Facile Preparation of an Enzyme-Immobilized Microreactor using a Cross-Linking Enzyme Membrane on a Microchannel Surface
Post on 06-Jun-2016
Facile Preparation of an Enzyme-Immobilized Microreactor usinga Cross-Linking Enzyme Membrane on a Microchannel Surface
Takeshi Honda,a Masaya Miyazaki,a,b,* Hiroyuki Nakamura,a
and Hideaki Maedaa,b,*a Nanotechnology Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 807-1Shuku, Tosu, Saga 841-0052, JapanPhone: (+81)-942-81-3676, Fax: (+81)-942-81-3657; e-mail: firstname.lastname@example.org or email@example.com
b Department of Molecular and Material Science, Interdisciplinary Graduate School of Engineering Sciences, KyushuUniversity, 61 Kasuga Koen, Kasuga, Fukuoka 816-8580, JapanPhone/Fax: (+81)-92-583-8891
Received: May 19, 2006; Accepted: August 15, 2006
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The enzyme microreactor has considerablepotential for use in biotechnological syntheses andanalytical studies. Simplifying the procedure ofenzyme immobilization in a microreactor is attrac-tive, and it is achievable by utilizing enzyme immobi-lization techniques and taking advantage of the char-acteristics of microfluidics. We previously developeda facile and inexpensive preparation method for anenzyme-immobilized microreactor. The immobiliza-tion of enzymes can be achieved by the formation ofan enzyme-polymeric membrane on the inner wall ofthe microchannel through cross-linking polymeri-zation in a laminar flow. However, this method is un-suitable for use in conjunction with electronegativeenzymes. Therefore, a novel preparation methodusing poly-l-lysine [poly ACHTUNGTRENNUNG(Lys)] as a booster and anadjunct for the effective polymerization of electro-negative enzymes was developed in this study. Using
aminoacylase as a model for an electronegativeenzyme, the reaction conditions for the enzyme-cross-linked aggregation were optimized. On thebasis of the determined conditions, an acylase-immo-bilized tubing microreactor was successfully preparedby cross-linking polymerization in a concentric lami-nar flow. The resulting microreactor showed a higherstability against heat and organic solvents comparedto those of the free enzyme. The developed methodusing poly ACHTUNGTRENNUNG(Lys) was applicable to various enzymeswith low isoelectric points, suggesting that this micro-reactor preparation utilizing a cross-linked enzymein a laminar flow could be expanded to microreac-tors in which a broad range of functional proteinsare employed.
Keywords: cross-linked enzyme aggregation; enzymemicroreactor; immobilization; surface modification
Microreactors are potentially powerful tools in thefields of chemistry and biotechnology. The excellentperformance of microreaction systems is achieved bytaking advantage of microchannel system featuressuch as rapid heat and mass transfer, and their largersurface/interface area. These attractive features arefavorable for conducting catalytic reactions throughcatalysts that are immobilized on the reactor. En-zymes are useful catalysts for the production of cer-tain types of chemicals and analysis, which are fre-quently applied to various reactors. There have beendemands for enzymatic microreaction devices includ-ing enzyme-immobilized microreactors in several
fields, especially for high-throughput biotechnologicalsyntheses and bioanalyses. In such microreactorpreparations, nanostructure fabrication and chemicalmodification of the microchannel are useful technolo-gies. We have also developed a method for immobi-lizing enzymes on a glass-capillary surface using achemical modification containing sol-gel techniques.
However, these methods require high-level techni-ques and multi-step procedures, leading to low costperformance. More simple methods using carrierswith immobilized enzymes have also been proposed.The carriers were constructed by forming monolithsof a sol-gel and a polymer, and simply incorporatingcommercial beads and membranes in a microchan-nel. These structures are often responsible for the
Adv. Synth. Catal. 2006, 348, 2163 2171 E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim 2163
generation of high pressure, which is unfavorable formicrosynthesis and -analysis systems with multipleprocesses based on highly-controlled microfluids,
and are energy-intensive to operate. Therefore, in ourprevious study, we designed an enzyme-immobilizedmicroreactor which was simply and inexpensively pre-pared and does not require high pressure. In our mi-croreactor process, a substrate membrane covers theinternal surface of the microchannel. The cylindricalmembrane is composed of the cross-linked polymer-ized enzyme product. The hollow structure of this re-actor prevents high pressure, compared to carrier-fill-ing microreactors.[6,8] In addition, this carrier-freemethod would be expected to have the advantage ofan absence of interactions between the enzyme andthe carrier material, which often leads to enzyme in-activation. This bottom-up method is suitable forthe formation of miniature structures, and has recent-ly been utilized in various nanotechnological applica-tions such as enzyme-polymerized nanoparticles.
The enzyme-polymerized membrane was based on across-linked enzyme aggregate (CLEA) formedusing a cross-linker with aldehyde groups which reactwith amino groups of the enzyme (Scheme 1). There-fore, CLEAs could be not obtained for electronega-tive enzymes due to a lack of reactive amino groups(relatively lower number of lysine residues). In thepresent study, to demonstrate the applicability of theCLEA-based enzyme-microreactor (CEM) methodfor various enzymes, we developed a novel methodfor preparing a CEM that is applicable to electroneg-ative enzymes utilizing poly-l-lysine [poly ACHTUNGTRENNUNG(Lys)]. Thesuccessful preparation was found to realize an effi-cient and fast CLEA formation in electronegative en-zymes by using poly ACHTUNGTRENNUNG(Lys).
