- Email: [email protected]
Biosensor for lactate determination in biological fluids. I. Construction and properties of the biosensor
Clinicu
Chinlicu
Acru,
162 (1987)
129-139
129
Elsevier CCA
03628
Biosensor for lactate determination in biological fluids. I. Construction and properties of the biosensor Jaroslav Racek and Jan Musil Deportment of Clinical Biochemistry, and Biochemistry, Facul!v of Medicine
(Received
7 February
1986: revision Key words:
Facul
received
Lactate;
4 August
Yeast
biosensor:
1986:
accepted
after
revision
Flavocytochrome
5 August
1986)
b,
Summary The preparation of a biosensor for lactate determination is described. The biosensor is based on an immobilized suspension of the aerobic yeast Hansenula anomala, containing flavocytochrome b, in high activity. The conditions for yeast cultivation were optimized to gain a sufficiently high activity of this enzyme converting lactate in the cells. The properties of the biosensor are compared with those of a sensor based on immobilized enzyme flavocytochrome b2. The yeast lactate biosensor has a sufficient sensitivity and linearity and short time of response. The precision and accuracy of lactate determination as well as the results of comparisons using an enzyme electrode and the spectrophotometric UV-test, enables this biosensor to be used in routine work. Analysis can be performed in blood plasma or whole blood. The stability of the biosensor makes it possible to work for 4 weeks with one yeast cell pellet.
Introduction Accumulation of lactate in body fluids and tissues results in lactate acidosis, the diagnosis of which requires quantitative determination of lactate in blood. This is especially the case in the early stages of diseases and in combined disorders of acid-base balance [1,2]. Knowledge of blood lactate concentration serves not only for the diagnosis of lactate acidosis, but it may also have prognostic value in Correspondence Czechoslovakia. 0009-8981/87/$03.50
to: J. Racek,
Department
0 1987 Elsevier
Science
of Clinical
Publishers
Biochemistry.
B.V. (Biomedical
Faculty
Hospital,
Division)
305 99 Pizeti.
130
critically ill patients, especially in shock [3]. For routine determination of blood lactate concentration it is desirable that the assays can be rapidly performed in whole blood without previous blood plasma separation. Enzyme sensors with electrochemical detection can fullfill these requirements. The most commonly described ‘enzyme lactate electrode’ uses flavocytochrome b, (L-lactate: ferricytochrome c oxidoreductase, EC 1.1.2.3), immobilized on a platinum anode [4-61. This enzyme catalyses the conversion of lactate to pyruvate in the presence of an electron acceptor - ferricyanide; the resulting ferrocyanide is reoxidized at the platinum electrode. The current thus produced is linearily related to the lactate concentration. This estimation is rapid, simple and inexpensive. Blood plasma, whole blood and other fluids may be used. On the other hand the instability of the biosensor due to inactivation of the immobilized flavocytochrome b2 is a disadvantage of the method [7]. The sensitivity of this enzyme to various chemical and physical influences makes it practically impossible to stabilise by chemical immobilization [8,9]. We now report studies designed to overcome these difficulties by replacing the isolated enzyme with the aerobic yeast Hansenula anomala, containing flavocytochrome b, of a sufficiently high activity. These cells represent a naturally immobilized enzyme. Materials and methods Enzyme isolation
The flavocytochrome b2 was extracted from baker’s yeast Saccharomyces cerevisiae according to the procedure of Dixon [lo]. The catalytic activity of the isolated enzyme, estimated spectrophotometrically with potassium ferricyanide at 25OC [a], was 217 pkat/l. Preparation of yeast cell suspension
The aerobic yeast Hansenula anomala, purchased from the Institute of Chemical Technology in Praba, was cultured on a Sabouraude’s agar and then inoculated into the liquid medium described by Bat&as and Spyridakis [ll], in which L-lactate was the only source of carbon. Trace elements were added to this solution as follows: H,BOs (100 pmol/l), CuSO, (2.5 pmol/l), MnSO., (25 pmol/l), KI (5 pmol/l), Na,MoO, (10 rmol/l), ZnSO, (25 pmol/l). The culture was kept at 25OC for 48 h; then sodium lactate buffer (0.2 mol/l), pH 5.4 was added (1: 4) and the incubation continued for 12 h. During growth the culture was shaken. The cell suspension was centrifuged at 1000 X g for 15 min. The supematant was discarded; the yeast cells in the pellet were resuspended in the same volume of sodium phosphate buffer (0.2 mol/l), pH.7.2. This suspension was either used directly for preparation of the yeast biosensor or freeze dried and kept at 4°C. The cells retained their enzyme activity for at least 4 mth. Apparatus and procedures
r. The lactate amperometric biosensor contained a platinum electrode with a platinum disc 2 mm in a diameter at one end, polarized at 0.35 V. The reference
131
semipermeable /
/
membrane
reference
P!-electrode
electrode
Ag/AgCI
/
chamber with cell SuSpenSiOn
Fig. 1. Scheme of a yeast lactate biosensor.
