of NAD(P)+under air conditions [60].The only byproduct of this system was environmental-friendly water.Subsequently,glucose dehydrogenase (GDH) was coupled for the transformation of D-glucose to D-gluconolactone.Glutamate dehydrogenase was also combined to prepare α-ketoglutarate from L-glutamate (Fig.15)[61].
Although there have been some examples of metals coupled with enzymes for cofactor regeneration,in some cases,the enzymes and metals might coordinate,resulting in mutual inactivation.The Hollmann group explained this kind of inactivation between ADH and [Cp*Rh(bpy)(H2O)]2+[62].If such coordination occurs with amino acid residues of the catalytic domain,it will hinder the binding of the substrate to the catalytic site or interfere with the tertiary or quaternary structure of the biocatalyst,thus impairing the activity.Considering the inactivation problem,the Ward group integrated iridium complexes with biotin to construct an artificial metalloenzyme for the regeneration of nicotinamide cofactors which could be compatible with natural enzymes [63].This solution breaks through the limitation of the combination of metal catalyst and biocatalyst,and provides an effective approach and useful reference for improving the compatibility of metal and enzyme.
Fig.14.The regeneration of NAD(P)H catalyzed by [Cp*Rh(bpy)(H2O)]2+.
Flavin,another common cofactor,also plays a crucial role in enzymatic redox reactions.As mentioned above,[Cp*Rh(bpy)(H2O)]2+could not only catalyze the regeneration of nicotinamide cofactor,but also be involved in the regeneration of flavin or porphyrin (Fig.16).The Schmid group reported that [Cp*Rh(bpy)(H2O)]2+regenerated flavin adenine dinucleotide (FAD) for the styrene monooxygenase-catalyzed epoxidation [64].The traditional pathway of two-component monooxygenase needs flavin reductase to regenerate FAD and another artificial NADH regeneration system such as formate dehydrogenase,GDH,and phosphite dehydrogenase.The constructed Rh complex-based system replaced this artificial NADH regeneration system and eliminated the need for long and redundant electron transport chains;thus,the efficiency of electron utilization was improved(Fig.16,b).Subsequently,the Hollmann group reported the use of [Cp*Rh(bpy)(H2O)]2+for the regeneration of flavin adenine mononucleotide(FMN).Coupled with chromate reductase (Crs),α,β-unsaturated ketones were enantioselectively reduced with high efficiency(Fig.16,c) [65].This research group also applied [Cp*Rh(bpy)(H2O)]2+for the regeneration of reduced heme [66,67](Fig.16,d).
In addition to DKR and cofactor regeneration,metal combined with enzyme have been involved in C-C coupling,C-N/O coupling and redox reactions.The Turner group has performed some fascinating work on the Suzuki coupling reaction catalyzed by metal combined with enzyme.For example,D-amino acid dehydrogenase(DAADH)was coupled with palladium to catalyze asymmetric amination and C-C coupling arylation,and L-and D-biarylalanines were effectively synthesized in aqueous medium under mild conditions (Fig.17) [68].A chemoenzymatic method consisting of a Rh complex and cyclohexylamine oxidase(CHAO)for the synthesis of tetrahydroquinoline was also established.The Rh complex catalyzed the C-C coupling of alkyl vinyl ketones and aminophenylboronic acids.CHAO and NH3·BH3were then coupled for the racemization of 3,4-substituted dihydroquinolines (Fig.18) [69].Gold was combined with MAO-N to catalyze C-C bond formation and oxidation.These two catalysts could work cooperatively in one pot under mild conditions,and a range ofN-alkyl tetrahydroisoquinolines were alkynylated effectively (Fig.19) [70].