The Effect of a Cationic Polymer on CLEAFormation of an Electronegative Protein
The efficient formation of a CLEA is essential for thesuccessful preparation of a CEM. We first exploredan efficient method in a batchwise system. Glutaral-dehyde (GA) and paraformaldehyde (PA) were usedas cross-linkers, as previously described. Whenmixing the cross-linker and enzyme, CLEAs for elec-tropositive enzymes such as chymotrypsin and trypsin,which contain a high number of lysine residues, arereadily formed. On the other hand, no CLEAs wereobserved in the case of electronegative enzymes. Therelatively low content of lysine residues in these en-zymes would cause a low degree of cross-linking,leading to inefficient aggregation. To achieve effectivecross-linking polymerization, we proposed the closepacking of electronegative enzyme molecules byenzyme-trapping using a cationic polymer matrixprior to the cross-linking reaction. The polymermatrix, possessing multiple basic functional groups,and electronegative enzymes will be adsorbed by elec-trostatic interactions, and form an enzyme-enzymecomplex bridged by the matrix. Actually, poly ACHTUNGTRENNUNG(Lys)immobilized on a carrier is frequently utilized to cap-ture electronegative proteins.[11,12] In this study, weused poly ACHTUNGTRENNUNG(Lys) as coupling agent for a carrier-free ag-gregate of the electronegative enzyme.The effect of poly ACHTUNGTRENNUNG(Lys) on CLEA formation was
compared between chymotrypsin and aminoacylase(hereafter referred to as acylase) as electropositiveand electronegative model enzymes, respectively. Theturbidity in the reaction mixture increases with in-creasing amounts of insoluble aggregates. Thus, theprogress of CLEA formation could be evaluated bythe turbidity of the solution, as estimated by measur-ing the absorbance at 630 nm. As shown in Figure 1,the cross-linker consisted of GA (0.25%, v/v) and PA(4%, v/v) agglutinated chymotrypsin, but not acylase.No turbidity was observed when concentrated acylase(5-fold; 100 mgmL1) and/or cross-linker (5-fold;1.25% GA and 20% PA) was used (data not shown).In contrast, the addition of the cross-linker into themixture of acylase and poly ACHTUNGTRENNUNG(Lys) (Mn=20 kDa) led toa remarkable increase in the turbidity of the reactionsolution. When mixing acylase and poly ACHTUNGTRENNUNG(Lys) withoutcross-linker, in which an acylase-poly ACHTUNGTRENNUNG(Lys) complexwould be formed as a result of electrostatic interac-tions, the solution contained negligible suspendedsolids (Figure 1). Poly ACHTUNGTRENNUNG(Lys) and the cross-linker mix-ture without acylase, in which the cross-linking reac-tion would occur between poly ACHTUNGTRENNUNG(Lys) molecules,showed no turbidity (Figure 1). These results can beattributed to the high hydrophilicity of poly ACHTUNGTRENNUNG(Lys)which enables each polymeric complex to prevent in-
Scheme 1. Reaction of enzyme amino groups and/or poly-ACHTUNGTRENNUNG(Lys) with aldehyde groups of the cross-linker.
2164 www.asc.wiley-vch.de E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim Adv. Synth. Catal. 2006, 348, 2163 2171
FULL PAPERS Takeshi Honda et al.
solubilization. It is clear that the turbidity of the acy-lase-polyACHTUNGTRENNUNG(Lys)-cross-linker mixture is a result of theconcerted reaction of the three reagents. A reactionusing lysine instead of poly ACHTUNGTRENNUNG(Lys) showed no evidenceof CLEA formation, indicating the importance of thepolymeric structure of poly ACHTUNGTRENNUNG(Lys) for effective CLEAformation. The addition of lysine to a mixture of chy-motrypsin-cross-linker greatly reduced the turbidityof the reaction mixture. This deleterious effect mightbe due to competition between amino groups of theenzyme and lysine, which possess little cross-linkingability.In order to achieve broad applicability for CEM for
various enzymes, we examined whether CLEAs wereformed using poly ACHTUNGTRENNUNG(Lys), using several industriallyuseful enzymes with distinct isoelectric points. Table 1shows the relationship between the isoelectric pointsof each enzyme and CLEA formation. Electropositiveenzymes, including chymotrypsin, cucumisin, papainand trypsin easily formed a CLEA without poly ACHTUNGTRENNUNG(Lys).In other enzymes, with low isoelectric points (pI
Figure 2. A stable laminar flow is important for thesuccessful formation of a CLEA-based membrane onthe internal surface of a CEM-tube, and appropriateflow rates should be selected for each solution. Weset the pumping rate of the cross-linker (correspond-ing to a central stream) to a value 12.5 times higherthan that of the acylase-poly ACHTUNGTRENNUNG(Lys) mixture (outerstream). The formation of a concentric laminar flowwas confirmed by confocal laser scanning microscopy(see Supporting Information, Figure S3). Calculatingfrom the two distinct pumping rates and microfluidicvolumes estimated from the confocal data, averagelinear velocities of the central and outer streams were187.6 mmmin1 and 6.4 mmmin1, respectively. In theCEM-tube, the residence time of the central streamwas much shorter (0.6 min) than that of the outerstream (39.1 min). According to the kinetics of CLEAformation in a batch (see Supporting Information,Figure S4), this residence time for the acylase-poly-ACHTUNGTRENNUNG(Lys) mixture solution was sufficient to form theCLEA.The resulting CEM was washed with distilled
water, dried and the enzyme-membrane observed.Figure 3 shows CCD images of the cylindricalenzyme-membrane (dry state) exposed from thePTFE tube and a sectional view of the CEM. The cy-
lindrical membrane was shrunk by drying, and thethickness of the obtained membrane was 3560 mm.When a suspension of CLEA prepared in a batchwisesystem was passed through a PTFE tube, no mem-brane was observed on the inner wall (data notshown). This result indicates that polymerization in amicrofluid enables the enzyme-polymer product toadopt a membrane structure on the inner wall of theCEM tube. The amount of acylase, estimated by aquantitative amino acid analysis as described previ-ously, was 3.250.17 nmol per acylase-CEM.