Ag/AgCl electrode was immersed in sodium phosphate buffer (0.2 mol/l), pH 7.2 containing KC1 (0.1 mol/l). The Pt-electrode was slightly (to the depth of about 0.3 mm) buried into the body of the sensor: a small chamber thus obtained was filled with the cell suspension (or enzyme solution) and covered with a semipermeable Viking membrane by means of an O-ring (Fig. 1). During measurements the top of the biosensor was put into the glass vessel, the temperature of which was maintained at 25OC (Fig. 2). The vessel was filled with 3.0 ml of sodium phosphate buffer (0.2 mol/l), pH 7.2 containing potassium ferricyanide (2.0 mmol/l). The current between the platinum anode and the reference electrode was measured using a mirror galvanometer (Karl Zeiss, Jena; sensitivity 30 nA-7 PA). 0.3 ml of blood plasma was then added (being so diluted 11 times) and’a second reading was taken after the current had stabilised. The solution was mixed with a magnetic stirrer. One measurement lasted for about 2 min (according to lactate concentrations) and could be repeated without delay. After finishing the measurement the electrode was immersed in sodium phosphate buffer (0.2 mol/l), pH 7.2 and stored at 4°C.
Fig. 2. Yeast lactate biosensor put in the measuring vessel.
132
The results obtained the results obtained by Lactate, fully enzymatic, used; the measurements 32, Finland).
with the yeast and enzyme biosensors were compared with a UV-test for blood plasma lactate determination. The kit Monotest 19 X 5 (No. 149.993, Boehringer Mannheim) was were made on an OLLI 3000 analyzer (Ollituote OY, Espoo
Reagents
Standard sodium L-lactate solutions were prepared from a stock solution of sodium lactate (1.0 mol/l) (Spofa, CSSR) by dilution with distilled water. When venous blood or plasma was used, 1 drop of ‘Fluorid/EDTA Lbung’ (No. 243.710, Boehringer Mannheim) was added to 1.0 ml of blood immediately after blood collection. All other reagents were purchased from Lachema Brno (CSSR) and were of analytical grade.
Changes in catalytic activity of jlavocytochrome 6, in the yeast cells during repeated cultivations
The first culture of the yeast cells in the liquid medium (with lactate) lead to a marked increase in enzyme activity. However, further cultures resulted in a fall in activity (Fig. 3). Thus the cell suspension from the fifth subculture produced (after immobilization on the electrode) a current response of less than 10% of that from the first culture.
current response to lactate concenlrotlon 20 mmolll
1 mmolll
0
1
2
3
4
El
5
of cultivations
Fig. 3. Changes of flavocytochrome b, activity in the yeast cells during repeated cultures. expressed as a yeast biosensor response to L-lactate. 0, yeast cells cultivated on a Sabouraude’s agar; l-5, number of cultures in a liquid medium with L-lactate.
133
Sensitivity Figure 4 shows the calibration curve of the yeast lactate biosensor. The shape of this curve was the same using a fresh suspension or a suspension obtained from lyophilized yeast cells, up to 4 wk after lyophilization. The lactate concentration in the measuring vessel of 1.0 mmol/l (corresponding to 11 mmol/l in biological material) increased the current by 0.55 PA. Linearity The response of the yeast biosensor was linear to a lactate concentration mmol/l in biological material (Fig. 4).
of 11.0
Time to reach the constant response Several seconds after adding the sample (containing lactate) to the reagent solution, the current started to rise becoming constant within approximately the next 2 min (according to the lactate concentration, Fig. 5). Influence of the ferricyanide concentration The response of the biosensor was tested with different ferricyanide concentrations (0.33-10.0 mmol/l). The measured current increased with this electron acceptor concentration up to the twofold lactate concentration. An increase in ferricyanide concentration over this limit failed to increase the sensitivity of the method (Fig. 6). Precision and accuracy of the assay Precision in the series was calculated from 20 repeated estimations of lactate concentration in pooled blood plasma and serum (Table I).
Fig. 4. Calibration curve concentration in biological
of a yeast lactate material is Il-times
biosensor (lactate concentration in measuring higher due to the dilution in the vessel).
,vessel;
134
1 mmolll
0.5 mmolll 0.25 mmolll
t 1
2 Iminl
Fig. 5. Increase of the current with time (at various lactate concentrations).
Fig. 6. Dependence of a current response on ferricyanide concentration (at various lactate concentrations).
TABLE I Precision of lactate determination with a yeast biosensor compared with the same enzyme biosensor ”
Pooled plasma
20
Pooled serum
20
o Enzyme biosensor.
1+2s (mmol/l) 1.21 f 0.09 1.18*0.08 a 3.38k0.16 3.25*0.15 a
cv w 3.12 3.39 D 2.43 2.31 ’
135
I
.r
hll0l/l] lactate (yeast
A
blosensorl
.-
J’
l-l=38 r = 0,990 y.-0,165*1,059x x=y= 2.79 mmolll
0
.
.
5.
.
