Biosensors % ,Bioelmmnics 7 (1992) 645-652
Covalent binding of urease *on ammonium-selective potent~ometric membraiks* M. Ii. Gil+ & A. P. Piedade Department
of Chemistry,
University
of Coimbra,
P-3049 Coimbra,
S. Alegret, J. Aionso, E. Ma~inez-F~b~gas Department
of Chemistry,
Autonomous
University
of Barcelona,
Portugal
& A. Oretlana
E-08193 Bellaterra, Catalonia,
Spain
(Received 26 May 1992: revised version received and accepted 27 July 1992)
Abstract: As part of the development of disposable urea bioselective probes, the covalent binding of urease on ammonium-selective potentiometric membranes has been assessed. Nonactin/bis( 1-butylpentyl)adipate/poly(vinylchloride) (PVC) membranes, directly applied to an internal solid contact (conductive epoxy-graphite composite), has been used as a support for covalent immobilization of urease. Two types of all-solid-state construction process have been assayed: thin layers of cellulose acetate (CA) were coated on the PVC ammonium-selective membranes (type 1) and blends of PVC and CAat various ratios were used as ammonium-selective membrane matrices (type 2). Urease was covalently attached to CA via aldehyde groups. These groups were created on the polysaccharide with sodium periodate to which the enzyme was immobilized through a spacer (hexamethylenediamine). The viability of both types of probe for the dete~ination of ammonium ions was assessed after each step of the activation process. Results indicated that type 2 potentiometric probes are altered after the treatment with sodium periodate. Good results were obtained with type 1 probes. Their dynamic concentration range of response to urea was from 2 X 10m5to 0.01 M with a sensibility of 50 mV/decade. Keywords: disposable immobilization.
urea
probe,
urea potentiometric
biosensor.
urease
Potent~omet~c biosensors usually comprise a membrane featuring biological materials which have been immobilized by physical or chemical
means. Such a membrane is cast directly on a potentiometric sensor based on an ion-selective membrane (H+, NH$’ etc.). An external dialysis membrane is usually employed in such an assembly.
*Paper presented at Biosensors ‘92, Geneva, Switzerland, 20-22 May 1992. +To whom correspondence should be addressed.
A different approach, rarely used in the construction of potentiometric biosensors, is to bind covalently the biological material (i.e.
INTRODUCTION
#56-5663/92/$05.00
@ 1992 Elsevier Science Publishers Ltd.
645
M. HOGil et al.
enzymes) to the transducer surface. This approach has been used in other types of electrochemical transducer (amperometric transducers, ISFETs etc.). This method brings about two innovations when applied to potentiometric biosensors: (a) it simplifies potential mass production by making the use of printing techniques feasible; and(b) the response of the resulting biosensor is made faster by speeding up the transfer of chemical species between the sample solution and the ion-selective membrane. To attain a covalent bond between the enzyme and the potentiometric membrane, the membrane matrix has to be modified by adding functional groups. However, the addition of these functional groups should not affect the response characteristics of the membrane (sensitivity, selectivity). This modification may also be useful for: (a) binding the ionophore covalently to the membrane matrix, thus preventing the ionophore from leaching when it is highly soluble in aqueous solutions; (b) attaching components that render the membrane biocompatible; (c) helping to improve adhesion between the membrane and the conductive substrate in all-solid-state electrodes (where the internal reference solution is absent): and (d) reducing the electrical resistance of the membrane. Examples of modified potentiometric membranes can be found in the literature. chloride) (PVC) Carboxyated poly(viny1 (Lindner et al., 1988; Ma et al., 1988) aminated PVC (Ma et al., 1988) and modified cellulose acetate (CA) (Cha & Meyerhoff, 1989) are examples of these matrices. CA has been used as a potentiometric membrane matrix where the enzyme is directly bound to it. The present work is focused on the development of urea-selective disposable potentiometric bioprobes. Covalent binding of urease on conventional ammonium-selective membranes has been PVC potentiometric assessed. Two strategies have been followed: (a) thin layers of CA have been coated on the ammonium-selective PVC membrane (asymmetric membrane); and (b) blends of PVC and CA have been used as an ammonium-selective membrane matrix. Urease was covalently attached to CA. The resulting bioselective membranes were cast in situ on conductive plastic substrates based on epoxy-graphite composites, thus producing simple potentiometric all-solidstate bioprobes. 646
Biosensors& Bioelectronics
EXPERIMENTAL Reagents Urease (from jack beans) was obtained from Sigma Chemical Co, Poole, Dorset, UK. Bis( lbutylpentyl) adipate and nonactin were supplied by Fluka Chemie AG, Switzerland. Poly(viny1 chloride) (PVC) and tetrahydrofuran (THF) were obtained from BDH, Poole, Dorset, UK Cellulose acetate (CA) was purchased from Kodak. Urea, ammonium chloride and glutaraldehyde (25% in water) were supplied by Merck, Germany. All other chemicals were of analytical grade and were obtained commercially.
Procedure The working and reference electrodes were connected to a digital potentiometer (Crison micropH 2000) and the potential variations were registered in a recorder (type BDl 11 from Kipp & Zonen, The Netherlands). The reference electrode used was a double junction Ag/AgCl electrode (90-02-00 from Orion, USA) with the outer compartment containing 0.1 M Tris-HCl buffer, pH 7.5. The reaction cell was always kept at 25 f O*l”C.
