Novel creatine biosensors based on all solid-state contact ammonium-selective membrane electrodes

Abstract

Novel creatine bienzymatic potentiometric biosensors were prepared by immobilizing urease and creatinase on all solid-state contact PVC-containing palmitic acid and carboxylated PVC matrix membrane ammonium-selective electrodes without inner reference solution. Potentiometric characteristics of biosensors were examined in physiological model solutions at different creatine concentrations. The linear working range and long-term sensitivity of the biosensors were also determined. The creatine biosensors prepared by using the carboxylated PVC membrane electrodes showed more effective performance than those of the PVC containing palmitic acid membrane electrodes. Creatine assay in serum samples was successfully carried out by using the standard addition method.

Keywords::

• Creatine
• all solid-state biosensors
• potentiometry
• carboxylated PVC
• palmitic acid

Introduction

Creatine plays an important role in the metabolism of proteins. Therefore, it appears to increase lean body mass and strength in humans (Stefan et al. 2006). Creatine is an important diagnostic substance in the assay of renal, thyroid, and muscular functions, and testing its clearance could also offer a quick and relatively simple biomedical diagnosis of acute myocardial infarction. Therefore determination of creatine in biological fluids is of great interest in clinical diagnosis (Kim et al. 1999, Wang et al. 2006, Stefan et al. 2003, Lakshmi et al. 2007, Karakus et al. 2006). The typical human reference range for serum/plasma creatine is between 35–140 µM; it can rise to greater than 1000 µM during kidney dysfunction (Stefan et al. 2003, Sena et al. 1988, Lad et al. 2008, Pan et al. 2011).

The following techniques are proposed for the assay of creatine; HPLC (Samanidou et al. 2010, Yang 1988), mass spectroscopy (Schwedhelm 2000), IR spectroscopy (Pezzanti et al. 2001), capillary zone electrophoresis (Clark et al. 2001, Burke et al. 1999), and flow injection analysis systems with electrochemical and spectrometric detection (Del Campo 1995). Among them, electrochemical enzymatic biosensors seem to be promising tools for such analysis, and various types of such biosensors for creatine detection have been developed (Stefan et al. 2003, 2006, Lakshmi et al. 2007, Patel et al. 2010, Motonaka et al. 1988, Koncki et al. 1996). Due to simplicity and low cost, potentiometric enzyme sensors are most widely applied among electrochemical enzymatic biosensors (Elmosallamy 2006, Soldatkin 2002, Shin et al. 1998, Koncki et al. 2000). In the literature, some of these biosensors are based on the catalysis of creatine at the surface of an ammonium-selective electrode. However, the biosensor prepared by immobilizing enzymes on ammonium-selective electrode membrane containing nonactin as ammonium ionophore and carboxylated PVC as the polymer membrane matrix material were expressed to be low sensitivities against creatine biomolecules (Koncki et al. 1996, Ma et al. 1988).

In recent years, attempts were made to develop miniaturized potentiometric biosensors, which are particularly important to reduce the amount of enzyme and reagents needed (Soldatkin et al. 2002, Bilitewski 1992). The miniaturization of a biosensor is possible by using an all-solid-state contact polymer membrane ion-selective electrode (Chou et al. 2009, Pookaiyaudom et al. 2011, Suzuki et al. 2001. Using the all-solid-state structure, the sensing devices provide many advantages, including simplicity of fabrication, being in the solid state form, ease of packaging, dry storage, and low cost (Alegret et al. 1989, Tinkilic 2002).

In the present work, we, for the first time, report all-solid-state contact creatine biosensors based on PVC containing palmitic acid and carboxylated PVC membrane NH₄⁺ -selective electrodes without an inner reference solution. The all-solid-state contact creatine biosensors functioned based on the catalysis of creatine by creatinase and urease enzymes at the surface of the NH₄ ⁺ -selective electrodes. The performance characteristics (e.g. pH influence, reproducibility, selectivity, stability, response time, optimum working range, etc.) of the prepared all-solid-state creatine biosensors were determined. Furthermore, we investigated whether these biosensors can be used to determine creatine level in human serum samples.

