sensors

Article Development of Ratiometric Fluorescent Biosensors for the Determination of Creatine and Creatinine in Urine

Hong Dinh Duong and Jong Il Rhee *

School of Chemical Engineering, Research Center for Biophotonics, Chonnam National University, Yong-Bong Ro77, 61186 Gwangju, Korea; [email protected] * Correspondence: [email protected]

Received: 21 September 2017; Accepted: 6 November 2017; Published: 8 November 2017

Abstract: In this study, the oxazine 170 perchlorate (O17)-ethylcellulose (EC) membrane was successfully exploited for the fabrication of creatine- and creatinine-sensing membranes. The sensing membrane exhibited a double layer of O17-EC membrane and a layer of enzyme(s) entrapped in the EC and polyurethane hydrogel (PU) matrix. The sensing principle of the membranes was based on the hydrolytic catalysis of urea, creatine, and creatinine by the enzymes. The reaction end product, ammonia, reacted with O17-EC membrane, resulting in the change in fluorescence intensities at two emission wavelengths (λem = 565 and 625 nm). Data collected from the ratio of fluorescence intensities at λem = 565 and 625 nm were proportional to the concentrations of creatine or creatinine. Creatine- and creatinine-sensing membranes were very sensitive to creatine and creatinine at the concentration range of 0.1–1.0 mM, with a limit of detection (LOD) of 0.015 and 0.0325 mM, respectively. Furthermore, these sensing membranes showed good features in terms of response time, reversibility, and long-term stability. The interference study demonstrated that some components such as amino acids and salts had some negative effects on the analytical performance of the membranes. Thus, the simple and sensitive ratiometric fluorescent sensors provide a simple and comprehensive method for the determination of creatine and creatinine concentrations in urine.

Keywords: creatine; creatinine; fluorescent sensing membrane; O17-EC membrane; ratiometric calculation

1. Introduction Creatine, a non-essential nutrient, plays an important role by supplying energy to muscle cells [1]. In the early 1990s, there was a great interest among consumers and researchers concerning therapeutic applications of creatine and benefits of using creatine as a dietary supplement [2]. In healthy adults, creatine levels in biological fluids such as serum or plasma are typically around 0.04–0.15 mM which may rise to above 1 mM under certain pathological conditions such as muscle disorders. Creatine is taken as an ergogenic supplement at a daily dose of 7–30 mM (or up to 140 mM) by athletes to increase their body mass [3]. During oral supplementation, some part of the consumed creatine may be excreted through urine. The difference between the amount of urinary creatine and ingested creatine dose indicates the amount of creatine absorbed by the muscles. Therefore, monitoring creatine levels is important in clinical diagnosis [4]. To our knowledge, few studies have been directed toward the development of creatine sensors/biosensors and the commercially available creatine kits are very expensive. Creatine biosensors have been studied for several decades and are based on the hydrolysis of creatine by creatinase and urease to produce ammonia. Ammonia is usually detected by any pH or ion-selective electrodes [5,6]. Alternatively, a second enzyme-catalyzed hydrolysis of creatine mediated by sarcosine oxidase results in the production of hydrogen peroxide (H2O2) as the end product, which can be

Sensors 2017, 17, 2570; doi:10.3390/s17112570 www.mdpi.com/journal/sensors Sensors 2017, 17, 2570 2 of 17 detected with a chronoamperometric technique [7–9]. Creatine detection has been also performed with flow-injection analysis (FIA) systems combined with immobilized enzyme reactors [10] and more expensive analytical tools such as high-performance liquid chromatography [10,11], capillary electrophoresis [12], and mass spectrometry [13]. However, the use of creatine sensors and other analytical tools is associated with practical issues related to sensitivity, reproducibility, stability, and interferences. Creatinine is the end product of creatine metabolism in mammalian cells [1]. Muscular creatine is converted to creatinine, which diffuses out of the cells and is excreted by the kidney into the urine. Therefore, creatinine levels in biological fluids are clinically important for the diagnosis of renal, thyroid, and kidney dysfunctions. The normal concentration of creatinine in human blood ranges from 0.04 to 0.15 mM and these values may rise to 1 mM or more in chronic kidney disease patients. A typical urine sample of an adult contains creatinine at 2.5–23 mM concentration [14]. The ratio of creatinine in human serum and urine sample is an important indicator of the kidney function. Therefore, it is important to monitor creatinine concentrations in biological fluids, including urine, for clinical diagnosis [15]. There has been a tremendous improvement in the quality of creatinine biosensors. Some review articles on creatinine biosensors have been published in recent years [16–20]. Creatine and creatinine may interact in a reversible pathway depending on the type of enzymes involved. Therefore, creatine and creatinine biosensors are based on the same principles. Ammonium ion or H2O2, the final product of the hydrolysis reaction of creatinine catalyzed by one or three enzymes, is often measured to determine creatinine concentrations. Transducers such as amperometry [21,22], potentiometry [23,24], ion-sensitive field effect transistor [25–27], or spectrophotometry [28] may be used based on the type of the final reaction products. Among these biosensors, the amperometric method was the first system to be successfully commercialized [17]. The use of other biosensors for the determination of creatine and creatinine is still limited. So far, traditional photometric methods based on Jaffe’s reaction have been widely used [29], however, colorimetric methods suffer from interferences during analysis. A diamond paste-based biosensor for creatine and creatinine detection is a rapid and simple sensor, but its measurement ranges are too low, 1–500 pM for creatine and 0.01–100 nM for creatinine [30]. The ratiometric fluorescence method offers some advantages over photometric and amperometric methods in terms of high sensitivity, good selectivity, and reproducibility, while enzyme immobilization allows high specificity, stability, and wide detection range to biosensors. Thus, we have developed ratiometric fluorescence biosensors using immobilized enzymes for the detection of creatine and creatinine. Creatine detection is based on the hydrolysis of urea and creatine catalyzed by two enzymes, urease and creatinase, to produce ammonia in the following reactions:

