cofactor regeneration for sustainable enzymatic bio synthesis

Edited by Foxit PDF Editor Copyright (c) by Foxit Software Company, 2004 For Evaluation Only.

http://www.paper.edu.cnBiotechnology Advances 25 (2007) 369 384

Research review paper

Cofactor regeneration for sustainable enzymatic biosynthesisWenfang Liu a , Ping Wang b,a

National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100080, China b Department of Bioproducts and Biosystems Engineering, University of Minnesota, St. Paul, MN 55108, United States Received 11 January 2007; received in revised form 3 March 2007; accepted 12 March 2007 Available online 23 March 2007

Abstract Oxidoreductases are attractive catalysts for biosynthesis of chiral compounds and polymers, construction of biosensors, and degradation of environmental pollutants. Their practical applications, however, can be quite challenging since they often require cofactors such as nicotinamide adenine dinucleotide (NAD) and nicotinamide adenine dinucleotide phosphate (NADP). These cofactors are generally expensive. Efficient regeneration of cofactors is therefore critical to the economic viability of industrialscale biotransformations using oxidoreductases. The chemistry of cofactor regeneration is well known nowadays. The challenge is mostly regarding how to achieve the regeneration with immobilized enzyme systems which are preferred for industrial processes to facilitate the recovery and continuous use of the catalysts. This has become a great hurdle for the industrialization of many promising enzymatic processes, and as a result, most of the biotransformations involving cofactors have been traditionally performed with living cells in industry. Accompanying the rapidly growing interest in industrial biotechnology, immobilized enzyme biocatalyst systems with cofactor regeneration have been the focus for many studies reported since the late 1990s. The current paper reviews the methods of cofactor retention for development of sustainable and regenerative biocatalysts as revealed in these recent studies, with the intent to complement other reviewing articles that are mostly regeneration chemistry-oriented. We classify in this paper the methods of sustainable cofactor regeneration into two categories, namely membrane entrapment and solidattachment of cofactors. 2007 Elsevier Inc. All rights reserved.Keywords: Oxidoreductases; Enzyme immobilization; Cofactor regeneration; Biocatalysis; Biosynthesis; Industrial biotechnology

Contents 1. 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . Membrane entrapment of free cofactors . . . . . . . . . . 2.1. Native cofactors . . . . . . . . . . . . . . . . . . . 2.2. Chemically modified cofactors . . . . . . . . . . . 2.2.1. Chemical modification of cofactors . . . . 2.2.2. Retention of chemically modified cofactors . . . . . . . . . . . . . . . . . . . . . . . . . by using . . . . . . . . . . UF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . membranes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370 372 372 374 374 375

Corresponding author. Tel.: +1 612 624 4792. E-mail address: [email protected] (P. Wang). 0734-9750/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.biotechadv.2007.03.002

370

Edited by Foxit PDF Editor Copyright (c) by Foxit Software Company, 2004 For Evaluation Only.W. Liu, P. Wang / Biotechnology Advances 25 (2007) 369384

http://www.paper.edu.cn

2.2.3. Retention of chemically modified cofactors by using microcapsules . 2.2.4. Retention of chemically modified cofactors on electrodes . . . . . . 3. Solid-phase cofactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Physically immobilized cofactors . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Physical attachment of cofactor for organic synthesis . . . . . . . . 3.1.2. Physical attachment of cofactors on electrodes . . . . . . . . . . . . 3.2. Chemical incorporation of cofactors . . . . . . . . . . . . . . . . . . . . . . 4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

375 376 376 376 376 377 377 380 381 381

1. Introduction Oxidoreductases represent about one quarter of the known enzymes (Kula and Kragl, 2000). Awide spectrum of applications have been explored so far for this group of enzymes, including synthesis of chiral compounds, such as chiral alcohols, aldehydes and acids; preparation and modification of polymers, especially biodegradable or biocompatible polymers; biosensors for a variety of analytical and clinical applications; and degradation of organic pollutants (Sheldon and Stephen, 1983; Hummel, 1999). Due to their high efficiency and specificity, these enzymes are particularly attractive for biosynthesis. Oxidoreductases generally require a non-protein chemical group, a cofactor, to catalyze reactions. Although certain oxidoreductases possess prosthetic groups to facilitate reactions, the majority of the enzymes explored for biosynthesis needed to interact with cofactors that are not permanently tethered to the enzymes. The most widely involved cofactors are several organic compounds, which are also often referred to as coenzymes, such as NAD(H), NADP(H) and ATP. In particular, NAD(H) and NADP(H) have been examined extensively in recent years for chemical processing applications. Unlike enzymes, cofactors act as stoichiometric agents in biotransformation reactions and undergo chemical reactions with substrates. Often they are much more expensive than the desired products. Accordingly, efficient regeneration and reuse of the cofactors are essential to large-scale synthetic applications (Chenault et al., 1988; Wichmann and Vasic-Racki, 2005). A total turnover number (TTN) of the cofactors, defined as mol product produced/mol cofactor applied, in the order of hundreds up to thousands is usually desired to make the biocatalytic processes economically viable (Chenault et al., 1988). Methods including chemical, electrochemical, photochemical, microbial and enzymatic reactions have all been developed for cofactor regeneration (Chenault and Whitesides, 1987). Among others, enzymatic approach

is particularly preferred for industrial processes due to its high selectivity and efficiency. It also affords the feasibility of coupling more than one valuable chemical production routes. There are two different ways to achieve enzymatic regeneration (Fig. 1). One is through the use of substrate-coupled reaction systems, in which one enzyme that uses both the reduced and oxidized forms of a cofactor is applied to catalyze both the desired synthesis of the product from one substrate and the cofactor regeneration reaction with a second substrate. One example for that is the alcohol dehydrogenase (ADH)-catalyzed organic synthesis reaction. Acetophenone was reduced enantioselectively into (S)1-phenylethanol by an NADPH-dependent ADH from Thermoanaerobacter sp., and the conversion of acetophenone could reach 98% when 2-propanol was used as the secondary substrate to drive the regeneration of NADPH catalyzed by the same ADH in a batch reactor (Findrik et al., 2005). Since the same enzyme is required to catalyze two separated reactions simultaneously, it is usually difficult to achieve thermodynamically-favorite reaction conditions for both reactions in the same reaction medium. The other way is through the use of a second enzyme to catalyze the cofactor regeneration reaction. The use of a second enzyme, which has been adapted for the majority of cofactor regeneration processes, usually affords broader options of substrates for the cofactor regeneration reaction, and thus makes it much easier to achieve large thermodynamic driving

Fig. 1. Enzymatic regeneration of cofactors (for substrate-coupled regeneration, the two enzymes are the same, E1 = E2; for enzymecoupled regeneration, E1 and E2 represent two different enzymes).

http://www.paper.edu.cnW. Liu, P. Wang / Biotechnology Advances 25 (2007) 369384 371

Fig. 2. Cofactor regeneration with electrochemical reactions. (a) Electro-enzymatic cofactor regeneration: the cofactor is regenerated through an enzyme-catalyzed reaction; redox reactions on electrodes regenerate the second substrate, the mediator; Medred and Medox represent the reduced and oxidized forms of mediator, respectively. (b) Electrochemical cofactor regeneration: the cofactor undergoes redox reactions directly on the electrode.

forces for both reactions. In either way, it is desirable to make value-added products from the second substrates applied for cofactor regeneration; otherwise, the second substrates have to be either very cheap or can be regenerated easily for reuse. Electrochemical reactions are usually introduced for the latter case, i.e., the second substrate applied for cofactor regeneration is regenerated electrochemically (Scheper et al., 1987; Montagn and Marty, 1995; Nakamura et al., 1996; Leca and Marty, 1997a,b; Noguer and Marty, 1997; Maines et al., 2000; Leonida, 2001). For such a purpose, the second substrates, usually referred to as mediators, are required to possess both good electrochemical reactivity on electrodes and good reactivity with the enzyme and cofactor (Fig. 2(a)).

Cofactors can be regenerated directly on the surface of electrodes without the involvement of mediators and enzymes (Fig. 2(b)). We may define this method as electrochemical regeneration of cofactors. In the case of mediator-assisted regeneration as discussed above, the cofactor is regenerated by an enzyme-catalyzed reaction, while the mediator is regenerated on electrodes. The mediators are usually much more active than cofactors for electrochemical reactions on electrodes, and thus provide the possibility for higher overall reaction rates. As mentioned earlier, efficient cofactor regeneration is highly desired in biosynthesis. A high TTN is difficult to achieve, however, if the products accumulate in the reaction media and build thermodynamic resistance

Fig. 3. Typical configurations for sustainable cofactor retention in continuous-flow bioreactors.