Metal-catalyzed C-hetero bond reactions have also been combined with biocatalysis.For example,a Cu-catalyzed cycloaddition reaction was combined with ADH-catalyzed reduction for the synthesis of different enantioenriched 1,2,3-triazole-derived diols(Fig.20).This concurrent chemoenzymatic system for the construction of C-N bonds could be processed in aqueous medium,with high yields andeevalues [71].The Bergman group reported a chemoenzymatic method for the formation of C-O bonds.Gold or ruthenium was used as the metal catalyst and encapsulated in tetrahedral supramolecular Ga4L6(L=N,N′-bis(2,3-dihydroxyben zoyl)-1,5-diamino-naphthalene)clusters to prevent enzymes inactivation (Fig.21).This research provided a general and feasible strategy for combining classic organic catalysis with biocatalysis[72].
The Redox reaction was applied for the metal and enzymecatalyzed chemoenzymatic reactions.The Gr?ger group used polydimethylsiloxane (PDMS) as a film to separate the metal catalyst PdCl2from ADH,avoiding the contact inhibition.Thus,PdCl2-catalyzed oxidation and ADH-catalyzed reduction were successfully combined to prepare (R)-1-phenylethanol (Fig.22) [73].The Berglund group reported a one-pot chemoenzymatic method for the synthesis of capsaicin derivatives from lignin derivatives.This pathway involved multiple steps in which vanillin was first oxidized to vanillyl alcohol by a nano Pd catalyst.Then,vanillin was converted to vanillylamine by transaminase and subsequently reacted with nonanoic acid to produce capsaicin through lipase.L-Alanine dehydrogenase and GDH were conjugated as cofactor regeneration system,and the final yield was 92% (Fig.23) [74].The Beifuss group developed a chemoenzymatic method that combined laccase with the lewis acid Gd(OTf)3to prepare 3-methyl-6,7-dihydrobenzofuran-4(5H)-ones;laccase oxidatively cleaves 2,5-dimethylfuran to produce (Z)-3-hexene-2,5-dione,and then undergoes a domino 1,2-addition/1,4-addition/elimination with 1,3-dicarbonyls under the catalysis of lewis acids to obtain the final product with yields of up to 82%(Fig.24)[75].The Ma group established a two-step chemoenzymatic method to convert formaldehyde into lactic acid,this system used an engineered formolase and NaOH as catalysts at temperatures ≤30 °C with favorable atom efficiency and overall yield (Fig.25) [76].In this study,the engineered formolase variant altered the main product from twocarbon glycolaldehyde (GA) to three-carbon dihydroxyacetone(DHA),followed by the NaOH-promoted rearrangement of DHA into lactic acid.
Fig.15.The regeneration of NAD+ catalyzed by Iron (III) porphyrin coupled with L-glutamate dehydrogenase.
Fig.16.The regeneration of nicotinamide,flavin,and iron porphyrin catalyzed by [Cp*Rh(bpy)(H2O)]2+.
Fig.17.Chemoenzymatic synthesis of chiral arylalanine derivatives.
Fig.18.[Rh(cod)Cl]2was coupled with CHAO to chemoenzymatic synthesize enantiomerically pure tetrahydroquinoline.
Fig.19.HAuCl4was coupled with MAO-N to chemoenzymatic synthesize N-Alkyl tetrahydroisoquinolines.
Fig.20.Chemoenzymatic synthesis of chiral 1,2,3-triazole-derived diols.
Fig.21.Chemoenzymatic synthesis of substituted tetrahydrofuran.
Fig.22.Chemoenzymatic synthesis of (R)-1-phenylethanol.
Fig.23.Chemoenzymatic synthesis of capsaicin analogues.
Fig.24.Chemoenzymatic synthesis of 3-methyl-6,7-dihydrobenzofuran-4(5H)-ones.
Fig.25.Two-step chemoenzymatic synthesis of lactate from formaldehyde.
Although several chemoenzymatic systems catalyzed by metals or their complexes have shown good selectivity and stability,there are still some problems to be solved,including (1) most of the metal catalysts are precious metals,whereas less inexpensive metal catalysts are involved in chemoenzymatic method,(2) the mutual inactivation of free metals or metal complexes with enzymes,(3) the wide gap in reactivity between metal catalysts and enzyme under the same reaction conditions,and(4)the recovery and reuse of these catalysts.Some optimization strategies should be adopted,such as developing inexpensive metal catalysts,simplified metal ligands,and encapsulation or immobilization technology to improve the capabilities of combined metals and enzymes.