Experimental

Reagents and solutions

Creatinase (Pseudomonas sp., 8.7 U/mg, E.C.3.5.3.3), nonactin, palmitic acid, bis-(2-ethyl)hexylsebacate (DOS), poly(vinylchloride) (PVC), carboxylated poly(vinylchloride) (PVC–COOH), tetrahydrofuran (THF) were obtained from Fluka. Urease (Jack Beans, 5 U/mg, E.C.3.5.1.5) was purchased from Merck (Darmstadt, Germany). 1-Ethyl-3-(3- dimetylaminopropyl) carbodiimide (EDC) was obtained from Sigma Chem. Co. (St. Louis, USA). Epoxy resin (Macroplast Su 2227) and hardener (Desmodur RFE), used in the preparation of conductive all solid contact, were purchased from Henkel (Istanbul, Turkey) and Bayer AG (Darmstadt, Germany), respectively. All of the reagents used were analytical reagent grade and deionized water was used throughout.

Apparatus

Potentiometric measurements were carried out at room temperature (20 ± 1°C) by using a laboratory-made, computer-controlled, high-input impedance, multi-channel potentiometric measurement system. Throughout the measurements, a micro-sized solid silver/silver chloride electrode was used as reference electrode (RE) with the ammonium-selective electrode and creatine biosensor. The reference electrode was obtained from Isedo Med. Ltd. Sti. (Samsun, Turkey). To investigate the potentiometric characteristics of the prepared biosensor, the emf measurements were taken in the following cell assembly:

Micro sized solid silver/silver chloride RE | test solution | Creatine biosensor membrane/Ammonium-selective electrode | all solid state contact material | Cu wire.

The steady-state measurements were taken while stirring at a constant rate by immersing the biosensor and reference electrodes to the same depth in 10 mL of tested solution. The reference electrode and biosensor were washed with deionized water and dried with adsorbent tissue before each measurement of the test solution.

The pH of the buffer solutions was adjusted by using a glass pH electrode (Russell) with a Jenway 3040 model Ion Analyser. Solutions at required concentrations were homogenized using a Ultrasonic LC30 (Germany) stirrer.

Preparation of all-solid-state ammonium-selective membrane electrode

The membrane compositions of ammonium-selective electrodes prepared by using PVC containing palmitic acid and carboxylated PVC are shown in Table 1. Each membrane, according to the compositions in Table 1, was prepared by dissolving the components in 3 mL of THF. The construction of all-solid-state contact PVC containing palmitic acid and carboxylated PVC matrix membrane ammonium selective electrodes without an inner reference solution was carried out as described below. The epoxy resin mixture used to bind the graphite in preparing the all-solid-state contact was made from epoxy and hardener in THF in the proportions 1.0:0.5 w/w. The powdered graphite was mixed with the epoxy resin mixture in the proportions 1.0:1.0 w/w. After mixing, the solution was allowed to stand for 5–10 min in air. When the appropriate viscosity was attained, a shielded copper wire end with the length of 5 cm and radius of 0.5 mm was dipped into the mixture several times to obtain a coating thickness of about 0.3 mm. The uniformly coated copper wire end was allowed to stand overnight in an oven at 40°C. The all-solid-state contact was dipped into each membrane solution (given in Table 1) at least three times to obtain a membrane thickness of about 0.2 mm (see Figure 1). The coated membrane was left to be dried at laboratory conditions for at least 5 h. Finally, the dried all-solid-state contact ammonium-selective membrane electrode was conditioned by soaking into 0.1 mol/L ammonium chloride solution for at least 6 hours before use. When not in use, all the electrodes were stored in laboratory conditions. Before the individual measurement process, the electrodes were reconditioned for at least 10 minutes in 0.1 mol/L ammonium chloride solution. The potentiometric performance characteristics of the electrodes were tested in steady-state conditions.

Figure 1. Potentiometric measurement system.

Table 1. The membrane compositions of the all-solid-state contact ammonium-selective electrodes.