.!, 5
20
15
10
[mmallll lactate 15
lactate biosensor)
lenlyme
(yeast
[mmalll]
s .
biasensorl
0
D
/
r = 0.976 y = 0.663 l 0.660 x x = y = 4.72 mmolll
5,
5
10
15
Fig. 7. Comparison of a yeast lactate ric UV-test for lactate.
TABLE Accuracy
lactate IUV-test1 20 lmmollll
biosensor
with:
(a) enzyme
lactate
biosensor;
(b) spectrophotomet-
II of lactate
determination
with Lactate
a yeast
concentration
compared
(mmol/I)
Declared
Measured 1.55 1.60 b 4.96 4.98 b
Moni-trol”
I E
1.42k0.15
Moni-trol’
II E
4.77 f 0.48
’ CB, coefficient of bias. b Enzyme biosensor.
biosensor
with
enzyme
biosensor D
+ 9.15 + 12.70 b + 3.98 + 4.40 b
ro Fig.
8. Comparison
of
lactate
15
concentration
i0
values
lmmollll
in whole
blood
and
blood
plasma
using
a yeast
biosensor.
Control sera Moni-trol I E and Moni-trol II E (Merz and Dade, 3186 Dtidingen), as well as a standard lactate solution (2.0 mmol/l) were used for the evaluation of the assay accuracy. Each serum was assayed five times; mean values and variation coefficients were calculated. The results are given in Table II. Precision and accuracy of the yeast biosensor were compared with those of the enzyme biosensor (Tables I and II). Comparison with the enzyme biosensor assay and UV-test
Lactate concentration was estimated in 38 blood plasma samples collected from patients admitted to the Faculty hospital in Plzeli and sportsmen after maximal physical exercise (Fig. 7a, b). Determination of lactate concentration in whole blood
All determinations of lactate concentration in 38 blood samples (see above) were made nor only in blood plasma, but also in whole blood, and the results were compared (Fig. 8). The comparison shows that the whole blood may be used instead of blood plasma with good results. Stability
The stability of the yeast biosensor was compared with that of the enzyme biosensor. While the response of the enzyme biosensor to lactate fell down to the practically inmeasurable values within 7 days, the response of the yeast biosensor was almost the same even after 18 days. It was found that the yeast biosensor could be used for 4 wk. Thereafter the yeast pellet of the biosensor should be replaced. Interferences
Interference of glucose could be eliminated by sodium fluoride addition to the reagent solution. A small positive interference was observed with uric acid above its
137
blood plasma concentration of 500 pmol/l. Further potentially interfering substances were studied in detail and the results are given in the second part of this work. However, no further important interferences affecting routine lactate determination in biological fluids were found.
The immobilization of living cells has a number of advantages which include the removal of a need to isolate and purify the enzyme, often with complicated procedures, and of enzyme inactivation during isolation or immobilization procedures. In addition it becomes unnecessary to add coenzymes and activators (the cells often containing them in sufficient amounts), multi-step systems in the cells can be used and enzymes in cells are usually well stabilized. All these advantages have been described for several microbial or tissue biosensors [12-171. The substrate, which is essential for a given microbe, can be determined with such a biosensor, if the electrode is sensitive to the product of enzyme reaction (NH,, CO,, H+ etc.) [12-141 or reflects respiratory activity of the cells [16,17]. The main reason for the construction of the yeast lactate biosensor was the poor stability of the enzyme biosensor with flavocytochrome b, [7]. Some authors have tried to improve the stability of enzyme biosensor by using an enzyme with a high specific catalytic activity; this requires a complicated procedure for enzyme extraction and purification, and is not suitable for a routine laboratory [5]. Flavocytochrome b, is an inductive enzyme [18] and we succeeded in inducing its synthesis by L-lactate in a culture medium. This method of enzyme induction by its substrate is a common procedure in preparation microbial sensors [15,16]. We observed the same phenomenon as described by Di Paolantonio and Rechnitz [15] in the case of induction of tyrosin phenol-lyase synthesis in microbial cells: after an increasing number of cultures the specific enzyme activity was falling. These authors have suggested that this is due to development of an alternative metabolic pathway for substrate conversion, which appears during the later cultures. For this reason we keep the yeast cells on a Sabouraude’s agar and only cells from the first culture with L-lactate were used for immobilization in the sensor. There are two methods for cell immobilization in the biosensor: (1) to fix cell suspension with a semipermeable membrane [15] and (2) to immobilize cells in a porous medium (paper, cellulose acetate or collagen membrane, agar gel or a nylon netting) [13,14,17]. We used the first to eliminate the diffusion barrier of a porous material. The principle of the yeast biosensor function is not different from that described for the enzyme biosensor. The cell membrane of the yeast Hansenula anomala is not a serious barrier for the substrate permeation and enables its diffusion. This can be evaluated by comparing the time, necessary for the same electrode response using enzyme biosensor and yeast biosensor. The values are similar. The sensitivity of the yeast biosensor is also comparable with data published for
