- Email: [email protected]
Determination of free and total cholesterol in human bile samples using an enzyme electrode
7
Clinica Chrmrca Acta, 132 ( 1983) 7- 15 Elsevier
CCA 2566
Determination of free and total cholesterol in human bile samples using an enzyme electrode Marco
Mascini
Institute
*, Mauro
Tomassetti
ofAnalytical Chemistry,
(Received
January
University
and Mario
Iannello
of Rome, Rome (Italy)
10th; revision March
18th, 1983)
A reliable cholesterol electrode was assembled with a Clark type oxygen electrode and cholesterol oxidase immobilised on a nylon net fixed on the electrode surface. Determination of free (unesterified) and total (esterified and unesterified) cholesterol in human bile samples was performed. Results were compared with the enzymaticspectrophotometric procedure based on the Roeschlau method. The electrode has a usable life of more than two weeks, and more than 300 determinations could be performed. The CV values for free and total cholesterol in bile samples is 56%. The comparison between the proposed procedure and the spectrophotometric procedure is fair for the half of the results (A < 5%) and acceptable in the other half (A < 25%).
Introduction Enzyme electrodes have recently been proposed for many determinations of clinical interest. One of the most valuable applications, because of its clinical importance and analytical difficulty, is cholesterol assay. Recently some papers dealing with cholesterol determination have been published [1,2] and reactors with cholesterol oxidase immobilised for use in continuous flow analysis have been proposed [3-61. All reports published so far on the electrochemical determination of cholesterol in biological samples deal with applications in serum or blood samples. Reports on cholesterol determination in bile samples are lacking. However, quantitative determination of cholesterol in bile is useful because of its close relation to the formation of gallstones [7,8]. Several studies have defined the saturation conditions of human bile and a numerical index, often referred as lithogenic index has been proposed; this index is a function of the total concentration of cholesterol in bile, but not phospholipids and bile salts. Moreover, bile samples are considered very
* Correspondence should be addressed to Prof. Marco Mascini, di Chimica Analitica, CittP Universitaria, 00185 Roma, Italy.
0009-8981/83/$03.00
0 1983 Elsevier Science Publishers
B.V.
UniversitA
degli Studi di Roma,
Istituto
x complex due to the presence of bile salts and of some coloured compounds: some calorimetric procedures give unsatisfactory results in these samples [9,10]. The recent assessment of a simple procedure [l l] for immobilising on nylon net oxidase enzymes with an high specific immobilised activity permitted us to produce a cholesterol enzyme electrode. The object of this report is the application of such a cholesterol electrode for the quantitative determination of free and total cholesterol in bile samples. The results are compared with a spectrophotometric method considered the most suitable for bile samples [ 10.121. Materials and methods Chemicals Cholesterol oxidase (EC 1.1.3.6. from Nocardiu erythropolis, 25 U/ml) and cholesterol esterase (EC 3.1.1 .I3 from microorganisms, 50 U/ml) were obtained from Boehringer, Mannheim, FRG. These enzymes and all other reagents were used without further purification. Cholesterol, Triton X- 100, dimethylsulphate, phosphate buffers and other chemicals were of the highest purity available from Farmitalia-Carlo Erba, Milano, Italy. Glutaraldehyde was a 25% solution obtained from Merck, Darmstadt, FRG. Nylon net was from A. Bozzone, Appian0 Gentile, Italy with 108 threads/cm2 with a thickness of 120 pm. The enzymatic colour test for cholesterol was obtained from Boehringer, Mannheim (No. 124079). Cholesterol standards, 7.8 mmol/l (= 3.0 g/l) or 12.9 mmol/l (= 5.0 g/l): dissolve 300 mg or 500 mg of cholesterol in 100 ml of 10 g/l solution of Triton X-100 in deionised water. Subjects The bile samples came from ten patients from 25 to 55 years old, four men and six women, most having cholesterol stones. Nine samples were duodenal or hepatic aspirates, and one was obtained during surgery. Apparatus The oxygen electrode was obtained from the Instrumentation Laboratory and the unit for reading the current was the IL Model 213. The internal solution of the oxygen electrode is KCl, 0.1 mol/l, and against the platinum cathode a teflon membrane 6 or 12 pm thick was used; the nylon net with the immobilised enzyme was fixed over the teflon membrane; the two membranes were secured by the same 0 ring. The spectrophotometric procedure was performed with a Beckman DK2-A model with a l-cm cell. Immobilisation procedure The chemical procedure for immobilising cholesterol oxidase on a nylon net derives from the method of Hornby and Morris [ 131. From the nylon net sheet small discs of 16 mm diameter were obtained and fixed around a glass rod with nylon thread for easy and rapid handling. The rod was immersed in a stoppered tube containing dimethyl sulphate and this was immersed in a boiling water bath for exactly 5 min; the tube was then transferred to an ice bath to stop the reaction; after
9
that, the glass rod was immersed in a second test tube containing anhydrous methanol for washing; then single discs were released from the glass rod and immersed as quickly as possible in 50 ml of lysine, 0.5 mol/l, at pH 9.0 for 2 h. The lysine acts as a spacer between the nylon structure and the enzyme molecule. Then the discs were washed thoroughly in NaCl 0.1 mol/l. Thereafter the discs were immersed in glutaraldehyde 12.5% in saturated borate buffer (0.1 mol/l) at pH 8.5 for about 45 min, and after thorough washing they were dipped in the enzyme solution in phosphate buffer 0.1 mol/l at pH 7.0 for 2 h at room temperature. Then the discs were washed and fixed directly on the oxygen electrode or stored in a phosphate buffer at 4°C. Measurement of immobilised enzyme activity To measure the activity of oxidase enzymes immobilised on membrane was immersed in 20 ml of 0.1 g/l of cholesterol in buffer, pH 7.0, and the decrease of oxygen was followed by under constant magnetic stirring. A stable constant decrease which the activity was calculated.