372

http://www.paper.edu.cnW. Liu, P. Wang / Biotechnology Advances 25 (2007) 369384

Fig. 4. Membrane entrapment of native cofactors. (a) Retention via size exclusion using uncharged membranes; (b) retention with charged membranes; (c) retention with uncharged UF membranes with the aid of polyelectrolytes.

against the forward reactions. Accordingly, continuousflow reactors (Fig. 3) that allow continuous feeding of substrates and removal of products are always preferred in large-scale synthetic applications. While continuousflow reactors have been applied widely and successfully with enzymes that do not require cofactors, it has been a daunting task with cofactor-associated reactions as the cofactors tend to leave the reactors along with the substrates and products. How to achieve the immobilization of the cofactors while maintaining their functionality has proved to be considerably challenging. Earlier methods for cofactor immobilization have been reviewed previously by Lowe (Lowe, 1978), while the broader biochemical potentials of cofactors have been reviewed more recently (Chenault and Whitesides, 1987; Peters, 1999; Zayats et al., 2000). The current review focuses on recent advances in developing immobilized cofactors for sustainable biocatalysis with cofactor regeneration. Herein we classify the methods for sustainable cofactor regeneration for continuousflow processes into two categories: (1) membrane entrapment of dissolved free cofactors, which could be either native cofactors or chemically modified, with various forms of membranes including nano-filtration (NF) membranes, ultrafiltration (UF) membranes, dialysis membranes, microcapsules, etc.; and (2) solidattachment of cofactors, including both physical and chemical incorporation of cofactors into insoluble solid supports. Most of the works reviewed are publications in 1990s and later, with several papers on modification chemistry of cofactors dating back to 1970s. Since the focus is on the retention of cofactors in bioreactors, many publications concerning the chemical aspects of cofactor regeneration reactions are not included. All the methods

reviewed here are those published in literature and examined through lab-scale experiments. Little is known regarding which methods have been practiced in industry. In fact, according to recent reviews of industrial biocatalysis (Liese et al., 2000), only several enzymatic processes involving cofactor regeneration have been practiced in industry. These include the lactate dehydrogenase (LDH)/NADH/formate dehydrogenase (FDH), alcohol dehydrogenase (ADH)/NADH/ FDH, ADH/NADH/glucose dehydrogenase (GDH), ADH/NADPH/GDH, and leucine dehydrogenase (LeuDH)/NADH/FDH systems (Liese et al., 2000). While the chemistry for these processes was reported, little is known regarding how the cofactors were handled in the industrial bioreactors. 2. Membrane entrapment of free cofactors 2.1. Native cofactors Compared to enzymes, cofactors are small molecules. That makes it difficult to choose an appropriate filter membrane to retain the cofactors while allowing the products and substrates to pass through (Fig. 4). In the study for production of l-glutamate and l-carnitine catalyzed by glutamate dehydrogenase (GLDH) and lcarnitine dehydrogenase (CDH), Lin and coworkers used a NF membrane, UTC-20, to partially retain NAD+ or NADP in the reactor (Lin et al., 1997, 1999). GDH was applied for the regeneration of NAD+ and its TTN was reported to be up to 10,000. The cofactors could still penetrate through the membrane and a certain amount of cofactors must be continuously amended along with the substrates for a continuous production. Membranes with smaller pores will improve the retention of cofactors;


however, that will lead to a reduced flow rate of substrates (Obn et al., 1998). How to prevent the loss of cofactors from reactors while maintaining high substrate flow rates has been a long-standing challenge. Over the years, different methodologies have been conceived and tested. One approach is to employ charged membranes (Fig. 4(b)), which help to retain cofactors via electrostatic repulsion instead of size exclusion (Kitpreechavanich et al., 1985; Howaldt et al., 1990; Ikemi et al., 1990a; Ikemi et al., 1990b; Kulbe et al., 1990; Rthig et al., 1990; Nidetzky et al., 1994; Nidetzky et al., 1996). For example, a negatively charged NF membrane, NTR 7410, was used for reactions involving NADP(H), which possesses negatively charged phosphate moieties (Ikemi et al., 1990b). The conversion of fructose into sorbitol coupled with GDH-catalyzed cofactor regeneration was operated continuously for over 800 h, and the TTN of NADPH was 106,000. In a continuous process for the synthesis of sulcatol using ADH, the cofactor NADP(H) was retained by a charged UF membrane and regenerated by oxidizing isopropanol to acetone (Rthig et al., 1990). It was claimed that the overall efficiency of NADP(H) with charged UF membrane was even better than that for poly(ethylene glycol) (PEG)-coupled NADP(H) with an uncharged UF membrane. The application of a sulfonated polysulfone membrane for the retention of NADP (H) and NAD(H) was also investigated (Kitpreechavanich et al., 1985; Howaldt et al., 1990). An alternative approach is to use polyelectrolyte like polyethyleneimine (PEI) with uncharged UF membranes (Fig. 4(c)). The polyelectrolyte can complex with the charged cofactors and enzymes and thus help to maintain them in reactors and allows the use of filter membranes with larger pores. This approach was demonstrated for the production of l-lactate, gluconate and glutamate (Obn et al., 1996, 1998). Although better retention of cofactor can be generally expected, the effectiveness of the electrostatic repulsion method is also subject to the effects of several other factors, including the charges carried by the products, substrates, salts, and other chemicals in the reaction media (Howaldt et al., 1990; Kulbe et al., 1990). The volume and time productivity of bioreactors with membrane-retained cofactors is largely determined by the effective area of the membranes. Microcapsules, which provide much more membrane area per unit volume than fixed membranes, were therefore developed for the retention of cofactors. Chang and coworkers developed microcapsules using lipidpolymer complex membranes for capsulation of cofactors in the early 1980s (Yu and Chang, 1981a,b, 1982; Chang et al.,

1982; Ilan and Chang, 1986). By adjusting the ratio of cholesterol to lecithin, they showed that such complex membranes could afford very low permeability to lipidinsoluble substances including cofactors, but at the same time allow lipid-soluble substrates such as ammonia to penetrate very freely. In one example, lipidpolyamide membrane capsulated NADH, urease, GLDH and ADH were applied for the production of amino acids from urea or ammonia. NADH was regenerated by ADH in the presence of ethanol and a maximum TTN of 40 was achieved for NADH in 3 h (Yu and Chang, 1982). Similarly, a liquid surfactant membrane was applied to capsulate NADH-mediated reaction systems including leucine dehydrogenase (LeuDH)FDH and ADHFDH (Scheper et al., 1987; Orlich and Schomcker, 1999; Orlich et al., 2000). A detailed review on these works was presented by Orlich and Schomcker (2001). Membrane entrapment has been applied widely for the preparation of enzyme-based electrodes. Most of these efforts were made for the development of biosensors or biofuel cells, for which the lifetime and reliability of electrodes are mostly concerned while high TTN of cofactors is usually not a focus. Many reviews on biosensors using enzyme-based electrodes have been published (for example, see Gilmartin and Hart (1995) for carbon electrode biosensors, and Cosnier (2003) and Vidal et al. (2003) for more recent publications). Several examples from recent publications are provided in the following to demonstrate the methodologies of cofactor retention on electrodes, which can be potentially applied for electrochemical regeneration of cofactors for biosynthetic purposes. A size-exclusion cellulose acetate (CA) membrane was applied to contain lactate dehydrogenase (LDH) and its cofactor NAD+ to the surface of a Meldola's blue (acting as a mediator)-modified carbon electrode for the construction of a disposable amperometric biosensor for lactic acid. By adjusting the composition of CA membrane, the leaching of cofactor and other biomolecules could be minimized (Sprules et al., 1995). Polylysine was also applied for the capsulation of NAD+, GDH and mediator on the electrode for use as biosensors (Mano and Kuhn, 2005). Similarly, an amperometric biosensor for alcohols was obtained by coating a carbon electrode with a solution containing ADH, NAD+ and poly(ester-sulfonic acid). After drying, a polymeric membrane layer was applied to cover the enzyme/cofactor layer. Ruthenium particles dispersed on the electrode provided an efficient electrocatalytic activity for the regeneration of NAD+ (Wang et al., 1995). In order to improve the biocompatibility of electrodes and improve their performance, a doublemembrane technique was developed to attach enzyme and cofactor onto the electrode. A layer of poly(vinyl chloride)