Compared with metal,small organic molecule can be easily prepared with a lower cost.Moreover,metal complexes might be sensitive to water or air and sometimes only work under high temperature or pressure,while small organic molecules are more stable,and can be coupled with enzymes to perform catalysis in a more cooperative state.In the following,we discuss chemoenzymatic catalysis catalyzed by small organic molecule combined with enzyme.
Compared with those catalyzed by metals,there have been relatively few DKR reactions catalyzed by small organic molecules combined with enzymes.The Turner group combined MAO-N with borane to catalyze the DKR of racemic amine,and various types of chiral amines were obtained by this strategy[77–79].For example,a monoamine oxidase/borane morpholine system was applied for the DKR of racemic benzyl tetrahydroisoquinoline,and the resulting product (S)-benzyl tetrahydroisoquinoline was then catalyzed by berberine bridging enzyme (BBE) to prepare chiral berberine,ultimately obtaining 90% conversion and 97%ee(Fig.26) [80].The Bertrand group prepared 2,2,2-trifluoroethanthiol compounds as racemic catalysts [81–84].Using these in-situ generated molecules,the DKR of 4-phenylbutan-2-amine was achieved,and alkaline protease andN-octanoyl alanine trifluoroethyl ester were used as biocatalyst and acylation reagent,respectively.Finally,enantioenriched amides was prepared with highee(Fig.27) [82].The Servi group reported the DKR of amino acid thioester by the combination of organic base trioctylamine with subtilisBacillus subtilisprotease.The chemoenzymatic system was suitable for the racemization of a range of amino acid thioesters.Optically active amino acids were finally obtained through a hydrolysis reaction[85].Subsequently,this group reported the combination ofBacillus subtilisprotease and 1,8-diazabicyclo [5.4.0]undec-7-ene (DBU) for the DKR of amino acid derivatives (Fig.28) [86,87].
Fig.26.DKR of amine catalyzed by monoamine oxidase/borane morpholine system.
Fig.27.DKR of amine catalyzed by alkaline protease combined with CF3CH2SH.
Small organic molecules involved in cofactor regeneration,such as methyl viologen,2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfo nate),1,10-phenanthroline-5,6-dione,generally contain conjugated structures(Fig.29,a).The Hollmann group reported a cofactor regeneration reaction using diammonium 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) as an organic catalyst[88].In this system,laccase indirectly oxidized NAD(P)H to NAD(P)+using molecular oxygen as oxidant and ABTS as electron mediator;furthermore,ADH was integrated with the regenerated NAD(P)+to catalyze the oxidation of alcohols (Fig.30).
Inspired by natural cofactors,biomimetic molecules have also been developed for the natural cofactor regeneration (Fig.29,b).For example,the Steckhan group constructed a NAD+regeneration system with 1,10-philolin-5,6-diketone (PDMe+) as the framework of the regeneration reagent,and ADH was coupled for the synthesis of ketones [89].PDMe+mimics the natural cofactor methoxatin and functions as a redox electron transporter(Fig.31).Another more typical biomimic was flavin.The regeneration pathway of NAD(P)+involves natural flavin (FMN) obtaining H-(H++2e-) from the NAD(P)H to form the reduced-state flavin(FMNH2).Then FMNH2is oxidized by oxygen to recover to the original oxidized state (FMN) so that a recycle is formed and NAD(P)+is regenerated.However,the electron transfer efficiency between NAD(P)H and FMN was very low(several days to complete a cycle)[90],thereby hindering the application of this regeneration system.The Hollmann group tried to promote this catalytic activity by means of excitation with special light sources,but no obvious improvement in catalytic efficiency was found [91].Considering the limitation of natural cofactors,the Ying group proposed replacing natural flavins with artificial bridged flavins to improve the regeneration efficiency of NAD(P)+[92].It was found that the TOF value of artificial flavin was 700 times higher than that of FMN.In subsequent study,a new strategy of regulating cofactor balance in vivo using synthetic flavin analogs was further reported,which was the first example of artificial flavin manipulating intracellular nicotinamide cofactor balance(Fig.32)[93].Similar to natural flavin,natural nicotinamide cofactor was also re-engineered,omitting the adenine and nucleotide groups that lacked catalytic activity and thus forming artificial nicotinamide molecules [94–97].Using these simplified and inexpensive molecules,natural flavin (FAD,FMN) was regenerated for monooxygenases-catalyzed epoxidation,sulfur oxidation,and halogenation (Fig.33) [98,99].Thus,the traditional catalytic system of two-component flavindependent monooxygenases was improved,removing flavin reductase and nicotinamide cofactor regeneration system.