a nylon net, a single 0.1 mol/l phosphate an oxygen electrode, was recorded, from
Procedure with cholesterol electrode Several samples in the range (0.2 to 1.0) ml of a standard solution of cholesterol (7.8 or 12.9) mmol/l were added to 10 ml of phosphate buffer 0.1 mol/l, pH 7.0, in which the cholesterol electrode was immersed and the reading on the IL Model 213 was positioned to 200 (relative current). The buffer phosphate should contain 0.1% of Triton X-100 to ensure solubility of cholesterol. After each addition, the current decreases slowly. Generally after 60 s it stopped. The current value was read after 120 s to obtain the maximum repeatibility. The cholesterol electrode between measurements was stored in phosphate buffer at 4°C. Procedure for free cholesterol determination in bile samples By adding 200 ~1 of bile sample to 10 ml of phosphate buffer (0.1 mol/l; pH 7.0) with 0.1% of Triton X- 100 in which the cholesterol electrode is dipped at (25 + O.l)‘C, the new value of current is recorded. In the same solution 200 ~1 of standard cholesterol solution are added and the new value of current is recorded. The unknown concentration of cholesterol in a bile sample is obtained from the following:
where C, is the concentration of the sample, C,, the concentration of the standard solution, Ai, and Ai,, the current variation relative to the sample and the standard addition, and V,, V,, the volume of the sample and the standard added, and V, the initial volume of the buffer. Procedure for total choleirterol determination in bile samples After adding 200 ~1 of bile sample to 200 ~1 of buffer
solution
(phosphate
0.1
IO
mol/l; pH 7.0) and 50 ~1 of soluble cholesterol esterase (1 U/50 ~1). we incubated the solution for 1 h at 37°C. Thereafter, the solution was processed in the same way as for free cholesterol determination. Procedwe for spectrophotometric determination of ,free and total cholesterol in bile sumples Each sample of bile was diluted 1 : 5 with saline solution (0.8% NaCl); 0.10 ml of this solution was analysed by Roeschlau et al’s [ 141 method adapted for bile samples [ 10,121; the reagents for this procedure are now commercially available (Boehringer, Mannheim, FRG). We used the procedure with or without addition of cholesterol esterase, depending on whether the amounts of total or free cholesterol were to be determined. Results The enzyme immobilised on nylon net had an immobilised specific activity of 600 nmol.min-‘.cmP2 (cf. the 2-3 mmol . mini ’ . cm-’ obtained by binding the same enzyme to collagen [ 151). In Fig. 1 the aging effect over a 20-days period is shown. In Table I the equations obtained by regression analysis of Fig. 1 values are reported with the relative CV values. The coefficient of correlation is around 0.999 for the first 15 days; the sensitivity is higher during the first 10 days and adequate for about 15 days after assembling the electrode. Therefore the usable life of the electrode is almost 2 weeks. In this period we were able to perform around 300 determinations in order to evaluate from an analytical point of view the procedure using the enzyme electrode. The aging effect seems unaffected by the measurements and probably derives from a slow denaturation of the immobilised enzyme. In Table II we give the
200
-
195
2 & k
20 d
190
15 d
2
\ 180
I OM
I
I
04c
060
0
10 d 6d 083
Cholest.( m moi/I)
Fig. 1. Lifetime of cholesterol electrode. Calibration curves in relative units obtained with cholesterol standard solutions (7.8 mmol/l) on different days after assembly of the electrode (d, days after assembly).
11
TABLE
I
Regression
lines of the calibration
Days after the assembly the electrode
curves of the cholesterol
Parameters
of the regression
of
line b
1 6 10 15 20
~ua.l.mmol-i)
(ua)
- 19.4 -25.6 - 24.8 - 13.8 -8.8
199.8 199.7 199.4 199.9 199.7
Analytical form y = nx + b; y = relative current mmol/l; f = correlation coefficient.