374


(PVC) membrane was first cast on a platinum electrode. Native NAD+ and ADH were then deposited on the PVC membrane, followed by covering with a cellulose triacetate (CTA) membrane. It was reported that such electrodes had good retention of NAD+, and demonstrated very good sensitivity and operational stability for 30 times of measurements within 4 weeks (Gotoh and Karube, 1994). In the development of a malate biosensor, cofactor NAD+ and enzymes including malate dehydrogenase (MDH) and diaphorase (DI) were first adsorbed on a high proteinbinding nitrocellulose membrane, which was subsequently covered by a second polymeric membrane. It was found that both a Tween-80 modified CA (5%CA + 5% Tween80) membrane and an un-plasticised spin-coated PVC/ polycarbonate (PC) resin membrane were effective to retain cofactor and mediator while provide good permeability to malate (Maines et al., 2000). 2.2. Chemically modified cofactors 2.2.1. Chemical modification of cofactors Increasing the size of the cofactors would allow the use of filters with larger pores and thus reduce the mass transfer resistance of substrates/products. This can be achieved by attaching chemical groups, such as polymers, onto cofactors. The modifying groups should be generally hydrophilic to maintain the aqueous solubility of the cofactors, and should have appropriate size to balance the reactivity and permeability of cofactors. Common water-soluble macromolecular modifiers that have been applied include PEI (Zappelli et al., 1975, 1976, 1977; Riva et al., 1986), dextran (Larsson and Mosbach, 1974; Grunwald and Chang, 1979, 1981, 1988a; Adachi et al., 1984; Gu and Chang, 1990a,b,c) and PEG (Furukawa et al., 1980; Bckmann et al., 1981; Okuda et al., 1985; Riva et al., 1986; Ottolina et al., 1990). Polylysine (Wykes et al., 1975; Zappelli et al., 1975, 1976), poly(acrylic acid) (PAA)

(Zappelli et al., 1977) and other soluble copolymers (Fuller et al., 1980) were also examined. Compared to PEI- and PAA-modified cofactors, PEG-NAD+ showed better results with several dehydrogenases (Riva et al., 1986). PEG-NADH also showed higher efficiency than dextran-NADH for biosensor applications (Noguer and Marty, 1997). In some cases, the enzymes such as ADH that require cofactors can serve as the modifiers for cofactors. Among others, the complexes of ADHNAD+ (Mnsson et al., 1978; Goulas, 1987; Sekhar and Plapp, 1988; Vanhommerig et al., 1996; LeBrun and Plapp, 1999; Leskovac et al., 2003) and LDHNAD+ (Gacesa and Venn, 1979; Schafer et al., 1986) were the most extensively studied enzymecofactor conjugates. It was reported that the reaction rate of a covalently linked GDHPEG-NAD+ conjugate was even higher than that observed for native enzyme and native NAD+ (Nakamura et al., 1986). A similar approach was adopted for MDH-catalyzed reactions (Eguchi et al., 1986). Genetic engineering technique has also been applied for the production of enzymecofactor conjugates. The production of NAD+ tethered to specific sites of GDH from Bacillus subtilis using site-directed mutagenesis by a disulfide bridge has been reported (Mnsson et al., 1991). The attachment of chemical groups can be achieved by using spacers between the cofactor and modifier (Srere et al., 1973). Zappelli et al. prepared a NAD+ analogue carrying a -carboxyalkyl side-chain (2hydroxy-3-carboxypropylamino) at exocyclic adenine amino group, and then coupled it to PEI or polylysine via the carbodiimide group (Zappelli et al., 1975). These macromolecular NAD+ derivatives showed 2560% efficiency of free NAD+ with LDH, while only 27% with yeast ADH (YADH) and alanine dehydrogenase (AlaDH). When carboxyalkyl group was introduced at the adenine C-8 position, PEI-NAD+ showed 47%

Fig. 5. Typical chemical modification routes of cofactors. (a) Multi-step; (b) single-step.


efficiency with YADH while only 3% with LDH, and polylysine-NAD+ showed 56% efficiency with both enzymes (Zappelli et al., 1976). Riva et al. had done a systematic investigation on the influence of coupling site on the properties of cofactors when they were linked to PEG, PEI or PAA (Riva et al., 1986). The results showed that the N6 position of the adenine ring was the optimum site, giving the most satisfactory results for many dehydrogenases. In addition, NAD+ pre-modified with succinyl group before coupling with PEI was reported to retain better activity (Wykes et al., 1975). The conjugation reaction of NADH with alkyl groups may include three steps: (1) alkylation of NAD+ at N-1 of adenine ring; (2) reduction of the nicotinamide moiety with dithionite; and (3) Dimroth rearrangement of alkyl linkage from the N-1 to the C-6 amino position (Fig. 5(a)). Following this procedure, the preparation of PEG-N6-(2aminoethyl)-NADH and dextran-N 6 -(2-aminoethyl)NADH has been reported (Bckmann et al., 1981). By incubating acrylic copolymer containing epoxy group with NAD+ at pH 10, a one-step cofactorpolymer coupling was also achieved (Fig. 5(b)) (Fuller et al., 1980). 2.2.2. Retention of chemically modified cofactors by using UF membranes Chemically modified and thus physically larger cofactors generally retain better in membrane reactors as compared to native cofactors. Succinyl-NAD + coupled to polylysine was applied for a membrane reactor along with ADH and LDH. Continuous production of lactate was performed, and the half-life of the system was found to be 10 days (Yamazaki et al., 1976). Loss of activity may come from either enzyme deactivation or leakage of enzymes and cofactor. By coupling N6-(2-carboxylethyl)-NAD+ to monoaminoPEG (Mw 30003700) with water-soluble carbodiimide, PEG-NAD+ showed activity that was up to 77% of native NAD+, depending on the dehydrogenases applied. PEG (30003700)-NAD+ could keep active for 200 h when it was applied for lactate production with enzymes LDH and YADH in a reactor with an UF membrane (Furukawa et al., 1980). It was also reported recently that PEG (10,000)-NAD+ was applied for continuous synthesis of l-tert-leucine in a small-scale UF-membrane reactor. The regeneration of the modified cofactor was catalyzed by FDH, and a TTN of 125,000 was achieved (Filho et al., 2003). Dextran-modified cofactors also showed promising results. Larsson and Mosbach prepared dextran-NAD+ by coupling N6-[N(6-aminohexyl)-acetamide]-NAD+ with cyanogen bromide activated dextran. Dextran-NAD+ remained about

15% of the cofactor's original activity with either YADH or LDH (Larsson and Mosbach, 1974). DextranNAD+ had also been applied for continuous production of alanine from pyruvate in an UF-membrane reactor. The modified cofactor was repeatedly used for 90 times with regeneration catalyzed by lactose dehydrogenase (Davies and Mosbach, 1974). NAD+ coupled with an enzyme, GDH from B. subtilis, was applied for a hollow fibermembrane reactor with LDH as a regenerating enzyme for continuous production of l-lactate and gluconic acid (Mnsson et al., 1991). A TTN of 135,000 was achieved for the cofactor within the first 2.5 days. 2.2.3. Retention of chemically modified cofactors by using microcapsules Chang and coworkers conducted a series of studies for the encapsulation of polymer-attached cofactors, especially NAD+, with native enzymes entrapped with ultra thin semi-permeable membranes (Campbell and Chang, 1976; Grunwald and Chang, 1979, 1981; Chang, 1985; 1987; Wahl and Chang, 1987; Gu and Chang, 1988a,b, 1990a,b,c). For examples, microcapsules of cellulose nitrate membranes have been prepared by using ether as the solvent and Tween-20 as an oil/ water emulsifier; while polyamide membrane microcapsules have prepared by interfacial polymerization of terephthaloyl and diamine-PEI (Chang, 1985, 1987; Chang and Prakash, 2001). In their work, dextran seems to be the most popularly employed modifier for cofactors (Grunwald and Chang, 1979; Gu and Chang, 1988a, 1990a,b,c), while PEI (Campbell and Chang, 1976), PEG (Stengelin and Patel, 2000), albumin and hemoglobin (Wahl and Chang, 1987) were also used. Among the several multi-enzyme systems that have been encapsulated, YADH/MDH/polymer-NAD + (Campbell and Chang, 1976; Grunwald and Chang, 1979; Wahl and Chang, 1987) and urease/LeuDH/ dextran-NAD+ (Gu and Chang, 1990a,b,c) systems have been studied extensively. Following a similar strategy, urease/GLDH/YADH/dextran-NAD+ (Gu and Chang, 1988a) and phenylalanine dehydrogenases(PheDH)/ FDH/PEG-NADH (Stengelin and Patel, 2000) systems were also examined. Unfortunately, some encapsulated multi-enzyme systems did not show good stability. Cellulose nitratecapsulated YADH, MDH and dextran-NAD+ only retain 31% of its original activity after being stored for 7 days (Grunwald and Chang, 1979). Interestingly, it was found that the addition of protein (typically, 10 g/dL purified hemoglobin) in the capsules could stabilize the enzyme systems. It was believed that the added protein provided