enzyme biosensors with immobilized flavocytochrome b2 [5,7]. The enzyme activity in the yeast cells is sufficiently high that it is not a limiting factor in the rate of enzyme reaction. Higher current response for the same lactate concentration could be reached with a larger surface area of a Pt-electrode [4,6]. The range of the linear current response is sufficient for most blood plasma (or blood) samples. If the lactate concentration still exceeds 11 mmol/l, the determination must be repeated with a smaller volume of biological material. The precision and accuracy of this method is comparable with enzyme lactate electrodes or the spectrophotometric UV-test [19,20]. Figure 7a shows that enzyme and yeast biosensors afford practically the same results and the agreement between the yeast biosensor and UV-test is also very good (Fig. 7b). The positive error in testing accuracy and a positive intercept in comparison with a UV-test may be due to interference of reducing substances. Interference caused by increased glucose levels can be eliminated by NaF addition to the reagent solution. A small positive interference by uric acid (above its blood plasma concentration of 500 pmol/l) was observed. The specificity of the yeast electrochemical biosensor is however sufficient; it will be discussed in detail in the second part of this work. For this reason the biosensor may be used for routine determination. Whole blood can be used instead of blood plasma (as in the case of enzyme lactate biosensor) with the same results. The semipermeable membrane protects the sensor from the deposition of plasma proteins and erythrocytes. The stability of microbe biosensors is usually between 1 and 4 weeks when living cells are used [12-171. Our yeast lactate biosensor with living cells could be used for 4 wk; it is easy to regenerate it with a new cell suspension prepared from lyophilized cells. References 1 2 3 4 5 6 7 8 9 10 11
K&berg RA. Lactate homeostasis and lactic acidosis. Ann Intern Med 1980; 92: 227-237. Cohen RD. Simpson, R. Lactate metabolism. Anesthesiology 1975; 43: 661-673. Luft D. Klinische Bedeutung der Hyperlaktat’knie. Stuttgart: G. Thieme Verlag, 1981. Williams DL, Doig AR Jr, Korosi A. Electrochemical-enzymatic analysis of blood glucose and lactate. Anal Chem 1970; 42: 118-121. Racine P. Rapid lactate determination with an electrochemical enzymatic sensor: Clinical usability and comparative measurements. J Clin Chem Clin Biochem 1975; 13: 533-539. Durliat H, Comtat M, Baudras A. Spectrophotometric and electrochemical determination of L( +)lactate in blood by’ use of lactate dehydrogenase from yeast. Clin Chem 1976; 22: 1802-1805. Racine P, Mindt W. On the role of substrate diffusion in enxyme electrodes. Experientia 1971; Suppl. 18: 525-534. Prats M. Etude de l’importance des paramttres temperature et lumiere naturelle sur la stabilid de deux types de L( +)lactate: cytochrome c oxydorkductases (cytochrome ha) extraites de la levure Hansenula anomala. CR Acad Sci Paris 1980; 291: 1037-1040. Kulis JJ, &irmickas GJS, AntanaviTjus VS, Vaptkjaviejus RK. Ingibirovanije citochroma b, akrilamidom. Biochimiya 1982; 47: 582-585. Dixon M. In: Bergmeyer, HU ed. Methoden der enxymatischen Analyse. 2nd ed., Vol. 1. Berlin: AkademieVerlag, 1970: 1448-1449. Baudras A, Spyridskis A. Etude de la L( +)lactate: cytochrome c oxydoreductase (cytochrome b,) de la levure Hansenula anomala. Biochimie (Paris) 1971; 53: 942-955.
139 12 Arnold, MA, Rechnitz GA. Comparison of bacterial, mitochondrial, tissue and enzyme biocatalysts for glutamine selective membrane electrodes. Anal Chem 1980: 52: 1170-1174. 13 Hikuma M, Obana H, Yasuda T. Ammonia electrode with immobilized nitrifying bacteria. Anal Chem 1980; 52: 1020-1024. I4 Matsumoto K, Seijo H, Watanabe T. Immobilized whole cell-based flow-type sensor for cephalosporins. Anal Chim Acta 1979; 105: 429-432. I5 Di Paolantonio CL, Rechnitz GA. Induced bacterial electrode for the potentiometric measurement of tyrosine. Anal Chim Acta 1982; 141: 1-13. 16 Neujahr I-IY, Kjellen KG. Bioprobe electrode for phenol. Biotechnol Bioeng 1979; 21: 671-678. 17 Karube I, Suzuki S, Okada T, Hikuma M. Microbial sensors for volatile compounds. Biochimie (Paris) 1980; 62: 567-573. 18 Somlo M. Induction des lactico-cytochromes c reductases (D- et L-) de la levure aerobie par les lactates (D- et L-). Biochim Biophys Acta 1965; 97: 183-201. 19 Guillot Ch, Vanuxem D, Grimaud Ch. Nouvelle methode d’analyse rapide des lactates sanguins par une electrode enzymatique specifique. Path Biol 1976; 24: 431-433. 20 Kragenings I, Rackwitz R. Bestimmung von Laktat nach enzymatisch-elektrochemischem Prinzip im Vergleich mit drei Modifikationen der enzymatischen Methode. Ant1 Lab 1977; 23: 549-554.