Creatinase Creatine + H2O −−−−−→ Urea + Sarcosine

Urease + Urea + H2O −−−→ NH4 + CO2 The enzyme creatinine deiminase catalyzes the conversion of creatinine to ammonia and N-methylhydantoin, as depicted in the following reaction:

Creatinine deiminase + Creatinine + H2O −−−−−−−−−−−→ NH4 + N-methylhydantoin

Ratiometric fluorescence biosensor is based on a fluorescent sensing membrane, which is a double-layer membrane of oxazine 170 perchlorate (O17) and ethylcellulose (EC) as an optical transducer and a layer of enzymes entrapped in the matrix of ethylcellulose (EC) and polyurethane hydrogel (PU) [31,32]. In this study creatine- and creatinine-sensing membranes were prepared (Scheme1) and their properties were investigated. The sensing membranes were also characterized in terms of their response Sensors 2017, 17, 2570 3 of 17 Sensors 2017, 17, 2570 3 of 17 toresponse creatine to and creatine creatinine and using creatinine ratiometric using calculation. ratiometric These calculation. creatine These and creatinine creatine biosensorsand creatinine were evaluatedbiosensors forwere their evaluated ability tofor determinetheir ability the to concentrationdetermine the ofconcentration creatine and ofcreatinine creatine and dissolved creatinine in artificialdissolved urine in artificial solution urine (AUS). solution (AUS).

Scheme 1. Structure of the creatine sensorsensor ((leftleft)) andand creatininecreatinine sensorsensor ((rightright).).

2.2. Materials and Methods

2.1.2.1. Materials OxazineOxazine 170170 perchlorateperchlorate (O17),(O17), ethylcelluloseethylcellulose (EC),(EC), ureaseurease (59,400(59,400 U/gU/g solid,solid, fromfrom CanavaliaCanavalia ensijormisensijormis (Jack bean)), bean)), creatinase creatinase (15 (15 U/mg U/mg solid, solid, from from ActinobacillusActinobacillus sp.),sp.), creatinine creatinine deiminase deiminase (20 (20U/mg U/mg solid, solid, microbialmicrobial), creatine,), creatine, creatinine, creatinine, glycine, glycine, histidin histidine,e, phenylalanine, phenylalanine, tryptophan, tryptophan, and urea werewere purchasedpurchased from from Sigma-Aldrich Sigma-Aldrich Chemical Chemical Co. Co. (Seoul, (Seoul, Korea). Korea). Tris Tris buffer buffer was was obtained obtained from from USB Co.USB (Cleveland, Co. (Cleveland, OH, USA)OH, USA) and PU and from PU AdvanceSourcefrom AdvanceSource Biomaterials Biomaterials Co. (Willmington, Co. (Willmington, MA, USA). MA, OtherUSA). analytical-gradeOther analytical-grade chemicals chemicals such as such sodium as phosphate,sodium phosphate, potassium potassium phosphate, phosphate, sodium chloride, sodium potassiumchloride, chloride,potassium sodium chloride, hydroxide, sodium hydrochloric hydroxide, acid, hydrochloric sodium bicarbonate, acid, sodium magnesium bicarbonate, sulfate, sodiummagnesium sulfate, sulfate, and calciumsodium sulfate, chloride and were calcium used without chloride further were used purification. without further purification.

2.2.2.2. Preparation of Creatine-Creatine- andand Creatinine-SensingCreatinine-Sensing MembranesMembranes µ WeWe preparedprepared O17-EC membranes as as previously previously de describedscribed [31] [31 ]by by mixing mixing O17 O17 stock stock (15 (15 µL,L, 2 µ 2mg/mL) mg/mL) with with EC (300(300 L,µL, 10 10 wt %)wt in%) ethanol. in ethanol. The mixture The mixture was incubated was incubated for 4 h at roomfor 4 temperature h at room andtemperature coated on and the bottomcoated ofon each the well bottom of a 96-well of each microtiter well of platea 96-well (NUNC microtiter Co. Copenhagen, plate (NUNC Denmark). Co. ◦ µ O17-ECCopenhagen, membrane Denmark). was dried O17-EC at 60 membraC for 12ne h.was In dried the next at 60 step, °C 20for 12L of h. the In mixturethe next ofstep, 10 wt20 %µL EC of andthe mixture 10 wt % PUof 10 in wt ethanol % EC and and water 10 wt (9:1 % in PU v/v%) in ethanol was coated and water onto O17-EC (9:1 in membrane,v/v%) was followedcoated onto by theO17-EC addition membrane, of the solution followed of a by given the amount addition of ureaseof the andsolution creatinase of a dissolvedgiven amount in 10 mMof urease phosphate and buffercreatinase (pH 7.4).dissolved The enzymes in 10 mM were phosphate entrapped buffer in the (pH polymer 7.4). matrixThe enzymes of EC and were PU entrapped and immobilized in the overpolymer O17-EC matrix membrane. of EC and The PU creatine-sensing and immobilized membrane over immobilizedO17-EC membrane. with enzymes The creatine-sensing was incubated ◦ formembrane 24 h at 4 immobilizedC. Surface morphologies with enzymes of was O17-EC incubate and creatine-sensingd for 24 h at 4 membrane°C. Surface were morphologies identified byof atomicO17-EC force and microscopycreatine-sensing (AFM). membrane The preparation were iden of thetified creatinine-sensing by atomic force membrane microscopy was (AFM). similar The to thatpreparation of the creatine-sensing of the creatinine-sen membrane.sing Onlymembrane one enzyme, was creatininesimilar to deiminase, that of the dissolved creatine-sensing in 10 mM phosphatemembrane. buffer Only (pHone 7.4)enzyme, was immobilizedcreatinine deiminase, on O17-EC dissolved membrane. in 10 mM phosphate buffer (pH 7.4) was immobilized on O17-EC membrane. 2.3. Measurements of Creatine and Creatinine