Fig.28.DKR of amine amino acid thioesters catalyzed by DBU combined with subtillsin.
Fig.29.Some small organic molecule catalysts applied for cofactor regeneration.
Fig.30.Cofactor regeneration catalyzed by ABTS.
Fig.31.Cofactor regeneration catalyzed by 1,10-philolin-5,6-diketone.
In addition to catalyzing cofactor regeneration and DKR,the combination of small organic molecule and enzyme can facilitate other types of reactions.In terms of organic acid/base catalysis,the Ragauskas group reported the combination of oxidation reaction catalyzed by laccase and Michael addition reaction catalyzed by a lewis base pyridine.A series of benzofuran compounds were synthesized from 3-methylphenylcatechol(Fig.34)[100].The Pietruszka group combined laccase with a lewis acid (S)-2-[diphenyl(trimethylsilyloxy)-methyl]pyrrolidine for oxidation and Michael addition in one pot.The chemoenzymatic reaction occurred in a mixture solvent of MeCN and H2O at room temperature,and the products were well separated with high yields andeevalues(Fig.35) [101].
Chiral proline is another biofriendly small organic molecule catalyst and is used to catalyze chemoenzymatic reactions.The Gr?ger group coupled a chiral proline derivative-catalyzed Aldol reaction with ADH-catalyzed reduction for the stereoselective synthesis of 1,3-diol[102].The results showed that the reaction type changed from kinetic control to thermodynamic control increasing the dosage of the proline derivative chemocatalyst.Subsequently,this group further demonstrated that the reaction system could be completely carried out in organic solvents without the separation of intermediates [103],and the final conversion reached 89%,withee>99% (Fig.36).
Fig.32.Artificial flavin cofactor manipulated cofactor balance in vitro/vivo.
Fig.33.Common seen artificial nicotinamide cofactors.
Fig.34.Laccase-catalyzed oxidation was combined with pyridine-catalyzed Michael addition.
Small organic molecules combined with enzyme-catalyzed redox reactions have been reported in many cases.For instance,the Nascimento group coupled CALB with peracid to catalyze epoxidation of β-carylene and studied the effects of different enzymes,oxidation reagents,acyl donors and organic solvents on the reaction yield.The results showed that the ratios of epoxidation and bi-epoxidation products in different organic solvents were quite disparate.In hexane,the products were epoxy compounds;in the case of ethyl acetate or toluene,all the substrates were transformed to bi-epoxy products (Fig.37) [104].The Subileau group constructed a biphasic chemoenzymatic cascade reaction catalyzed by peracid and lipase [105].With hydrogen peroxide as the substrate,lipase catalyzed the hydrolysis of valerolactone to form 5-hydroxyperoxypentanoic acid which could catalyze the conversion of olefin to epoxide;thus,valerolactone was recycled.
Fig.35.Laccase was combined with (S)-2-[diphenyl (trimethylsilyloxy)-methyl]pyrrolidine for the reaction of oxidation and Michael addition.
Fig.36.Chemoenzymatic synthesis of chiral 1,3-diol through chiral proline derivative combined with ADH.