Performance of the Acylase-CEM
The enzymatic performance of the obtained acylase-CEM was evaluated using the hydrolysis of acetyl-d,l-phenylalanine (Ac-d,l-Phe) at various conditions.The CEM was cut off at both ends to a length of12.2 cm, corresponding to 24 mL of inner volume. Thehydrolysis at 4 mM of substrate was complete at thisflow rate (1 mL min1) (Figure 4). The MichaelisMenten constant (Km) and maximal velocity (Vmax),calculated from a LineweaverBurke plot, were4.58 mM and 1.10 mMmin1, respectively. All prod-ucts showed high enantiopurities (99.30.8%) com-pared to those obtained in a batch process (99.20.4%). When treating 100 mM substrate at a flowrate of 4 mL min1, the highest amount of l-Phe pro-duction was 45 nmol per minute. Under similar reac-tion conditions, the Km and Vmax values for the freeacylase were estimated to be 2.33 mM and4.69 mMmin1, respectively. The Vmax value wassomewhat lower than that of the free acylase while ahigher Km value was observed for the immobilizedacylase. These results suggest that chemical modifica-tion with the cross-linker somewhat suppressed thehydrolytic ability of the acylase.When a PTFE tube was passed through the acy-
lase-polyACHTUNGTRENNUNG(Lys) solution without a cross-linker, no en-
Figure 2. Schematic illustration of the procedure used to prepare an acylase-CEM. The cross-linking polymerization was per-formed in a concentric laminar flow. A silica capillary was fitted to the outer diameter of the T-shape connector by attachingto a PTFE tube using heat-shrink tubing. The capillary was set in the connector located at the concentric position of theCEM tube. The cross-linker solution was supplied to the substrate PTFE tube through the silica capillary, corresponding to acentral stream in the concentric laminar flow. A solution of acylase-polyACHTUNGTRENNUNG(Lys) mixture was poured from the other inlet of theT-shape connector, and formed an outer stream of the laminar flow.
Figure 3. CCD images of a) cylindrical enzyme-membrane(dry state) exposed from the PTFE tube, which forms onthe inner wall of the tube and b) a sectional view of the ob-tained CEM.
2166 www.asc.wiley-vch.de E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim Adv. Synth. Catal. 2006, 348, 2163 2171
FULL PAPERS Takeshi Honda et al.
zymatic activity was observed. This result indicatesthat the enzymatic activity of the acylase-CEM wasdue to the acylase immobilized in the CLEA-mem-brane, and not by non-specific adsorbed free acylaseand/or a water soluble acylase-poly ACHTUNGTRENNUNG(Lys) complex.It is very important for enzymes incorporated in a
microreaction system to have a high stability againstconditions such as pH, temperature, and solvent. ThepH activity profile of the free acylase and acylase-CEM was obtained by incubating the enzyme in50 mM phosphate buffer for pH 57.5, 50 mM Tris-HCl buffer for pH 89 at 40 8C for 30 min. The acy-lase in the CEM showed a maximum activity atpH 7.0, and was found to be more acid-tolerant than
the free acylase (the optimal pH 8.0) (Figure 5a). Thethermostability of the enzyme reactor is an importantfactor regarding its use for industrial applications. Toexamine the effect of temperature on enzymatic activ-ity, an enzymatic assay was performed after incubat-ing the CEM and free acylase solution in the absenceof substrate at various temperatures for 30, 60, and120 min. The reaction was performed at 40 8C for30 min. The thermostability of the immobilized acy-lase in the CEM was superior to that of the free acy-lase (Figure 5b). In an enzymatic synthesis system, ahighly hydrophobic substrate is often unsuitable dueto its insolubility. The stability of the enzyme-reactorto organic solvents is very important because organicsolvents improve the solubility of certain substrates.Organic solvent tolerances for free acylase and acy-lase-CEM were estimated using the hydrolytic effi-ciency in the presence of N,N-dimethylformamide(DMF). The reaction was performed at 40 8C for30 min. The acylase-CEM showed a higher resistanceto DMF, as compared to the free enzyme (Figure 5c).These stabilizations were observed in a-chymotryp-sin-CEM as well.