TABLE
electrode
in arbitrary
-
0.9985 0.9994 0.9979 0.998 9 0.9955
units (ua); x = cholesterol
concentration
in
II
Reproducibility of measurements after standard additions of cholesterol solution; all measurements were performed with the same electrode (Ai values are differences of relative current in arbitrary units (ua)) Cholesterol concentration in mmol/l
Ai (mean of six determinations)
SDS
C”M.
0.26 0.49 0.72 0.96 1.16
5.7 9.0 11.4 13.4 15.5
2.5 3.4 2.9 1.9 2.3
2.9
__----------------------Reproducibility are in mmoI/l) Sample number
of free cholesterol
Species of sample
1 gb. 2 g.b. 4 h.b. 8 gb. ___________--------------Reproducibility are in mmol/l)
of total cholesterol
determination
of human
bile samples
by the enzyme electrode
Value found (mean of 5 determinations)
SD’%
C”,,,
8.7 14.3 1.4 4.8
2.7 5.6 4.2 5.1 _--_________
6.6
determination
of human
bile samples by the enzyme electrode
Sample number
Species of sample
Value found (mean of 5 determinations)
SD%
C”,,,
1
g.b. g.b. g.b. h.b.
20.2 22.0 12.8 3.8
2.3 6.0 3.4 4.8
5.6
2 6 9
g.b., gallbladder bile; h.b., hepatic bile; C”,,,. residue variation
coefficient,
called by IUPAC:
‘pooled
standard
deviation’
% [ 181.
(values
(values
III
gb. gb. gb. h.b. h.b. gb. gb. gb. h.b. h.b.
1 2 3 4 5 6 1 8 9 10
g.b., gallbladder bile; h.b., hepatic bile.
Soecies of sample
Samole number
9.3 14.5 6.9 1.2 2.5 11.0 3.5 3.8 2.8 0.1
Free cholesterol found by Roeschlau method 8.7 14.3 6.9 1.4 2.4 10.7 4.2 4.7 3.4 -0
Free cholesterol found by enzyme electrode - 6.5 - 1.4 0.0 16.7 - 4.0 - 2.7 20.0 23.7 21.4 _
A%
and Roeschlau
20.7 23.0 7.2 1.4 2.9 13.0 3.7 3.8 3.0 0.1
Total cholesterol found by Roeschlau method
Comparison of results of free and total cholesterol concentration obtained by the enzyme electrode bile samples (values are in mmol/l; each value is the mean of at least three determinations)
TABLE
procedure
9.0 1.4 3.6 12.8 4.4 4.7 3.7 -0
20.2 22.0
Total cholesterol found by enzyme electrode
spectrophotometric
- 2.4 - 4.3 25.0 0.0 24. I - 1.5 18.9 23.7 23.3 _
3%
of 10 human
13
5.0
10.0
15.0
Roeschlau
20.0 m mol/l
Fig. 2. Correlation of concentration values of free (0) and total (0) cholesterol obtained by the enzyme electrode and the Roeschlau procedure for the 10 samples. Equations of linear regressions for free and total cholesterol were found to be respectively: y = 0.950x + 0.154 (r = 0.9954), and y = 0.942x +0.297 (r = 0.9965), where y is the concentration found using the Roeschlau procedure; r = correlation coeffi-
cient.
values of current decrease (in relative units) for five additions of cholesterol standard solution. The calibration curve was repeated six times on the same day to obtain the standard deviation and CV values. In the same table we evaluated the reproducibility of the electrode procedure in four bile samples, for both free and total cholesterol. The determination was repeated five times on the same sample. The CV was 5-675. In Table III the results of the proposed procedure are compared, for ten human bile samples, with those obtained by an enzymatic spectrophotometric procedure considered the most suitable for bile samples [10,12,14]. In Fig. 2 the correlation is reported between the two procedures. The values obtained for total cholesterol in almost all samples were barely higher than the values for free cholesterol, in accord with existing knowledge that the amount of cholesterol esterified in bile is small. According to the behaviour of the immobilised enzyme the cholesterol electrode is responsive only to the free cholesterol, i.e. the unesterified cholesterol and the total can be obtained only by preliminary hydrolysis of the esterified cholesterol. Immobilisation of cholesterolesterase could be attempted in the future, thus permitting all the enzyme reactions to take place on the electrode surface.