376


an intracellular environment that is comparable to red blood cells (Chang, 1985). Addition of 10% PEI also showed a certain stabilizing effect (Chang, 1987). Further stabilization could be obtained by cross-linking with glutaraldehyde (GA) (Chang, 1971, 1985, 1987; Grunwald and Chang, 1979). 2.2.4. Retention of chemically modified cofactors on electrodes Polymer-conjugated cofactors were also often immobilized on electrodes for biosensors. For example, PEGNAD+ was entrapped in a photo-polymerized poly(vinyl alcohol) (PVA) matrix with FDH, salicylate hydroxylase (SHL) between a dialysis membrane and a gas permeable PE membrane on an electrode for the determination of formate. PEG-NAD+ could be effectively regenerated by SHL using sodium salicylate and oxygen as the substrates. However, the working stability was limited to 7 days due to the inactivation of the enzymes (Scheper et al., 1987). In another report, dextran-NAD+ was entrapped with NADH oxidase and LDH in a photo cross-linked PVA bearing styrylpyridinium groups on a platinum electrode and was then covered with a dialysis membrane of cut-off Mw of 10,000. This sensor was reported to function for months for the detection of d-lactate (Montagn and Marty, 1995). An ethanol biosensor was also developed following a similar procedure (Leca and Marty, 1997a, b). A polyurethane hydrogel layer deposited on the electrode was also investigated for retaining PEGNADH and ADH followed by the covering of a dialysis membrane. The regeneration of PEG-NADH was catalyzed by a series of the qiunonoid cationic redox dyes immobilized in graphite-epoxy composite electrodes (Grndig et al., 1995). 3. Solid-phase cofactors Membrane entrapment is faced with the technical dilemma of balancing the retention of catalyst system and the feeding of substrates. That is probably a dilemma for any immobilization strategies. Solidimmobilized catalysts do not need membranes, but basically has to face the same problem, although the overall performance of the reactors can be leveraged via a different set of factors. Nevertheless, solid supportattached insoluble cofactors are probably easier to reuse and may afford more flexible reactor design as compared to dissolved free cofactors, and for that, it has been explored about as vigorously as membrane entrapment. Both physical and chemical approaches have been developed for solid-phase cofactors.Fig. 6. Configurations of physical attachment of cofactors to insoluble solid supports. (a) Surface-adsorbed cofactors; (b) cofactor entrapment in porous materials; (c) entrapment in cross-linked polymeric networks.

3.1. Physically immobilized cofactors 3.1.1. Physical attachment of cofactor for organic synthesis Cofactors physically immobilized to insoluble solids have been used by many researchers for organic synthesis. Similar to enzyme immobilization, cofactors can be absorbed to the outer surface of solid supports, entrapped inside porous materials, or contained in crosslinked polymeric networks (Fig. 6). Surface adsorption may work best in nonaqueous reaction medium where the cofactor does not have a good solubility. It was shown that NAD(H) along with horse liver ADH (HLADH) coated onto glass beads and were then used as catalysts for the asymmetric reduction of 2-phenylpropionaldehyde and 2-chlorocyclohexanone to their corresponding ketones in a water-immiscible organic solvent such as isopropyl ether and ethyl acetate (Grunwald et al., 1986). The cofactor was regenerated by the oxidation of ethanol or the reduction of isobutyaldehyde with TTN of up to 106. Parida et al. employed partially hydrated porous silica particles (with a water content corresponding to 70% of pore volume) for the same enzyme/cofactor system, and it was found that the system catalyzed the reduction of 2-methylvaleraldehyde with a TTN of 3.4 105 and a reaction rate that was about 6-fold higher than that with nonporous glass beads (Parida et al., 1992). Wong et al. entrapped enzymes and cofactor in the matrix of a cross-linked neutral polyester, XAD-8. The entrapped enzymes and cofactor was then suspended in an organic solvent such as butyl acetate containing the substrates for the synthesis of chemicals. In one example, coentrapped HLADH and NADH catalyzed the production of l-lactaldehyde dimethyl acetal from pyruvaldehyde dimethylacetal with cyclohexanol as a co-substrate for cofactor regeneration. A TTN of 80 was achieved in 5 days (Wong, 1986). Recently, NADPH and ADH from L. kefir (LKADH) were entrapped in PVA gel beads and were used to transform a number of hydrophobic


ketones to their corresponding enantiomerically pure (R)-alcohols. Cofactor regeneration was achieved within the beads with isopropanol as the co-substrate catalyzed by the same enzyme. In the synthesis of (R)-phenylethanol from acetophenone, the TTN of the cofactor was reported to be up to 103 (De Temio et al., 2005). 3.1.2. Physical attachment of cofactors on electrodes Cofactors have been physically immobilized onto biochemical electrodes through various methods. The method that applied membranes for retention of free cofactors has been discussed in Section 2. In those cases, the cofactors are still mobile locally once its surrounding environment is filled with water. Here we review the methods that apply cofactors physically attached to a solid-phase support that is insoluble in a solution. Surface-attachment is one of the popular methods. For certain disposable biosensor, enzyme and cofactors can be physically adsorbed onto the surface of electrodes (usually modified to enhance their affinity to enzymes and cofactors) and thus construct simple biosensors. By a simple drop-coating procedure, GLDH, 2-oxoglutarate and NADH were coated on an MBdoped screen-printed carbon electrode (SPCE), and the + resulting device could be used as a disposable NH4 biosensor, which could keep active for 29 days when stored at 4 C in a desiccator (Hart et al., 1999). Similarly, biosensors using NAD+ and corresponding enzymes for lactate (Sprules et al., 1996) and blood ketones (Li et al., 2005) were reported. Nakamura et al. (1996) applied PEI-ferrocene on the surface of graphite carbon electrode first, then adsorbed glutathione reductase (GR), G6PDH, and sodium alginate modified NADP+ (Alg-NADP+) onto the electrode surface. Matrix-entrapment is another popular physical immobilization method. The typical strategy is to absorb enzymes and cofactors onto solid powders, and then form a matrix by physical compression or chemical reactions. Carbon powder has been used extensively in the construction of enzyme electrodes. Typically, enzymes, cofactors, as well as mediators can be mixed into a carbon paste with the aid of pasting liquids, and thereby the paste matrix can act as a reservoir of cofactors. This approach offers good flexibility of incorporating cofactors, enzymes and other possible components (Mello and Kubota, 2002). Electron transfer mechanism in such modified carbon paste electrodes is not quite clear yet (Habermller et al., 2000). Nevertheless, cofactors contained in the paste were generally still mobile and could diffuse away, leading to short lifetime of biosensors (Boujtita and El

Murr, 1995). For example, a glucose biosensor based on carbon paste with GDH, NAD+ and a redox mediator retained only 10% of its original activity after 1 day (Gorton et al., 1991). To overcome such a drawback, a polymeric membrane was usually applied to maintain the integrity of the paste and prevent loss of cofactors and mediators. Cation-exchanging poly(ester-sulfonic acid) (Bremle et al., 1991; Persson et al., 1993) and charged poly(vinyl pyridine) (Fernndez et al., 1998) films were applied for glucose and ethanol sensors. Membranes bearing negative charges generally do not work well for the biosensors detecting negatively charged substrates such as d-lactic acid under certain pH condition due to electrostatic expulsion (Bremle et al., 1991). In addition, membrane coverage of carbon pastes usually led to significant loss of the sensitivity of the sensors due to the great mass transfer resistance of the membranes (Chi and Dong, 1994; Shu et al., 1995). Various other methods have also been developed to keep the carbon powder paste integral on electrodes. Polyelectrolytes, such as PEI, were applied to hold enzymes and cofactors together within carbon powder pastes (Domnguez et al., 1993). Cross-linking of enzymes can stabilize the enzymes significantly, and help to prevent leakage. It was shown that a carbon paste electrode consisting NAD+ entrapped within a gel formed by cross-linking ADH and bovine serum albumin (BSA) with GA retained 95% of its original activity after 7 months (Boujtita and El Murr, 1995). More recently, this method was applied to a lactate biosensor (Pereira et al., 2006). Formation of powder polymer composites is another approach to stabilize the enzyme/cofactor matrix (Katrlk et al., 1998, 1999). Reviews on the application of rigid carbon-polymer composite materials in electrochemical sensing are available (Alegret, 1996; Cespedes and Alegret, 2000). 3.2. Chemical incorporation of cofactors Tethering cofactors to solid supports can effectively prevent leakage of cofactors. However, chemical modifications may lead to poor cofactor activity as it may alter the binding affinities of cofactors to the active sites of enzymes. In one of the earlier works, it was demonstrated that NAD+ immobilized on diazotized glass beads exhibited redox activity with YADH for the production of acetaldehyde from ethanol. Regeneration of NAD+ was achieved by using acetaldehyde with the same enzyme, however, TTN of NAD+ was only around 1 (Weibel et al., 1971). NAD+-N6-[N-(6-aminohexyl)acetamide] was attached to Sepharose 4B through cyanogen bromide group and showed an activity with

378


Fig. 7. Chemical attachment of enzyme and cofactor on the surface of nylon tube.