In this study, we describe a novel and facile methodfor preparing a CEM that is applicable to a broadrange of enzymes. The most important factor for theCEM preparation is the efficiency of enzyme-enzymecross-linking through amino groups. Thus, it is reason-ably difficult to prepare a CEM for amino group-poor(electronegative) enzymes. Chemical modification ofsuch an enzyme, for example, by amination using car-bodiimide, could facilitate the cross-linking, but it isa rather intricate and time-consuming process. In thisstudy, it was found that the simple addition of a cat-
Figure 5. (a) Effect of pH on the enzymatic activity of the acylase-CEM (*) and the free acylase (~). The activities at thepH optima were assigned values of 100%. (b) Temperature dependence of the enzymatic activity of the acylase-CEM (solidlines) and free acylase (dashed lines) at 40 8C (circles), 50 8C (triangles), and 60 8C (squares). The activities without pre-incu-bation were assigned values of 100%. (c) Effect of organic solvent on the enzymatic activity of the acylase-CEM (*) andthe free acylase (~). The activities without organic solvent were assigned values of 100%. In all enzyme assays, 4 mM Ac-d,l-Phe was used as substrate.
Figure 4. Hydrolysis of Ac-d,l-Phe in the acylase-CEM. Asolution of substrate in Tris buffer (pH 8.0) was charged intoa CEM tubing (inner volume=24 mL) for various residencetimes. The reactions were carried out at three substrate con-centrations [4 mM (*), 20 mM (~), and 100 mM (&)] at40 8C. This acylase-based reaction converted the l-isomeronly. A PTFE tube passed through a mixture of acylase andpoly ACHTUNGTRENNUNG(Lys) had no enzymatic activity (^).
Adv. Synth. Catal. 2006, 348, 2163 2171 E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.asc.wiley-vch.de 2167
FULL PAPERSFacile Preparation of an Enzyme-Immobilized Microreactor
ionic polymer, poly ACHTUNGTRENNUNG(Lys), effectively facilitated CLEAformation for an electronegative enzyme (Figure 1,Table 1).A previous study showed that a cationic polymer
could effectively trap an acylase as a result electro-static interactions. In a similar manner, poly ACHTUNGTRENNUNG(Lys)would bridge and closely pack (trap) acylase mole-cules. GA- and PA-conjugated poly ACHTUNGTRENNUNG(Lys) would alsobehave as a longer cross-linker possessing more reac-tion sites per molecule, which would augment thedegree of cross-linking as compared to GA and PA.These factors would lead to the effective promotionof an acylase-polymerization based on the covalentcross-linking. A lower-cost method for acylase-CLEAformation has been reported, in which the acylasewas precipitated using an inexpensive low molecularprecipitant. However, such precipitation (suspen-sion) would be unsuitable for preparing a CEM dueto generating unevenness in the membrane and tubeobstruction. In contrast, the acylase-polyACHTUNGTRENNUNG(Lys) com-plex showed no precipitation, indicating that poly-ACHTUNGTRENNUNG(Lys) is better suited to the CEM preparation systemas a booster of CLEA formation.When applying this method to various enzymes, the
influence on CLEA formation and enzymatic activityshould be carefully examined because poly ACHTUNGTRENNUNG(Lys) oftenhas different characteristics depending on the poly-mer size. For instance, there are interesting re-ports that the aggregation of an acidic material withpoly ACHTUNGTRENNUNG(Lys) and activation of a protease by poly ACHTUNGTRENNUNG(Lys)were sensitive to the size of the poly ACHTUNGTRENNUNG(Lys).[23,24] In ouracylase-CLEA, there was little influence of the poly-ACHTUNGTRENNUNG(Lys) size on enzymatic activity. On the other hand,the degree and the kinetics of CLEA formation weredependent on poly ACHTUNGTRENNUNG(Lys) size (see Supporting Informa-tion, Figure S2). These size effects can be attributedto several factors such as electrostatic properties andthe number of interaction sites of poly ACHTUNGTRENNUNG(Lys).The cross-linker also plays an important role in
CEM preparation. It was found that the use of a com-bination of GA and PA was suitable for acylase-CEMpreparation, and that the acylase was more sensitiveto the GA/PA than chymotrypsin. For applicationsof various enzymes to CEM preparations, the totalamount of aldehyde reagents and the GA/PA ratiomust be optimized for each enzyme.In this study, a CEM was prepared by utilizing a
combination of membrane-formation and microfluidicmanipulation. A concentric laminar flow is geometri-cally appropriate to form the reaction interface alongthe internal wall of the CEM-tube. In comparison toour previous system, in which a laminar flow consist-ing of three parallel streams using a simple cross-shaped connector was used, the present system pos-sesses several advantages such as the simplification ofpumping work by decreasing the number of inlets, thenon-occurrence of aggregates in the inner space of
the connector because polymerization began in theCEM-tube (leading to non-obstruction at the connec-tor region), and a more homogeneous and stable for-mation of the CLEA-based membrane occurred alongthe inner wall of the CEM-tube.A previous study reported on a polymer-membrane
formation in a microchannel utilizing interfacial poly-merization and multilayer flow inside a microchan-nel. In a similar manner, the enzyme polymeri-zation reaction was considered to mainly occur at theinterface between the central and the outer flows, andalso at the outer side, in preference to the centralstream. When the acylase-poly ACHTUNGTRENNUNG(Lys) complex diffusesfrom the outer stream into the central stream andpolymerizes at the central region, the complex beginsto polymerize and would be unable to have sufficientreaction time (at least 510 min from Supporting In-formation, Figure S4) to mature into an insoluble ag-gregate due to the shorter residence time (0.6 min) inthe central stream. In addition, the diffusion rate ofthe acylase-poly ACHTUNGTRENNUNG(Lys) complex from the outer streamto the central stream is lower than the cross-linkerfrom the central to the outer stream. (Roughly calcu-lating based on the literature, the diffusion coeffi-cients are 910 mm2 sec1 for the cross-linker and70 mm2 sec1 for acylase.) These factors would lead tothe enzyme-polymer being promoted more preferen-tially in the vicinity of the inner wall of CEM-tubethan at the central region.It is highly possible that the enzyme of the CEM
can have altered enzymatic characteristics comparedto the native state. The obtained acylase-CEMshowed an decrease in hydrolytic efficiency and an in-crease in acid-, heat-, and organic solvent-tolerance(Figure 4 and Figure 5). These results were consistentwith previous reports on acylase immobilization[11,12,27]
and similar results have been observed for other en-zymes. Immobilization-based conformational rigidi-fication would allow the acylase to avoid conforma-tional collapse which is responsible for inactivation byheat, pH, and organic solvents. The observed changein pH-sensitivity might also be caused by the modifi-cation of amino groups in the acylase and/or the buf-fering action of poly ACHTUNGTRENNUNG(Lys) matrix in the microenviron-ment of the enzyme as described in a previousreport.