Fig.37.Epoxidation of carylene catalyzed by CALB combined with peracid.
2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO) can trap free radicals,quench singlet oxygen,perform selective oxidation,etc.It is often used as a chemocatalyst for the oxidation of various alcohols.The Gotor group combined laccase with TEMPO to catalyze the oxidation of halohydrins to the corresponding ketones,and then,ADH was used to reduce haloketones to optically pure halohydrins (Fig.38) [106].The Lavandera group reported another chemoenzymatic system employing TEMPO for the synthesis of several chiral saturated ketones with racemic allyl secondary alcohols as substrates.In this system,ene reductase was coupled for the enantioselective reduction of α,β-unsaturated ketone [107].N-Oxyl-2-azaadamantane (AZADO),another stable nitroxyl free radical catalyst,was also applied to oxidize alcohols with high efficiency.The Rebolledo group combined AZADO with transaminase for oxidation and amine transfer.First,the racemic alcohol was transformed to a ketone by AZADO,and then,transaminase catalyzed the conversion of the ketone to the corresponding chiral amine (Fig.39) [108].
Fig.38.TEMPO mediated oxidation was coupled with ADH to synthesize chiral alcohol.
There are relatively few reports on chemoenzymatic methods through the catalysis of enzymes combined with small organic molecules.However,these flexible molecules have advantages such as good stability,easy synthesis,and good compatibility,and are therefore essential for the development of chemoenzymatic method.At present,the problems of this kind of reaction are that there are very few small organic molecules that can be integrated with enzymes,and separating and recycling these molecules remain challenging.Designing more efficient,biofriendly,and stable small organic molecules and developing reasonable reaction systems for the recycling of these interesting molecular catalysts are future research directions.
In the last decade,photochemistry and electrochemistry have been developed as complementary and unique methods for chemical activation or electron transfer in traditional chemocatalysis.In the following,cofactor regeneration and other reactions catalyzed by photo/electrocatalysts combined with enzymes are discussed.
Fig.39.AZADO mediated oxidation was coupled with TA to synthesize chiral amine.
Photochemical cofactor regeneration is a novel method and has attracted much attention from researchers [109].Due to its usage of inexpensive,rich and clean solar energy,this process plays a very important role in biosynthesis [110],biosensors [111]and energy storage[112].Some effective photosensitizers and electron transport mediators [113]including metalloporphyrin [114,115],thionine [116],toluidine blue [117],and cobalt oxime [118],were designed for cofactor regeneration.The Lian group reported a visible light-driven method for regenerating FMN using CdSe quantum dots (QD) and methyl violet (MV) as electron mediators.The reduced flavin FMNH2was then integrated with the old yellow enzyme to catalyze the reduction of 2,6,6-trimethylcyclohexane-2-ene-1,4-dione.Transient absorption studies showed that the main step limiting the FMN reduction efficiency was the process of hole-filling in the QD by TEOA.Screening more effective electron donors,preparing suitable nanostructures,and enhancing the hole removal ability are promising strategies to increase the efficiency(Fig.40) [119].The Karube group proposed a photochemical NADPH regeneration method [120],and oligomeric thiophene,methyl violet and ethylenediamine tetraacetic acid (EDTA) were employed as photosensitizer,electron acceptor and electron donor respectively.Ferredoxin reductase (FDR) was coupled to oxidize free radical MV+to MV2+,the electron was further transferred to NADP+,and thus,NADPH was regenerated.The results suggested that AT oligomer had an efficiency of 0.04% in converting NADP+to NADPH,while CT oligomer had an efficiency of 0.025% after 40 minutes of light exposure(Fig.41).The Dunn group used immobilized thionine as photocatalyst to regenerate cofactor NADP+in a SiO2sol–gel matrix.Isocitrate dehydrogenase was coupled to employ NADP+for the catalytic transformation of isocitrate to αketoglutaric acid(Fig.42)[116].The Baeg group synthesized semiconductor photocatalyst W2Fe4Ta2O17and constructed a heterogeneous NADH regeneration system using EDTA as the electron donor and a rhodium complex as the electron transfer mediator.The regeneration efficiency of the semiconductor photocatalyst was 12 times higher than that of the TiO2-based photocatalyst(Fig.43) [121].The Cha group combined photocatalyst platinum cadmium sulfide quantum dots (Pt@CdS) with ADH to synthesize 2-ethylhexenal fromn-butanol [122].In the reaction,NAD+was regenerated by Pt@CdS,ADH used NAD+to catalyze the oxidation ofn-butanol ton-butyraldehyde,and 2-ethylhexenal was finally synthesized by β-alanine-catalyzed Aldol condensation ofnbutyraldehyde (Fig.44).