Chinlicu
Acru,
162 (1987)
129-139
129
Elsevier CCA
03628
Biosensor for lactate determination in biological fluids. I. Construction and properties of the biosensor Jaroslav Racek and Jan Musil Deportment of Clinical Biochemistry, and Biochemistry, Facul!v of Medicine
(Received
7 February
1986: revision Key words:
Facul
received
Lactate;
4 August
Yeast
biosensor:
1986:
accepted
after
revision
Flavocytochrome
5 August
1986)
b,
Summary The preparation of a biosensor for lactate determination is described. The biosensor is based on an immobilized suspension of the aerobic yeast Hansenula anomala, containing flavocytochrome b, in high activity. The conditions for yeast cultivation were optimized to gain a sufficiently high activity of this enzyme converting lactate in the cells. The properties of the biosensor are compared with those of a sensor based on immobilized enzyme flavocytochrome b2. The yeast lactate biosensor has a sufficient sensitivity and linearity and short time of response. The precision and accuracy of lactate determination as well as the results of comparisons using an enzyme electrode and the spectrophotometric UV-test, enables this biosensor to be used in routine work. Analysis can be performed in blood plasma or whole blood. The stability of the biosensor makes it possible to work for 4 weeks with one yeast cell pellet.
Introduction Accumulation of lactate in body fluids and tissues results in lactate acidosis, the diagnosis of which requires quantitative determination of lactate in blood. This is especially the case in the early stages of diseases and in combined disorders of acid-base balance [1,2]. Knowledge of blood lactate concentration serves not only for the diagnosis of lactate acidosis, but it may also have prognostic value in Correspondence Czechoslovakia. 0009-8981/87/$03.50
to: J. Racek,
Department
0 1987 Elsevier
Science
of Clinical
Publishers
Biochemistry.
B.V. (Biomedical
Faculty
Hospital,
Division)
305 99 Pizeti.
130
critically ill patients, especially in shock [3]. For routine determination of blood lactate concentration it is desirable that the assays can be rapidly performed in whole blood without previous blood plasma separation. Enzyme sensors with electrochemical detection can fullfill these requirements. The most commonly described ‘enzyme lactate electrode’ uses flavocytochrome b, (L-lactate: ferricytochrome c oxidoreductase, EC 1.1.2.3), immobilized on a platinum anode [4-61. This enzyme catalyses the conversion of lactate to pyruvate in the presence of an electron acceptor - ferricyanide; the resulting ferrocyanide is reoxidized at the platinum electrode. The current thus produced is linearily related to the lactate concentration. This estimation is rapid, simple and inexpensive. Blood plasma, whole blood and other fluids may be used. On the other hand the instability of the biosensor due to inactivation of the immobilized flavocytochrome b2 is a disadvantage of the method [7]. The sensitivity of this enzyme to various chemical and physical influences makes it practically impossible to stabilise by chemical immobilization [8,9]. We now report studies designed to overcome these difficulties by replacing the isolated enzyme with the aerobic yeast Hansenula anomala, containing flavocytochrome b, of a sufficiently high activity. These cells represent a naturally immobilized enzyme. Materials and methods Enzyme isolation
The flavocytochrome b2 was extracted from baker’s yeast Saccharomyces cerevisiae according to the procedure of Dixon [lo]. The catalytic activity of the isolated enzyme, estimated spectrophotometrically with potassium ferricyanide at 25OC [a], was 217 pkat/l. Preparation of yeast cell suspension
The aerobic yeast Hansenula anomala, purchased from the Institute of Chemical Technology in Praba, was cultured on a Sabouraude’s agar and then inoculated into the liquid medium described by Bat&as and Spyridakis [ll], in which L-lactate was the only source of carbon. Trace elements were added to this solution as follows: H,BOs (100 pmol/l), CuSO, (2.5 pmol/l), MnSO., (25 pmol/l), KI (5 pmol/l), Na,MoO, (10 rmol/l), ZnSO, (25 pmol/l). The culture was kept at 25OC for 48 h; then sodium lactate buffer (0.2 mol/l), pH 5.4 was added (1: 4) and the incubation continued for 12 h. During growth the culture was shaken. The cell suspension was centrifuged at 1000 X g for 15 min. The supematant was discarded; the yeast cells in the pellet were resuspended in the same volume of sodium phosphate buffer (0.2 mol/l), pH.7.2. This suspension was either used directly for preparation of the yeast biosensor or freeze dried and kept at 4°C. The cells retained their enzyme activity for at least 4 mth. Apparatus and procedures
r. The lactate amperometric biosensor contained a platinum electrode with a platinum disc 2 mm in a diameter at one end, polarized at 0.35 V. The reference
131
semipermeable /
/
membrane
reference
P!-electrode
electrode
Ag/AgCI
/
chamber with cell SuSpenSiOn
Fig. 1. Scheme of a yeast lactate biosensor.