The method of enzyme immobilisation coupled with the oxygen electrode described seems to be valuable for cholesterol determination in biological samples, and especially in bile samples, which can be considered complex due to the number of interfering substances, the chromophore dissolved and the high density of some of them. The use of the oxygen electrode seems to be necessary for the elimination of
14
interfering substances dissolved in bile samples. However, the use of detergents should be limited because of the negative effect on the permeability of the gas membrane [l]. The immobilisation procedure on nylon net has several advantages: the structure on the nylon net permits easy diffusion from the environs of the electrode surface to the immobilised enzyme and vice versa without forming a barrier to the diffusion of the reagent or product of the enzymatic reaction. Furthermore, the procedure for immobilisation is very simple and a large amount of activity remains after the immobilisation process. The mechanical structure of nylon net is practically unchanged by the immobilisation procedure. This allows the membrane to be assembled or disassembled from the electrode surface several times without a decrease in enzyme performance. The reproducibility with cholesterol standards and bile samples never exceeded 6%, which can be considered a reasonable value taking into account that a single analysis (2 additions) is performed in 4 min. The comparison between the proposed procedure and the spectrophotometric procedure on ten samples reported in detail in Table III is satisfactory for half of the results (A < 5%) and acceptable for the other half (A < 25%). Further studies, e.g. a comparison with gas liquid chromatography [16] might clarify the cause of the discrepancies in the latter samples. The determination of cholesterol in bile samples is important for evaluation of the risk of gallstone formation and lithogenic index calculation [17]; the procedure using the cholesterol electrode realised with a nylon net immobilised enzyme appears to be a potentially valuable tool for the task. Acknowledgement This work was supported
by the National
Research
Council
(CNR)
of Italy.
References 1 Noma A, Nakayama K. Polarographic method for rapid microdetermination of cholesterol with cholesterol esterase and cholesterol oxidase. Clin Chem 1976; 22: 336-340. 2 Clark LC Jr, Duggam CA, Grooms TA, Hart LM, Moore ME. One minute electrochemical enzimic assay for cholesterol in biological materials. Clin Chem 1981; 27: 1978-1982. 3 Huang H, Kuan SS, Guilbault GG. Amperometric determination of total cholesterol in serum with use of immobilized cholesterol ester hydrolase and cholesterol oxidase. Clin Chem 1977; 23: 671-676. 4 Bertrand C, Coulet PR, Gautheron DC. Enzyme electrode with collagen-immobilized cholesterol oxidase for the microdetermination of free cholesterol. Anal Lett 1979; 12: 1477-1488. 5 Karube I. Hara K, Matsuoka H, Suzuki S. Amperometric determination of total cholesterol in serum with use of immobilized cholesterol esterase and cholesterol oxidase. Anal Chim Acta 1982: 139: 127-132. 6 Tabata M. Endo J, Murachi T. Automated analysis of total cholesterol in serum using coimmobilized cholesterol ester hydrolase and cholesterol oxidase. J Appl Biochem 1981; 3: 84-92. 7 Admirand WH, Small DM. Physicochemical basis of cholesterol gallstone formation in man. J Clin Invest 1968; 47: 1043-1052. 8 Carey MC, Small DM. The physical chemistry of cholesterol solubility in bile. J Clin Invest 1978; 6 I : 998- 1026. 9 Watson D. A simple method for the determination of serum cholesterol. Clin Chim Acta 1960; 5: 637-643. 10 Biader Ceipidor U. Curini R, D’Ascenzo G, Tomassetti M. Alessandrini A. Evaluation of two
15
enzymatic
methods
for determination
of total cholesterol
in bile. G Ital Chim Clin 1979; 4: 355-369. with improved mechanical and analytical characteristics obtained by binding enzymes to nylon nets. Anal Chim Acta 1983; 146: 135- 148. Roda A, Festi D, Sama C et al. Enzymatic determination of cholesterol in bile. Clin Chim Acta 1975; 64: 337-341. Hornby WE, Morris DL. Modified nylons in enzyme immobilization and their use in analysis. In: Weetall HH, ed. Immobilized enzymes, antigens, antibodies and peptides. New York: M. Dekker Inc., 1975: 141-169. Roeschlau P, Bernt E, Gruber W. Enzymatic determination of total cholesterol in serum. Z Klin Chem Klin Biochem 12: 1974; 403-407. Bertrand C, Coulet PR, Gautheron DC. Multipurpose electrode with different enzyme systems bound to collagen films. Anal Chim Acta 1981; 126: 23-34. Bolton CH, Low-Beer TS, Pomare EW, Wicks ACB, Yeates J, Heaton KW. A simplified procedure for the analysis of cholesterol, phospholipids and bile salts in human bile. Clin Chim Acta 1978; 83: 177-181. Metger AL, Heymsfield S, Grundy SM. The lithogenic index - a numerical expression for the relative lithogenicity of bile. Gastroenterology 1972; 62: 499-500. IUPAC, Compendium of analytical nomenclature. Definitive rules 1977, prepared by Irving HMNH, Freiser H, West TS. Oxford: Pergamon Press, 1978.
