Technology for regenerable biosensor probes based on enzyme

Vancouver, British Columbia, Canada V6T 1Z3, Department of Chemical ... University of British Columbia, 300-6174 University Boulevard,Vancouver, Briti...
0 downloads 0 Views 3MB Size
Biotechnoi. Rug. 1004, 10, 433-440

Technology for Regenerable Biosensor Probes Based on Enzyme-Cellulose Binding Domain Conjugates Michael R. Phelps,tJ John B. Hobbs,t*oDouglas G. Kilburn,tJvs and Robin F. B. Turner*st*ll Biotechnology Laboratory, The University of British Columbia, 237-6174 University Boulevard, Vancouver, British Columbia, Canada V6T 123, Department of Chemical Engineering, The University of British Columbia, 306-2216 Main Mall, Vancouver, British Columbia, Canada V6T 124, Department of Microbiology, The University of British Columbia, 300-6174 University Boulevard, Vancouver, British Columbia, Canada V6T 123, and Department of Electrical Engineering, The University of British Columbia, 434-2356 Main Mall, Vancouver, British Columbia, Canada V6T 124

The application of enzyme-basedbiosensors for on-line bioprocess monitoring and control has been slowed by problems relating to the in situ sterilizability of the probe and the stability of the enzyme component. A novel technology with the potential to address both of these difficulties is presented here. The approach is based on the reversible immobilization of enzymes conjugated with the cellulose binding domain (CBD) of cellulases from Cellulomonas fimi. A regenerable biosensor electrode can be configured with a cellulose matrix onto which the enzyme-CBD conjugate can be repeatedly loaded (bound by the CBD) and subsequently eluted by perfusing the cellulose matrix with the appropriate solution. Glucose oxidase (GOx) conjugated to CBD with glutaraldehyde was used in an experimental glucose biosensor to demonstrate the feasibility of multiple cycles of loading and elution of the conjugate. Michaelis-Menten enzyme kinetics provided an empirical model for the calibration of the experimental biosensor. The development of a computer-controlled prototype glucose biosensor and a fermentation monitoring system based on this approach is discussed.

Introduction The optimization of the performance of bioprocesses depends on the ability to monitor and control the parameters describing the bioreactor environment. At present, only a few parameters can be reliably monitored on-line (e.g., temperature, pH, pOz, and stir rate) without the use of highly sophisticated and costly instrumentation. The analysis of fermentation substrates, products, and metabolites is usually achieved by off-linemethods (Brooks et al., 1987/88). However, optimal control of a bioprocess requires that measurable parameters be determined as frequently as possible, which in turn requires frequent sampling that increases the risk of contamination (Huang et al., 1991). Furthermore, off-line methods usually are too slow to be used in a closed-loop control system, and it is often difficult to ensure that samples are not significantly degraded or changed during the sampling/ analysis procedure. A sensor system based on an in situ probe that could provide continuous, real-time analysis should therefore allow more effective control of fed-batch bioprocesses. The research and development of new, reliable, on-line sensors for fermentation monitoring has been stimulated by the potential for the development of new feedback control strategies and adaptive control techniques to achieve high cell density in fed-batch processes. It should be possible to control a fermentation using the regulation of glucose feed to maintain the optimum glucose concen-

* Author to whom correspondence should be addressed. + Biotechnology Laboratory.

Department of Chemical Engineering. Department of Microbiology. I( Department of Electrical Engineering.

t

8756-7938/94/30 10-0433$04.50/0

tration for growth or recombinant protein production (Bradley et al., 1991;Brooks et al., 1987/88). For example, it has been reported that glucose is a limiting factor for both cell growth and protein synthesis in the cultivation of BHK cells for the production of human antithrombin I11 protein (Wirth et al., 1989). However, it is also known that, in baker’s yeast cultivation, catabolite repression can be prevented by maintaining glucose levels below 0.1 g/L (Cleland and Enfors, 1983). In a typical batch culture of Escherichia coli, glucose ranges from 20 to 0 g/Lover a period of 8 h (Stamm et al., 19921, but acetate production at high concentrations of glucose can limit the maximum cell density that can be achieved. Using a fed-batch protocol where the glucose concentration is controlled at a lower level, acetate production could be minimized, thereby achieving higher cell densities and higher process yields. Biosensors have enormous potential, in principle, for the analysis of complex fermentation media due to the specificity and selectivity of the biological component of the sensor. Numerous examples of amperometricenzyme electrode biosensors have been described in the literature (Brooks et al., 19911, and a wide variety of analytes could be monitored using different enzyme-based systems. We have focused our efforts on the development of a glucose biosensor using glucose oxidase (GOx) as the biological component because of the industrial importanceof glucose as the main carbon source and growth-limiting substrate in many fermentations (Filippini et al., 1991;Freitag, 1993; Huang et al., 1991). Problems relating to the in situ sterilizability of the probe and the stability of the enzyme component have been addressed in a variety of ways (Enfors and Molin, 1978;Enfors and Nilsson, 1979;Cleland and Enfors, 1983,

0 1994 American Chemical Society and American Instkute of Chemical Engineers

Biotechnol. Rog., 1994, Vol. 10, No. 4

434

-

Proline Threonine Linker

I 1

I Catalytic Domain

316 335

I

443

Cellulose Binding Domain

Figure 1. Schematicrepresentationof the cellulaseexoglucanase from C. fimi. The diagram shows the two structurally distinct and independent domains that can be cleaved at the PT linker.