	Membrane composition (%)
Membrane components	I	II
PVC	30.0	–
Palmitic acid	3.0	–
Carboxylated PVC	–	32.3
Nonactin	3.0	3.0
Bis-(2-ethyl)hexyl sebacate (DOS)	64.0	64.5

Preparation of creatine biosensor

Enzymes were chemically immobilized on the all-solid-state ammonium-selective membrane electrode surfaces by using three different steps as described in procedures given by Koncki et al. (1994) and Karakus et al. (2005, 2006).

(A) 4 mg creatinase, 6 mg urease, and 1 mg 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide were dissolved in 0.2 mL deionized water. 10 μl of this solution was dropped onto the membranes of ammonium-selective electrodes and left overnight (biosensor A).

(B) 10 μl of 2.5% glutaraldehyde solution was deposited on the surface of biosensor A and the biosensor was left for 0.5 h. Then the surface was washed with deionized water to remove the excess of glutaraldehyde (biosensor B).

(C) 6 mg urease and 4 mg creatinase were dissolved in 0.2 mL deionized water. 10 μl of this solution was deposited on the biosensor B surface. To remove the excess of unbounded enzymes all types of biosensors were left in the vigorously stirred phosphate buffer (pH 7.0, 10 mM) for 1 h. When not in use, biosensors were stored in a refrigerator at 4°C. The schematic diagram of all-solid-state contact PVC membrane ammonium-selective electrode with immobilized enzyme layer and the set-up of the measurement system are illustrated in Figure 1.

Potential measurements for creatine biosensors were carried out by varying creatine concentration in steady-state condition. Ten millimolar of TRIS buffer was used as a working buffer solution (pH 7.0). The calibration curves were obtained by plotting the potential values of a series of standard creatine solutions against the logarithm of creatine concentration.

Procedure for determination of creatine in serum

The creatine levels in the serum samples in a concentration range of 0.97–6.81 mg/dL were succesfully determined by using the standard addition method (Premanode and Toumazou, 2007. The creatine biosensor was immersed in 10 mM TRIS buffer solution (at pH 7.0) containing a certain amount of serum creatine. To check our results obtained from the proposed biosensor method, the Jaffe (1886) method was also used for determining creatine levels in serum samples.

Results and Discussion

Potentiometric performances of the all-solid-state ammonium-selective membrane electrodes

The potentiometric performance characteristics of the all-solid-state contact ammonium-selective membrane electrodes are summarized in Table 2. Both electrodes based on PVC containing palmitic acid and carboxylated PVC exhibited almost 47 ± 5 and 50 ± 5 mV slopes between 1 × 10^–1 and 1 × 10^–5 M concentration levels of ammonium, respectively. The detection limits of both electrodes were about 2 × 10^–6 and 3 × 10^–6 M and the response times of the electrodes were less than 15 and 10 s, respectively. The ion-selective membrane electrodes with slopes in that vicinity range value and sensitivity could be used for analytical purposes such as biosensor development (Karakus et al. 2006, Buck 1972, Wegmann et al. 1984, Evans 1991).

Table 2. Potentiometric performance characteristics of the all-solid-state contact ammonium-selective membrane electrodes.

	NH4^+-Selective electrode
Parameter	membrane I	membrane II
Slope (mV/decade)	47 ± 5	50 ± 5
Correlation coefficient (V)	0.9889	0.9953
Limit of linear range (mol/L)	L × 10^–1–l × 10^–5	1 × 10^–1–1 × 10^–5
Response time for 10^–2 mol/L (s)	< 15 s	< 10s
Limit of detection (mol/L)	2 × 10^–6	3 × 10^–0

Although the sensitivity and the response time of both electrodes were comparable, the slope of the ammonium-selective electrode prepared by using carboxylated PVC (membrane II) was a bit better than that of the electrode prepared by using PVC containing palmitic acid (membrane I), as shown in Table 2. However, in respect to overall potentiometric performance, the present results obtained with the ammonium-selective electrode prepared by using carboxylated PVC can be comparable and more useful than those of the ammonium selective electrodes employed in other studies (Koncki et al. 1996, Chou et al. 2009).