11 Mascini M, Iannello M, Palleschi G. Enzymes electrodes 12 13
14 15 16
17 18
Clinica Chrmrca Acta, 132 ( 1983) 7- 15 Elsevier
CCA 2566
Determination of free and total cholesterol in human bile samples using an enzyme electrode Marco
Mascini
Institute
*, Mauro
Tomassetti
ofAnalytical Chemistry,
(Received
January
University
and Mario
Iannello
of Rome, Rome (Italy)
10th; revision March
18th, 1983)
A reliable cholesterol electrode was assembled with a Clark type oxygen electrode and cholesterol oxidase immobilised on a nylon net fixed on the electrode surface. Determination of free (unesterified) and total (esterified and unesterified) cholesterol in human bile samples was performed. Results were compared with the enzymaticspectrophotometric procedure based on the Roeschlau method. The electrode has a usable life of more than two weeks, and more than 300 determinations could be performed. The CV values for free and total cholesterol in bile samples is 56%. The comparison between the proposed procedure and the spectrophotometric procedure is fair for the half of the results (A < 5%) and acceptable in the other half (A < 25%).
Introduction Enzyme electrodes have recently been proposed for many determinations of clinical interest. One of the most valuable applications, because of its clinical importance and analytical difficulty, is cholesterol assay. Recently some papers dealing with cholesterol determination have been published [1,2] and reactors with cholesterol oxidase immobilised for use in continuous flow analysis have been proposed [3-61. All reports published so far on the electrochemical determination of cholesterol in biological samples deal with applications in serum or blood samples. Reports on cholesterol determination in bile samples are lacking. However, quantitative determination of cholesterol in bile is useful because of its close relation to the formation of gallstones [7,8]. Several studies have defined the saturation conditions of human bile and a numerical index, often referred as lithogenic index has been proposed; this index is a function of the total concentration of cholesterol in bile, but not phospholipids and bile salts. Moreover, bile samples are considered very
* Correspondence should be addressed to Prof. Marco Mascini, di Chimica Analitica, CittP Universitaria, 00185 Roma, Italy.
0009-8981/83/$03.00
0 1983 Elsevier Science Publishers
B.V.
UniversitA
degli Studi di Roma,
Istituto
x complex due to the presence of bile salts and of some coloured compounds: some calorimetric procedures give unsatisfactory results in these samples [9,10]. The recent assessment of a simple procedure [l l] for immobilising on nylon net oxidase enzymes with an high specific immobilised activity permitted us to produce a cholesterol enzyme electrode. The object of this report is the application of such a cholesterol electrode for the quantitative determination of free and total cholesterol in bile samples. The results are compared with a spectrophotometric method considered the most suitable for bile samples [ 10.121. Materials and methods Chemicals Cholesterol oxidase (EC 1.1.3.6. from Nocardiu erythropolis, 25 U/ml) and cholesterol esterase (EC 3.1.1 .I3 from microorganisms, 50 U/ml) were obtained from Boehringer, Mannheim, FRG. These enzymes and all other reagents were used without further purification. Cholesterol, Triton X- 100, dimethylsulphate, phosphate buffers and other chemicals were of the highest purity available from Farmitalia-Carlo Erba, Milano, Italy. Glutaraldehyde was a 25% solution obtained from Merck, Darmstadt, FRG. Nylon net was from A. Bozzone, Appian0 Gentile, Italy with 108 threads/cm2 with a thickness of 120 pm. The enzymatic colour test for cholesterol was obtained from Boehringer, Mannheim (No. 124079). Cholesterol standards, 7.8 mmol/l (= 3.0 g/l) or 12.9 mmol/l (= 5.0 g/l): dissolve 300 mg or 500 mg of cholesterol in 100 ml of 10 g/l solution of Triton X-100 in deionised water. Subjects The bile samples came from ten patients from 25 to 55 years old, four men and six women, most having cholesterol stones. Nine samples were duodenal or hepatic aspirates, and one was obtained during surgery. Apparatus The oxygen electrode was obtained from the Instrumentation Laboratory and the unit for reading the current was the IL Model 213. The internal solution of the oxygen electrode is KCl, 0.1 mol/l, and against the platinum cathode a teflon membrane 6 or 12 pm thick was used; the nylon net with the immobilised enzyme was fixed over the teflon membrane; the two membranes were secured by the same 0 ring. The spectrophotometric procedure was performed with a Beckman DK2-A model with a l-cm cell. Immobilisation procedure The chemical procedure for immobilising cholesterol oxidase on a nylon net derives from the method of Hornby and Morris [ 131. From the nylon net sheet small discs of 16 mm diameter were obtained and fixed around a glass rod with nylon thread for easy and rapid handling. The rod was immersed in a stoppered tube containing dimethyl sulphate and this was immersed in a boiling water bath for exactly 5 min; the tube was then transferred to an ice bath to stop the reaction; after
9
that, the glass rod was immersed in a second test tube containing anhydrous methanol for washing; then single discs were released from the glass rod and immersed as quickly as possible in 50 ml of lysine, 0.5 mol/l, at pH 9.0 for 2 h. The lysine acts as a spacer between the nylon structure and the enzyme molecule. Then the discs were washed thoroughly in NaCl 0.1 mol/l. Thereafter the discs were immersed in glutaraldehyde 12.5% in saturated borate buffer (0.1 mol/l) at pH 8.5 for about 45 min, and after thorough washing they were dipped in the enzyme solution in phosphate buffer 0.1 mol/l at pH 7.0 for 2 h at room temperature. Then the discs were washed and fixed directly on the oxygen electrode or stored in a phosphate buffer at 4°C. Measurement of immobilised enzyme activity To measure the activity of oxidase enzymes immobilised on membrane was immersed in 20 ml of 0.1 g/l of cholesterol in buffer, pH 7.0, and the decrease of oxygen was followed by under constant magnetic stirring. A stable constant decrease which the activity was calculated.