free YADH or LDH that was less than 1% of the free cofactor (Larsson and Mosbach, 1971). Regeneration of such immobilized cofactor could be achieved with LDH using pyruvate as the substrate (Lindberg et al., 1973). While most of the solid-attached cofactors were active with free enzymes, it has been difficult to achieve activity with systems that have cofactor and enzymes both immobilized. Wykes et al. reported that Agaroseattached NAD+ was active with free ADH and LDH, whereas no activity was observed with ADH and LDH immobilized separately on DEAE-cellulose matrices (Wykes et al., 1975). NADH and HLADH coimmobilized on Sepharose 4B showed certain activity, but the immobilized cofactor could not be regenerated either enzymatically with a pyruvate/LDH system or chemically due to steric hindrance (Gestrelius et al., 1975). Mazid et al. co-immobilized NAD+ and YADH on the surface of nylon tube (Fig. 7) with the cofactor was regenerated chemically in the presence of an indophenol dye (Mazid and Laidler, 1982). It was claimed that the enzyme retained about 10% of its activity. Success in regeneration of co-immobilized cofactors was achieved by using a coupled-substrate regeneration strategy with single-enzyme systems. HLADH and

NAD(H) was covalently incorporated into cross-linked albumin porous particles by GA cross-linking and used for the oxidation of alcohols, the cofactor was regenerated by supplying different substrates (Pulvin et al., 1986; Lortie et al., 1989). Similarly, cross-linked HLADHNADH crystalline complex was used in the production of cinnamyl alcohol from cinnamaldehyde, and ethanol was supplied as a co-substrate for cofactor regeneration (Lee et al., 1986). Recently, the same enzyme and cofactor complex was applied to the reduction of 6-methyl-5-hepten-2-one and several ketones with isopropanol replacing ethanol as the cofactor regenerating substrate (Clair et al., 2000). The single enzymecofactor strategy was applied for development of biosensors where the cofactor can be regenerated electrochemically. An NAD+-analog was co-immobilized with LDH on the surface of a porous glass carbon electrode to give an amperometric lactate biosensor (Khayyami et al., 1996). NAD+ was regenerated by direct electrochemical oxidation or with a soluble mediator, MB. In another work, N6-(2-aminoethyl)-NAD+ was covalently attached to the pyrroloquinoline quinone (PQQ)-monolayer-functionalized electrode, and then NAD+-dependent LDH or ADH was associated to the PQQ-NAD+ monolayer through

Fig. 8. Chemical attachment procedures of enzyme and cofactor on pyrroloquinoline quinone-functionalized electrode.


Fig. 9. Chemical incorporation of enzymes and cofactor into nano-porous silica glass. Although being tethered to the wall of silica glass pores, enzymes and co-enzyme can still share interactions via thermal vibration due to the close vicinity within the nanopores.

affinity interactions. Upon cross-linking, an integrated electrically-wired enzyme electrode was constructed (Fig. 8). The cross-linked enzyme layer exhibited high stability for alcohol detection (Bardea et al., 1997). In the development of glucose sensors, cofactor FAD was first wired to electrode via conductive tethers followed by affinity binding of apo-glucose oxidase (Katz et al., 1997; Willner et al., 1996; Xiao et al., 2003). A similar immobilization and regeneration strategy of cofactor was applied in an integrated NAD+-dependent enzyme field-effect transistor device for the biosensing of lactate (Zayats et al., 2000). Much less work has been reported for the regeneration of immobilized cofactors with co-immobilized multi-enzyme systems. Yamazaki and Maeda reported a work with covalently immobilized NAD but with

physically entrapped enzymes (Yamazaki and Maeda, 1982). Pre-modified NAD+ was co-polymerized into a poly(acrylamide) hydrogel matrix in the presence of native enzymes. In this way, co-immobilized multienzyme systems, FDHMDHNAD+ or HLADHDI NAD+, were constructed. Regeneration of the cofactor was achieved, and its activity could keep for several days. Ukeda et al.'s applied GA to co-cross-link NAD+, YADH and DI onto Sepharose, and a mean cycling rate of 8 h 1 was obtained for the cofactor (Ukeda et al., 1989a,b). The use of nanostructures provides a new route to realize co-immobilized multi-enzyme/cofactor bioactive systems. Recently El-Zahab et al. applied nano-porous silica glass as the support of both enzymes and cofactor NADH (Fig. 9) (El-Zahab et al., 2004). LDH and a regenerating enzyme GDH were co-

Fig. 10. Nanoparticle-driven multi-enzyme catalytic systems in form of an artificial cell. Nanoparticles carrying surface-attached enzymes and cofactor will provide enzymecofactor interactions within the porous capsule (100 m in diameter) through particle collision.

380


Table 1 Characteristics of cofactor regeneration strategies Methods of cofactor retention Membrane entrapment Free cofactor via size exclusion Concerns Advantages

4. Conclusion Biocatalysis can offer highly selective reactions, environmentally friendly processes, and energy-effective operations, as compared to traditional chemical processing. The widespread public concerns about environmental quality and energy resources have made bioprocessing increasingly desirable to many industrial applications such as chemical production, drug synthesis, environmental remediation, and fuel refining. Biosynthesis may be achieved with either whole cells or isolated enzymes. Often the cellular processes are associated with slow reactions and complicated culturing conditions. Significant advances have been made through genetic and/or metabolic engineering for industrial microbial technology; but that approach raises concerns over the release of non-natural organisms into the environment. On the other hand, enzymatic biocatalysis is faster, cleaner and simpler. The current arts of enzymatic biosynthesis are primarily for single-enzyme systems. How to apply multi-enzyme systems to carry out complex reactions, especially those involve cofactors, represents one great challenge in bioprocessing. Apparently this will also be a long-standing challenge. Although decades of research and development efforts have been made, no ideal solutions have emerged. All the strategies reviewed in this paper have mixed advantages and disadvantages. Table 1 summarizes a simple comparison of the characteristics of different methods for cofactor retention. Generally speaking, membrane entrapment has the advantage to allow the use of free cofactors that have excellent mobility and high activity, but is compromised by small membrane pore size as determined by the size of the cofactor. A similar problem is faced with solid-phase cofactors when the cofactors are physically entrapped or adsorbed. Such configurations will lead to mass transfer-limited reaction kinetics, and should be considered for reaction systems that have low intrinsic enzyme activities. Immobilization of cofactors to solid supports through covalent bonds may exhibit advantages of materials with much more open structures to minimize mass transfer resistance, but the chemical modification may significantly reduce the activity of the catalyst system. Such a strategy can lead to reactionlimited kinetics, and should be considered for catalyst systems that have high intrinsic activities. The microcapsulation of nanoparticle-attached cofactor and enzymes was designed in an attempt to combine the advantages of membrane entrapment and solid-attachment (Wang et al., 2005). It can be expected that new approaches and concepts will continue to emerge, and most probably will through the introduction of more

Free cofactor via electrostatic interaction

Chemically modified cofactor

Easy to practice; Cofactor leaching; limited reactor options; maintain the high mass transfer intrinsic activity resistance of the catalyst system pH and ionic strength Reduce mass of solution may impact transfer rates overall retention of substrates effectiveness; while maintain limited mass transfer the intrinsic of polar chemicals activity of the catalyst Activity loss of Allow the use modified cofactor of membranes of larger pores for reduced mass transfer resistance Cofactor leaching; Flexible reactor mass transfer design; easy to resistance of substrates practice; easy replacement of lost catalyst Activity loss of Flexible reactor modified cofactor ; design; may difficult to achieve stabilize the shuttling between catalyst system; two co-immobilized reduced mass enzymes transfer limitations

Solid attachment Physical adsorption or entrapment

Chemical tethering

immobilized along with the cofactor via spacers of different sizes. The coupled reactions catalyzed by LDH and GDH were achieved, indicating the immobilized NAD(H) shuttled between the two co-immobilized enzymes. It was also found that reaction rate increased when longer spacer and smaller pores were employed. It was believed that the pores helped to bring together all the catalytic components, while thermal vibration of molecules facilitated the shuttling of the cofactor between the two enzymes. In a different approach, enzymes and cofactor chemically attached onto nanoparticles were encapsulated into polymeric capsules (Wang et al., 2005). The membrane of the capsules was porous with pore size controlled to be in nanometer scale to allow substrates and products to penetrate while retain the particle-attached enzymes and cofactor. Since the nanoparticles were mobile inside the capsules when they were filled with water, collision between particles could enable interactions among the enzymes and cofactor, thus allowing multiple reactions to take place inside the capsules (Fig. 10).


sophisticated structures and materials for improved activity and retention of the cofactors. Acknowledgements The authors thank the National Natural Science Foundation of China (NSFC#20576135) and the National Knowledge Innovation Project from the Chinese Academy of Sciences (# KSCX2-YW-G-019) and 863 Project (#2006AA02Z217) for the financial support. P.W. acknowledges the support of a CAREER award from the US National Science Foundation (BES# 0348412).