There were little differences in enzymatic perfor-mance between batchwise CLEA and CEM. Howev-er, the handling ability of CEM was obviously higherthan the batchwise CLEA. For example, in reactionassay and washing procedures, batchwise systems re-quire a series of operations including mechanical agi-tation, centrifugation, and removal of the supernatant.In the CEM system, such procedures were simply re-placed by pumping, leading to more simple, efficient,and accurate handling. Dealing with multiple micro-scale samples in a batchwise CLEA system is also
2168 www.asc.wiley-vch.de E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim Adv. Synth. Catal. 2006, 348, 2163 2171
FULL PAPERS Takeshi Honda et al.
considered to be technically difficult and troublesome.The CEM would lead to the establishment of a dis-posable enzyme-reactor system which enables easymicroscale-handling in a multi-parallel manner.A combination of such advantageous handling abil-
ity of CEM and the enzymatic characteristics of acy-lase has strong potential for the high throughput mi-croprocess of optical separation in a combinatorialenantiopure amino acid production system. Thus, theestablishment of such system would be attractive infields such as basic research of proteins and drug(functional biomaterial) discovery because diverseunnatural amino acids are required on-demand.In this study, we propose the applicability of the
CEM preparation method for other enzymes active invarious biotechnological fields as well as acylases. Atpresent, a great interest in such fields is being direct-ed towards the development of combinatorial synthe-sis and high throughput analysis systems. CEM couldhave significant potentials as a platform technology ofenzyme-microreactions which could be used in the ex-ploration of new applications.
A novel CEM preparation method applicable to elec-tronegative enzymes previously incapable of CLEAformation by our previous method is described. Anacylase-CEM model was successfully constructed withan acylase-CLEA using poly ACHTUNGTRENNUNG(Lys) and cross-linkercontaining GA and PA under optimal conditions. Acationic bio-polymer poly ACHTUNGTRENNUNG(Lys) is a useful booster forCLEA formation in the case of electronegative pro-teins, which closely packs such proteins as a result ofelectrostatic interactions, and forms a soluble protein-poly ACHTUNGTRENNUNG(Lys) complex. The simple CLEA preparationmethod developed here is applicable to various en-zymes with distinct isoelectric points, suggesting thatthis CEM preparation technology could be expandedto a broad range of functional proteins including en-zymes. Such broad applicability would lead to theconstruction of a flexible technology platform forscreening and designing a potential protein-immobi-lized microreactor for broad applications.
Poly(tetrafluoroethylene) (PTFE) microtubes (250 mm innerdiameter (i.d.), 1.59 mm outer diameter (o.d.) and 500 mmi.d., 1.59 mm o.d.), PTFE adapter, and heat-shrink tubingwere obtained from Flon Chemical Inc. (Osaka, Japan). GLScience Co., Ltd. (Tokyo, Japan) provided a silica-fused mi-crocapillary (100 mm i.d., 350 mm o.d.) and a stainless-steelT-shaped connector (Union Tee, SUS-316). When using the
microcapillary, the polyimide coating was removed (310 mmo.d.). Glutaraldehyde (GA) and paraformaldehyde (PA) ascross-linker agents, aminoacylase I, d-aminoacylase, a-chy-motrypsin, b-d-galactosidase, and other reagents for aminoacid analysis were obtained from Wako Pure Chemical Ind.,Ltd (Osaka, Japan). Poly-l-lysine hydrobromides of a seriesof distinct molecular weights were synthesized from Ne-ben-zyloxycarbonyl-l-lysine N-carboxyanhydride as describedpreviously. High molecular weight poly ACHTUNGTRENNUNG(Lys) (>300 kDa,Mn=360 kDa), cucumisin, esterase, lipase, papain, pepsin A,thermolysin, sodium cyanoborohydride (NaCNBH3), andother reagents for buffers were obtained from SigmaAl-drich Co. (St. Louis, MO, USA). Trypsin, tyrosinase, and V8protease were purchased from Worthington BiochemicalCo., Ltd. (Lakewood, NJ, USA). Alkaline phosphatase wasobtained from Biozyme Laboratories Co., Ltd. (San Diego,CA, USA). N-Acetyl d,l-phenylalanine (Ac-d,l-Phe), l-phenylalanine (l-Phe), and d-phenylalanine (d-Phe) werepurchased from Watanabe Chemical Industries Ltd. (Hir-oshima, Japan). All enzymes were ultrafiltered using a mi-crocon YM-10 (molecular weight cut-off=10 kDa) filter ob-tained from the Millipore Co. (Bedford, MA, USA). Thepoly ACHTUNGTRENNUNG(Lys)s were dialyzed with a dialysis tube (molecularweight cut-off=500 Da) against phosphate buffer (PB) orgel-filtered in PB for smaller poly ACHTUNGTRENNUNG(Lys)s.