a nylon net, a single 0.1 mol/l phosphate an oxygen electrode, was recorded, from
Procedure with cholesterol electrode Several samples in the range (0.2 to 1.0) ml of a standard solution of cholesterol (7.8 or 12.9) mmol/l were added to 10 ml of phosphate buffer 0.1 mol/l, pH 7.0, in which the cholesterol electrode was immersed and the reading on the IL Model 213 was positioned to 200 (relative current). The buffer phosphate should contain 0.1% of Triton X-100 to ensure solubility of cholesterol. After each addition, the current decreases slowly. Generally after 60 s it stopped. The current value was read after 120 s to obtain the maximum repeatibility. The cholesterol electrode between measurements was stored in phosphate buffer at 4°C. Procedure for free cholesterol determination in bile samples By adding 200 ~1 of bile sample to 10 ml of phosphate buffer (0.1 mol/l; pH 7.0) with 0.1% of Triton X- 100 in which the cholesterol electrode is dipped at (25 + O.l)‘C, the new value of current is recorded. In the same solution 200 ~1 of standard cholesterol solution are added and the new value of current is recorded. The unknown concentration of cholesterol in a bile sample is obtained from the following:
where C, is the concentration of the sample, C,, the concentration of the standard solution, Ai, and Ai,, the current variation relative to the sample and the standard addition, and V,, V,, the volume of the sample and the standard added, and V, the initial volume of the buffer. Procedure for total choleirterol determination in bile samples After adding 200 ~1 of bile sample to 200 ~1 of buffer
solution
(phosphate
0.1
IO
mol/l; pH 7.0) and 50 ~1 of soluble cholesterol esterase (1 U/50 ~1). we incubated the solution for 1 h at 37°C. Thereafter, the solution was processed in the same way as for free cholesterol determination. Procedwe for spectrophotometric determination of ,free and total cholesterol in bile sumples Each sample of bile was diluted 1 : 5 with saline solution (0.8% NaCl); 0.10 ml of this solution was analysed by Roeschlau et al’s [ 141 method adapted for bile samples [ 10,121; the reagents for this procedure are now commercially available (Boehringer, Mannheim, FRG). We used the procedure with or without addition of cholesterol esterase, depending on whether the amounts of total or free cholesterol were to be determined. Results The enzyme immobilised on nylon net had an immobilised specific activity of 600 nmol.min-‘.cmP2 (cf. the 2-3 mmol . mini ’ . cm-’ obtained by binding the same enzyme to collagen [ 151). In Fig. 1 the aging effect over a 20-days period is shown. In Table I the equations obtained by regression analysis of Fig. 1 values are reported with the relative CV values. The coefficient of correlation is around 0.999 for the first 15 days; the sensitivity is higher during the first 10 days and adequate for about 15 days after assembling the electrode. Therefore the usable life of the electrode is almost 2 weeks. In this period we were able to perform around 300 determinations in order to evaluate from an analytical point of view the procedure using the enzyme electrode. The aging effect seems unaffected by the measurements and probably derives from a slow denaturation of the immobilised enzyme. In Table II we give the
200
-
195
2 & k
20 d
190
15 d
2
\ 180
I OM
I
I
04c
060
0
10 d 6d 083
Cholest.( m moi/I)
Fig. 1. Lifetime of cholesterol electrode. Calibration curves in relative units obtained with cholesterol standard solutions (7.8 mmol/l) on different days after assembly of the electrode (d, days after assembly).
11
TABLE
I
Regression
lines of the calibration
Days after the assembly the electrode
curves of the cholesterol
Parameters
of the regression
of
line b
1 6 10 15 20
~ua.l.mmol-i)
(ua)
- 19.4 -25.6 - 24.8 - 13.8 -8.8
199.8 199.7 199.4 199.9 199.7
Analytical form y = nx + b; y = relative current mmol/l; f = correlation coefficient.