Fig.40.Photocatalytic cofactor regeneration system using CdSe QD as electron mediator for the stereoselective reduction of oxoisophorone.
Fig.41.Photosensitizer oligomeric thiophene was combined with FDR to regenerate NADPH.
Fig.42.Photocatalyst thionine was coupled with ICDH to regenerate NADPH.
Fig.43.Photocatalyst W2Fe4Ta2O17was coupled with [Cp*Rh(bpy)(H2O)]2+ to regenerate NADH.
Fig.44.Photocatalyst Pt@CdS using hydrogen as reduced agent to regenerate NADH.
Fig.45.Electrode modified by Ru complex to regenerate NADH.
Fig.46.The products of direct electrochemical reduction of NAD(P)+.
Fig.47.Preparation of saturated chiral alkane by photocatalysis combined with olefin reductase.
Electrochemical cofactor regeneration has long been recognized as a potentially powerful technology,because it uses low-cost electricity as the regenerative reagent,and the reaction process is easy to monitor,more importantly,the redox potential is simple to control[123].In 2007,the Omanovic group studied the direct regeneration of NADH from NAD+on metal electrodes.It was found that the gold electrode had a higher NADH yield (75%) but a poor NAD+reduction rate at a low negative potential.Under a high negative potential,the reduction rate of NAD+was favorable,but the yield of active NADH decreased significantly (28%),and most of the products were inactive dimers form NAD(P)2.When the surface of Au electrode was modified with platinum nanoparticles,the yield of active NADH was increased at high negative potential.The reason for this result was not only that submonolayer structure of platinum nanoparticles acted as an effective hydrogenreleasing catalyst to accelerate the protonation of NAD free radicals but also that platinum nanoparticles physically prevented the contact of two newly formed and adjacent free radicals leading to dimerization[124].Later,the Park group synthesized a conductive vanadium-silica sol–gel matrix with vanadium trioxide(VOTP)and tetraethyl protosilicate(TMOS)as precursors[125].This conductor could effectively improve the efficiency of electrochemical NADH regeneration.When the VOTP ratio in the matrix was increased,the NADH regeneration efficiency was greatly promoted.Then,this electrochemical NADH regeneration system was combined with glutamate dehydrogenase to catalyze α-ketoglutarate to Lglutamate.The Yoo group prepared a porous stannic oxide electrode with a large surface area.It was demonstrated that tin oxide has its own redox potential by electrochemical analysis.Therefore,NAD+could be synthesized from NADH by electrochemical reaction on the surface of tin oxide,which avoided the usage of electron mediators in the regeneration reaction [126].The Etienne groupcopolymerized(2,2′-bipyridinyl)(pentamethyl cyclopentadienyl)-rhodium complex on the surface of carbonbased polyporous electrodes to regenerate NADH [127].Galactol dehydrogenase (GatDH) was further coupled to reduce hydroxyacetone to 1,2-propanediol.The final regeneration efficiency of NADH reached 87% (Fig.45).
Although the abovementioned examples showed that the electrochemical NADH regeneration system could exhibit steady operation,limitations still existed.Typically,the inactive byproduct 1,6-NAD(P)H or dimer NAD(P)2are often formed along with the active product 1,4-NAD(P)H.Approximately 40% of the nicotinamide cofactors are inactivated during each regeneration cycle(Fig.46) [128].