1984a,b;Brooks et al., 1987/88;Bradley et al., l988,1989a,b, 1990,1991). The most promising approach employs an autoclavable, in situ electrode housing with a dialysis membrane as a sterile barrier between the bioreactor medium and the interior of the sensor. The enzyme, or the electrode with immobilized enzyme, is inserted into the housing following sterilization of the bioreactor. Dilution of the analyte and periodic recalibration of the sensor are accomplished using an internal buffer flow, which flows through the enzyme chamber formed by the electrode housing, the dialysis membrane, and the electrode. The useful operating lifetime of the sensor can be prolonged, in some cases, by periodic recalibration of the sensor and/or replacement of the enzyme component. In this work, a novel technology is presented that incorporates the essential features of the designs above, where all can be implemented under computer control, including automated replacement of the enzyme. The approach used here is based on the reversible immobilization of enzymesconjugated to the cellulosebinding domain (CBD) of cellulases from Cellulomonas fimi. These cellulases have a modular structure consisting of two or more structurally distinct domains, as shown in Figure 1 (O'Neill et al., 1986; Gilkes et al., 1991). The cellulose binding domain (CBD) functions independently of the catalytic domain and has been expressed in E. coli to obtain the isolated CBD polypeptide (Ong et al., 1993). The CBD can be conjugated chemically or genetically to other proteins, which then acquire the capacity to bind reversibly to cellulose. Enzyme-CBD fusion proteins, such as a @-glucosidase-CBD fusion protein (Ong et al., 1991), retained the activity of the enzyme for its substrate, as well as the binding affinity of the CBD for cellulose. Binding to cellulose was stable for prolonged periods of time at temperatures up to at least 70 "C, at ionic strengths from 10 mM to greater than 1M, and a t pH values below 8. The fusion protein could be desorbed from cellulose by distilled water, 1M NaOH, or 8M guanidine hydrochloride (Kilburn et al., 1992). In the application presented here, the CBD from C. fimi exoglucanase (CBDcex) has been chemically conjugated to glucose oxidase using glutaraldehyde. An experimental glucose biosensor has been developed by incorporating a cellulose matrix adjacent to the surface of a platinum rotating disk electrode such that the GOxCBD conjugate can be reversibly immobilized on the electrode. Using the experimental system, the regeneration of a glucose biosensor during a fermentation is simulated to demonstrate the feasibility of this approach, and a sensor system based on the reversible immobilization of the enzyme using CBD technology is proposed for fermentation monitoring.

Materials and Methods Enzymes and Chemicals. The electrolyte used in all biosensor experiments was phosphate-buffered saline

(PBS; pH 7.4, p = 0.2 M; 0.1 M NaC1,5 mM NaH2P04, 30 mM Na2HP04, preserved with 1mM EDTA, and 5 mM sodium benzoate). Glucose oxidase (EC 1.1.3.4) was type X from Aspergillus niger (Sigma Chemical Co., St. Louis, MO). The cellulose binding domain from C. fimi exoglucanase (CBDh.) was harvested from recombinant E. coli and purified accordingto methods described elsewhere (Ong et al., 1993). CBDCex antiserum was produced in rabbits (Whittle et al., 1982). Grade 1 glutaraldehyde, 25% aqueous solution (Sigma Chemical Co.), was used for the GOx-CBD conjugation. Unpreserved 50 mM potassium phosphate buffer (pH 7) was used for the synthesis, purification, and storage of the GOx-CBD conjugate. Cellulose powder (Avicel,type PH101, FMC International Food & Pharmaceutical Products, Cork, Ireland), washed with distilled water and phosphate buffer, was used for purification of the GOx-CBD conjugates. The guanidine elution buffer for the experimental biosensor trials was 8 M guanidine hydrochloride (99+ % (C1) from Sigma Chemical Co.) in phosphate buffer. All other chemicals were analytical or reagent grade and used as received. Instrumentation. Electrochemicaldata were obtained using a Pine AFRDE4 bipotentiostat and a Pine AFMSRX analytical rotator (Pine Instrument Co., Grove City, PA). A Pine AFMDI1980 0.5 cm diameter platinum rotating disk electrode (RDE) was used for the experimental indicator electrode (0.196 cm2area). A platinum counter electrode was fabricated in-house from 1.0 mm diameter platinum wire (Aldrich, Milwaukee, WI) bound in flint glass tubing (5 mm outside diameter) with epoxy (Chemgrip, Norton Co., Wayne, NJ). A 25 X 25 mm2, 52 mesh platinum wire gauze (Aldrich) was crimped onto the platinum wire to increase the surface area of the counter electrode. The reference electrode was a saturated calomel electrode (SCE) (Fisher Scientific, Ottawa, Ontario, No. 13-620-52). A Kipp & Zonen Model BD112 strip chart recorder (Kipp & Zonen, Delft, The Netherlands) was used for recording sensor output, and a Kipp & Zonen Model BD91 XYY't recorder was used for recording cyclic voltammograms. The platinum working electrode (RDE) was prepared by polishing with 0.05 pm alumina and rinsing with distilled water. Immediately before each experiment, the working electrode was cycled in quiescent 0.5 M H&04 from -0.26 V to +1.2 V at 100 mV/s for 10 min, anodized at +1.8 V for 10 min, and then cycled again until a stable cyclic voltammogram was obtained. Enzyme-CBDConjugation. Glucoseoxidase was first activated with glutaraldehyde by adding a 50-fold excess of glutaraldehyde for each amino group (lysine residue or N-terminus) on the enzyme (Gibsonand Woodward, 1992). After incubation overnight at 4 "C, the excess glutaraldehyde was removed by dialysis versus phosphate buffer using Spectra/Por 2 dialysis tubing (MWCO 1200014 000). CBDbxwas added in a 1:l molar ratio based on glucose oxidase and incubated overnight at 4 "C. The CBD binds to the activated GOx either via the singlelysine residue present or via the N-terminus of the CBD polypeptide (Coutinho et al., 1992; Hansen and Mikkelsen, 1991): GOX-NH2