Potentiometric performances of all-solid-state contact creatine biosensors

As previously mentioned, the measurements with the creatine biosensors based on detection of ammonium ions was performed in 10 mM Tris buffer solution. The response times of the creatine biosensors were measured to observe the reaction of the enzymes immobilized on different electrode membranes. According to IUPAC recommendations (Richard and Erno 1994), t₉₅ (95% of the steady-state signal) values of the creatine biosensors based on the all-solid-state PVC-COOH and PVC containing palmitic acid membrane ammonium-selective electrodes were less than 35 and 60 s, respectively. The response times of the prepared biosensors were also found to be better than most similar biosensors reported in the literature (Karakus et al. 2006, Koncki et al. 1996, Walcerz et al. 1996, Radomska et al. 2006, Magalhães et al. 2002).

The detection limits of the creatine biosensors based on all-solid-state PVC-COOH and PVC containing palmitic acid membrane electrodes were about 30 μM with a sensitivity of 50 ± 0.8 mV/perdecade and 20 μM with a sensitivity of 35 ± 0.3 mV/perdecade, respectively. Moreover, the linear working range of both of the creatine biosensors was between 0.1–10 mM. From the obtained results it can be concluded that the creatine biosensor prepared by using carboxylated PVC exposed a wider linear working range and better sensitivity when compared to the creatine biosensor prepared by using PVC containing palmitic acid.

Effect of buffer concentration

Measurements of the creatine biosensors, based on detection of ammonium ions, were performed at different TRIS buffer concentrations varying from 0.1 to 50 mM. Most effective potentiometric responses of the creatine biosensors were somehow obtained at constant 10 mM creatine levels in TRIS buffer at pH 7.0 is given in Figure 2.

Figure 2. Effect of buffer concentration on the response of all-solid-state creatine biosensor prepared by using PVC containing palmitic acid (-■-) and carboxylated PVC (-♦-) membrane ammonium-selective electrodes.

Effect of pH

Effect of pH on the analytical signal of the creatine biosensors was investigated by measurements in 10 mM TRIS buffer at different pHs (6.0–8.5). The potentiometric responses versus pH are shown in Figure 3. As can be seen from Figure 3, the output signals of creatine biosensors increase up to pH 7.0 and decrease over pH 7.0. This can be due to reduced activities of urease and creatinase enzymes; those might be gradually denaturized at the higher and lower pH values rather than the pH of 7.0 (Adeloju et al. 1996, Lucio et al. 1999).

Figure 3. Effect of pH on the response of all-solid-state creatine biosensor prepared by using PVC containing palmitic acid (-■-) and carboxylated PVC (-♦-) membrane ammonium-selective electrodes.

Effect of temperature

Figure 4 shows the effect of temperature on the response of all-solid-state contact creatine biosensors. As can be seen from Figure 4, the biosensor’s responses remain almost constant between 20–25 and 20–30°C for both biosensors based on PVC containing palmitic acid and carboxylated PVC membrane ammonium-selective electrodes, respectively. The responses then decrease at higher temperatures for both biosensors because of the decrease in the activity of urease and creatinase at higher temperatures.

Figure 4. Effect of temperature on the response of all-solid-state creatine biosensor prepared by using PVC containing palmitic acid (-■-) and carboxylated PVC (-♦-) membrane ammonium-selective electrodes.

Selectivities of the biosensors

To simulate the application of the creatine biosensor to a real sample, the selectivity of the biosensor membrane to entities other than ammonium ion has to be considered. Therefore, the interference effects from potassium, sodium, calcium, and iron (III) ions with ascorbic acid, uric acid, and creatinine were studied. Their selectivity coefficients were evaluated by the separate solution method according to the IUPAC recommendation (Umezawa et al. 2000). As indicated in Table 3, interference effects of Na⁺, K ⁺, Ca^{2 +}, Fe³⁺ ions, ascorbic acid, uric acid, and creatinine on the response of the creatine biosensors were very low. On the other hand, as the all-solid-state contact ammonium-selective electrode solely exhibited just around 20 times higher selectivity toward ammonium ion over potassium ion. Therefore, the potassium ion was the most important interfering specie for the nonactin ionophore-membrane-based biosensors (Chou et al. 2009). The obtained results indicate a better selectivity than those of other potentiometric biosensors (Karakus et al. 2006, Walcerz et al. 1996). Ascorbic acid, uric acid, and creatinine that existed in serum could not expose the interference effect on the developed potentiometric creatine biosensors.