All-solid-state ammonium-selective PVC membrane probes The construction procedure involved the application of a layer of conductive epoxy resin, about 0.7 cm thick, to one of the ends of an PVC tube (electrode body) as described in an earlier paper (Alegret & Martinez-Fabregas, 1989). A shielded lead was fixed inside the tube to the conductive epoxy while still soft. After drying at 40°C for 24 h, the resin was drilled to form a 0.3 mm deep flat cavity. The solution of the sensing ammonium membrane material consisted of 0.1 g of PVC, 0.204 g of bis(lbutylpentyl) adipate and 0.0015 g of nonactin dissolved in 5 ml of tetrahydrofuran. 120 ~1 of this solution was added dropwise to the cavity and allowed to evaporate at room temperature. A convex membrane was formed, up to a central thickness of0.3-0.4 mm. After the assembly of the electrode had been completed, it was conditioned in a 0.1 M NHACl solution for 1 h.
Biosensors & Bioelectronics
All-solid-state ammonium-selective CA-PVC membrane probes The cellulose acetate-poly(viny1 chloride) membrane electrodes were prepared using two methods: Type 1. A cellulose acetate solution was prepared by dissolving the polysaccharide in THF (53 mg/2 ml) at room temperature. Aliquots of this solution (4 X 10 ,u1 each) were dropped on top of the sensing PVC layer and the solvent was evaporated at room temperature (Fig. 1, left). After 12 h the electrode was conditioned in a 0.1 M NH&l solution for 1 h. Type2, Mixtures of CA:PVC (25:75,50:50,75:25 and 100:0) were dissolved in THF (53 mg/2 ml). The PVC content in the sensing cocktail of the ammonium-selective PVC membrane electrodes was replaced by these mixtures and the probes were prepared in the usual manner (Fig. 1, right).
Covalent binding of w-ease
0.5 ml of 0.1 M phosphate buffer, pH 7.0, containing 6 mg/ml of urease. During previous work with CA in powder form (see Tables 4 and 5), the urease was linked to the aldehyde groups by introducing the polymer in a O-1 M phosphate buffer, pH 7.0 (5 ml), containing the enzyme (4 mg/ml). The amount of immobilized enzyme was determined by the method of Lowry et al. (1951). Electrochemical measurements For electrochemical measurements, the urea bioprobe and the reference electrode were introduced in 25 ml of 0.1 M Tris-HCl buffer, pH 7.5. or in 25 ml of the buffer under study. Aliquots of standard solution of urea and ammonium chloride were added. Determination of the aqueous solutions’ sorption capacity
Immobilization of urease The urease was immobilized on the cellulose acetate which contained hexamethylenediamine as a spacer. For the attachment of the amine, aldehyde groups were created on the CA by activation with a solution of 0.5 M sodium periodate (10 ml) at room temperature. The polymer containing the aldehyde groups was treated with a solution of hexamethylenediamine (10 ml of 1% (w/v) solution) for 18 h at room temperature, while stirring. The -NH2 groups were then activated with glutaraldehyde (Habeeb & Hiramoto, 1968). The electrodes with the activated cellulose membranes were immersed in
Samples (20 mg) of CA, PVC or the CA:PVC mixtures (25:75, 50:50 and 75:25) in powder or film were immersed in distilled water, in 0.1 M phosphate buffer, pH 7.0, or in 0.1 M Tris-HCl buffer, pH 7.5 (10 ml), at 25°C for 24 h. The excess water was removed with filter paper. Afterwards, the polymer was, in each instance, weighed every minute for 10 min. The initial sorption capacity was obtained graphically after extrapolation to zero time. Each sample was dried to constant mass under reduced pressure at 100°C. The percentage of solution uptake was given by % sorption
=
W(w) - Wr(c) x 1oo W,(c)
where Wi(w) is the initial wet mass at zero time and W,(c) is the final dry mass corrected with a blank. Determination of contact angles
Fig. 1. Type 1 (iej) and Type2 (right) CA-PVC membrane electrodes. I: electrode body: 2: shielded cable: 3: internal solid contact: 4: ion-selective PVC membrane: 5: CA membrane; 6: ion-selective PVC-CA membrane.
To determine the contact angles of the CA, PVC and CA:PVC mixtures, films were prepared. For this purpose, solutions of the polymeric supports in THF (25 mg/ml) were evaporated at room temperature on microscopy glass plates. Contact angles were determined with a Contact-&Meter (Livereel, Consett, Durham, UK). A drop of water, of 0.1 M phosphate buffer (pH 7-O), of 0 1 M Tris-HCl buffer (pH 7.5) or of 647
M H Gil et al.
First, the effectiveness of these electrodes when measuring ammonium ion was assessed. As shown in Table 1, both types of electrode are suitable for the determination of NH$. The urease was covalently linked to the CA after being treated with sodium periodate. This reaction produced aldehyde groups, to which units extension of hexamethylenediamine (HMD) were covalently bound through the development of Schiffs base complexes. The urease.was then immobilized on the -NH* groups of the extension units after these had been activated with glutaraldehyde. Fig. 2 summarizes the various steps of the activation of the cellulose acetate before the coupling of the urease. The viability of both types of electrode for the determination of NH2 was assessed after each step of activation. The results, indicated in Table 2, show that the Type 1 probes are suitable for the determination of NH4+ even after all the activation procedure has been completed. However, the response of the Type 2 probes is altered after the treatment with sodium periodate. In order to explain the lack of activity shown by this type of membrane, we assessed the effect of the periodate on nonactin. This compound was treated with the sodium periodate before being mixed with PVC and then the sensing membrane was prepared. We observed that, in this case, there was no response of the sensor to NH:. The results obtained with various PVC:CA mixtures show that for percentages of CA higher than 25% the membrane did not respond to NH$. In the light of these results, the work was carried out using the Type 1 electrodes. In order to get a good yield of activity of the immobilized urease, the reaction times for the different steps of activation shown in Fig. 2 were
an ethanol:water mixture (34:66; 48:52; 6040) was placed on the surface of the membrane and the contact angle was determined at 20°C. Determination of aldehyde aad amino groups The amount of aldehyde groups was determined by a potentiometric titration (Siggia & Maxcy, 1947). The amount of -NH2 free groups in the polymers was determined by an acid-base titration. Samples of the polymer (50 mg) were immersed in O-1 M HCl(l0 ml) for 12 h at room temperature. The HCl solutions were then backtitrated with 0.1 M NaOH solution, using 10 ml of O-1 M HCl as a blank. Chemical determination of urease activity The urease activity was determined using Nessler’s reagent (Vogel, 1978) (1 activity unit = 1 PM of NH: produced min-I). All measurements were done using a Jasco 7800 Spectrophotometer.