+ OHC(CH2),CH0 + H2N-CBD

-

GOx-N=CH-(CH2),-CH=N-CBD

(1) Excess CBD was removed by buffer exchange with phosphate buffer in an Amicon stirred cell using an Amicon PM30 (non-cellulose) membrane (Amicon Canada Ltd., Oakville,ON, Canada). The GOx-CBDconjugatewas then purified in a single step by binding to Avicel. The Avicelbound conjugate was washed with 1M NaCl in phosphate buffer and then with 5 mM phosphate buffer. The

435

Bbtednwl. Rug., 1994, Vol. 10, No. 4

A

B 1 2 3 4 5 6

1

2

3

4

5

6

7

97 kD 68 kD

43 kD Figure 2. (A) Gel electrophoresis (SDS-PAGE) of the GOx-CBD conjugate. A 7.5% polyacrylamide gel and Coomassie staining were used. Lane designations: 1,size markers; 2, GOx control; 3, raw conjugate after binding to cellulose; 4, raw conjugate before binding to cellulose; 5, purified conjugate after binding to cellulose; 6, purified conjugate before binding to cellulose. (B) Western blot of the gel in A. Primary antibody was rabbit-arCBDb.

conjugate was eluted from the Avicel using distilled, deionized water, and the purified conjugate was made up to a 50 mM phosphate-buffered solution with an aliquot of 0.5 M phosphate buffer. The GOx-CBD conjugate was stored in 50 mM phosphate buffer at 4 "C either in the soluble form or bound to Avicel for longer term storage. Glucose Biosensor Experiments. The RDE was modified using a retainer such that different cellulose matrices could be draped over the platinum disk and held in place. The retainer was either a Teflon ring or a rubber ring formed by cutting a slice off the end of a piece of Tygon tubing and sized to fit tightly over the RDE. Three different cellulose-basedmembranes were used during the experimental biosensor trials: (1)Whatman No. 1qualitative filter paper, (2) Spectra/Por 2 regenerated cellulose dialysis membrane (MWCO 12 000-14 OOO), and (3) nitrocelluloseprotein transfer membrane (0.45 pm pore size) from Schleicher & Schuell (Keene, NH). The GOx-CBD conjugate was loaded by immersingthe electrode assembly in a solution of GOx-CBD conjugate, followed by a single wash with phosphate buffer. The RDE was rotated at 500 rpm and potentiostated a t +0.7 V versus SCE. The sensor was calibrated in PBS by recording the steady-state RDE current in response to a series of aliquots of glucose standard. The sensor response to decreasing concentrations of glucose was recorded by removing aliquots of the cell electrolyte and replacing with fresh PBS. Elution of the GOx-CBD conjugate was achieved by immersing the electrode assembly in 8 M guanidine elution buffer, followed by a series of washes with phosphate buffer only before reloading fresh GOx-CBD conjugate. Total Protein and Enzyme Activity Assays. Total protein was assayed using the Bio-Rad protein assay (BioRad Laboratories, Richmond, CA). Glucose oxidase activity was assayed using the Sigma GOx activity assay (SigmaProcedure No. 510, Sigma Chemical Co., St. Louis, MO). Samples were assayed in triplicate on a 96-wellplate using the Molecular Devices Vmax kinetic microplate reader (Menlo Park, CA).

Results and Discussion GOx-CBD Conjugation. Figure 2 shows the results of gel electrophoresis (SDS-PAGE) of the GOx-GBD conjugate. A homogeneous, 7.5 % polyacrylamide gel was

used, suitable for a molecular weight range of 45000200 OOO. Bands containing CBDcex were identified by a Western blot using antibody for CBDa.. The GOx control in lane 2 shows a band between 68 OOO and 97 000 due to the subdivision of GOx (Mr155 000) (Wilson and Turner, 1992) into two equal subunits. The CBDCex control (M, 11000)(Ong et al., 1993) was not separated on this gel but was carried to the anode in the dye front. The raw GOxCBD conjugate before purification is shown in lane 4. Uncoupled GOx was present in the raw GOx-CBD conjugate, but not after purification of the sample by binding to cellulose, as shown in lane 3. SDS-PAGE of a sample of purified GOx-CBD conjugate both before (lane 6) and after (lane 5) binding to cellulose does not demonstrate any significant change due to binding to cellulose. The GOx-CBD conjugation product appears in bands at 200 000 and higher. The soluble conjugation product retained greater than 60% of the activity of the original, unconjugated enzyme. In addition, the GOx-CBD conjugate retained the binding affinity of the CBD for cellulose and also exhibited GOx activity when bound to cellulose that was comparable to that of GOx immobilized by other techniques (Harrison et al., 1988). The conjugate could be adsorbed and desorbed from cellulose repeatedly and could be used for up to 2 months when stored bound to Avicel in phosphate buffer at 4 "C. The cellulose-bound conjugate could also be dehydrated for storage a t room temperature and reconstituted with PBS (pH 7.4) without a significant loss of activity. The high molecular weight of the product is attributed to the cross-linking of glucose oxidase by glutaraldehyde reagent in the activation step of the conjugation protocol. That is, the mechanism as indicated in eq 1is probably simplistic. Although the quantity of glutaraldehyde added was calculated as a 50-fold molar excess over the enzyme amino group concentration, this was based on the assumption of a monomeric form of glutaraldehyde. It is quite possible, however, that the actual cross-links involve a polymeric form (Quiocho, 1976). This would, of course, decrease the effective molar excess of the cross-linking reagent and may partially account for the high retention of enzymic activity in the conjugate. This high degree of cross-linking of glucose oxidase may also cause a GOx:

Bbtechnol. Prog, 1994, Vol. 10, No. 4

436

CBD conjugation ratio greater than unity. This feature of the conjugate protein could be advantageous, in that a higher enzymic activity could be loaded onto a fixed number of CBD binding sites on a given cellulose matrix. On the other hand, the cross-linking of the glucose oxidase may cause a change in the conformation of the protein that could reduce the enzymic activity or change the enzyme specificity. It may be possible to tailor the GOx: CBD conjugation ratio by modifying the conjugation protocol in order to increase,reduce, or eliminate this crosslinking. A potential alternative to chemical conjugation would be to develop an appropriate genetic construct that yields an active GOx-CBD fusion protein. Apropos of this, the gene encoding the glucose oxidase protein of Aspergillus niger has been cloned and expressed in yeast (Frederick et al., 1990;DeBaetselier et al., 1991). The main advantages of such “genetic conjugates” would be greater uniformity and homogeneity of the conjugate reagent, as well as considerable simplification of large-scaleconjugate production. A possible disadvantage, however, might be that the size and G0x:CBD ratio of the resulting conjugates could not be as easily tailored when compared to the chemical conjugation method, which may limit the efficacy of the GOx-CBD reagent. The feasibility of a genetically engineered GOx-CBD fusion protein is being investigated. A t present, the chemically conjugated protein is suitable for demonstration of the concept of a renewable biosensor. The conjugation protocol has not been modified to investigate the sensitivityto different concentration ratios of GOx and CBD or to optimize the conjugation procedure. Further characterization of the GOx-CBD conjugate will follow the optimization of the final conjugation protocol. Experimental Glucose Biosensor. An experimental glucose biosensor was assembled using an RDE system in order to investigate the feasibility of loading, eluting, and reloading the enzyme-CBD conjugate on a platinum electrode with an immobilized cellulose matrix. The experimental system was intended to simulate the automated process. In the prototype system, the appropriate reagent solutions will be pumped to the cellulose matrix enclosed in the enzyme chamber of the probe via internal tubing in the probe body. The system will be automated using computer-controlled pumps and valves for the reagents. The RDE system was advantageous during the first stages of development because experimental parameters could be easily manipulated. In addition, the rotation of the RDE created well-defined hydrodynamic conditions at the surface of the electrode, which contributed to a stable, reproducible sensor signal, as well as reproducible conditions for loading/eluting the GOx-CBD conjugate. In the experimental system, this was essential for evaluating and comparing different procedural modifications and conditions. In order to enable easy access to the cellulose matrix/enzyme membrane, no coveringdialysis membrane was used in the work reported here. Within the scope of the present feasibility investigation, the analyte medium was limited to well-defined solutions of PBS and glucose, and the use of a covering membrane was not imperative. However, a suitable covering dialysis membrane will be required for monitoring glucose in fermentation broths. The addition of a covering dialysis membrane should extend the linear range of the biosensor, although with the result of a slower response time. A number of candidate membrane materials currently are being evaluated. Figure 3 shows a typical calibration curve obtained for the experimental glucose biosensor. Relatively large

1

t

I

I

/

00

0

5

10

15

20

Glucose Concentration (mM)

Figure 3. Typical calibration data for the experimentalglucose biosensor. The GOx-CBD conjugatewas boundtoa nitrocellulose matrix on the Pt RDE. The RDE was rotated at 500 rpm and potentiostated at +0.7 V versus SCE. Data points represent increasing (H) and decreasing (+) changes in glucose concentration. The solid line represents a nonlinear parametric fit of the calibration data (concentration changes in the increasing direction) using the Michaelis-Menten model for enzyme kinetics. Apparent kinetic parameters were 1- = 4.4 MAand K,’ = 21.3 mM.

electrode currents were obtained in response to glucose, indicating that a substantial amount of enzyme had been immobilized. The sensor was calibrated using increasing and decreasing concentrations of glucose in order to demonstrate the reversibility of the sensor, a necessary characteristic for fermentation monitoring. The response time for decreasing concentrations of glucose (60-80 s) was longer than the response time for increasing glucose concentrations (50 s), suggesting a storage effect due to the cellulose matrix. This effect was not considered to be critical at this stage of development, as a different system configuration will be used in the prototype sensor design. A different response time, signal magnitude, and durability were observed for each cellulose matrix used (see Table 1). The response time was measured as the time required for the sensor signal to reach steady state following the injection of an aliquot of glucose standard into the electrolyte. The signal attenuation due to the cellulosematrix was determined relatively by injecting an aliquot of H202 into the electrolyte and recording the magnitude of the sensor current for each cellulose matrix. The filter paper matrix had the fastest response time (1550 s) and the highest signal magnitude, but disintegrated over time when wet and when subjected to stress, such as the shear stress encountered when the RDE was rotated at high rpm. The regenerated cellulose matrix was more durable, but the response time was much slower (2.5-5 min) and the signal was severely attenuated, indicative of the substantially larger mass transport resistance of the regenerated cellulose when compared to the filter paper. The nitrocellulose matrix with the 0.45 pm pore size was equally as durable as the regenerated cellulose matrix when exposed to the 8 M guanidine elution buffer, but it hardened and cracked when 1 M NaOH was used as the elution buffer. Nevertheless, the response time when the nitrocellulose matrix was used was 20-50 s, and the signal attenuation was not as severe as that of the regenerated cellulose. For the best combination of response time, signal magnitude, the durability, the nitrocellulose matrix was used in conjunction with the 8 M guanidine elution buffer. To investigate the possibility of desorption of the enzyme-CBD conjugate during the experiment, the back-

Blotechnol. Rog., lW4, Vol. 10, No. 4

437

Table 1. Properties of the Cellulose Matrices Used in the Experimental Glucose Biosensor System* sensor current in response time response to 50 p L of 0.03 5% HzOz in following glucose cellulose PBS W) standard injection matrix fiiter paper 0.188 16-60 8 0.050 2.5-5 min regenerated cellulose nitrocellulose 0.110 20-50 8 a The cellulose matrix was held on the Pt RDE by a retaining ring. For the glucose response time, the cellulose matrix was loaded with GOx-CBD conjugate.