2.3. MeasurementsConcentrations of ofCreatine creatine and and Creatinine creatinine for measurements were in the range of 0.1 to 10 mM. Data were collected from the fluorescence intensity of the sensing membranes at two emission wavelengths Concentrations of creatine and creatinine for measurements were in the range of 0.1 to 10 mM. (λ = 565 and 625 nm) with an excitation wavelength of 460 nm (λ = 460 nm). The fluorescence Dataem were collected from the fluorescence intensity of the sensingex membranes at two emission spectra for the detection of creatine and creatinine were measured using a multifunctional fluorescence wavelengths (λem = 565 and 625 nm) with an excitation wavelength of 460 nm (λex = 460 nm). The microtiter plate reader (Safire2, Tecan Austria GmbH, Wien, Austria). fluorescence spectra for the detection of creatine and creatinine were measured using a multifunctional fluorescence microtiter plate reader (Safire2, Tecan Austria GmbH, Wien, Austria).

Sensors 2017, 17, 2570 4 of 17

2.3.1. Creatine Measurements The optimization of creatinase amount for immobilization was performed with 3.5, 7, 14, 28, and 42 unit (U) of creatinase and a fixed amount of urease (10 U). The immobilization efficiency of urease and creatinase in the creatine-sensing membrane was calculated by dividing the amount of immobilized enzymes with the total amount of enzymes used for immobilization. The amount of the immobilized urease and creatinase was determined by subtracting the amount of the un-immobilized urease and creatinase from the total amount of the enzymes used. The un-immobilized urease and creatinase were separated from the immobilized urease and creatinase in one well by washing several times with 10 mM phosphate buffer (pH 7.4). The protein values of the washed, un-immobilized urease and creatinase were determined by Bradford method. The kinetic parameters, maximal reaction rate (Vmax) and Michaelis-Menten constant (Km), of the enzymes co-immobilized on the supporting material of EC and PU were determined from the Hanes plot using the ratio of fluorescence intensities at λem = 565 and 625 nm. The sensitivity of creatine-sensing membranes with different amounts of creatinase was evaluated through the slope value (SI), i.e., the ratio of the fluorescence intensities at two emission wavelengths (λem = 565 and 625 nm) with respect to creatine. The reversibility of the creatine-sensing membrane was tested in distilled water and 1.0 mM creatine prepared in Tris buffer. The creatine-sensing membrane was first exposed to distilled water and then to 1.0 mM creatine solution and the fluorescence measurements were performed on the microtiter plate reader at an interval of 30 s. In addition, the effects of pH and temperature on the creatine-sensing membrane were investigated. The creatine-sensing membrane was exposed to 1.0 mM creatine solutions at a pH range of 5.0–9.0. The membrane was also tested at different temperatures (25, 30, 33, 35, 37, and 40 ◦C) at creatine concentrations of 0.1–10 mM. The long-term stability of the creatine-sensing membrane in the presence of creatine at various concentrations was evaluated through the determination of its repeatability by measuring the fluorescence intensity obtained initially and after 3 months. The interfering effects of some components in urine samples on the creatine-sensing membrane were investigated using 1.0 mM creatine and 0.2 mM of glycine, histidine, phenylalanine, tryptophan, and creatinine as well as their mixture.

2.3.2. Creatinine Measurements The optimal amount of creatinine deiminase for immobilization on O17-EC membrane was tested with 2.5, 5, and 7.5 U of creatinine deiminase. The immobilization efficiency of creatinine deiminase in the creatinine-sensing membrane was evaluated with Bradford method, as described to measure the immobilization efficiency of enzymes in part Section 2.3.1. The sensitivity of the creatinine-sensing membrane was also evaluated through SI. The kinetic parameters (Vmax and Km) of the immobilized creatinine deiminase were determined from the Lineweaver-Burk plot based on the ratio of the fluorescence intensities at λem = 565 and 625 nm. The reversibility of the creatinine-sensing membrane was examined with 0.1 and 1.0 mM creatinine dissolved in Tris buffer and the fluorescence measurements were performed on a microtiter plate at an interval of 30 s. The effects of pH and temperature on the creatinine-sensing membrane were investigated with 0.4 mM creatinine solution at a pH range of 5.0 to 10.0 and temperature of 30, 33, 35, 37, and 40 ◦C using creatinine concentrations from 0.1 to 10 mM. The long-term stability of the creatinine-sensing membrane at various creatinine concentrations was evaluated by measuring the fluorescence intensity obtained initially and after 1.5 months. The interference effects of some components on the creatinine-sensing membrane were also investigated using 1.0 mM creatinine and 0.2 mM of glycine, histidine, phenylalanine, tryptophan, urea, and creatinine as well as their mixture.

2.3.3. Artificial Urine Solution (AUS) Artificial urine solution containing various concentrations of creatine and creatinine was prepared, and the concentrations of creatine and creatinine were determined with the creatine- Sensors 2017, 17, 2570 5 of 17

and creatinine-sensing membranes, respectively. The AUS comprised 2.5 mM CaCl2, 45 mM NaCl, 3.5 mM KH2PO4, 3.5 mM K2HPO4, 2.5 mM NaHCO3, 1 mM MgSO4, 2.5 mM Na2SO4, and creatine or creatinine at concentration range of 0.1–10 mM and pH 7.2.