Table 3. The selectivity coefficients of creatine biosensors based on the all-solid-state contact ammonium-selective membrane electrodes

	The selectivity coefficient, Kk£
Interfering agent	Membrane I	Membrane II
K ^+	l.l × 10^–2	3.2 × 10^–2
Na ^+	2.4 × 10^–3	1.2 × 10^–3
Ca^2 +	5.3 × 10–^4	3.1 × 10^–4
Fe^3 +	7.5 × 10^–4	3.2 × 10^–4
Ascorbic acid	3.8 × 10–^4	6.3 × 10^–4
Uric acid	6.1 × 10^–4	5.0 × 10^–4
Creatinine	1.9 × 10^–4	6.1 × 10^–4

Operational and storage stability of the creatine bioensors

Operational and storage stability of the creatine biosensors were determined by reading the potential values of the calibration solutions and by plotting the calibration curves for a period of 80 days (Figure 5). Studies corresponding to storage stability indicate that the biosensor did not lose any activity for almost 30 days. After 30 days, the biosensor response retained about 32 and 58% of the initial activity of the creatine biosensors based on a PVC containing palmitic acid and carboxylated PVC membrane electrodes, respectively. The decrease in response could be attributed to the denaturizing of the surface of immobilized enzyme (Chou et al. 2009). The responses were also almost constant during 12 times at operation at 10 mM TRIS buffer, pH 7.0-. The relative standard deviations of the biosensor responses were found to be less than 1,2 % (n:12) for both PVC containing palmitic acid and carboxylated PVC membrane electrodes. These results were better than those of amperometric and potentiometric creatine biosensors in some other studies (Stefan et al. 2006, 2003, Karakus et al. 2006, Koncki et al. 1996).

Figure 5. Operational and storage stability test for creatine biosensors; PVC containing palmitic acid (-■-) and carboxylated PVC (-♦-) membrane ammonium-selective electrodes (Measurements were done in 10 mM TRIS buffer, pH 7.0).

Application to creatine determination in human serum

The all-solid-state contact creatine biosensor based on carboxylated PVC was applied to human serum samples for the determination of creatine. Creatine levels in human serum samples obtained with the biosensor method by the standard addition method were also correlated with the Jaffe method (Koncki et al. 2000). The obtained result is given in Table 4. As indicated in Table 4, good correlation between the results was obtained using both the Jaffe method and proposed creatine biosensor method. The obtained results prove the usefulness of the biosensor for determination of creatine in serum samples.

Table 4. Determination of creatine in serum samples (R.S.D. values are lower than 0.1%).

Sample	Jaffe method (mg/dL)	Proposed method (mg/dL)
1	0.83	0.97
2	1.23	1.19
3	6.70	6.81
4	2.81	2.97

Conclusions

The novel all-solid-state contact creatine biosensor based on a carboxylated PVC ammonium-selective electrode showed good analytical parameters such as high selectivity and sensitivity (20 μM), storage stability over 30 days with less decrease of sensitivity, and short response time (0.5 min). Some other useful properties of the biosensor developed here are low cost, ease of preparation and use, and good operational stability in static conditions. The all-solid-state creatine biosensor can easily be miniaturized and applied to direct creatine detection in human serum. Ascorbic acid, uric acid, and creatinine did not exhibit interference effect on the response of the proposed potentiometric creatine biosensor. This is an advantage for creatine determination in serum samples when it is compared with other amperometric biosensors (Stefan et al. 2003, 2006, Lakshmi et al. 2007).

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.