ground signal was monitored using a bare, GOx-free, platinum electrode, and the electrolyte in the electrochemical cell was assayed for GOx activity following the experiment. No GOx activity could be detected, indicating that the enzyme was sufficiently strongly bound to the sensor. Figure 4 shows calibration curves obtained for five complete cycles of enzyme loading, elution, and replacement. The GOx-CBD conjugate sample was divided into equal aliquots before the experiment, and a fresh aliquot was used for each loading cycle. Total protein assays of the aliquots following the experiment showed an approximately 20% variation in the amount of protein loaded. The sensor was calibrated following each GOxCBD conjugate loading procedure (loads 1-5). After each calibration curve was recorded, the cellulose membrane was washed with 8 M guanidine elution buffer to remove the attached enzyme. A typical calibration curve following the elution step is included in Figure 4, demonstrating that the sensor does not respond to glucose after the elution step. The elution procedure was found to be effective in clearing all active enzyme from the cellulose matrix; however, the possibility exists that some denatured enzyme may have remained attached. No attempt was made to quantify residual denatured enzyme that remained bound to the cellulose matrix following the elution procedure; the absence of a clear, sequence-dependent trend in the calibration curves reported in Figure 4 suggests that very little enzyme remained attached. The repeatability of the calibration curves following each loading step is attributed to differences in the amount of enzyme activity loaded and is not unlike the typical repeatability observed for enzyme immobilization by other methods. For a given experiment, the specific activity of the GOx-CBD conjugate sample and the protein loading data could be used to calculate the amount of enzyme activity loaded in each loading cycle. Results were typically in the range of 0.5-3 units. The shape of the calibration curves suggested that the Michaelis-Menten equation for enzyme kinetics might be useful as an experimental model. In the case of an amperometric enzyme electrode, the sensor current can be taken to be representative of the enzyme rate, since the electron transfer at the electrode is relatively fast, and the sensor current and substrate concentration data can be used to determine the kinetic parameters of the MichaelisMenten equation. The kinetic parameters determined for the immobilized enzyme are not necessarily the same as those for the soluble enzyme, however. The kinetic parameters determined for the immobilized enzyme will include unresolved mass transport resistances due to the immobilization matrix and the configuration of the sensor system (e.g., covering membrane, sensor geometry, etc.) and, hence, must be denoted as “apparent” kinetic parameters (Leypoldt and Gough, 1984). The apparent Michaelis constant, &’, will be intrinsic to the enzyme

0

5

10

15

20

5

Glucose Concentration (mM)

Figure 4. Calibration data for multiple cycles of loading and elution of the GOx-CBD conjugate (seeFigure 3 for experimental conditions). The GOx-CBD conjugate sample was divided into equal aliquota before the experiment, and a fresh aliquot was used for each loading cycle. The elution buffer was 8 M guanidine in phosphate buffer. A typical glucose response curve following the elution of the GOx-CBD conjugate is shown. Calibration curve8 are for load 1 ( 0 ) ;load 2 (0);load 3 (A);load 4 (m); load 5 (+); and after elution (e).

and will also be dependent on the sensor configuration used. The maximum sensor current, I-, will be analogous to the maximum reaction velocity, V-, and can be used as an indicator of the amount of enzyme loaded or the amount of immobilized enzyme activity. The Michaelis-Menten model is useful for ita simplicity, and it describes the experimental data very well, as shown in Figure 3. This observation was supported by the linearity of a Lineweaver-Burk plot of the data from the multiple cycle experiment (Figure 5) and the correlation coefficients from linear regression analysis of the Lineweaver-Burk data (Table 2). Table 2 shows the apparent enzyme kinetic data determined from the LineweaverBurk plot in Figure 5 and from a nonlinear parametric fit of the data in Figure 4 using the Michaelis-Menten equation. The possibility of a trend in the apparent kinetic data of column 6 (Table 2) was considered to be artifactual, as no trends were observed in any other kinetic data. Using an accurate empirical these two parameters (I- and L’), model for the calibration curve can be obtained, which enables operation of the sensor over a working range much greater than the linear range normally reported in biosensor literature (Cleland and Enfors, 1984a,b; Bradley and Schmid, 1991). Application to Fermentation Processes. For fermentation applications, enzyme electrode sensors in general should have the following characteristics. 1. The capability for continuous, on-line operation is necessary for process control and to avoid the possibility of changes in the sample during measurement by off-line methods. 2. The measuring range and response time of the sensor must be suitable for monitoring changes in the analyte concentration encountered in the fermentor with sufficient speed and sensitivity for process control. 3. The sensor must be resistant to electrochemical interferenta and enzymatic denaturantsfinhibitors, etc., in the analyte medium, and must be insensitive to changes in medium chemistry, such as dissolved oxygen tension, pH, and ionic strength. 4. The sensor system should allow for recalibration during fermentor operation. Periodic recalibration is

Bbtechnol. Prog., 1994, Vol. 10, No. 4

438

l/(Glucose Concentration) (mM-')

Figure6. Lineweaver-Burkplot of the calibration data in Figure 4. Curves are for load 1 (a);load 2 (0);load 3 (A); load 4 (a); load 5 (+). Table 2. Apparent Enzyme Kinetic Data Derived from the Calibration Data of Figure 4 linear regression of the nonlinear parametric Lineweavel-Burk plot fit of the calibration data (Figure 5) of the calibration in Figure 4 using the Michaelis-Menten equation data in Figure 4

Iload 1 2 3 4 5

(MA) 7.0 5.2 7.6 7.4 8.5

Km' (mM) 18.3 15.1 18.0 15.9 17.4

rz

0.9999 0.9999 0.9998 0.9999 0.9998

1&A) 6.4 5.3 5.9 6.0 6.3

Km'