RESULTS AND DISCUSSION Immobilization of urease The purpose of this work was to immobilize urease onto ammonium-selective PVC membranes. In order to obtain reactive groups on these membranes, cellulose acetate was used as one of their membrane components. The preparation of two types of potentiometric probe, Type 1 and Type 2, has already been described (see Fig. 1).
TABLE 1 Response
Electrode
Standard Type 1 Type 2 Type 2 Type 2 Type 2
PVC
characteristics of CA-PVC mination of NH$
C&PVC
Potential response (mV/de~de)
-
63 62 61 61 61 57
25:75 5050 75:25 loo:0
“Lower limit of linear response
electrodes
LLLRa (MI 2 2 2 2 2 2
x x x x x x
IO-” 10-5 lo-’ lo-” 10-S 10-5
in the deter-
Response time Is) 5 5 5 5 5 5
Biosensors & ~i~l~~nics
Covalent binding of urease
step 1
+ NH2~CH2~~~2 (KMDJ
H2N&HC’)N
step 2
+ CHO(CH,),CHO
step 3
W-raldehyW H,N,@WN
N(CH,),NH,
WCH,l,NH,
$&HC)N
N(CH&# YH tCH& I
T (C&)3 I
COH
COH
Fig. 2. Schematic representation of the various steps of activation of cellulose acetate before enzyme coupling.
optimized using the cellulose acetate in powder form. The ensuing coupling reactions were realized by using the optimized times indicated in Table 3. Urease was then covalently linked to the electrode and the activity was determined using chemical and eIect~hemica1 methods. The results, indicated in Table 4, show that: (i)
the yield of activity of the immobilized enzyme is higher when cellulose acetate is used in powder form;
(ii)
the amount of reactive groups formed during the activation procedure is not directly related to the activity of the immobilized enzyme; (iii) the results obtained with these systems are promising; (iv) the Kh value of the immobilized enzyme is very high compared to the KM of the free enzyme; (v) the optimum pH value does not change when the urease is immobilized.
TABLE 2 Response of the CA-PVC electrodes to NH$ before and after each step of activation
Electrode (C&PVC)
Potential response (mV/decade) before activation
Type 1 Type Trpe Type Type
2 2 2 2
(2975) (50:50) (75:25) (1oO:O)
after step 1 after step 2
after step 3
62
63
63
63
61 61 61 57
43 0 0 0
43 0 0 0
43 0 0 0
M H. Gil et al.
Biosensorsh Bioelectronics
TABLE 3
Optimized
Time of activation with IO; (h)
times of reaction for the various steps of urease immobilization onto cellulose acetate powder
Time of reaction with Time of activation HMD (h) -NH2 (h) 18
0.75
enzyme (h)
3
18
[IO;] = 0.5 M; [HMD] = 1% (w/v); ]glu~mldehyde]
= 5% (v/v); [enzyme] = 4 mgfml.
TABLE 4
of urease onto cellulose acetate
Results of the immobilization
Determination
procedure
Cellulose acetate powder
Cellulose acetate Free enzyme electrode (Type 1)
mmol aldehyde groups/g (after step 1)
5.7
5.7
mmol -NH2 groups/g (after step 2)
0.13
0.19
mmol aldehyde groups/g (after step 3)
3.8
3-5
mg ureaselg
8.5
500.0
mg ureaselelectrode Immobilization Yield of activi@
0.04
yield” (%) (%)
Units/g
5.1
0.81
2.8
O-07
26 1
Units/electrode
V 1 chemical det. (U/mgE) max electrochem det. (U/electrode) pH 1 chemical de&. electrochemical a
h
1596 0.011
chemical det. (mM) Km ! electrochemical det. (mM)
312.5
8.0 16.7
IO.0
26.0 100.0
7.5
7.5
7.0
det.
mg of bound enzyme x 100 mg total enzyme in solution mg of active bound enzyme x 100 mg of total active enzyme in solution
Fig. 3 shows the calibration curves obtained with the urease biosensor on day 1. These results can be seen as promising, as well as the ones in Table 5, which show the calibration curve characteristics obtained with the same biosensor after several periods of time had elapsed. When not in use, the electrode was kept in distilled water at 4°C.
650
of Time of reaction with
Characterization of the cellulose acetate (in film and powder forms) In order to relate the immobilization results to the chemical and physical properties of the supports, CA, PVC and CA:PVC mixtures, in either film or powder form. were characterized before and after their activation by measuring the contact angles
Biosensors & Bioelectronics
-60
.