Typical Preparation of CLEAs in Batchwise Systemand Measurement of Turbidity
The enzyme and poly ACHTUNGTRENNUNG(Lys) were dissolved in phosphate-buf-fered saline; PBS (10 mM, pH 7.2) at final concentrations of20 and 10 mgmL1, respectively. The cross-linker reagentscontaining GA (0.0160.25%) and PA (0.254%) in PB(0.1M, pH 8.0) were used at a constant ratio (1:16, v/v). Allsolutions were filtered (pore size 0.2 mm). The enzyme, poly-ACHTUNGTRENNUNG(Lys), and cross-linker solutions were mixed at a volumeratio (50 mL:50 mL:100 mL) in a 96-microwell plate (Nunc,Kamstrup, Denmark). After incubation at 4 8C, the turbidityof the mixture was estimated by measuring the absorbanceat 630 nm using a microplate reader Multiscan JX (ThermoElectron Co., Milford, MA, USA).
Acylase-CEM Preparation using a Laminar FlowSystem
The CEM preparation system was assembled as shown inFigure 2. A fused silica microcapillary (5 cm length) and aPTFE tube (250 mm i.d., 5 cm length) for introducing re-agents and a PTFE tube (500 mm i.d., 13 cm length) corre-sponding to the CEM substrate were attached to a T-shapedconnector as shown in Figure 2. In order to form a concen-tric laminar flow in the CEM preparation, the silica capillarywas set in the T-shape connector and positioned at the con-centric position of the CEM tube. The enzyme and poly-ACHTUNGTRENNUNG(Lys) solutions were mixed at 1:1 volume ratios on ice. Allsolutions were filtered and charged into 1 mL plastic syring-es (Terumo Co., Tokyo, Japan). The cross-linker andenzyme/polyACHTUNGTRENNUNG(Lys) solutions were supplied to the CEM tubethrough the capillary and another inlet of the T-shape con-nector using a PTFE tube, respectively. Solution introduc-tion was performed at different pumping rates (6.25 mLmin1 for the cross-linker and 0.5 mL min1 for the enzyme)
Adv. Synth. Catal. 2006, 348, 2163 2171 E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.asc.wiley-vch.de 2169
FULL PAPERSFacile Preparation of an Enzyme-Immobilized Microreactor
by a Harvard 11 Pico Plus Syringe Pump (Harvard Appara-tus, Inc., Holliston, MA, USA). After the polymerizationwas performed for 3 h, the obtained CEM was rinsed withTris-HCl (1M, pH 9.0), which simultaneously quenched anyactive aldehyde groups remaining on the formed membrane.In order to reduce the resulting Schiff base, CEM was treat-ed with NaCNBH3 in borate buffer (50 mM, pH 9.0). TheCEM was extensively washed with PBS (pH 7.5) and no elu-tion of protein was confirmed by measurement of absorb-ance at 220 nm. All preparations were performed at 4 8C.The obtained CEM was filled with PBS and stored at 4 8C.An enzyme-polymerizing membrane of the obtained CEMwas observed using a color CCD digital microscope system(HI-SCOPE DH-2700; Hirox Co., Ltd, Tokyo, Japan).
Estimation of the Amount of Enzyme
To estimate the amount of the immobilized enzyme in theCLEA, a quantitative amino acid analysis was carried out.The tube was lightly washed with distilled water, and theenzyme-membrane was then dried using a vacuum pump,and peeled from the tube by means of a forceps. The ob-tained membrane was hydrolyzed by HCl (6M) at 110 8Cfor 24 h. The CLEA obtained in a batchwise system wasalso treated in a similar manner. The hydrolyzed productwas reacted to phenyl isothiocyanate (PITC), the PITC-la-beled amino acids were analyzed on an HPLC system usinga Wakosil-PTC (4.0 mmP200 mm, Wako Pure ChemicalInd., Ltd) with a linear gradient of acetonitrile (60%, v/v)over 15 min at a flow rate of 1 mLmin1 at 25 8C. The pro-tein content was quantified by amino acid analysis using thestandard PITC procedure. The amount of the immobilizedacylase was calculated by comparison with the amino acidcomposition of the acylase protein. In the immobilized poly-ACHTUNGTRENNUNG(Lys), it is very difficult to estimate the precise amount dueto the difficulty of detecting cross-linker-conjugated lysineresidues by amino acid analysis.