2.4. Ratiometric Method The ratiometric method for creatine and creatinine biosensors was based on the ratio of the fluorescence intensities of the creatine- and creatinine-sensing membranes at an excitation wavelength of 460 nm (λex = 460 nm) and two emission wavelengths (λem = 565 nm [FI565] and 625 nm [FI625]) as follows: R = FI565/FI625

2.5. Data Analysis The differences in fluorescence intensity of the sensing membranes at different pH, and temperature values were assessed by one-way analysis of variance (ANOVA). A value of p < 0.05 was considered as a statistically significant. Statistical tests were performed using the software InStat (v.3.01, GraphPad Software Inc., San Diego, CA, USA).

3. Results and Discussion

3.1. Enzyme Immobilization for Creatine-Sensing Membrane The response of the creatine-sensing membrane in the presence of different amounts of immobilized creatinase is shown in Figure1a. The creatine-sensing membrane immobilized with 14 U creatinase showed the highest SI (SI14U = 0.415) at creatine concentrations from 0.1 to 1.0 mM. We failed to see any increase in the sensitivity of the membrane at higher amounts (28 and 42 U) of creatinase, as evident from the lower SI at 28 and 42 U (SI28U = 0.104 and SI42U = 0.127) as compared with 14, 7, and 3.5 U creatinase. Results of Bradford protein assay showed that the immobilization efficiency of urease and creatinase on the second layer of the creatine-sensing membrane was very high at all creatinase amounts. The immobilization efficiency for the two enzymes was 83.7%, 75.8%, 75.5%, 85.1%, and 91.9% with 3.5, 7, 14, 28, and 42 U creatinase, respectively. The high immobilization efficiency resulted in the good enzyme-capturing capacity of EC and PU matrix. The immobilization efficiency of urease (10 U) into the matrix of EC and PU was higher than 90% in preliminary studies. Based on the high immobilization efficiency of urease (10 U), the immobilization efficiency for creatinase into EC and PU matrix may be estimated to be higher than 70%. However, the excess of creatinase immobilization + into the matrix of EC and PU may obstruct NH4 transport in O17-EC membrane. The high density + of the enzyme-supporting layer obstructed the passage of NH4 in O17-EC membrane below the enzyme layer. Based on these observations, we chose 14 U creatinase to fabricate the creatine-sensing membrane for further work. The creatine-sensing membrane included two layers. The first layer was O17-EC membrane, while the second layer was the membrane comprising urease and creatinase co-immobilized in the matrix of EC and PU. As seen in AFM images (Figure1b), O17-EC membrane displayed a smooth surface with a surface mean roughness (Ra) and root mean square roughness (Rq) of 1.07 and 1.343 nm, respectively, whereas the surface of the second layer immobilized with urease and creatinase displayed higher values of Ra and Rq (9.514 and 12.304 nm, respectively). The increase in the roughness of the sensing membrane resulted from the successful immobilization of the enzymes into EC-PU matrix and the subsequent formation of PU hydrogel during the reaction with the aqueous enzyme solution. Sensors 2017, 17, 2570 6 of 17 Sensors 2017, 17, 2570 6 of 17

(a)

Creatinase : 3.5 U, 7 U, 14 U, 28 U, 42 U

0.7 100 SI=0.415 625 0.6

/FI SI=0.279 80 565

0.5 60

SI=0.265 0.4 40 SI=0.127 0.3 SI=0.104 20 Immob. efficiency, [%] Ratio of Fl. intensities, FI 0.2 0 0.0 0.2 0.4 0.6 0.8 1.0 5 7 4 8 2 3. se e1 e2 e4 se ra as as as ra C Cr Cr Cr Creatine [mM] UC U U U U Immobilized Enzymes (10 Urease)

(b)

Figure 1. (a) The response of the creatine-sensing membrane immobilized with 3.5, 7, 14, 28, and 42 Figure 1. (a) The response of the creatine-sensing membrane immobilized with 3.5, 7, 14, 28, and 42 U U creatinase and fixed amount (10 U) of urease in the presence of 0.1 to 1.0 mM creatine (top). SI creatinase and fixed amount (10 U) of urease in the presence of 0.1 to 1.0 mM creatine (top). SI presents presents slope value in the linear detection range of 0.1 to 1.0 mM. The immobilization efficiency for slope value in the linear detection range of 0.1 to 1.0 mM. The immobilization efficiency for various various amounts of creatinase and 10 U of urease (top); (b) AFM images of O17-EC membrane (top) amounts of creatinase and 10 U of urease (top); (b) AFM images of O17-EC membrane (top) and the and the second layer containing entrapped enzyme(s) (below). second layer containing entrapped enzyme(s) (below). Sensors 2017, 17, 2570 7 of 17

3.2. Characterization of the Creatine-Sensing Membrane As shown in Figure2, the creatine-sensing membrane was very sensitive to ammonia produced during the hydrolysis reactions of urea and creatine in the presence of urease and creatinase, respectively. The detection range of creatine based on the ratiometric calculation method could be divided into two linear ranges of 0.1–1.0 mM and 1.0–10 mM with high regression coefficient 2 2 values of r 0.1–1.0 mM = 0.92 and r 1.0–10 mM = 0.96, respectively. The relative standard deviation for 1.0 mM creatine was 2.4% (n = 7) based on five repetitive measurements, while the detection limit (LOD, S/N = 3) was 0.015 mM. According to the response of O17-EC membrane to different ammonia concentrations [31], the fluorescence intensity of the creatine-sensing membrane would increase at λem = 565 nm and decrease at λem = 625 nm in the presence of the increasing concentrations of creatine. However, a small increase in the fluorescence intensity was observed at λem = 625 nm at low creatine concentrations. This observation may be attributed to the low amount of ammonia, which reacted with O17 dye as per the following equation