Ag/AgCl electrode was immersed in sodium phosphate buffer (0.2 mol/l), pH 7.2 containing KC1 (0.1 mol/l). The Pt-electrode was slightly (to the depth of about 0.3 mm) buried into the body of the sensor: a small chamber thus obtained was filled with the cell suspension (or enzyme solution) and covered with a semipermeable Viking membrane by means of an O-ring (Fig. 1). During measurements the top of the biosensor was put into the glass vessel, the temperature of which was maintained at 25OC (Fig. 2). The vessel was filled with 3.0 ml of sodium phosphate buffer (0.2 mol/l), pH 7.2 containing potassium ferricyanide (2.0 mmol/l). The current between the platinum anode and the reference electrode was measured using a mirror galvanometer (Karl Zeiss, Jena; sensitivity 30 nA-7 PA). 0.3 ml of blood plasma was then added (being so diluted 11 times) and’a second reading was taken after the current had stabilised. The solution was mixed with a magnetic stirrer. One measurement lasted for about 2 min (according to lactate concentrations) and could be repeated without delay. After finishing the measurement the electrode was immersed in sodium phosphate buffer (0.2 mol/l), pH 7.2 and stored at 4°C.
Fig. 2. Yeast lactate biosensor put in the measuring vessel.
132
The results obtained the results obtained by Lactate, fully enzymatic, used; the measurements 32, Finland).
with the yeast and enzyme biosensors were compared with a UV-test for blood plasma lactate determination. The kit Monotest 19 X 5 (No. 149.993, Boehringer Mannheim) was were made on an OLLI 3000 analyzer (Ollituote OY, Espoo
Reagents
Standard sodium L-lactate solutions were prepared from a stock solution of sodium lactate (1.0 mol/l) (Spofa, CSSR) by dilution with distilled water. When venous blood or plasma was used, 1 drop of ‘Fluorid/EDTA Lbung’ (No. 243.710, Boehringer Mannheim) was added to 1.0 ml of blood immediately after blood collection. All other reagents were purchased from Lachema Brno (CSSR) and were of analytical grade.
Changes in catalytic activity of jlavocytochrome 6, in the yeast cells during repeated cultivations
The first culture of the yeast cells in the liquid medium (with lactate) lead to a marked increase in enzyme activity. However, further cultures resulted in a fall in activity (Fig. 3). Thus the cell suspension from the fifth subculture produced (after immobilization on the electrode) a current response of less than 10% of that from the first culture.
current response to lactate concenlrotlon 20 mmolll
1 mmolll
0
1
2
3
4
El
5
of cultivations
Fig. 3. Changes of flavocytochrome b, activity in the yeast cells during repeated cultures. expressed as a yeast biosensor response to L-lactate. 0, yeast cells cultivated on a Sabouraude’s agar; l-5, number of cultures in a liquid medium with L-lactate.
133
Sensitivity Figure 4 shows the calibration curve of the yeast lactate biosensor. The shape of this curve was the same using a fresh suspension or a suspension obtained from lyophilized yeast cells, up to 4 wk after lyophilization. The lactate concentration in the measuring vessel of 1.0 mmol/l (corresponding to 11 mmol/l in biological material) increased the current by 0.55 PA. Linearity The response of the yeast biosensor was linear to a lactate concentration mmol/l in biological material (Fig. 4).
of 11.0
Time to reach the constant response Several seconds after adding the sample (containing lactate) to the reagent solution, the current started to rise becoming constant within approximately the next 2 min (according to the lactate concentration, Fig. 5). Influence of the ferricyanide concentration The response of the biosensor was tested with different ferricyanide concentrations (0.33-10.0 mmol/l). The measured current increased with this electron acceptor concentration up to the twofold lactate concentration. An increase in ferricyanide concentration over this limit failed to increase the sensitivity of the method (Fig. 6). Precision and accuracy of the assay Precision in the series was calculated from 20 repeated estimations of lactate concentration in pooled blood plasma and serum (Table I).
Fig. 4. Calibration curve concentration in biological
of a yeast lactate material is Il-times
biosensor (lactate concentration in measuring higher due to the dilution in the vessel).
,vessel;
134
1 mmolll
0.5 mmolll 0.25 mmolll
t 1
2 Iminl
Fig. 5. Increase of the current with time (at various lactate concentrations).
Fig. 6. Dependence of a current response on ferricyanide concentration (at various lactate concentrations).
TABLE I Precision of lactate determination with a yeast biosensor compared with the same enzyme biosensor ”
Pooled plasma
20
Pooled serum
20
o Enzyme biosensor.
1+2s (mmol/l) 1.21 f 0.09 1.18*0.08 a 3.38k0.16 3.25*0.15 a
cv w 3.12 3.39 D 2.43 2.31 ’
135
I
.r
hll0l/l] lactate (yeast
A
blosensorl
.-
J’
l-l=38 r = 0,990 y.-0,165*1,059x x=y= 2.79 mmolll
0
.
.
5.
.