TABLE
electrode
in arbitrary
-
0.9985 0.9994 0.9979 0.998 9 0.9955
units (ua); x = cholesterol
concentration
in
II
Reproducibility of measurements after standard additions of cholesterol solution; all measurements were performed with the same electrode (Ai values are differences of relative current in arbitrary units (ua)) Cholesterol concentration in mmol/l
Ai (mean of six determinations)
SDS
C”M.
0.26 0.49 0.72 0.96 1.16
5.7 9.0 11.4 13.4 15.5
2.5 3.4 2.9 1.9 2.3
2.9
__----------------------Reproducibility are in mmoI/l) Sample number
of free cholesterol
Species of sample
1 gb. 2 g.b. 4 h.b. 8 gb. ___________--------------Reproducibility are in mmol/l)
of total cholesterol
determination
of human
bile samples
by the enzyme electrode
Value found (mean of 5 determinations)
SD’%
C”,,,
8.7 14.3 1.4 4.8
2.7 5.6 4.2 5.1 _--_________
6.6
determination
of human
bile samples by the enzyme electrode
Sample number
Species of sample
Value found (mean of 5 determinations)
SD%
C”,,,
1
g.b. g.b. g.b. h.b.
20.2 22.0 12.8 3.8
2.3 6.0 3.4 4.8
5.6
2 6 9
g.b., gallbladder bile; h.b., hepatic bile; C”,,,. residue variation
coefficient,
called by IUPAC:
‘pooled
standard
deviation’
% [ 181.
(values
(values
III
gb. gb. gb. h.b. h.b. gb. gb. gb. h.b. h.b.
1 2 3 4 5 6 1 8 9 10
g.b., gallbladder bile; h.b., hepatic bile.
Soecies of sample
Samole number
9.3 14.5 6.9 1.2 2.5 11.0 3.5 3.8 2.8 0.1
Free cholesterol found by Roeschlau method 8.7 14.3 6.9 1.4 2.4 10.7 4.2 4.7 3.4 -0
Free cholesterol found by enzyme electrode - 6.5 - 1.4 0.0 16.7 - 4.0 - 2.7 20.0 23.7 21.4 _
A%
and Roeschlau
20.7 23.0 7.2 1.4 2.9 13.0 3.7 3.8 3.0 0.1
Total cholesterol found by Roeschlau method
Comparison of results of free and total cholesterol concentration obtained by the enzyme electrode bile samples (values are in mmol/l; each value is the mean of at least three determinations)
TABLE
procedure
9.0 1.4 3.6 12.8 4.4 4.7 3.7 -0
20.2 22.0
Total cholesterol found by enzyme electrode
spectrophotometric
- 2.4 - 4.3 25.0 0.0 24. I - 1.5 18.9 23.7 23.3 _
3%
of 10 human
13
5.0
10.0
15.0
Roeschlau
20.0 m mol/l
Fig. 2. Correlation of concentration values of free (0) and total (0) cholesterol obtained by the enzyme electrode and the Roeschlau procedure for the 10 samples. Equations of linear regressions for free and total cholesterol were found to be respectively: y = 0.950x + 0.154 (r = 0.9954), and y = 0.942x +0.297 (r = 0.9965), where y is the concentration found using the Roeschlau procedure; r = correlation coeffi-
cient.
values of current decrease (in relative units) for five additions of cholesterol standard solution. The calibration curve was repeated six times on the same day to obtain the standard deviation and CV values. In the same table we evaluated the reproducibility of the electrode procedure in four bile samples, for both free and total cholesterol. The determination was repeated five times on the same sample. The CV was 5-675. In Table III the results of the proposed procedure are compared, for ten human bile samples, with those obtained by an enzymatic spectrophotometric procedure considered the most suitable for bile samples [10,12,14]. In Fig. 2 the correlation is reported between the two procedures. The values obtained for total cholesterol in almost all samples were barely higher than the values for free cholesterol, in accord with existing knowledge that the amount of cholesterol esterified in bile is small. According to the behaviour of the immobilised enzyme the cholesterol electrode is responsive only to the free cholesterol, i.e. the unesterified cholesterol and the total can be obtained only by preliminary hydrolysis of the esterified cholesterol. Immobilisation of cholesterolesterase could be attempted in the future, thus permitting all the enzyme reactions to take place on the electrode surface.