(mM) 15.7 15.2 12.8 11.8 11.2

X2

0.0010 0.0004 0.0008 O.OOO9

0.0016

desirable in order to correct for changes in the calibration constants of the sensor that may occur during a fermentation due to fouling,poisoning, deactivating or other intemal chemical changes in the probe (Kok and Hogan, 1987/88). 5. Sufficient long-term stability of the enzyme component of the sensor is necessary for it to be useful in a typical fermentation. This may dictate that periodic recalibration of the sensor is required if the enzyme activity degrades. Alternatively, this may require the capacity for replacement of the enzyme component when the activity has degraded to an unsatisfactory level. This regeneration of the probe must be possible without interrupting the fermentation. 6. In situ sterilization of the probe by conventional methods, such as autoclaving, must be possible. Steam Sterilizationis preferred by industry over other sterilization methods, such as ethanol, chloroform, radiation, etc. One possible hardware configuration for the application of CBD technology for enzyme immobilization to in situ fermentation monitoring is illustrated schematically in Figure 6. The basic design consists of a platinum electrode, a porous cellulose matrix, and a protective dialysis membrane, all incorporated into a stainless steel probe for insertion into the bioreactor. Inlet and outlet tubes in the probe body allow for perfusion of the cellulosematrix with the-enzyme-CBD conjugate solution and the elution buffer. The internal electrode can be raised for complete perfusion of the cellulose matrix with the appropriate reagents. The concept is similar to that employed in the commercial COz electrode from Ingold (1990). A prototype glucose biosensor using this principle is being built using the probe body from the Ingold COZ electrode, and glucose-

permeable outer membranes are being evaluated by considering not only the sensor performance but also the permeability and fouling of the membrane of components of the bioprocess. The process flow diagram in Figure 7 outlines the principle of operation of the sensor system for continuous fermentation monitoring. Following steam sterilization of the bioreactor and the probe body, the enzyme component is loaded into the sensor by perfusion of the cellulose matrix with the enzyme-CBD conjugate solution, resulting in immobilization of the enzyme onto the cellulose. The sensor is initially calibrated during the addition of glucose to the fermentation broth. Response to internal calibration standards can then be measured and compared to the external calibration response. Bradley and Schmid (1991) reported that the relationship between the response to internal calibrant and external glucose concentration was constant and depended on the membrane and internal buffer flow rate selected. The relationship between the internal calibration and the initial external calibration constants can be used later during the fermentation for periodic calibration checks using the internal calibration standards. If the calibration check is satisfactory, then fermentation monitoring continues. Otherwise, the sensor calibration constants may be updated, or if the enzyme activity or electrochemical performance of the sensor is diagnosed as having degraded to an unsatisfactory degree, the sensor regeneration cycle is started. The sensor can be regenerated without interrupting the fermentation by perfusing the cellulose matrix with the elution buffer to remove the attached enzyme. To continue fermentation monitoring, the sensor is reloaded with enzyme-CBD conjugate and the sensor calibration constants are reset using the internal calibration standards. The sensor calibration can be verified, if desired, by comparison of the sensor signal with the analysis of a sample of the fermentor broth using an off-line glucose analyzer. The sensor could also be perfused with the appropriate solutionsfor in situ membrane cleaning or electrode cycling before the cellulose matrix is reloaded with fresh enzymeCBD conjugate. It would also be possible to clean the external surface of the membrane at this point, without breaching sterility. Two approaches have been described: Biihler and Ingold (1976) designed a probe that can be retracted completely from the fermentor and then isolated from the fermentor contents by a ball valve. Kok and Hogan (1987/88) reported a method for mechanically isolating the probe tip from the fermentation broth, using two half-cylinders that are forced together pneumatically to form a small chamber around the probe tip. Once the probe tip has been isolated from the fermentation broth, the membrane can be cleaned by washing or spraying with a jet of cleaning solution, followed by resterilization of the probe tip with steam before reexposure to the fermentor medium. If the regeneration process is limited to a simple exchange of the enzyme component and a two-point internal calibration, then a regeneration cycle could realistically be completed in less than 5 min, since the absorption and desorption kinetics of the CBD are very fast. Previously published CBD binding studies indicated that adsorption of the CBD to cellulose was complete within the shortest incubation time feasible under the conditions of the experiment (i.e., 0.2 min) (Gilkes et al., 1992),but the actual adsorption kinetics are likely much faster. The advantage of this approach is that the complete process of diagnosis, regeneration, and recalibration could

Rug., 1994, Vol. 10, No. 4

Bbt&mol.

439

( Enzyme-CBD Reagent < Inlet Conduit

Figure 6. Schematic diagram of a biosensor incorporating a cellulose matrix for immobilizationof the enzyme-CBD conjugate. One proposed sensor hardware configuration is shown. Load Enzyme-CBD

Sterilize

Run

Calibrate

I

Check Calibration Clean ok? Electrode

ym

I

Elute Enzyme-CBD

Figure 7. Process flow diagram demonstrating the principle of operation of a fermentation monitoring system using the renewable biosensor probe.

be performed in situ and under computer control. The proposed system allows for (a) conventional autoclaving of the electrode housing exclusive of the enzyme, (b) periodic internal calibration verification, (c) long-term sensor operation by replacing the enzyme when necessary, in situ, and without interrupting the fermentation, (d) recalibration followingenzyme replacement using internal standards, and (e)the possibilityof cleaningthe membrane and the electrode during the enzyme replacement cycle. In addition, it should be straightforward to conjugate other oxidase-type enzymes to CBD, such that the same sensor hardware could be used to measure different analytes in different fermentations or a t different stages of the same fermentation, depending on which enzymeCBD conjugatesolution was perfused through the cellulose matrix. A multiple enzyme sensor for analytes requiring a multienzyme system could also be realized by incorporating two or more different enzyme-CBD conjugates in the same perfusate.