+ − + − + − NH4 OH + H Dye ⇔NH4 Dye + H2O

A large amount of the dye was in its free state. The reaction mechanism of O17 dye and ammonia produced from the catalytic breakdown of creatine was explained in our previous research [32]. Some modifications in the original sensor may result in unexpected observations, as evident from the behavior of the peak at λem = 625 nm, which was different from that observed for the original O17 dye. However, the ratiometric method may normalize the change in the fluorescence intensities that are unrelated to the change in the target concentration. Therefore, the values for the ratio of fluorescence intensities at λem = 565 and 625 nm corresponded well with creatine concentrations. For fluorescence-based sensors, measurements of fluorescence intensity at a single band edge are known to be problematic for practical applications [33]. The ratiometric fluorescence method is useful for the correction of a variety of analyte-independent factors in fluorescent sensors, wherein the temporal and spatial distribution of the measured fluorescence intensity may typically fluctuate owing to the unequal distribution of fluorophores within the sensor, variation in dynamics of fluorophores in different media, and noise in the measurement system (e.g., variations in the illumination intensity). The self-calibration property of the ratiometric method has led to the development of a wide range of ratiometric fluorescent sensors that provide precise quantitative analysis. Nakata et al. described real-time monitoring of saccharide conversion pathway using a seminaphthorhodafluor-conjugated lectin-based ratiometric fluorescent biosensor [34], whereas Xie et al. exploited the enzymatic reaction of hyaluronidase and hyaluronan bound to two fluorescent dyes for fluorescence quenching and dequenching of these dyes before and after enzyme reaction to obtain the proportion between the ratiometric fluorescence intensity and hyaluronidase level [35]. Other researchers have used ratiometric fluorescence biosensors for the detection of DNA [36] and nitric oxide [37]. Kinetic parameters of creatinase (14 U) and urease (10 U) co-immobilized on the supporting material of EC and PU were evaluated with Michaelis-Menten kinetics using the ratio of fluorescence app intensities at λem = 565 and 625 nm. An apparent maximal reaction rate (Vmax ) of 0.0406 1/min and app apparent Michaelis-Menten constant (Km ) of 3.441 mM were obtained using Hanes plot. The value app of Km was greater than that obtained for creatinase immobilized with chitosan-SiO2-multiwall app app carbon nanotubes nanocomposite (Km = 0.58 mM, [38]). Km is usually dependent upon the app supporting material and immobilization method. The large value of Km in this work indicates the low affinity of creatinase to EC and PU matrix over EC-O17 membrane and may be associated with the conformation and arrangement of the enzyme during immobilization. The slow reaction rate of the + creatine-sensing membrane may be attributed to the retardation of the reaction product (NH4 OH) toward O17-EC layer through the thick enzyme-immobilized layer. The response time of the creatine-sensing membrane was approximately t95(0.1–0.4 mM) = 1 to 2.5 min, t95(0.6–2 mM) = 2.5 to 3.5 min, and t95(4–10 mM) = 0.5 to 1.5 min. Moreover, the addition of PU to Sensors 2017, 17, 2570 8 of 17 Sensors 2017, 17, 2570 8 of 17

the supporting material for(a) enzyme immobilization(b) offered a convenient environment to shorten the 18,000 transport time of ammonia passing625 through nm the 0.1 mM enzyme-immobilized membrane to contact O17-EC 0.2 mM 0.9 membrane. The use of PU as a supportingmaterial 0.4 mM for enzyme immobilization may improve the 15,000 565 nm 0.6 mM Sensors 2017, 17, x 0.8 mM 625 8 of 17 response time of the enzyme-immobilized membrane. 0.8 1.0 mM /FI

10 mM 2.0 mM 565 12,000 4.0 mM (b) (a) 6.0 mM 0.7 18,000 8.0 mM 0.1 mM 625 nm 10.0 mM 9,000 0.2 mM 0.9 0.4 mM 15,000 565 nm 0.6 mM 0.6

0.8 mM 625 6,000 0.1 mM 0.8 1.0 mM /FI

10 mM Creatine 2.0 mM 0.5 565 12,000 4.0 mM Fluorescence intensity [a.u.] intensity Fluorescence 3,000 6.0 mM

0.7 Ratio Fl.intenisties, of FI 8.0 mM 9000 10.0 mM 0.4

0 0.6 525 550 575 600 625 650 675 700 0246810 6000 Emission wavelength0.1 mM [nm] Creatine [mM] Creatine 0.5

Fluorescence intensity [a.u.] intensity Fluorescence 3000 Figure 2. (a) Fluorescence emission spectra of the creatine-sensing membrane in theRatio Fl.intenisties, of FI presence of creatine at 0.1 to 10 mM concentration monitored at an excitation wavelength0.4 of 460 nm. (b) 0 Calibration curve for 525creatine 550 575 calculated 600 625 650 by 675 the 700 ratio0246810 of two fluorescence intensities measured at emission wavelengths of 565Emission and 625wavelength nm. [nm] Creatine [mM]