0
Fig. 3. Calibration
Covalent binding of urease
. .o 0. O 0 O0
0
curves obtained biosensor.
urea
ammonium
l
with the
urease
and the aqueous solutions’ sorption capacity. The results of the contact angle determinations are shown in Table 6. It was also possible, from the results of the contact angles of the films with ethanol/water mixtures, to evaluate the surface
energy of the films before and after the treatment with periodate, HMD and glutaraldehyde. These results are obtained by plotting the surface tension of the ethanol/water systems used against the cosine of the contact angles obtained. For each film, the intersection of the plot with the cos 8 = 1 corresponds to the critical wetting tension of the polymeric films. The results obtained show that (a) the variation of the liquid affinity for the studied surfaces is not significant; and (b) the critical wetting tension increases slightly after the activation treatment of the various studied supports. The results shown in Fig. 4 demonstrate that the sorption capacity of the mixture of CA and PVC is dependent on the presence of CA and also on the form of the polymer. The results shown in Fig. 5 indicate that the cellulose acetate activation process does not affect the sorption capacity of the polymer for the aqueous solutions.
CONCLUSIONS TABLE 5 Variations of the response characteristics of the urea biosensor (in urea determination) with time Day
1 10 20 30
Potential response (mV/decade)
LLLR (M)
Response time
50 50 47 45
2.0 x 10-5 6-3 X 1O-5 6.3 X 1O-5
30 37
TABLE 6
f95%
(6)
60 -
1*0x 10-4
A new type of disposable all-solid-state potentiometric bioselective probe for urea has been developed. The construction process is based on the in situ deposition of an asymmetric ammonium-selective membrane (a poly(viny1 chloride) matrix coated with cellulose acetate layers) on a conductive epoxy-graphite composite. Cellulose acetate acts as a support for the in situ covalent binding of urease.
Contact angles and critical wetting tensions of the polymers (films)
Polymer
Contact angle (0) with Distilled water
cellulose acetate (CA) PVC 25 CA: 75 PVC 50 CA: 50 PVC 75 CA: 25 PVC activated CA 25 activated CA: 75 PVC 50 activated CA: 50 PVC 75 activated CA: 25 PVC
75 >90 >90 >90 85 -
0.1 M phosphate 0.1 M Tris-HCl buffer, pH 74 buffer, pH 7.5
75 >90 >90 >90 86
80 >90 >90 >90 88
Ethanol
Critical wetting % in water tension 34 48 60 (dynkm) 56 78 44 64 57 40 37 38 43
48 56 35 43 45 25 24 27 29
35 44 23 35 34 9 14 14 -
25.1 24 1 25.0 24.9 25.0 28-2 27.1 27.0 27.5
651
iM H Gil et
al.
3
300
.S i Q
200
5 .i
6x1
8
tW
13 distilled war a 1 PVC
400 300
25:75
50:50
75:25
0
CA
1
2
3
4
400
0,lM phosphatebuffer pH 7.0 0,lM Tris-HClbuffer pH 7.5
PVC
25:75
50:50
75:25
CA
Fig. 4. Sorption capacities of CA-PVC mixtures (a) in powderform, (b) in film form.
The main innovation of the developed allsolid-state planar methodology based on the in situ deposition, modification and immobilization of different polymeric materials is that the proposed const~ction process simplifies potential mass production by making the use of printing techniques feasible.
ACKNOWLEDGEMENTS We gratefully acknowledge the grant from JNICT (Junta National de Investiga@o Cientifica e Tecnologica, Portugal) to Ana Paula da Fonseca Piedade. This work was partially supported by CICYT (Comision Interministerial de Ciencia y Tecnologia, Spain). Financial support for exchange of researchers received under a ‘Portuguese-Spanish Integrated Action’ programme is also gratefully acknowledged.
REFERENCES Alegret, S. & Martinez-Fabtegas, E. (1989). Biosensors based on conducting filled polymer all-solid-state 652
100
8
300
b .2: 1l g
200
g .$
100
0,lM phosphatebufferpH 7.0 O.lM Tris-HCIbuffer pH 7.5
0
0-4
1
2
3
4
Fig. 5. Sorption capacitiesof CA (a) in powderfonn, (b) in film form, aftereach stepof activationprocedure (seeFig 2). 1, b@ore activation:2, after step 1; 3, afrer step 2; 4, after step 3.
PVC matrix membrane electrodes. Biosensors, 4, 287-97. Cha, G. S. & Meyerhoff, M. E. (1989). Po~ntiome~c ion- and bio-selective electrodes based on asym metric cellulose acetate membranes. Talanta, 36, 271-8. Habeeb, A. F. & Hiramoto, R (1%8). Reaction of proteins with glutaraldehyde. Arch. Biochem. Biophys., 126, 16-22. Lindner, E., Gmf, E., Niegreisz, Z., T&h, K. & Pungor, E. (1988). Responses of site-controlled, plasticized membrane electrodes. Anal. Chem., 60, 295301. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R J. (1951). Protein measurement with the folin phenol reagent. J. Bioi. Chem., 193,26572. Ma, S. C., Chaniotakis, N. A. & Meyerhoff, M. E, (1988), Response properties of ion-selective polymeric membrane electrodes prepared with aminated and carboxylated poly(viny1 chloride). Anal. Chem., 40, 2293-9. Siggia, S. & Maxcy, W. (1947). Improved procedure for the determination of aldehydes. Ind. & Eng. Chem., Anal. Ed., 19, 1023-S. Vogel, A, I. (1978). Textbook of QuantitativeInorganic Analysis. Longman, New York, 4th edn., p. 730.