.!, 5
20
15
10
[mmallll lactate 15
lactate biosensor)
lenlyme
(yeast
[mmalll]
s .
biasensorl
0
D
/
r = 0.976 y = 0.663 l 0.660 x x = y = 4.72 mmolll
5,
5
10
15
Fig. 7. Comparison of a yeast lactate ric UV-test for lactate.
TABLE Accuracy
lactate IUV-test1 20 lmmollll
biosensor
with:
(a) enzyme
lactate
biosensor;
(b) spectrophotomet-
II of lactate
determination
with Lactate
a yeast
concentration
compared
(mmol/I)
Declared
Measured 1.55 1.60 b 4.96 4.98 b
Moni-trol”
I E
1.42k0.15
Moni-trol’
II E
4.77 f 0.48
’ CB, coefficient of bias. b Enzyme biosensor.
biosensor
with
enzyme
biosensor D
+ 9.15 + 12.70 b + 3.98 + 4.40 b
ro Fig.
8. Comparison
of
lactate
15
concentration
i0
values
lmmollll
in whole
blood
and
blood
plasma
using
a yeast
biosensor.
Control sera Moni-trol I E and Moni-trol II E (Merz and Dade, 3186 Dtidingen), as well as a standard lactate solution (2.0 mmol/l) were used for the evaluation of the assay accuracy. Each serum was assayed five times; mean values and variation coefficients were calculated. The results are given in Table II. Precision and accuracy of the yeast biosensor were compared with those of the enzyme biosensor (Tables I and II). Comparison with the enzyme biosensor assay and UV-test
Lactate concentration was estimated in 38 blood plasma samples collected from patients admitted to the Faculty hospital in Plzeli and sportsmen after maximal physical exercise (Fig. 7a, b). Determination of lactate concentration in whole blood
All determinations of lactate concentration in 38 blood samples (see above) were made nor only in blood plasma, but also in whole blood, and the results were compared (Fig. 8). The comparison shows that the whole blood may be used instead of blood plasma with good results. Stability
The stability of the yeast biosensor was compared with that of the enzyme biosensor. While the response of the enzyme biosensor to lactate fell down to the practically inmeasurable values within 7 days, the response of the yeast biosensor was almost the same even after 18 days. It was found that the yeast biosensor could be used for 4 wk. Thereafter the yeast pellet of the biosensor should be replaced. Interferences
Interference of glucose could be eliminated by sodium fluoride addition to the reagent solution. A small positive interference was observed with uric acid above its
137
blood plasma concentration of 500 pmol/l. Further potentially interfering substances were studied in detail and the results are given in the second part of this work. However, no further important interferences affecting routine lactate determination in biological fluids were found.
The immobilization of living cells has a number of advantages which include the removal of a need to isolate and purify the enzyme, often with complicated procedures, and of enzyme inactivation during isolation or immobilization procedures. In addition it becomes unnecessary to add coenzymes and activators (the cells often containing them in sufficient amounts), multi-step systems in the cells can be used and enzymes in cells are usually well stabilized. All these advantages have been described for several microbial or tissue biosensors [12-171. The substrate, which is essential for a given microbe, can be determined with such a biosensor, if the electrode is sensitive to the product of enzyme reaction (NH,, CO,, H+ etc.) [12-141 or reflects respiratory activity of the cells [16,17]. The main reason for the construction of the yeast lactate biosensor was the poor stability of the enzyme biosensor with flavocytochrome b, [7]. Some authors have tried to improve the stability of enzyme biosensor by using an enzyme with a high specific catalytic activity; this requires a complicated procedure for enzyme extraction and purification, and is not suitable for a routine laboratory [5]. Flavocytochrome b, is an inductive enzyme [18] and we succeeded in inducing its synthesis by L-lactate in a culture medium. This method of enzyme induction by its substrate is a common procedure in preparation microbial sensors [15,16]. We observed the same phenomenon as described by Di Paolantonio and Rechnitz [15] in the case of induction of tyrosin phenol-lyase synthesis in microbial cells: after an increasing number of cultures the specific enzyme activity was falling. These authors have suggested that this is due to development of an alternative metabolic pathway for substrate conversion, which appears during the later cultures. For this reason we keep the yeast cells on a Sabouraude’s agar and only cells from the first culture with L-lactate were used for immobilization in the sensor. There are two methods for cell immobilization in the biosensor: (1) to fix cell suspension with a semipermeable membrane [15] and (2) to immobilize cells in a porous medium (paper, cellulose acetate or collagen membrane, agar gel or a nylon netting) [13,14,17]. We used the first to eliminate the diffusion barrier of a porous material. The principle of the yeast biosensor function is not different from that described for the enzyme biosensor. The cell membrane of the yeast Hansenula anomala is not a serious barrier for the substrate permeation and enables its diffusion. This can be evaluated by comparing the time, necessary for the same electrode response using enzyme biosensor and yeast biosensor. The values are similar. The sensitivity of the yeast biosensor is also comparable with data published for