Figure 2. (a) Fluorescence emission spectra of the creatine-sensing membrane in the presence of creatine TheFigure response 2. (a) Fluorescence time of the emission creatine-s spectraensing of membranethe creatine-sensing was approximately membrane in t 95(0.1–0.4the presence mM) = 1 of to 2.5 at 0.1 to 10 mM concentration monitored at an excitation wavelength of 460 nm. (b) Calibration curve min, creatinet95(0.6–2 mM) at =0.1 2.5 to to 10 3.5 mM min, concen and trationt95(4–10 mM) monitored = 0.5 to at1.5 an min. excitation Moreover, wavelength the addition of 460 ofnm. PU (b to) the supportingforCalibration creatine material calculated curve forfor by enzymecreatine the ratio calculatedimmobilization of two fluorescence by the offeredratio intensities of twoa convenient fluorescence measured environment at intensities emission wavelengthsmeasured to shorten at ofthe transport565emission and time 625 wavelengths nm. of ammonia of 565passing and 625 through nm. the enzyme-immobilized membrane to contact O17-EC membrane. The use of PU as a supporting material for enzyme immobilization may improve the responseAsThe shown timeresponse inof the our time enzyme-immobilized previous of the creatine-s study [32ensing ],membrane. O17-EC membrane membrane was approximately showed high t95(0.1–0.4 reversibility mM) = 1 to despite 2.5 beingmin, coated Ast95(0.6–2 shown mM) with = in 2.5 a our second to 3.5previous min, layer and study of polymert95(4–10 [32], mM) O17-EC (e.g.,= 0.5 to EC) membrane1.5 and min. urease. Moreover, showed The highthe fluorescent addition reversibility O17of PU dyedespite to couldthe supporting material for enzyme immobilization+ − offered a convenient environment to shorten the offerbeing a protoncoated inwith the a formationsecond layer of of NH polymer4 OH (eby.g., theEC) protonation and urease. The of ammonia fluorescent in O17 water dye and could react transport time of ammonia passing+ through− + the- enzyme-immobilized membrane to contact O17-EC reversiblyoffer a proton upon reductionin the formation of NH4 ofOH NH.4 Herein,OH by the a sequence protonation of hydrolysis of ammonia reactions in water of creatineand react and membrane. The use of PU as a supporting+ - material for enzyme immobilization may improve the ureareversibly resulted upon in the reduction production of NH of4 ammonia,OH . Herein, which a sequence reacted of with hydrolysis O17-EC reactions membrane of creatine and changed and response time of the enzyme-immobilized membrane. theurea fluorescence resulted in intensity the production of O17 of dye. ammonia, This change which wasreacted easy with to recognize O17-EC membrane upon repeated and changed exposure As shown in our previous study [32], O17-EC membrane showed high reversibility despite ofthe fluorescence creatine-sensing intensity membrane of O17 dye. to 1.0 This mM change creatine was andeasy distilled to recognize water upon (DW). repeated Figure exposure3 shows of the thebeing creatine-sensing coated with a secondmembrane layer to of 1.0 polymer mM creatine (e.g., EC) and and distilled urease. water The (DW).fluorescent Figure O17 3 showsdye could the reversibility of the creatine-sensing membrane+ - as the ratio of the fluorescence intensities at λem = 565 reversibilityoffer a proton of thein thecreatine-sensing formation of membrane NH4 OH byas thethe ratioprotonation of the fluorescence of ammonia intensities in water at and λem =react 565 and 625 nm. The sensing membrane exhibited a very low relative standard deviation of 0.78% and andreversibly 625 nm. upon The reduction sensing membrane of NH4+OH exhibited-. Herein, a avery sequence low relative of hydrolysis standard reactions deviation of ofcreatine 0.78% and 2.69% in DW and 1.0 mM creatine, respectively. 2.69%urea resulted in DW andin the 1.0 productionmM creatine, of respectively.ammonia, which reacted with O17-EC membrane and changed

the fluorescence intensity 0.8of O17 dye. This change was easy to recognize upon repeated exposure of the creatine-sensing membrane to 1.0 mM creatine and distilled water (DW). Figure 3 shows the

em

reversibility of the creatine-sensing625 membrane as the ratio of the fluorescence intensities at λ = 565 0.7 1.0 mM creatine and 625 nm. The sensing/FI membrane exhibited a very low relative standard deviation of 0.78% and 2.69% in DW and 1.0 mM565 creatine, respectively. 0.6 0.8

625 0.5 0.7 1.0 mM creatine /FI 565 DW 0.4 0.6 Ratio of Fl.intensities, FI Fl.intensities, of Ratio

0.3 0.5 0 20406080100 Time, [min] DW 0.4 Figure 3. Reversibility of the creatine-sensing membrane in the presence of 1.0 mM creatine and

Figure 3. ReversibilityFI Fl.intensities, of Ratio of the creatine-sensing membrane in the presence of 1.0 mM creatine and distilleddistilled water water (DW). (DW). 0.3 0 20406080100 Time, [min]

Figure 3. Reversibility of the creatine-sensing membrane in the presence of 1.0 mM creatine and distilled water (DW). Sensors 2017, 17, 2570 9 of 17