Conclusions Glucose oxidase and the cellulose binding domain have been successfullyconjugatedchemically,and the conjugate protein has been used in an experimental glucosebiosensor with a cellulose matrix. The nitrocelluloseprotein transfer membrane with 0.45 pm pore size was shown to be the

cellulose matrix that provided the best combination of response time, signal magnitude, and durability in the experimental system. Loading and eluting protocolsusing 8 M guanidine elution buffer were developed to simulate multiple cycles of loading and eluting that would be encountered during a typical fermentation. These results are the first to demonstrate the concept and feasibility of a renewable biosensor based on reversible immobilization of the enzyme using CBD technology. This represents a significant step toward a practical, industrially acceptable probe design and also toward better instrumentation for fermentation monitoring and control. The realization of a reliable, on-line system for glucose analysis has enormous potential for the evolution of new fermentation control strategies. Experimental work continues toward the development of a prototype renewable glucose biosensor based on the Ingold CO1 electrode (Ingold, 1990). The sensor system will be used in pilotscale fermentations for the investigation and optimization of the growth rate and product production of industrial/ commercial fermentations, based on glucose control. Experimental work also continues on the further characterization of the glucose oxidase-CBD conjugates, the feasibilityand utility of developinga genetically engineered fusion protein to replace chemical conjugation, and the application of the CBD immobilization approach to other oxidase enzyme systems.

Acknowledgment The authors acknowledgefinancial support of this work provided by the Natural Sciences and Engineering Research Council of Canada (NSERC). We are also grateful to Dr. E. Ong and Dr. A. Wierzba for their valuable technical assistance in purifying and characterizing the GOx-CBD conjugate.

Literature Cited Bailey, J. E.; Ollis, D. F. In Biochemical Engineering Fundamentals, 2nd ed.; McGraw-Hill: New York,1986;pp 105-108.

Bbtechtwl. Frog., 1994, Vol. 10, No. 4

440

Bradley, J.; Schmid, R. D. Biosensors for in situ Fermentation Monitoring of Glucose Concentration. Process Biosens. 1990, 90 (May, Singapore), 255. Bradley, J.; Schmid, R. D. Optimisation of a Biosensor for insitu Fermentation Monitoring of Glucose Concentration. Biosens. Bioelectron. 1991, 6, 669-674. Bradley, J.; Anderson, P. A,; Dear, A. M.; Ashby, R. E.; Turner, A. P. F. Glucose Biosensorsfor the Study and Controlof Baker’s Compressed Yeast Production. In Computer Applications in Fermentation Technology: Modelling and Control of Biotechnological Processes; Fish, N. M., Fox, R. I., Thornhill, N. F., Eds.; Elsevier Applied Science: New York, 1988; pp 47-51. Bradley, J.; Kidd, A. J.; Anderson, P. A.; Dear, A. M.; Ashby, R. E.; Turner, A. P. F. Rapid Determination of the Glucose Content of Molasses Using a Biosensor. Analyst 1989a, 114, 375-379. Bradley, J.; Turner, A. P. F.; Schmid, R. D. An in situ Fermenter Probe for Baker’s Yeast Propagation Monitoring. In GBF Monographs. Biosensors: Application in Medicine, Environmental Protection and Process Control; Schmid, R. D., Scheller, R., Eds.; VCH New York, 1989b; Vol. 13, pp 85-88. Bradley, J.; Stocklein, W.; Schmid, R. D. Biochemistry Baaed Analysis Systems for Bioprocess Monitoring and Control. Process Control Qual. 1991, 1, 157-183. Brooks, S. L.;Ashby, R. E.;Turner, A. P. F.; Calder, M. R.; Clarke, D. J. Development of an On-Line Glucose Sensor for Fermentation Monitoring. Biosensors 1987/88,3, 45-56. Brooks, S. L.; Higgins, I. J.; Newman, J. D.; Turner, A. P. F. Biosensors for Process Control. Enzyme Microbiol. Technol. 1991,13,946-955. Biihler, H.; Ingold, W. Measuring pH and Oxygen in Fermenters. Process Biochem. 1976, 11 (3), 19-24. Cleland, N.; Enfors, S. 0. Control of Glucose Fed-Batch Cultivations of E. coli By Means of an Oxygen Stabilized Enzyme Electrode. Eur. J.Appl. Microbiol. Biotechnol. 1983,18,141147. Cleland, N.; Enfors, S. 0.Externally Buffered Enzyme Electrode for Determination of Glucose. Anal. Chem. 1984a, 56,18801884. Cleland, N.; Enfors, S. 0. Monitoring Glucose Consumption in an Escherichia coli Cultivation with an Enzyme Electrode. Anal. Chim. Acta 1984b, 163,281-285. Coutinho, J. B.;Gilkes, N. R.; Warren, R. A. J.; Kilburn, D. G.; Miller, R. C., Jr. The Binding of Cellulomonus fimi endoglucanase C (CenC)to Cellulose and Sephadex is Mediated by the N-terminal Repeats. Mol. Microbiol. 1992,6,1243-1252. De Baetaelier, A.; Vasavada, A.; Dohet, P.; Ha-Thi, V.; De Beukelaer, M.; Erpicum, T.; De Clerck, L.; Hanotier, J.; Rosenberg, S. Fermentation of a Yeast Producing A. Niger GlucoseOxidase: Scale-up, Purification and Characterization of the Recombinant Enzyme. BiolTechnology 1991,9,559561. Enfors, S. 0.Oxygen-StabilizedEnzyme Electrode for D-Glucose Analysis inFermentation Broths. Enzyme Microbiol. Technol. 1981, 3, 29-32. Enfors, S. 0.; Molin, N. Enzyme Electrodes for Fermentation Control. Process Biochem. 1978, 13, 9-11. Enfors, S. 0.;Nilsson, H. Design and Response Characteristics of an Enzyme Electrode for Measurement of Penicillin in Fermentation Broth. Enzyme Microbiol. Technol. 1979, 1, 260-264. Filippini, C.; Sonnleitner, B.; Fiechter, A; Bradley, J.; Schmid, R. On-Line Determination of Glucose in Biotechnological Processes: Comparison Between FIA and an in situ Enzyme Electrode. J. Biotechnol. 1991, 18, 153-160. Frederick, K. R.; Tung, J.; Emerick, R. S.; Masiarz, F. R.; Chamberlain, S. H.; Vasavada, A.; Rosenberg, S.;Chakraborty, S.; Schopter, L. M.;Massey, V. Glucose Oxidase From Aspergillus niger: Cloning, Gene Sequence, Secretion From Saccharomyces cerevisiae and Kinetic Analysis of a YeastDerived Enzyme. J. Biol. Chem. 1990, 265, 3793-3802. Freitag, R. Applied Biosensors. Curr. Opin. Biotechnol. 1993, 4 , 75-79. Gibson,T. D.; Woodward, J. R. Protein Stabilization in Biosensor Systems. In Biosensors & Chemical Sensors: Optimizing Performance Through Polymeric Materials; Edelman, P. G.,