Photocatalysis has been coupled with other types of enzymatic reactions in addition to cofactor regeneration.For example,the Zhao group performed some pioneering work.In 2018,this group constructed a cooperative chemoenzymatic system for the preparation of chiral saturated alkanes from racemic olefins,and visible light-induced olefin isomerization and ene reductase-catalyzed double-bond reduction were combined (Fig.47) [129].The Park group developed a photosynthesis mimicking system combining photocatalysis with biocatalysis.Under visible light irradiation,water was oxidized at the Co-Pi photoanode and acted an as electron donor to transfer to the photocathodes α-Fe2O3and BiFeO3.The electron was then transferred to NAD+through a rhodium complex.Finally,NADH was regenerated and participated in three enzymes cascade reactions to synthesis methanol from CO2(Fig.48) [130].
Fig.48.Photochemical cascade was combined with three enzymes to synthesize methanol from CO2.
Compared with metal catalysis and small organic molecule catalysis,photo-,electro-and chemocatalysis employ cleaner and more available energy,which is the main reason why photocatalysis and electrocatalysis are receiving increasing attention.However,there are some problems,such as low catalytic efficiency,the need for complex light or electron energy transfer systems,and a decrease in enzyme activity.Some strategies to promote the development of this kind of chemoenzymatic reaction such as the design of a stable photosensitizer,the improvement of the light or electron energy transfer system,and the exploitation of various types of artificial electronic mediators or electrode materials should be developed.
As one of the incompatibility challenges between chemocatalysis and biocatalysis,their mismatched substrate scope has prevented the establishment of chemoenzymatic approaches.The development of protein engineering can ameliorate this issue.Similarly,using this newly developed strategy,the reconstructed biocatalyst will be better adapted to the relatively harsh conditions of chemoprocess [131].The Janey group expanded the active site space of a transaminase by directed evolution,enabling this enzyme to accept the macromolecule ketone substrate (Fig.49)[132].In the process of preparing the drug sitagliptin,the established chemoenzymatic method containing this transaminase mutant and phosphoric acid greatly improved the overall yield and shortened the reaction steps compared with the traditional metal-catalyzed chemoprocess [133].The Turner group modified the monoamine oxidase MAO-N fromAspergillus niger,expanding its substrate scope.Several engineered MAO-N biocatalysts were obtained by the combination of rational structure-guided engineering and high-throughput screening.Combining these excellent biocatalysts with BH3-NH3,several enantioenriched amines were obtained (Fig.50) [134].
Another approach to improve the stability or catalytic activity of biocatalysts in chemoenzymatic systems is reaction engineering.The Micklefield group combined halogenase-catalyzed C-H activation with a Pd-catalyzed cross coupling reaction to synthesize aryl indole products (Fig.51) [135].Due to the mismatched reaction conditions,PDMS membrane was applied to isolate the metalcatalyzed process and enzyme-catalyzed process.As PDMS is a hydrophobic polymer,nonpolar compounds (reaction substrates and intermediates) can pass through the PDMS membrane,while charged reagents (palladium catalysts,enzymes,and cofactors)will not permeate;thus,the biocatalyst and chemocatalyst can be divided into two parts to mediate the reaction without interference.
Deep eutectic solvents (DESs) are a type of green eutectic mixtures composed of hydrogen bond acceptors such as cheap and nontoxic choline chloride and hydrogen bond donors such as glycerin,urea,and saccharides [136].DESs have been proven to be an effective approach for promoting the catalysis of chemoenzymatic reaction[137].For example,the Javier group reported the one-pot reduction involving Ru-catalyzed isomerization of allylic alcohols combined with an enantioselective bioreduction in DES-buffer medium.Using a buffer containing of the DES mixtures choline chloride/glycerol or choline chloride/sorbitol,the performance of the biocatalyst was enhanced,and the corresponding secondary alcohols synthesized by this chemoenzymatic method were achieved up to >99% conversion and up to >99%ee(Fig.52) [138].