Covalent binding of urease *on ammonium-selective potent~ometric membraiks* M. Ii. Gil+ & A. P. Piedade Department
of Chemistry,
University
of Coimbra,
P-3049 Coimbra,
S. Alegret, J. Aionso, E. Ma~inez-F~b~gas Department
of Chemistry,
Autonomous
University
of Barcelona,
Portugal
& A. Oretlana
E-08193 Bellaterra, Catalonia,
Spain
(Received 26 May 1992: revised version received and accepted 27 July 1992)
Abstract: As part of the development of disposable urea bioselective probes, the covalent binding of urease on ammonium-selective potentiometric membranes has been assessed. Nonactin/bis( 1-butylpentyl)adipate/poly(vinylchloride) (PVC) membranes, directly applied to an internal solid contact (conductive epoxy-graphite composite), has been used as a support for covalent immobilization of urease. Two types of all-solid-state construction process have been assayed: thin layers of cellulose acetate (CA) were coated on the PVC ammonium-selective membranes (type 1) and blends of PVC and CAat various ratios were used as ammonium-selective membrane matrices (type 2). Urease was covalently attached to CA via aldehyde groups. These groups were created on the polysaccharide with sodium periodate to which the enzyme was immobilized through a spacer (hexamethylenediamine). The viability of both types of probe for the dete~ination of ammonium ions was assessed after each step of the activation process. Results indicated that type 2 potentiometric probes are altered after the treatment with sodium periodate. Good results were obtained with type 1 probes. Their dynamic concentration range of response to urea was from 2 X 10m5to 0.01 M with a sensibility of 50 mV/decade. Keywords: disposable immobilization.
urea
probe,
urea potentiometric
biosensor.
urease
Potent~omet~c biosensors usually comprise a membrane featuring biological materials which have been immobilized by physical or chemical
means. Such a membrane is cast directly on a potentiometric sensor based on an ion-selective membrane (H+, NH$’ etc.). An external dialysis membrane is usually employed in such an assembly.
*Paper presented at Biosensors ‘92, Geneva, Switzerland, 20-22 May 1992. +To whom correspondence should be addressed.
A different approach, rarely used in the construction of potentiometric biosensors, is to bind covalently the biological material (i.e.
INTRODUCTION
#56-5663/92/$05.00
@ 1992 Elsevier Science Publishers Ltd.
645
M. HOGil et al.
enzymes) to the transducer surface. This approach has been used in other types of electrochemical transducer (amperometric transducers, ISFETs etc.). This method brings about two innovations when applied to potentiometric biosensors: (a) it simplifies potential mass production by making the use of printing techniques feasible; and(b) the response of the resulting biosensor is made faster by speeding up the transfer of chemical species between the sample solution and the ion-selective membrane. To attain a covalent bond between the enzyme and the potentiometric membrane, the membrane matrix has to be modified by adding functional groups. However, the addition of these functional groups should not affect the response characteristics of the membrane (sensitivity, selectivity). This modification may also be useful for: (a) binding the ionophore covalently to the membrane matrix, thus preventing the ionophore from leaching when it is highly soluble in aqueous solutions; (b) attaching components that render the membrane biocompatible; (c) helping to improve adhesion between the membrane and the conductive substrate in all-solid-state electrodes (where the internal reference solution is absent): and (d) reducing the electrical resistance of the membrane. Examples of modified potentiometric membranes can be found in the literature. chloride) (PVC) Carboxyated poly(viny1 (Lindner et al., 1988; Ma et al., 1988) aminated PVC (Ma et al., 1988) and modified cellulose acetate (CA) (Cha & Meyerhoff, 1989) are examples of these matrices. CA has been used as a potentiometric membrane matrix where the enzyme is directly bound to it. The present work is focused on the development of urea-selective disposable potentiometric bioprobes. Covalent binding of urease on conventional ammonium-selective membranes has been PVC potentiometric assessed. Two strategies have been followed: (a) thin layers of CA have been coated on the ammonium-selective PVC membrane (asymmetric membrane); and (b) blends of PVC and CA have been used as an ammonium-selective membrane matrix. Urease was covalently attached to CA. The resulting bioselective membranes were cast in situ on conductive plastic substrates based on epoxy-graphite composites, thus producing simple potentiometric all-solidstate bioprobes. 646
Biosensors& Bioelectronics
EXPERIMENTAL Reagents Urease (from jack beans) was obtained from Sigma Chemical Co, Poole, Dorset, UK. Bis( lbutylpentyl) adipate and nonactin were supplied by Fluka Chemie AG, Switzerland. Poly(viny1 chloride) (PVC) and tetrahydrofuran (THF) were obtained from BDH, Poole, Dorset, UK Cellulose acetate (CA) was purchased from Kodak. Urea, ammonium chloride and glutaraldehyde (25% in water) were supplied by Merck, Germany. All other chemicals were of analytical grade and were obtained commercially.
Procedure The working and reference electrodes were connected to a digital potentiometer (Crison micropH 2000) and the potential variations were registered in a recorder (type BDl 11 from Kipp & Zonen, The Netherlands). The reference electrode used was a double junction Ag/AgCl electrode (90-02-00 from Orion, USA) with the outer compartment containing 0.1 M Tris-HCl buffer, pH 7.5. The reaction cell was always kept at 25 f O*l”C.