Wang, J., Eds.; American Chemical Society: Washington, DC, 1992; pp 40-55. Gilkes, N. R.; Henrissat, B.; Kilburn, D. G.; Miller, R. C., Jr.; Warren, R. A. J. Domains in Microbial fi-1,4-Glycanases: Sequence Conservation, Function, and Enzyme Families. Microbiol. Rev. 1991,55, 303-315. Gilkes, N. R.; Jervis, E.; Henrissat, B.; Tekant, B.; Miller, R. C., Jr.; Warren, R. A. J.; Kilburn, D. G. The Adsorption of a Bacterial Cellulaseand Ita Two Isolated Domains to Crystalline Cellulose. J. Biol. Chem. 1992,267 (lo),6743-6149. Hansen, E. H.; Mikkelsen, H. S. Enzyme-Immobilization by the Glutardialdehyde Procedure. An Investigation of the Effecta of Reducing the Schiff-BasesGenerated, as Based on Studying the Immobilizationof GlucoseOxidase to Silanized Controlled Pore Glass. Anal. Lett. 1991,24 (81, 1419-1430. Harrison, D. J.; Turner, R. F. B.; Baltes, H. P. Characteriation of Perfluorosulfonic Acid Polymer Coated Enzyme Electrodes and a Miniaturized Integrated Potentiostat for Glucose Analysis in Whole Blood. Anal. Chem. 1988,60,2002-2007. Huang, Y. L.; Li, S. Y.; Dremel, B.A. A.; Bilitewski, U.; Schmid, R. D. On-Line Determination of Glucose Concentration Throughout Animal Cell Cultures Based on Chemiluminescent Detection of Hydrogen Peroxide Coupled with Flow-Injection Analysis. J. Biotechnol. 1991, 18, 161-172. Ingold, W. COz Measuring System Product Brochure, Ingold Electrodes Inc., Wilmington, MA, 1990. Kilburn, D. G.; Turner, R. F. B.; Coutinho, J. B.; Din, N.; Gilkes, N. R.; Greenwood, J. M.; Hobbs, J. B.; Miller, R. C., Jr.; Ong, E.; Phelps, M. R.; Ramirez, C.; Warren, R. A. J. Cellulose Binding Domains: Applications in Biotechnology. Proc. Cellucon 92 1992, in press. Kok, R.; Hogan, P. The Development of an in situ Fermentation Electrode Calibrator. Biosensors 1987/88,3, 89-100. Leypoldt, J. K.; Gough, D. A. Model of a Two-substrate Enzyme Electrode for Glucose. Anal. Chem. 1984,56, 2896-2904. O’Neill, G.; Goh, S. H.; Warren, R. A. J.; Kilburn, D. G.; Miller, R. C., Jr. Structure of the Gene Encoding the Exoglucanase of Cellumonas fimi. Gene 1986,44,325-330. Ong,E.;Gilkes,N.R.;Miller,R.C.,Jr.;Warren,R.A. J.;Kilburn, D. G. Enzyme Immobilization Using a Cellulose-Binding Domain: Properties of a BGlucosidase Fusion Protein. Enzyme Microb. Technol, 1991, 13, 59-65. Ong, E.; Gilkes, N. R.; Miller, R. C., Jr.; Warren, R. A. J.; Kilburn, D. G. The Cellulose-Binding Domain (CBDax) of an Exoglucanase from Cellulomonas fimi: Production in Escherichia coli and Characterizatino of the Polypeptide. Biotechnol. Bioeng. 1993,42, 401-409. Quiocho, F. A. Immobilized Proteins in Single Crystale. In Methods in Enzymology: Immobilized Enzymes; Mosbach, K., Ed.; Academic Press: New York, 1976; Vol. 44, pp 546558. S t a ” , W. W.; Seidler, W.; Kahlert, W.; Kappel, W. On-Line Monitoring of Glucose in Batch Fermentations of Escherichia coli with a Commercially Available Biosensor AnalyzerFermentation Control System. Proceedings of the 2nd World Congress on Biosensors, May 1992, Geneva, Switzerland; Elsevier: Oxford, U.K., 1992; p 506. Whittle, D. J.; Kilburn, D. G.; Warren, R. A.J.; Miller, R. C., Jr. Molecular Cloning of a Cellumonas fimi Cellulase Gene in Escherichia coli. Gene 1982, 17, 139-145. Wilson, R.; Turner, A. P. F. Glucose Oxidase: An Ideal Enzyme. Biosens. Bioelectron. 1992, 7, 165-185. Wirth, M.; Li, S. Y.; Schumacher, L.; Lehmann, J. Screening for and Fermentation of High Producer Cell Clones from Recombinant BHK Cells. In Advances in Animal Cell Biology and Technology forBioprocesses; Spier, R. E., Griffiths, J. B., Stephenne, J., Gooy,P. J.,Eds.;Butterworth: Sevenoaks,Kent, England, 1989; pp 44-51. Accepted April 20, 1994.’ 0

Abstract published in Advance ACS Abstracts, July 1, 1994.