ReferencesAdachi S, Ogata M, Tobita H, Hashimoto K. Effects of molecular weight of dextran and NAD+ density on cofactor activity of high molecular weight NAD+ derivative covalently bound to dextran. Enzyme Microb Technol 1984;6:25962. Alegret S. Rigid carbon-polymer biocomposites for electrochemical sensing. Rev Anal 1996;121:17518. Bardea A, Katz E, Bckmann AF, Willner I. NAD+-dependent enzyme electrodes: electrical contact of cofactor-dependent enzymes and electrodes. J Am Chem Soc 1997;119:91149. Boujtita M, El Murr N. Biosensors for analysis of ethanol in foods. J Food Sci 1995;60:2014. Bremle G, Persson B, Gorton L. An amperometric glucose electrode based on carbon paste, chemically modified with glucose dehydrogenase, nicotinamide denine dinucleotide, and a phenoxazine mediator, coated with a poly(ester sulfonic acid) cation exchanger. Electroanalysis 1991;3:7786. Bckmann AF, Kula MR, Wichmann R, Wandrey C. An efficient synthesis of high-molecular-weight NAD(H) derivatives suitable for continuous operation with cofactor-dependent enzyme systems. J Appl Biochem 1981;3:30115. Campbell J, Chang TMS. The recycling of NAD+ (frese and immobilized) within semipermeable aqueous microcapsules containing a multi-enzyme system. Biochem Biophys Res Commun 1976;69:5629. Cespedes F, Alegret S. New materials for electrochemical sensing II. Rigid carbon-polymer biocomposites. Trends Anal Chem 2000;19. Chang TMS. Stabilisation of enzymes by microencapsulation with a concentrated protein solution or by microencapsulation followed by cross-linking with gutaraldehyde. Biochem Biophys Res Commun 1971;44:15316. Chang TMS. Artificial cells containing multienzyme systems. Methods Enzymol 1985;112:195203. Chang TMS. Recycling of NAD(P) by multienzyme systems immobilized by microencapsulation in artificial cells. Methods Enzymol 1987;136:6783. Chang TMS, Prakash S. Procedures for microencapsulation of enzymes, cells and genetically engineered microorganisms. Mol Biotechnol 2001;17:24960. Chang TMS, Yu YT, Grunwald J. Artificial cells immobilized multienzyme systems and cofactors. Enzyme Eng 1982;6:451561. Chenault HK, Whitesides GM. Regeneration of nicotinamide cofactors for use in enzymatic synthesis. Appl Biochem Biotechnol 1987;14:14797.

Chenault HK, Simon ES, Whitesides GM. Cofactor regeneratioin for enzyme-catalysed synthesis. Biotechnol Genet Eng Rev 1988;6:22170. Chi QJ, Dong SJ. Electrocatalytic oxidation of reduced nicotinamide cofactors at methylene green-modified electrodes and fabrication of amperometric alcohol biosensors. Anal Chim Acta 1994;285:12533. Clair NS, Wang Y-F, Margolin AL. Cofactor-bound cross-linked enzyme crystals (CLEC) of alcohol dehydrogenase. Angew Chem Int Ed Engl 2000;39:3803. Cosnier S. Biosensors based on electropolymerized films: new trends. Anal Bioanal Chem 2003;377:50720. Davies P, Mosbach K. The application of immobilized NAD+ in an enzyme electrode and in model enzyme reactors. Biochim Biophys Acta 1974;370:32938. De Temio DM-R, Hartmeier W, Ansorge-Schumacher MB. Entrapment of the alcohol dehydrogenase from Lactobacillus kefir in polyvinyl alcohol for the synthesis of chiral hydrophobic alcohols in organic solvents. Enzyme Microb Technol 2005;36:39. Domnguez E, Lan HL, Okamoto Y, Hale PD, Skotheim TA, Gorton L, et al. Reagentless chemically modified carbon paste electrode based on a phenothiazine polymer derivative and yeast alcohol dehydrogenase for the analysis of ethanol. Biosens Bioelectron 1993;8:22937. Eguchi T, Iizuka T, Kagotani T, Lee JH, Urabe I, Okada H. Covalent linking of poly(ethyleneglycol)-bound NAD with Thermus thermophilus malate dehydrogenase NAD(H)-regeneration unit for a coupled second-enzyme reaction. Eur J Biochem 1986;155: 41521. El-Zahab B, Jia H, Wang P. Enabling multienzyme biocatalysis using nanoporous materials. Biotechnol Bioeng 2004;87:17883. Fernndez JJ, Lpez JR, Correig X, Katakis I. Reagentless carbon paste phosphate biosensors: preliminary studies. Sens Actuators B Chem 1998;47:1320. Filho MV, Stillger T, Mller M, Liese A, Wandrey C. Is log Pa convenient criterion to guide the choice of solvents for biphasic enzymatic reactions? Angew Chem Int Ed Engl 2003;42:29936. Findrik Z, Vasic'-Racki D, Lutz S, Daussmann T, Wandrey C. Kinetic modeling of acetophenone reduction catalyzed by alcohol dehydrogenase from Thermoanaerobacter sp. Biotechnol Lett 2005;27:108795. Fuller CW, Rubin JR, Bright HJ. A simple procedure for covalent immobilization of NADH in a soluble and enzymically active form. Eur J Biochem 1980;103:42130. Furukawa S, Katayama N, IZUKA T, Urabe I, Okada H. Preparation of polyethylene glycol-bound NAD and its application in a model enzyme reactor. FEBS Lett 1980;121:23942. Gacesa P, Venn RF. The preparation of stable enzymecofactor complexes with endogenous catalytic activity. Biochem J 1979;177:36972. Gestrelius S, Mansson M, Mosbach K. Preparation of an alcoholdehydrogenase-NAD(H)-sepharose complex showing no requirement of soluble cofactor for its activity. Eur J Biochem 1975;57:52935. Gilmartin MAT, Hart JP. Sensing with chemically and biologically modified carbon electrodes. Rev Anal 1995;120:102945. Gorton L, Bremle G, Csregi E, Jnsson-Pettersson G, Persson B. Amperometric glucose sensors based on immobilized glucoseoxidizing enzymes and chemically modified electrodes. Anal Chim Acta 1991;249:4354. Gotoh M, Karube I. Ethanol biosensor using immobilized cofactor. Anal Lett 1994;27:27384. Goulas P. Covalent binding of an NAD+ analogue to horse liver alcohol dehydrogenase in a ternary complex with pyrazole. Eur J Biochem 1987;168:46973.

382

http://www.paper.edu.cnW. Liu, P. Wang / Biotechnology Advances 25 (2007) 369384 Katrlk J, Pizzariello A, Mastihuba V, vorc J, Streansk M, Miertu S. Biosensors for l-malate and l-lactate based on solid binding matrix. Anal Chim Acta 1999;379:193200. Katz E, Heleg-Shabtai V, Willner B, Willner I, Bckmann AF. Electrical contact of redox enzymes with electrodes: novel approaches for amperometric biosensors. Bioelectrochem Bioenerg 1997;42:95104. Khayyami M, Garcia NP, Larsson PO, Danielsson B, Johansson G. An amperometric lactate sensor based on a NAD+-analog and lactate dehydrogenase coimmobilized on reticulated vitreous carbon. Electroanalysis 1996;9:5236. Kitpreechavanich V, Nishio N, Hayashi M, Nagai S. Regeneration and retention of NADP(H) for xylitol production in an ionized membrane reactor. Biotechnol Lett 1985;7:65762. Kula MR, Kragl U. Dehydrogenases in synthesis of chiral compounds. In: Ramesh NP, editor. Stereoselective Biocatalysis. New York: Marcel Dekker; 2000. p. 83966. Kulbe KD, Howaldt MW, Schmidt K, Rthig TR, Chmiel H. Rejection and continuous regeneration of the native cofactor NAD(P)H in a charged ultrafiltration membrane enzyme reactor. Ann NY Acad Sci 1990;613:8206. Larsson PO, Mosbach K. Preparation of a NAD(H)-polymer matrix showing coenzymic function of the bound pyridine nucleotide. Biotechnol Bioeng 1971;13:3938. Larsson PO, Mosbach K. The preparation and characterisation of a water-soluble coenzymically active dextran-NAD+. FEBS Lett 1974;46:11922. LeBrun LA, Plapp BV. Control of cofactor binding to horse liver alcohol dehydrogenase. Biochemistry 1999;38:1238793. Leca B, Marty JL. Reagentless ethanol sensor based on a NADdependent dehydrogenase. Biosens Bioelectron 1997a;12:10838. Leca B, Marty JL. Reusable ethanol sensor based on a NAD+dependent dehydrogenase without cofactor addition. Anal Chim Acta 1997b;340:1438. Lee KM, Blaghen M, Samama JP, Biellmann JF. Crosslinked crystalline horse liver alcohol dehydrogenase as a redox catalyst: activity and stability toward organic solvent. Bioorg Chem 1986;14:20210. Leonida MD. Redox enzymes used in chiral syntheses coupled to cofactor regeneration. Curr Med Chem 2001;8:34569. Leskovac V, Trivi S, Periin D. Isomerization of an enzymecofactor complex in yeast alcohol dehydrogenase-catalyzed reactions. J Serb Chem Soc 2003;68:7784. Li G, Ma NZ, Wang Y. A new handheld biosensor for monitoring blood ketones. Sens Actuators B Chem 2005;109:28590. Liese A, Seel bach K, Wandrey C. Industrial biotransformations. Weinheim: Wiley-VCH; 2000. Lin SS, Harada T, Hata C, Miyawaki O, Nakamura K. Nanofiltration membrane bioreactor for continuous asymmetric reduction of 2ketoglutarate to produce l-glutamate with NADH regeneration. J Ferment Bioeng 1997;83:548. Lin SS, Miyawaki O, Nakamura K. Continuous production of lcarnitine with NADH regeneration by a nanofiltration membrane reactor with coimmobilized l-carnitine dehydrogenase and glucose dehydrogenase. J Biosci Bioeng 1999;87:3614. Lindberg M, Larsson PO, Mosbach K. A new immobilized NAD+ analogue, its application in affinity chromatography and as a functioning cofactor. Eur J Biochem 1973;40:18793. Lortie R, Villaume I, Legoy M, Thomas D. Enzymatic production of long-chain aldehydes in a fixed bed reactor using organic solvents and cofactor regeneration. Biotech Bioeng 1989;33:22932. Lowe CR. Immoblised cofactors. Trends Biochem Sci 1978;3:1347. Maines A, Prodromidis MI, Tzouwara-Karayanni SM, Karayannis MI, Ashworth D, Vadgama P. Reagentless enzyme electrode for