All-solid-state ammonium-selective PVC membrane probes The construction procedure involved the application of a layer of conductive epoxy resin, about 0.7 cm thick, to one of the ends of an PVC tube (electrode body) as described in an earlier paper (Alegret & Martinez-Fabregas, 1989). A shielded lead was fixed inside the tube to the conductive epoxy while still soft. After drying at 40°C for 24 h, the resin was drilled to form a 0.3 mm deep flat cavity. The solution of the sensing ammonium membrane material consisted of 0.1 g of PVC, 0.204 g of bis(lbutylpentyl) adipate and 0.0015 g of nonactin dissolved in 5 ml of tetrahydrofuran. 120 ~1 of this solution was added dropwise to the cavity and allowed to evaporate at room temperature. A convex membrane was formed, up to a central thickness of0.3-0.4 mm. After the assembly of the electrode had been completed, it was conditioned in a 0.1 M NHACl solution for 1 h.
Biosensors & Bioelectronics
All-solid-state ammonium-selective CA-PVC membrane probes The cellulose acetate-poly(viny1 chloride) membrane electrodes were prepared using two methods: Type 1. A cellulose acetate solution was prepared by dissolving the polysaccharide in THF (53 mg/2 ml) at room temperature. Aliquots of this solution (4 X 10 ,u1 each) were dropped on top of the sensing PVC layer and the solvent was evaporated at room temperature (Fig. 1, left). After 12 h the electrode was conditioned in a 0.1 M NH&l solution for 1 h. Type2, Mixtures of CA:PVC (25:75,50:50,75:25 and 100:0) were dissolved in THF (53 mg/2 ml). The PVC content in the sensing cocktail of the ammonium-selective PVC membrane electrodes was replaced by these mixtures and the probes were prepared in the usual manner (Fig. 1, right).
Covalent binding of w-ease
0.5 ml of 0.1 M phosphate buffer, pH 7.0, containing 6 mg/ml of urease. During previous work with CA in powder form (see Tables 4 and 5), the urease was linked to the aldehyde groups by introducing the polymer in a O-1 M phosphate buffer, pH 7.0 (5 ml), containing the enzyme (4 mg/ml). The amount of immobilized enzyme was determined by the method of Lowry et al. (1951). Electrochemical measurements For electrochemical measurements, the urea bioprobe and the reference electrode were introduced in 25 ml of 0.1 M Tris-HCl buffer, pH 7.5. or in 25 ml of the buffer under study. Aliquots of standard solution of urea and ammonium chloride were added. Determination of the aqueous solutions’ sorption capacity
Immobilization of urease The urease was immobilized on the cellulose acetate which contained hexamethylenediamine as a spacer. For the attachment of the amine, aldehyde groups were created on the CA by activation with a solution of 0.5 M sodium periodate (10 ml) at room temperature. The polymer containing the aldehyde groups was treated with a solution of hexamethylenediamine (10 ml of 1% (w/v) solution) for 18 h at room temperature, while stirring. The -NH2 groups were then activated with glutaraldehyde (Habeeb & Hiramoto, 1968). The electrodes with the activated cellulose membranes were immersed in
Samples (20 mg) of CA, PVC or the CA:PVC mixtures (25:75, 50:50 and 75:25) in powder or film were immersed in distilled water, in 0.1 M phosphate buffer, pH 7.0, or in 0.1 M Tris-HCl buffer, pH 7.5 (10 ml), at 25°C for 24 h. The excess water was removed with filter paper. Afterwards, the polymer was, in each instance, weighed every minute for 10 min. The initial sorption capacity was obtained graphically after extrapolation to zero time. Each sample was dried to constant mass under reduced pressure at 100°C. The percentage of solution uptake was given by % sorption
=
W(w) - Wr(c) x 1oo W,(c)
where Wi(w) is the initial wet mass at zero time and W,(c) is the final dry mass corrected with a blank. Determination of contact angles
Fig. 1. Type 1 (iej) and Type2 (right) CA-PVC membrane electrodes. I: electrode body: 2: shielded cable: 3: internal solid contact: 4: ion-selective PVC membrane: 5: CA membrane; 6: ion-selective PVC-CA membrane.
To determine the contact angles of the CA, PVC and CA:PVC mixtures, films were prepared. For this purpose, solutions of the polymeric supports in THF (25 mg/ml) were evaporated at room temperature on microscopy glass plates. Contact angles were determined with a Contact-&Meter (Livereel, Consett, Durham, UK). A drop of water, of 0.1 M phosphate buffer (pH 7-O), of 0 1 M Tris-HCl buffer (pH 7.5) or of 647
M H Gil et al.
First, the effectiveness of these electrodes when measuring ammonium ion was assessed. As shown in Table 1, both types of electrode are suitable for the determination of NH$. The urease was covalently linked to the CA after being treated with sodium periodate. This reaction produced aldehyde groups, to which units extension of hexamethylenediamine (HMD) were covalently bound through the development of Schiffs base complexes. The urease.was then immobilized on the -NH* groups of the extension units after these had been activated with glutaraldehyde. Fig. 2 summarizes the various steps of the activation of the cellulose acetate before the coupling of the urease. The viability of both types of electrode for the determination of NH2 was assessed after each step of activation. The results, indicated in Table 2, show that the Type 1 probes are suitable for the determination of NH4+ even after all the activation procedure has been completed. However, the response of the Type 2 probes is altered after the treatment with sodium periodate. In order to explain the lack of activity shown by this type of membrane, we assessed the effect of the periodate on nonactin. This compound was treated with the sodium periodate before being mixed with PVC and then the sensing membrane was prepared. We observed that, in this case, there was no response of the sensor to NH:. The results obtained with various PVC:CA mixtures show that for percentages of CA higher than 25% the membrane did not respond to NH$. In the light of these results, the work was carried out using the Type 1 electrodes. In order to get a good yield of activity of the immobilized urease, the reaction times for the different steps of activation shown in Fig. 2 were
an ethanol:water mixture (34:66; 48:52; 6040) was placed on the surface of the membrane and the contact angle was determined at 20°C. Determination of aldehyde aad amino groups The amount of aldehyde groups was determined by a potentiometric titration (Siggia & Maxcy, 1947). The amount of -NH2 free groups in the polymers was determined by an acid-base titration. Samples of the polymer (50 mg) were immersed in O-1 M HCl(l0 ml) for 12 h at room temperature. The HCl solutions were then backtitrated with 0.1 M NaOH solution, using 10 ml of O-1 M HCl as a blank. Chemical determination of urease activity The urease activity was determined using Nessler’s reagent (Vogel, 1978) (1 activity unit = 1 PM of NH: produced min-I). All measurements were done using a Jasco 7800 Spectrophotometer.