The performance of a biosensor with an enzyme is usually influenced by pH and temperature. pH determines the activity of the enzyme and the consequent efficiency of the catalytic reaction in the biosensor. Similar to pH is the influence of temperature, which may increase or decrease the catalytic reaction rate of the enzyme. The creatine-sensing membrane preferred alkaline (pH range of 7.5 to pH 9.0) to weakly acidic (pH 5.0 to pH 7.0) medium. The ratio of the fluorescence intensities at λem = 565 and 625 nm obtained for the creatine-sensing membrane in the presence of 1.0 mM creatine failed to change significantly in the acidic pH range, but considerably increased with an increase in pH from 7.5 to 9.0 (data not shown). Creatine was difficult to dissolve in common aqueous solutions and, hence, was prepared in 10 mM Tris buffer at pH range of 7.5–8.5. Therefore, the creatine-sensing membrane worked well under alkaline conditions induced via ammonia produced from the hydrolysis of creatine and urea or interferences from other alkaline factors. The response of the creatine-sensing membrane to different temperatures was investigated at various creatine concentrations. The temperature range of 25 to 35 ◦C had no effect on the creatine-sensing membrane in the presence of 0.1 to 10 mM creatine (data not shown). However, the sensitivity of the membrane decreased at high temperatures (37–40 ◦C) and creatine concentration (1.0–10 mM). However, the sensitivity was unaffected by temperature at low concentrations of creatine (0.1–1.0 mM). Therefore, a temperature of approximately 33 ◦C was used for creatine measurements to extend the performance of the enzymes as well as the sensitivity of the creatine-sensing membrane for long-term use. The creatine-sensing membrane was tested after 3 months of use and storage; its sensitivity was found to be quite good (Figure4), as evident from SI of the linear curve with creatine concentration of 0.1 to 1.0 mM (SI0.1–1.0 mM) (0.22 at initial use and 0.242 after 3 months of use). Thus, the creatine-sensing membrane including O17-EC layer as a transducer and EC and PU matrix layer immobilized with two enzymes (urease and creatinase) showed excellent stability after long-time use. EC and PU were good supporting materials for the fluorescent dye O17 and enzyme immobilization. The layer of EC and PU matrix prevented enzyme leakage [39], but at the same time created an open environment, which improved the sensitivity of the membrane following ammonia production. The presence of PU in the enzyme-immobilized layer induced softening of the supporting layer for enzyme immobilization, thereby allowing higher amount of enzyme immobilization and offering a wider location for the passing ammonia to react with O7-EC membrane. In addition, we suggest that the covalent binding between the isocyanate groups of PU and amine groups of the enzymes increased the lifetime of the sensor. The increase in the background signal was associated with the increased opacity of the sensing membrane. The change probably occurred owing to the long-term soaking in 10 mM phosphate buffer solution that increased the reflection of the incident light and fluorescence emission during creatine measurements. The recovery capability of a biosensor may suffer from some interferences existing in the sample. Urine samples may contain several types of carbohydrates, metabolites, and electrolytes, which may interfere with the fluorescent creatine biosensor developed here. According to the results of one-way ANOVA with Tukey-Kramer multiple comparison post-test, the measurement for 1.0 mM creatine in the presence of 0.2 mM histidine showed a significant difference from the sample without histidine (p < 0.001). The presence of other components such as glycine, phenylalanine, tryptophan, and creatinine or the mixture of these components containing histidine had no significant effect on the measurement ability of the creatine-sensing membrane for 1.0 mM creatine (p > 0.05) (Figure5). In addition, no significant interference was observed with the mixture of cations such as Ca2+, Na+, K+, and Mg2+ (data not shown). For biological samples, the ratiometric fluorescence creatine biosensor showed good analytical performance as compared with other creatine biosensors (Table1). The linear detection range was wide and extended up to 10 mM creatine. The wide linear detection range of the sensing membrane allows quantification of the amounts of creatine in urine samples. The response time of the sensor was relatively fast and its stability was good, attributable to the structure of the supporting materials used for the immobilization of the two enzymes. We speculate the covalent binding between the Sensors 2017, 17, 2570 10 of 17

SensorsSensors 2017 2017, ,17 17, ,2570 x 1010 of of 17 17 isocyanate groups in the hydrogel PU and the amine groups of the enzymes, leading to the extension of thebetweenbetween shelf-life thethe isocyanateofisocyanate the biosensor groupsgroups for inin thethe several hydrogelhydrogel months PU PU and and without thethe amineamine any groups evidentgroups ofof loss thethe enzymes,inenzymes, the enzyme leadingleading activity. toto Thethe ratiometricthe extensionextension fluorescent ofof thethe shelf-lifeshelf-life sensing ofof thethe membrane biosensorbiosensor with foforr severalseveral good sensitivity monthsmonths withoutwithout and long-term anyany evidentevident stability lossloss inin maythethe be appliedenzymeenzyme to the activity.activity. high-throughput TheThe ratiometricratiometric analysis fluorescentfluorescent of creatine. sensingsensing membranemembrane withwith goodgood sensitivitysensitivity andand long-termlong-term stability stability may may be be applied applied to to the the high-throughput high-throughput analysis analysis of of creatine. creatine.

20,00020,000 afterafter 3 3 months months 1.01.0 15,00015,000 1010 mM mM 625 625 0.90.9 10,00010,000 /FI

/FI

565 565 0.10.1 mM mM 5,0005000 0.80.8 creatinecreatine

00

0.70.7 20,00020,000 initialinitial use use

15,00015,000 0.60.6 initial initial use use 1010 mM mM after after 3 3 months months 10,00010,000

Ratio of Fl.intensities, FI Ratio of Fl.intensities, FI Fluorescence intensity [a.u.] 0.50.5 Fluorescence intensity [a.u.]

0.10.1 mM mM 5,0005000 creatinecreatine 0.40.4 02468100246810 00 CreatineCreatine [mM] [mM] 550550 600 600 650 650 700 700 Emission wavelength [nm] Emission wavelength [nm] FigureFigureFigure 4. Long-term 4.4. Long-termLong-term stability stabilitystability of theofof thethe creatine-sensing creatine-sensingcreatine-sensing membrane memembranembrane in inin the thethe presence presencepresence of ofof creatine creatinecreatine 0.1 0.10.1 to toto 10 1010 mM creatinemMmM creatine uponcreatine initial upon upon use initial initial and use use after and and 3 after monthsafter 3 3 months months of use of of (left use use), ( (left andleft),), twoand and two fluorescencetwo fluorescence fluorescence emission emission emission spectra spectra spectra of the ofof the the creatine-sensing creatine-sensing membrane membrane with with respect respect to to creatine creatine concentrations concentrations at at λ λexex = = 460 460 nm. nm. creatine-sensing membrane with respect to creatine concentrations at λex = 460 nm.