Grndig B, Wittstock G, Rdel U, Strehlitz B. Mediator-modified electrodes for electrocatalytic oxidation of NADH. J Electroanal Chem 1995;395:14357. Grunwald J, Chang TMS. Continuous recycling of NAD+ using an immobilized system of collodion microcapsules containing dextran-NAD+, alcohol dehydrogenase, and malic dehydrogenase. J Appl Biochem 1979;1:10414. Grunwald J, Chang TMS. Immobilization of alcohol dehydrogenase, malic dehydrogenase and dextran-NAD+ within nylon-polyethyleneimine microcapsules: preparation and cofactor recylcing. J Mol Catal 1981;11:8390. Grunwald J, Wirz B, Scollar MP, Klibanov AM. Asymmetric oxidoreductions catalyzed by alcohol dehydrogenase in organic solvents. J Am Chem Soc 1986;108:67324. Gu KF, Chang TMS. Conversion of -ketoglutarate into l-glutamic acid with urea as ammonium source using multienzyme systems and dextran-NAD+ immobilized by microencapsulation within artificial cells in a bioreactor. Biotechnol Bioeng 1988a;32:3638. Gu KF, Chang TMS. Conversion of ammonia or urea into l-leucine, lvaline and l- isoleucine using artificial cell immobilising multienzyme system and dextran-NADH+. I. Glucose dehydrogenase for cofactor recycling. Trans Am Soc Artif Intern Organs 1988b;11:248. Gu KF, Chang TMS. Conversion of ammonia or urea into essential amino acids, l-leucine, l-valine, and l-isoleucine using artificial cells containing an immobilized multienzyme system and dextranNAD+. IV. Malate dehydrogenase for cofactor recycling. J Catal 1990a;62:3319. Gu KF, Chang TMS. Conversion of ammonia or urea into essential amino acids, l-leucine, l-valine, and l-isoleucine, using artificial cells containing an immobilized multienzyme system and dextranNAD+ 2. Yeast alcohol dehydrogenase for cofactor recycling. Biotechnol Appl Biochem 1990b;12:22736. Gu KF, Chang TMS. Production of essential l-branched-chain ammo acids in bioreactors containing artificial cells immobilized multienzyme systems and dextran-NAD+. Biotechnol Bioeng 1990c;36:2639. Habermller K, Mosbach M, Schuhmann W. Electron-transfer mechanisms in amperometric biosensors. Fresenius J Anal Chem 2000;366:5608. Hart JP, Abass AK, Cowell DC, Chappell A. Development of a disposable amperometric NH biosensor based on a chemically modified screenprinted carbon electrode coated with glutamate dehydrogenase, 2-oxoglutarate, and NADH. Electroanalysis 1999;11:40611. Howaldt M, Kulbe KD, Chmiel H. A continuous enzyme membrane reactor retaining the native nicotinamide cofactor NAD(H). Ann NY Acad Sci 1990;589:25360. Hummel W. Large-scale applications of NAD(P)-dependent oxidoreductases: recent developments. Trends Biotechnol 1999;17: 48792. Ikemi M, Ishimatsu Y, Kise S. Sorbitol production in charged membrane bioreactor with cofactor regeneration system: II. Theoretical analysis of a continuous reaction with retained and regenerated NADPH. Biotechnol Bioeng 1990a;36:15565. Ikemi M, Koizumi N, Ishimatsu Y. Sorbitol production in charged membrane bioreactor with cofactor regeneration system: I. Selective retainment of NADP(H) in a continuous reaction. Biotechnol Bioeng 1990b;36:14954. Ilan E, Chang TMS. Lipidpolyamidepolyethyleneimine microcapsules for immobilization of free cofactors and multienzyme system. Appl Biochem Biotechnol 1986;13:22130. Katrlk J, vorc J, Streansk M, Miertu S. Composite alcohol biosensors based on solid binding matrix. Biosens Bioelectron 1998;13:18191.


malate based on modified polymeric membranes. Anal Chim Acta 2000;408:21724. Mano N, Kuhn A. Molecular lego for the assembly of biosensing layers. Talanta 2005;66:217. Mnsson MO, Larsson PO, Mosbach K. Covalent binding of an NAD analogue to liver alcohol dehydrogenase resulting in an enzyme cofactor complex not requiring exogenous cofactor for activity. Eur J Biochem 1978;86:45563. Mnsson MO, Larsson PO, Blow L, Mosbach K. Continuous regenration of NAD(H) covalently bound to a cysteine genetically engineered into glucose dehydrogenase. Bio/Technology 1991;9:2804. Mazid MA, Laidler KJ. Kinetics of yeast alcohol dehydrogenase and its cofactor coimmobilized in a tubular flow reactor. Biotechnol Bioeng 1982;24:208797. Mello LD, Kubota LT. Review of the use of biosensors as analytical tools in the food and drink industries. Food Chem 2002;77: 23756. Montagn M, Marty JL. Bi-enzyme amperometric d-lactate sensor using macromolecular NAD+. Anal Chim Acta 1995;315:297302. Nakamura A, Urabe I, Okada H. Anchimeric assistance in the intramolecular reaction of glucose-dehydrogenasepolyethylene glycol NAD conjugate. J Biol Chem 1986;261:167924. Nakamura Y, Suye S, Kira J, Tera H, Tabata I, Senda M. Electrontransfer function of NAD+-immobilized alginic acid. Biochim Biophys Acta 1996;1289:2215. Nidetzky B, Haltrich D, Kulbe KD. Carry out cofactor conversions economically. Chemtech 1996;26:316. Nidetzky B, Schmidt B, Neuhauser W, Haltrich D, Kulbe KD. Application of charged ultrafiltration membranes in continuous, enzyme-catalyzed processes with cofactor regeneration. In: Pyle DL, editor. Separation for biotechnology, 3. Cambridge: Royal Society of Chemistry; 1994. p. 3517. Noguer T, Marty JL. Reagentless sensors for acetaldehyde. Anal Lett 1997;30:106980. Obn JM, Almagro MJ, Manjn A, Iborra JL. Continuous retention of native NADP(H) in an enzyme membrane reactor for gluconate and glutamate production. J Biotechnol 1996;50:2736. Obn JM, Manjn A, Iborra JL. Retention and regeneration of native NAD (H) in noncharged ultrafiltration membrane reactors: application to llactate and gluconate production. Biotechnol Bioeng 1998;57:5107. Okuda K, Urabe I, Okada H. Synthesis of poly(ethylene glycol)-bound NADP by selective modification at the 6-amino group of NADP. Eur J Biochem 1985;151:338. Orlich B, Berger H, Lade M, Schomcker R. Stability and activity of alcohol dehydrogenases in W/O-microemulsions: enantioselective reduction including cofactor regeneration. Biotechnol Bioeng 2000;70:63846. Orlich B, Schomcker R. Enzymatic reduction of a less water-soluble ketone in reverse micelles with NADH regeneration. Biotechnol Bioeng 1999;65:35762. Orlich B, Schomcker R. Enzyme catalysis in reverse micelles. Advances in Biochemical Engineering, BiotechnologyBerlin: Springer/Heidelberg; 2001. p. 185208. Ottolina G, Carrea G, Riva S, Bckmann A. Coenzymatic properties of low molecular-weight and macromolecular N6-derivatives of NAD+ and NADP+ with dehydrogenases of interest for organic synthesis. Enzyme Microb Technol 1990;12:596602. Parida S, Datta R, Dordick JS. Supported aqueous-phase enzymatic catalysis in organic media. Appl Biochem Biotechnol 1992;33:114. Pereira AC, Macedo DV, Santos AS, Kubota LT. Amperometric biosensor for lactate based on Meldola's blue adsorbed on silica gel modified with niobium oxide. Electroanalysis 2006;18:120814.