Enzymatic Enantioselective Deacetylation of N-Acetyl-d,l-amino Acids
Acylase activity was estimated using a hydrolytic assay withAc-d,l-Phe as a racemic substrate. The substrate was dis-solved in Tris buffer (50 mM), and adjusted the pH 8.0 withNaOH (6M). In a batchwise system using free acylase andacylase-CLEA, an equal amount of substrate solution wasmixed with free acylase dissolved or acylase-CLEA suspend-ed in Tris-HCl buffer (50 mM, pH 8.0) by vigorous stirringat 3,000 rpm in a microtest tube. The enzymatic reactionwas terminated by the addition of 0.2M HCl (final concen-tration of 0.1M pH 2.0). In the CEM system, the substratesolution was charged into a 1 mL syringe, and supplied tothe tubing CEM using a syringe pump. The flow rate affect-ed the reaction time which corresponded to the residencetime in which the substrate solution flowed through theCEM. Standard enzymatic reactions were performed at40 8C in a thermostated incubator. The reaction product wasanalyzed by an HPLC system using a Wakosil C18 ARcolumn (3.0P250 mm, Wako Pure Chemical Ind., Ltd) witha linear gradient of TFA (0.05%)-acetonitrile over 30 min ata flow rate of 0.5 mLmin1 at 25 8C. The reaction yieldswere calculated from the peak area calibrated with the stan-dard compound. Km and Vmax were computed by fitting the
data to the MichaelisMenten equation using a publicdomain program, GNUPLOT. The optical yields of the ob-tained products were calculated from the peaks of l-Pheand d-Phe which were analyzed by HPLC system using aChirobiotic T column (4.6P250 mm, Tokyo Kasei KogyoCo., Ltd) with an isocratic flow of 50% ethanol for 12 minat a flow rate of 0.8 mLmin1 at 30 8C.
 a) V. Hessel, S. Hardt, H. Lowe, Chemical Micro Pro-cess Engineering, Wiley-VCH, Weinheim, 2004 ;b) P. D. I. Fletcher, S. J. Haswell, E. Pombo-Villar, B. H.Warrington, P. Watts, S. Y. F. Wong, X. Zhang, Tetrahe-dron 2002, 58, 47354757; c) T. Chovan, A. Guttman,Trends Biotechnol. 2002, 20, 116122; d) J. Khanduri-na, A. Guttman, J. Chromatogr. A 2002, 943, 159183.
 a) K. Kusakabe, D. Miyagawa, Y. Gu, H. Maeda, S.Morooka, J. Chem. Eng. Jpn. 2001, 34, 441443; b) C.de Bellefon, N. Pestre, T. Lamouille, P. Grenouillet, V.Hessel, Adv. Synth. Catal. 2003, 345, 190193; c) X. Li,H. Wang, K. Inoue, M. Uehara, H. Nakamura, M.Miyazaki, E. Abe, H. Maeda, Chem. Commun. 2003,964965; d) J. Kobayashi, Y. Mori, S. Kobayashi, Adv.Synth. Catal. 2005, 347, 18891892.
 a) N. Lion, F. Reymond, H. H. Giraut, J. S. Rossier,Curr. Opin. Biotechnol. 2004, 15, 3137; b) P. L. Urban,D. M. Goodall, N. C. Bruce, Biotechnol. Adv. 2006, 24,4257.
 a) J. Drott, E. Rosengren, K. Lindstrom, T. Laurell,Mikrochim. Acta 1999, 131, 115120; b) M. Bengtsson,S. Ekstrom, G. Marko-Varga, T. Laurell, Talanta 2002,56, 341353; c) H. Qu, H. Wang, Y. Huang, W. Zhong,H. Lu, J. Kong, P. Yang, B. Liu, Anal. Chem. 2004, 76,64266433.
 a) M. Miyazaki, J. Kaneno, M. Uehara, H. Fujii, M. ,Shimizu, H. Maeda, Chem. Commun. 2003, 648649;b) J. Kaneno, R. Kohama, M. Miyazaki, M. Uehara, K.Kanno, M. Fujii, H. Shimizu, H. Maeda, New J. Chem.2003, 27, 17651768; c) M. Miyazaki, J, Kaneno, R.Kohama, K. Kanno, M. Fujii, H. Shimizu and H.Maeda, Chem. Eng. J. , 2004, 101, 277284; d) M.Miyazaki, J. Kaneno, S. Yamaori, T. Honda, M. P. P.Briones, M. Uehara, K. Arima, K. Kanno, K. Yamasita,Y. Yamaguchi, H. Nakamura, H. Yonezawa, M. Fujii,H. Maeda. Prot. Pept. Lett. 2005, 12, 207210.
 a) D. S. Peterson, T. Rohr, F. Svec, J. M. J. FrSchet,Anal. Chem. 2002, 74, 40814088; b) K. Sakai-Kato, M.Kato, T. Toyooka, Anal. Chem. 2003, 75, 388393;c) G. W. Slysz, D. C. Schriemer, Rapid Commun. MassSpectrom. 2003, 17, 10441050; d) J. W. Cooper, J.Chen, Y. Li, C. S. Lee, Anal. Chem. 2003, 75, 10671074.
 M. Tokeshi, Y. Kikutani, A. Hibara, K. Sato, H. Hisa-moto, T. Kitamori, Electrophoresis 2003, 24, 35833594.
 T. Honda, M. Miyazaki, H. Nakamura, H. Maeda,Chem. Commun. 2005, 50625064.
 L. Cao, L. Langen, R. A. Sheldon, Curr. Opin. Biotech-nol. 2003, 14, 387394.
2170 www.asc.wiley-vch.de E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim Adv. Synth. Catal. 2006, 348, 2163 2171
FULL PAPERS Takeshi Honda et al.