Fig.49.Comparison of traditional chemical method with the designed chemoenzymatic method to the synthesis of drug sitagliptin.
Fig.50.Application of the MAO-N mutant for the deracemization of alkaloid natural products (1 atm=101325 Pa).
Fig.51.Chemoenzymatic synthesis of biphenyl compounds using PDMS membrane.
Fig.52.One-pot Ru complex-catalyzed isomerization of allylic alcohols combined with an enantioselective bioreduction in DES-buffer medium.
The combination of biocatalysis and chemocatalysis can make full use of their separate advantages and provides a promising pathway for the production of drugs,chemicals and bioactive molecules [139–141].However,as seen from the above examples,chemoenzymatic catalytic methods are still in the primary development stage,with some work remaining before from large-scale application.For example,regarding DKR,although many metal catalysts such as rhodium,ruthenium and iron have been developed for the racemization of alcohols,problems such as low catalytic efficiency and harsh reaction conditions(high temperature or pressure,strong base)need to be solved.Compared with those of alcohols,the methods for racemization of amines are more challenging to develop due to the lower efficiency and dearth of racemic catalysts.Therefore,it is urgent to develop a variety of highly effective catalysts for the racemization of amines.Cofactor regeneration,another typical chemoenzymatic catalysis,is also confronted with many limitations;for example,compared with that for NAD(P)H,the method for NAD(P)+regeneration is lacking.Furthermore,these two kinds of regeneration strategies are mostly confined to the metal,and considering the diversity of redox enzymes,it is hard to find a metal species that is compatible with all nicotinamide-dependent oxidoreductases.Thus,developing a more reasonable and efficient small organic molecule catalyst instead of a metal to catalyze the regeneration of nicotinamide cofactors is highly desirable.
As with DKR and cofactor regeneration,the most important problem of chemoenzymatic catalysis is the compatibility of biocatalyst and chemocatalyst.In improving the suitability of biocatalyst,protein engineering is a kind of important means to broaden the substrate spectrum and improve the regio-,stereo-,and enantioselectivity,and site-specific mutation or directed evolution of some proteins can promote their catalytic activity in harsh environments (high temperature,acid and base,hyperosmosis)[131,142,143].Another effective method is reaction engineering,such as biphasic systems,coimmobilization or crosslinking of enzymes and chemocatalysts,and supramolecular cage fusion catalysis [13].In recent years,inspired by the concept of biomimetic catalysis,some scientists have also developed strategies of artificial metalloenzymes combined with chemocatalysts to improve the suitability of the biocatalysis and chemocatalysis.The Ward group has made some impressive progress including the multifunctional artificial hydrogenase.This remarkable metalloenzyme could be combined with monoamine oxidase to catalyze the racemization of amine,coupled with peroxidase to catalyze the hydrogen transfer reaction,and used to regenerate nicotinamide cofactor to facilitate the catalysis of monooxygenase [49,63,144].This group also summarized the preparation of artificial biomimetic enzymes and their applications in cascade catalysis[145,146].In conclusion,this review described the reactions of enzymes combined with different types of chemocatalysis in recent years based on the classification of chemocatalysts.These chemoenzymatic approaches were summarized especially from the perspective of DKR and cofactor regeneration.More green,efficient and coordinated chemoenzymatic methods will be developed to further address health,energy,environment,safety and other challenges through protein engineering,reaction engineering or artificial designed enzymes.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors thank the financial support by National Natural Science Foundation of China (21776132,21878142);Jiangsu Province Natural Science Foundation for Distinguished Young Scholars(BK20190035);National Key Research and Development Program of China (2019YFD1101202);Jiangsu Province Natural Science Foundation for Youths (BK20200685);China Postdoctoral Science Foundation (2019M660113).
Chinese Journal of Chemical Engineering2021年2期