RESULTS AND DISCUSSION Immobilization of urease The purpose of this work was to immobilize urease onto ammonium-selective PVC membranes. In order to obtain reactive groups on these membranes, cellulose acetate was used as one of their membrane components. The preparation of two types of potentiometric probe, Type 1 and Type 2, has already been described (see Fig. 1).
TABLE 1 Response
Electrode
Standard Type 1 Type 2 Type 2 Type 2 Type 2
PVC
characteristics of CA-PVC mination of NH$
C&PVC
Potential response (mV/de~de)
-
63 62 61 61 61 57
25:75 5050 75:25 loo:0
“Lower limit of linear response
electrodes
LLLRa (MI 2 2 2 2 2 2
x x x x x x
IO-” 10-5 lo-’ lo-” 10-S 10-5
in the deter-
Response time Is) 5 5 5 5 5 5
Biosensors & ~i~l~~nics
Covalent binding of urease
step 1
+ NH2~CH2~~~2 (KMDJ
H2N&HC’)N
step 2
+ CHO(CH,),CHO
step 3
W-raldehyW H,N,@WN
N(CH,),NH,
WCH,l,NH,
$&HC)N
N(CH&# YH tCH& I
T (C&)3 I
COH
COH
Fig. 2. Schematic representation of the various steps of activation of cellulose acetate before enzyme coupling.
optimized using the cellulose acetate in powder form. The ensuing coupling reactions were realized by using the optimized times indicated in Table 3. Urease was then covalently linked to the electrode and the activity was determined using chemical and eIect~hemica1 methods. The results, indicated in Table 4, show that: (i)
the yield of activity of the immobilized enzyme is higher when cellulose acetate is used in powder form;
(ii)
the amount of reactive groups formed during the activation procedure is not directly related to the activity of the immobilized enzyme; (iii) the results obtained with these systems are promising; (iv) the Kh value of the immobilized enzyme is very high compared to the KM of the free enzyme; (v) the optimum pH value does not change when the urease is immobilized.
TABLE 2 Response of the CA-PVC electrodes to NH$ before and after each step of activation
Electrode (C&PVC)
Potential response (mV/decade) before activation
Type 1 Type Trpe Type Type
2 2 2 2
(2975) (50:50) (75:25) (1oO:O)
after step 1 after step 2
after step 3
62
63
63
63
61 61 61 57
43 0 0 0
43 0 0 0
43 0 0 0
M H. Gil et al.
Biosensorsh Bioelectronics
TABLE 3
Optimized
Time of activation with IO; (h)
times of reaction for the various steps of urease immobilization onto cellulose acetate powder
Time of reaction with Time of activation HMD (h) -NH2 (h) 18
0.75
enzyme (h)
3
18
[IO;] = 0.5 M; [HMD] = 1% (w/v); ]glu~mldehyde]
= 5% (v/v); [enzyme] = 4 mgfml.
TABLE 4
of urease onto cellulose acetate
Results of the immobilization
Determination
procedure
Cellulose acetate powder
Cellulose acetate Free enzyme electrode (Type 1)
mmol aldehyde groups/g (after step 1)
5.7
5.7
mmol -NH2 groups/g (after step 2)
0.13
0.19
mmol aldehyde groups/g (after step 3)
3.8
3-5
mg ureaselg
8.5
500.0
mg ureaselelectrode Immobilization Yield of activi@
0.04
yield” (%) (%)
Units/g
5.1
0.81
2.8
O-07
26 1
Units/electrode
V 1 chemical det. (U/mgE) max electrochem det. (U/electrode) pH 1 chemical de&. electrochemical a
h
1596 0.011
chemical det. (mM) Km ! electrochemical det. (mM)
312.5
8.0 16.7
IO.0
26.0 100.0
7.5
7.5
7.0
det.
mg of bound enzyme x 100 mg total enzyme in solution mg of active bound enzyme x 100 mg of total active enzyme in solution
Fig. 3 shows the calibration curves obtained with the urease biosensor on day 1. These results can be seen as promising, as well as the ones in Table 5, which show the calibration curve characteristics obtained with the same biosensor after several periods of time had elapsed. When not in use, the electrode was kept in distilled water at 4°C.
650
of Time of reaction with
Characterization of the cellulose acetate (in film and powder forms) In order to relate the immobilization results to the chemical and physical properties of the supports, CA, PVC and CA:PVC mixtures, in either film or powder form. were characterized before and after their activation by measuring the contact angles
Biosensors & Bioelectronics
-60
.