enzyme biosensors with immobilized flavocytochrome b2 [5,7]. The enzyme activity in the yeast cells is sufficiently high that it is not a limiting factor in the rate of enzyme reaction. Higher current response for the same lactate concentration could be reached with a larger surface area of a Pt-electrode [4,6]. The range of the linear current response is sufficient for most blood plasma (or blood) samples. If the lactate concentration still exceeds 11 mmol/l, the determination must be repeated with a smaller volume of biological material. The precision and accuracy of this method is comparable with enzyme lactate electrodes or the spectrophotometric UV-test [19,20]. Figure 7a shows that enzyme and yeast biosensors afford practically the same results and the agreement between the yeast biosensor and UV-test is also very good (Fig. 7b). The positive error in testing accuracy and a positive intercept in comparison with a UV-test may be due to interference of reducing substances. Interference caused by increased glucose levels can be eliminated by NaF addition to the reagent solution. A small positive interference by uric acid (above its blood plasma concentration of 500 pmol/l) was observed. The specificity of the yeast electrochemical biosensor is however sufficient; it will be discussed in detail in the second part of this work. For this reason the biosensor may be used for routine determination. Whole blood can be used instead of blood plasma (as in the case of enzyme lactate biosensor) with the same results. The semipermeable membrane protects the sensor from the deposition of plasma proteins and erythrocytes. The stability of microbe biosensors is usually between 1 and 4 weeks when living cells are used [12-171. Our yeast lactate biosensor with living cells could be used for 4 wk; it is easy to regenerate it with a new cell suspension prepared from lyophilized cells. References 1 2 3 4 5 6 7 8 9 10 11
K&berg RA. Lactate homeostasis and lactic acidosis. Ann Intern Med 1980; 92: 227-237. Cohen RD. Simpson, R. Lactate metabolism. Anesthesiology 1975; 43: 661-673. Luft D. Klinische Bedeutung der Hyperlaktat’knie. Stuttgart: G. Thieme Verlag, 1981. Williams DL, Doig AR Jr, Korosi A. Electrochemical-enzymatic analysis of blood glucose and lactate. Anal Chem 1970; 42: 118-121. Racine P. Rapid lactate determination with an electrochemical enzymatic sensor: Clinical usability and comparative measurements. J Clin Chem Clin Biochem 1975; 13: 533-539. Durliat H, Comtat M, Baudras A. Spectrophotometric and electrochemical determination of L( +)lactate in blood by’ use of lactate dehydrogenase from yeast. Clin Chem 1976; 22: 1802-1805. Racine P, Mindt W. On the role of substrate diffusion in enxyme electrodes. Experientia 1971; Suppl. 18: 525-534. Prats M. Etude de l’importance des paramttres temperature et lumiere naturelle sur la stabilid de deux types de L( +)lactate: cytochrome c oxydorkductases (cytochrome ha) extraites de la levure Hansenula anomala. CR Acad Sci Paris 1980; 291: 1037-1040. Kulis JJ, &irmickas GJS, AntanaviTjus VS, Vaptkjaviejus RK. Ingibirovanije citochroma b, akrilamidom. Biochimiya 1982; 47: 582-585. Dixon M. In: Bergmeyer, HU ed. Methoden der enxymatischen Analyse. 2nd ed., Vol. 1. Berlin: AkademieVerlag, 1970: 1448-1449. Baudras A, Spyridskis A. Etude de la L( +)lactate: cytochrome c oxydoreductase (cytochrome b,) de la levure Hansenula anomala. Biochimie (Paris) 1971; 53: 942-955.
139 12 Arnold, MA, Rechnitz GA. Comparison of bacterial, mitochondrial, tissue and enzyme biocatalysts for glutamine selective membrane electrodes. Anal Chem 1980: 52: 1170-1174. 13 Hikuma M, Obana H, Yasuda T. Ammonia electrode with immobilized nitrifying bacteria. Anal Chem 1980; 52: 1020-1024. I4 Matsumoto K, Seijo H, Watanabe T. Immobilized whole cell-based flow-type sensor for cephalosporins. Anal Chim Acta 1979; 105: 429-432. I5 Di Paolantonio CL, Rechnitz GA. Induced bacterial electrode for the potentiometric measurement of tyrosine. Anal Chim Acta 1982; 141: 1-13. 16 Neujahr I-IY, Kjellen KG. Bioprobe electrode for phenol. Biotechnol Bioeng 1979; 21: 671-678. 17 Karube I, Suzuki S, Okada T, Hikuma M. Microbial sensors for volatile compounds. Biochimie (Paris) 1980; 62: 567-573. 18 Somlo M. Induction des lactico-cytochromes c reductases (D- et L-) de la levure aerobie par les lactates (D- et L-). Biochim Biophys Acta 1965; 97: 183-201. 19 Guillot Ch, Vanuxem D, Grimaud Ch. Nouvelle methode d’analyse rapide des lactates sanguins par une electrode enzymatique specifique. Path Biol 1976; 24: 431-433. 20 Kragenings I, Rackwitz R. Bestimmung von Laktat nach enzymatisch-elektrochemischem Prinzip im Vergleich mit drei Modifikationen der enzymatischen Methode. Ant1 Lab 1977; 23: 549-554.