The method of enzyme immobilisation coupled with the oxygen electrode described seems to be valuable for cholesterol determination in biological samples, and especially in bile samples, which can be considered complex due to the number of interfering substances, the chromophore dissolved and the high density of some of them. The use of the oxygen electrode seems to be necessary for the elimination of
14
interfering substances dissolved in bile samples. However, the use of detergents should be limited because of the negative effect on the permeability of the gas membrane [l]. The immobilisation procedure on nylon net has several advantages: the structure on the nylon net permits easy diffusion from the environs of the electrode surface to the immobilised enzyme and vice versa without forming a barrier to the diffusion of the reagent or product of the enzymatic reaction. Furthermore, the procedure for immobilisation is very simple and a large amount of activity remains after the immobilisation process. The mechanical structure of nylon net is practically unchanged by the immobilisation procedure. This allows the membrane to be assembled or disassembled from the electrode surface several times without a decrease in enzyme performance. The reproducibility with cholesterol standards and bile samples never exceeded 6%, which can be considered a reasonable value taking into account that a single analysis (2 additions) is performed in 4 min. The comparison between the proposed procedure and the spectrophotometric procedure on ten samples reported in detail in Table III is satisfactory for half of the results (A < 5%) and acceptable for the other half (A < 25%). Further studies, e.g. a comparison with gas liquid chromatography [16] might clarify the cause of the discrepancies in the latter samples. The determination of cholesterol in bile samples is important for evaluation of the risk of gallstone formation and lithogenic index calculation [17]; the procedure using the cholesterol electrode realised with a nylon net immobilised enzyme appears to be a potentially valuable tool for the task. Acknowledgement This work was supported
by the National
Research
Council
(CNR)
of Italy.
References 1 Noma A, Nakayama K. Polarographic method for rapid microdetermination of cholesterol with cholesterol esterase and cholesterol oxidase. Clin Chem 1976; 22: 336-340. 2 Clark LC Jr, Duggam CA, Grooms TA, Hart LM, Moore ME. One minute electrochemical enzimic assay for cholesterol in biological materials. Clin Chem 1981; 27: 1978-1982. 3 Huang H, Kuan SS, Guilbault GG. Amperometric determination of total cholesterol in serum with use of immobilized cholesterol ester hydrolase and cholesterol oxidase. Clin Chem 1977; 23: 671-676. 4 Bertrand C, Coulet PR, Gautheron DC. Enzyme electrode with collagen-immobilized cholesterol oxidase for the microdetermination of free cholesterol. Anal Lett 1979; 12: 1477-1488. 5 Karube I. Hara K, Matsuoka H, Suzuki S. Amperometric determination of total cholesterol in serum with use of immobilized cholesterol esterase and cholesterol oxidase. Anal Chim Acta 1982: 139: 127-132. 6 Tabata M. Endo J, Murachi T. Automated analysis of total cholesterol in serum using coimmobilized cholesterol ester hydrolase and cholesterol oxidase. J Appl Biochem 1981; 3: 84-92. 7 Admirand WH, Small DM. Physicochemical basis of cholesterol gallstone formation in man. J Clin Invest 1968; 47: 1043-1052. 8 Carey MC, Small DM. The physical chemistry of cholesterol solubility in bile. J Clin Invest 1978; 6 I : 998- 1026. 9 Watson D. A simple method for the determination of serum cholesterol. Clin Chim Acta 1960; 5: 637-643. 10 Biader Ceipidor U. Curini R, D’Ascenzo G, Tomassetti M. Alessandrini A. Evaluation of two
15
enzymatic
methods
for determination
of total cholesterol
in bile. G Ital Chim Clin 1979; 4: 355-369. with improved mechanical and analytical characteristics obtained by binding enzymes to nylon nets. Anal Chim Acta 1983; 146: 135- 148. Roda A, Festi D, Sama C et al. Enzymatic determination of cholesterol in bile. Clin Chim Acta 1975; 64: 337-341. Hornby WE, Morris DL. Modified nylons in enzyme immobilization and their use in analysis. In: Weetall HH, ed. Immobilized enzymes, antigens, antibodies and peptides. New York: M. Dekker Inc., 1975: 141-169. Roeschlau P, Bernt E, Gruber W. Enzymatic determination of total cholesterol in serum. Z Klin Chem Klin Biochem 12: 1974; 403-407. Bertrand C, Coulet PR, Gautheron DC. Multipurpose electrode with different enzyme systems bound to collagen films. Anal Chim Acta 1981; 126: 23-34. Bolton CH, Low-Beer TS, Pomare EW, Wicks ACB, Yeates J, Heaton KW. A simplified procedure for the analysis of cholesterol, phospholipids and bile salts in human bile. Clin Chim Acta 1978; 83: 177-181. Metger AL, Heymsfield S, Grundy SM. The lithogenic index - a numerical expression for the relative lithogenicity of bile. Gastroenterology 1972; 62: 499-500. IUPAC, Compendium of analytical nomenclature. Definitive rules 1977, prepared by Irving HMNH, Freiser H, West TS. Oxford: Pergamon Press, 1978.
11 Mascini M, Iannello M, Palleschi G. Enzymes electrodes 12 13
14 15 16
17 18