0
Fig. 3. Calibration
Covalent binding of urease
. .o 0. O 0 O0
0
curves obtained biosensor.
urea
ammonium
l
with the
urease
and the aqueous solutions’ sorption capacity. The results of the contact angle determinations are shown in Table 6. It was also possible, from the results of the contact angles of the films with ethanol/water mixtures, to evaluate the surface
energy of the films before and after the treatment with periodate, HMD and glutaraldehyde. These results are obtained by plotting the surface tension of the ethanol/water systems used against the cosine of the contact angles obtained. For each film, the intersection of the plot with the cos 8 = 1 corresponds to the critical wetting tension of the polymeric films. The results obtained show that (a) the variation of the liquid affinity for the studied surfaces is not significant; and (b) the critical wetting tension increases slightly after the activation treatment of the various studied supports. The results shown in Fig. 4 demonstrate that the sorption capacity of the mixture of CA and PVC is dependent on the presence of CA and also on the form of the polymer. The results shown in Fig. 5 indicate that the cellulose acetate activation process does not affect the sorption capacity of the polymer for the aqueous solutions.
CONCLUSIONS TABLE 5 Variations of the response characteristics of the urea biosensor (in urea determination) with time Day
1 10 20 30
Potential response (mV/decade)
LLLR (M)
Response time
50 50 47 45
2.0 x 10-5 6-3 X 1O-5 6.3 X 1O-5
30 37
TABLE 6
f95%
(6)
60 -
1*0x 10-4
A new type of disposable all-solid-state potentiometric bioselective probe for urea has been developed. The construction process is based on the in situ deposition of an asymmetric ammonium-selective membrane (a poly(viny1 chloride) matrix coated with cellulose acetate layers) on a conductive epoxy-graphite composite. Cellulose acetate acts as a support for the in situ covalent binding of urease.
Contact angles and critical wetting tensions of the polymers (films)
Polymer
Contact angle (0) with Distilled water
cellulose acetate (CA) PVC 25 CA: 75 PVC 50 CA: 50 PVC 75 CA: 25 PVC activated CA 25 activated CA: 75 PVC 50 activated CA: 50 PVC 75 activated CA: 25 PVC
75 >90 >90 >90 85 -
0.1 M phosphate 0.1 M Tris-HCl buffer, pH 74 buffer, pH 7.5
75 >90 >90 >90 86
80 >90 >90 >90 88
Ethanol
Critical wetting % in water tension 34 48 60 (dynkm) 56 78 44 64 57 40 37 38 43
48 56 35 43 45 25 24 27 29
35 44 23 35 34 9 14 14 -
25.1 24 1 25.0 24.9 25.0 28-2 27.1 27.0 27.5
651
iM H Gil et
al.
3
300
.S i Q
200
5 .i
6x1
8
tW
13 distilled war a 1 PVC
400 300
25:75
50:50
75:25
0
CA
1
2
3
4
400
0,lM phosphatebuffer pH 7.0 0,lM Tris-HClbuffer pH 7.5
PVC
25:75
50:50
75:25
CA
Fig. 4. Sorption capacities of CA-PVC mixtures (a) in powderform, (b) in film form.
The main innovation of the developed allsolid-state planar methodology based on the in situ deposition, modification and immobilization of different polymeric materials is that the proposed const~ction process simplifies potential mass production by making the use of printing techniques feasible.
ACKNOWLEDGEMENTS We gratefully acknowledge the grant from JNICT (Junta National de Investiga@o Cientifica e Tecnologica, Portugal) to Ana Paula da Fonseca Piedade. This work was partially supported by CICYT (Comision Interministerial de Ciencia y Tecnologia, Spain). Financial support for exchange of researchers received under a ‘Portuguese-Spanish Integrated Action’ programme is also gratefully acknowledged.
REFERENCES Alegret, S. & Martinez-Fabtegas, E. (1989). Biosensors based on conducting filled polymer all-solid-state 652
100
8
300
b .2: 1l g
200
g .$
100
0,lM phosphatebufferpH 7.0 O.lM Tris-HCIbuffer pH 7.5
0
0-4
1
2
3
4
Fig. 5. Sorption capacitiesof CA (a) in powderfonn, (b) in film form, aftereach stepof activationprocedure (seeFig 2). 1, b@ore activation:2, after step 1; 3, afrer step 2; 4, after step 3.
PVC matrix membrane electrodes. Biosensors, 4, 287-97. Cha, G. S. & Meyerhoff, M. E. (1989). Po~ntiome~c ion- and bio-selective electrodes based on asym metric cellulose acetate membranes. Talanta, 36, 271-8. Habeeb, A. F. & Hiramoto, R (1%8). Reaction of proteins with glutaraldehyde. Arch. Biochem. Biophys., 126, 16-22. Lindner, E., Gmf, E., Niegreisz, Z., T&h, K. & Pungor, E. (1988). Responses of site-controlled, plasticized membrane electrodes. Anal. Chem., 60, 295301. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R J. (1951). Protein measurement with the folin phenol reagent. J. Bioi. Chem., 193,26572. Ma, S. C., Chaniotakis, N. A. & Meyerhoff, M. E, (1988), Response properties of ion-selective polymeric membrane electrodes prepared with aminated and carboxylated poly(viny1 chloride). Anal. Chem., 40, 2293-9. Siggia, S. & Maxcy, W. (1947). Improved procedure for the determination of aldehydes. Ind. & Eng. Chem., Anal. Ed., 19, 1023-S. Vogel, A, I. (1978). Textbook of QuantitativeInorganic Analysis. Longman, New York, 4th edn., p. 730.