Persson B, Lan HL, Gorton L, Okamoto Y, Hale PD, Boguslavsky LI, et al. Amperometric biosensors based on electrocatalytic regeneration of NAD+ at redox polymer-modified electrodes. Biosens Bioelectron 1993;8:818. Peters J. 9 Dehydrogenasescharacteristics, design of reaction conditions, and applications. In: Rehm HJ, Reed G, editors. Biotechnology: A multi volume comprehensive treatise, vol. 8a. Weinheim: Wiley-VCH; 1999. p. 391460. Pulvin S, Legoy MD, Lortie R, Pensa M, Thomas D. Enzyme technology and gas phase catalysis: alcohol dehydrogenase example. Biotechnol Lett 1986;8:7834. Riva S, Carrea G, Veronese FM, Bckmann AF. Effect of coupling site and nature of the polymer on the coenzymatic properties of watersoluble macromolecular NAD derivatives with selected dehydrogenase enzymes. Enzyme Microb Technol 1986;9:55660. Rthig TR, Kulbe KD, Bckmann F, Carrea G. Continuous cofactor dependent stereoselective synthesis of sulcatol by alcohol dehydrogenase. Biotechnol Lett 1990;12:3536. Schafer HG, Jacobi T, Eichhorn H, Woenckhaus C. Covalent fixation of NAD+ to dehydrogenases and properties of the modified enzymes. Biol Chem Hoppe Seyler 1986;367:96980. Scheper T, Likidis Z, Makryaleas K, Nowottny C, Schgerl K. Three different examples of enzymatic bioconversion in liquid membrane reactors. Enzyme Microb Technol 1987;9:62531. Sekhar VC, Plapp BV. Mechanism of binding of horse liver alcohol dehydrogenase and nicotinamide adenine dinucleotide. Biochemistry 1988;27:50828. Sheldon WM, Stephen RP. Oxidoreductase enzymes in biotechnology: current status and future potential. Bio/Technology 1983;1:67786. Shu HC, Mattiasson B, Persson B, Nagy G, Gorton L, Sahni S, et al. A reagentless amperometric electrode based on carbon paste, chemically modified with d-lactate dehydrogenase, NAD+, and mediator containing polymer for d-lactic acidanalysis. I. Construction, composition, and characterization. Biotechnol Bioeng 1995;46:2709. Sprules SD, Hart JP, Wring SA, Pittson R. A reagentless, disposable biosensor for lactic acid based on a screen-printed carbon electrode containing Meldola's Blue and coated with lactate dehydrogenase, NAD+ and cellulose acetate. Anal Chim Acta 1995;304:1724. Sprules SD, Hart JP, Pittson R, Wring SA. Evaluation of a new disposable screen-printed sensor strip for the measurement of NADH and its modification to produce a lactate biosensor employing microliter volumes. Electroanalysis 1996;8:53943. Srere PA, Mattiasson B, Mosbach K. An immobilized three-enzyme system: a model for microenvironmental compartmentation in mitochondria. Proc Natl Acad Sci U S A 1973;70:25348. Stengelin M, Patel RN. Phenylalanine dehydrogenase catalyzed reductive amination of 6-(1,3-dioxolan-2-yl)-2-keto-hexanoic acid to 6-(1,3-dioxolan-2-yl)-2S-aminohexanoic acid with NADH regeneration and enzyme and cofactor retention. Biocatal Biotransform 2000;18:373400. Ukeda H, Imabayashi M, Matsumoto KM, Osajima Y. The coimmobilization of alcohol dehydrogenase, diaphorase, and NAD on glutaraldehyde-activated Sepharose. Agric Biol Chem 1989a;53: 22635. Ukeda H, Imabayashi M, Matsumoto KM, Yutaka O. Co-immobilization of alcohol dehydrogenase, diaphorase and NAD and its application to flow injection analytical system for ethanol. Agric Biol Chem 1989b;53:290915. Vanhommerig SA, Sluyterman LA, Meijer EM. Kinetic and modelling studies of NAD+ and poly(ethylene gycol)-bound in horse liver alcohol dehydrogenase. Biochim Biophys Acta 1996;1295: 12538.

384

http://www.paper.edu.cnW. Liu, P. Wang / Biotechnology Advances 25 (2007) 369384 Yamazaki Y, Maeda H. The co-immobilization of NAD and dehydrogenases and its application to bioreactors for synthesis and analysis. Agric Biol Chem 1982;46:157181. Yamazaki Y, Maeda H, Suzuki H. Application of polylysine bound succinyl-NA to a membrane reactor. Biotechnol Bioeng 1976;18: 176175. Yu YT, Chang TMS. Lipidpolymer membrane artificial cells containing. multienzyme systems, cofactors and substrates for the removal of ammonia and urea. Trans Am Soc Artif Intern Organs 1981a;27: 5358. Yu YT, Chang TMS. Ultrathin lipidpolymer membrane microcapsules containing multienzymes, cofactors and substrates for multistep enzyme reactions. FEBS Lett 1981b;125:946. Yu YT, Chang TMS. Immobilization of multienzymes and cofactors within lipidpolyamide membrane microcapsules for the multistep conversion of lipophilic and lipophobic substrates. Enzyme Microb Technol 1982;4:32731. Zappelli P, Rossodivita A, Re L. Synthesis of coenzymically active soluble and insoluble macromolecularized NAD+ derivatives. Eur J Biochem 1975;54:47582. Zappelli P, Rossodivita A, Prosperi G, Pappa R, Re L. New coenzymicallyactive soluble and insoluble macromolecular NAD+ derivatives. Eur J Biochem 1976;62:2115. Zappelli P, Pappa R, Rossodivita A, Re L. Preparation of coenzymic activity of soluble polyethyleneimine-bound NADP+ derivatives. Eur J Biochem 1977;72:30915. Zayats M, Kharitonov AB, Katz E, Bckmann AF, Willner I. An integrated NAD+-dependent enzyme-functionalized field-effect transistor (ENFET) system: development of a lactate biosensor. Biosens Bioelectron 2000;15:67180.

Vidal J-C, Garcia-Ruiz E, Castillo J-R. Recent advances in electropolymerized conducting polymers in amperometric biosensors. Microchim Acta 2003;143:93111. Wahl HP, Chang TMS. Recycling of NAD+ crosslinked to albumin or hemoglobin immobilized with multienzyme systems in artificial cells. J Mol Catal 1987;39:14754. Wang J, Chen Q, Pedrero M, Pingarrbn JM. Screen-printed amperometric biosensors for glucose and alcohols based on ruthenium-dispersed carbon inks. Anal Chim Acta 1995;300:1116. Wang P, Ma G, Liao L, Gao F. Construction of multienzyme bioactive systems using a multiscale design approach. China Particuology 2005;3(36):3049. Weibel MK, Weetall HH, Bright HJ. Insolubilized cofactor: the covalent coupling of enzymatically active NAD to glass surfaces. Biochem Biophys Res Commun 1971;44:34752. Wichmann R, Vasic-Racki D. Cofactor regeneration at the lab scale. Adv Biochem Eng Biotechnol 2005;92:22560. Willner I, Heleg-Shabtai V, Blonder R, Katz E, Tao G, Buckmann AF, et al. Electrical wiring of glucose oxidase by reconstitution of FAD-modified monolayers assembled onto Au-electrodes. J Am Chem Soc 1996;118:103212. Wong CH. Nicotinamide cofactor-requiring enzymatic synthesis in organic solventwater biphasic systems. Biocatalysis in organic media. Wageningen: The Netherlands; 1986. p. 197208. Wykes JR, Dunnill P, Lilly MD. Cofactor recycling in an enzyme reactor. A comparison using free and immobilized dehydrogenases with free and immobilized NAD. Biotechnol Bioeng 1975;17: 5168. Xiao Y, Patolsky F, Katz E, Hainfeld JF, Willner I. qPlugging into enzymesq: nanowiring of redox enzymes by a gold nanoparticle.q. Science 2003;299:1877.

cofactor regeneration for sustainable enzymatic bio synthesis

Documents