 G. Liu, Y. Lin, V. OstatnT, J. Wang, Chem. Commun.2005, 34813483.
 B. He, M. Li, D. Wang, React. Polym. 1992, 17, 341346.
 E. Topoglidis, E. Palomares, Y. Astuti, A. Green, C. J.Campbell, J. R. Durrant, Electroanalysis 2005, 17,10351041.
 A. Bommarius, K. Drauz, C. Wandrey, Ann. NY Acad.Sci. 1992, 672, 126136.
 T. Chibata, T. Tosa, T. Sato, T. Mori, Y. Matsuo, Prepa-ration and industrial application of immobilized amino-acylase, in: Fermentation technology today, Proceedingsof the Fourth International Fermentation Symposium,(Ed.: G. Terui), Japan, Society of Fermentation Tech-nology, 1972, pp. 383389.
 M. Takagi, T. Maki, M. Miyahara, K. Mae, Chem. Eng.J. 2004, 101, 269276.
 F. Lopez-Gallego, T. Montes, M. Fuentes, N. Alonso, V.Grazu, L. Betancor, J. M. Guisan, R. Fernandez-La-fuente, J. Biotechnol. 2005, 116, 110.
 T. Kawai, M. Nakamura, K. Sugita, K. Saito, T. Sugo,Biotechnol. Prog. 2001, 17, 872875.
 M. L. Bode, F. Rantwijk, R. A. Sheldon, Biotechnol.Bioeng. 2003, 84, 710713.
 A. Bargellesi, G. Damiani, W. M. Kuehl, M. D. Scharff,J. Cell. Physiol. 1976, 88, 247251.
 a) I. L. Shih, Y. T. Van, M. H. Shen, Mini Rev. Med.Chem. 2004, 4, 179188; b) J. P. Clarenc, G. Degols,J. P. Leonetti, P. Milhaud, B. Lebleu, Anticancer DrugDes. 1993, 8, 8194; c) W. T. Godbey, A. G. Mikos, J.Control. Release. 2001, 72, 115125.
 M. A. Wolfert, L. W. Seymour, Gene Ther. 1996, 3,269273.
 K. Tanaka, K. Ii, A. Ichihara, L. Waxman, A. L. Gold-berg, J. Biol. Chem. 1986, 261, 1519715203.
 R. A. Brown, T. M. Wells, J. Pharmacy Pharmacol.1992, 44, 467471.
 K. Tanaka, K. Ii, A. Ichihara, L. Waxman, A. L. Gold-berg, J. Biol. Chem. 1986, 261, 1519715203.
 H. Hisamoto, Y. Shimizu, K. Uchiyama, M. Tokeshi, Y.Kikutani, A. Hibara, T. Kitamori, Anal. Chem. 2003,75, 350354.
 S. Yamamoto, K. Nakanishi, R. Matsuno, Ion-ExchangeChromatography of Proteins, Marcel Dekker, Inc., NewYork and Basel, 1988.
 P. M. Lee, K. H. Lee, Y. S. Siaw, J. Chem. Tech. Biotech-nol. 1993, 58, 6570.
 M. N. Gupta, B. Mattiasson, in: Methods of Biochemi-cal Analysis: Bioanalytical Applications of Enzymes,Vol. 15. John Wiley, New Jersey, 1979, pp 134.
 a) T. L. Hendrickson, V. Crecy-Lagard, P. Schimmel,Annu. Rev. Biochem. 2004, 73, 147176; b) N. Hino, Y.Okazaki, T. Kobayashi, A. Hayashi, K. Sakamoto, S.Yokoyama, Nature Methods 2005, 2, 201; c) B. L. Nils-son, M. B. Soellner, R. T. Raines, Annu. Rev. Biophys.Biomol. Struct. 2005, 34, 91118; d) R. Jain, S. Cha-wrai, Mini Rev. Med. Chem. 2005, 5, 469477.
 T. Honda, M. Miyazaki, H. Nakamura, H. Maeda, LabChip 2005, 5, 812818.
 I. Gilles, H. Loffler, F. Schneider, Z. Naturforsch. C1984, 39, 10171020.
 G. Cathala, C. Brunel, J. Biol. Chem. 1975, 250, 60406045.
 M. Wakayama, M. Moriguchi, J. Mol. Catal. B: Enzym.2001, 12, 1525.
 H. T. Wright, J. Mol. Biol. 1968, 37, 363366. M. Kaneda, N. Tomonaga, J. Biochem. 1975, 78, 1287
1296. D. J. Horgan, J. K. Stoops, E. C. Webb, B. Zerner, Bio-
chemistry 1969, 8, 20002006. P. W. Goodenough, J. Owen, Phytochemistry 1987, 26,
7579. A. P. Ryle, Methods Enzymol. 1970, 19, 316336. D. Tsuru, T. Yoshimoto, Handbook of Microbiology,
2nd edn., Vol. 8, CRC Press, Florida, 1987, pp. 239283.
 S. J. Endo, J. Ferment. Technol. 1962, 40, 346353. K. A. Walsh, Methods Enzymol. , 1970, 19, 4163. H. J. Wichers, Y. A. M. Gerritsen, C. G. J. Chapelon,
Phytochemistry 1996, 43, 333337. O. Vesterberg, T. Wadstrm, K. Vesterberg, H. Svens-
son, B. Malmgren, Biochim. Biophys. Acta 1967, 133,435445.
 M. Huckel, H. J. Wirth, M. T. Hearn, J. Biochem. Bio-phys. Methods 1996, 31, 165179.
Adv. Synth. Catal. 2006, 348, 2163 2171 E 2006 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim www.asc.wiley-vch.de 2171
FULL PAPERSFacile Preparation of an Enzyme-Immobilized Microreactor