0.80.8 100%100% 625 625 90%90% /FI /FI 565 565 0.60.6

70%70%

0.40.4

0.20.2 Ratio of Fl.intensities,Ratio of FI Ratio of Fl.intensities,Ratio of FI

0.00.0 CCrr GGlyly +HHisis Phhee Trprp Crnrn Mixix CCr+r+ CCr+r CCr+r+P CCr+r+T CCr+r+C CCr+r+M ComponentsComponents

FigureFigure 5.5. ResponseResponse ofof thethe creatine-sensingcreatine-sensing membranemembrane toto aqueousaqueous solutionssolutions ofof 1.01.0 mMmM creatinecreatine (Cr)(Cr) Figure 5. Response of the creatine-sensing membrane to aqueous solutions of 1.0 mM creatine (Cr) andand 0.20.2 mMmM ofof eacheach ofof thethe followingfollowing components:components: GlGly;y; glycine,glycine, His;His; histidinhistidine,e, Phe;Phe; phenylalanine,phenylalanine, and 0.2 mM of each of the following components: Gly; glycine, His; histidine, Phe; phenylalanine, Trp; Trp;Trp; tryptophan, tryptophan, Crn; Crn; creatinine, creatinine, Mix: Mix: mixture mixture of of Gly, Gly, His, His, Phe, Phe, Trp, Trp, and and Crn. Crn. tryptophan, Crn; creatinine, Mix: mixture of Gly, His, Phe, Trp, and Crn. TableTable 1. 1. Summary Summary of of the the analytical analytical performance performance of of some some creatine creatine biosensors. biosensors. Table 1. Summary of the analytical performance of some creatine biosensors. DetectionDetection Method Method LimitLimit of of ResponseResponse Long-TermLong-Term LinearLinear Detection Detection Range Range Ref.Ref. (Enzymes(Enzymes Used) Used) DetectionDetection TimeTime StabilityStability Limit of Response Long-Term DetectionPotentiometryPotentiometry Method (Enzymes (CTN (CTN + + Used)URS) URS) Linear Detection 0.1–300.1–30 mM mM Range 1010 µM µM 1.5–4 1.5–4 min min 14 14 days days [5] [5] Ref. PotentiometryPotentiometry (CTN (CTN + + URS) URS) 0.01–10.01–1 mM mM Detection 1010 µM µM 1–2 1–2Time min min 2 2 months monthsStability [6] [6] PotentiometryAmperometryAmperometry (CTN (CTN (CTN + URS) + + SOD) SOD) 0.1–30 0–20–2 mM mM 3.4 103.4 µMµ µMM 1.5–43 3 min min min 3 3 months months 14 days [7] [7] [5] PotentiometryAmperometryAmperometry (CTN (CTN (CTN + URS) + + SOD) SOD) 0.01–1 0–3.50–3.5 mM mM mM 10 n.a.n.a.µM 1 11–2 min min min 8 8 days 2days months [8] [8] [6] AmperometryAmperometryAmperometry (CTN (CTN (CTN + SOD) + + SOD) SOD) 0.2–3.8 0.2–3.8 0–2µM µM mMand and 9–120 9–120 µM µM 0.2 3.4 0.2 µM µµMM 10 103 minss 30 30 days 3days months [9] [9] [7] Amperometry (CTN + SOD) 0–3.5 mM n.a. 1 min 8 days [8] RatiometricRatiometric fluorescence fluorescence (CTN (CTN + + URS) URS) 0.1–1 0.1–1 Mm Mm and and 1–10 1–10 mM mM 0.015 0.015 mM mM 3.5 3.5 min min >3 >3 months months This This work work Amperometry (CTN + SOD) 0.2–3.8 µM and 9–120 µM 0.2 µM 10 s 30 days [9] Ratiometric fluorescenceCCTN: (CTNTN: creatinase, creatinase, + URS) SOD: SOD: 0.1–1 sarcosine sarcosine Mm and 1–10 oxidase, oxidase, mM URS: URS: 0.015 urease, urease, mM n.a.: n.a.: not 3.5not minavailable. available. >3 months This work CTN: creatinase, SOD: sarcosine oxidase, URS: urease, n.a.: not available. SensorsSensors 20172017,, 1717,, 2570 2570 1111 of of 17 17

3.3. Enzyme Immobilization for the Creatinine-Sensing Membrane 3.3. Enzyme Immobilization for the Creatinine-Sensing Membrane The optimal amounts of creatinine deiminase for immobilization into the matrix of EC and PU are testedThe optimal and shown amounts in Figure of creatinine 6. The deiminase sensing formembrane immobilization immobilized into the with matrix 2.5 ofU ECcreatinine and PU are tested and shown in Figure6. The sensing membrane immobilized with 2.5 U creatinine deiminase deiminase showed the smallest SI (SI5U = 0.109) in the presence of 0.1–1.0 mM creatinine, whereas SI showed the smallest SI (SI = 0.109) in the presence of 0.1–1.0 mM creatinine, whereas SI of the of the membrane immobiliz5Ued with 5 or 7.5 U creatinine deiminase (SI10U = 0.833 and SI15U = 0.881) weremembrane much immobilizedhigher than withthose 5 orobtained 7.5 U creatinine for the membrane deiminase (SIimmobilized10U = 0.833 andwith SI 2.515U =U 0.881)creatinine were deiminase.much higher In than addition, those obtained data collected for the membrane from Bradford immobilized protein with assay 2.5 U creatinineindicated deiminase.that the immobilizationIn addition, data efficiency collected fromof creatinine Bradford deiminas protein assaye on indicatedthe second that layer the immobilization of the creatinine-sensing efficiency of membranecreatinine deiminasewas high onat theall amounts second layer of creatinine of the creatinine-sensing deiminase used. membrane The percentage was high immobilization at all amounts wasof creatinine 47.1%, 52.9%, deiminase and 41.3% used. for The 2.5, percentage 5, and 7.5 immobilizationU creatinine deiminase. was 47.1%, Based 52.9%, on andthese 41.3% observations, for 2.5, 5, 5and U creatinine 7.5 U creatinine deiminase deiminase. was used Based to catalyze on these the observations, hydrolysis of 5 Ucreatinine creatinine at deiminasea concentration was used range to ofcatalyze 0.1 to 1.0 the mM hydrolysis to produce of creatinine ammonia. at a concentration range of 0.1 to 1.0 mM to produce ammonia.

Creatinine deiminase (CD): 2.5 U, (a) 5 U, 7.5 U

1.2 (b) 100 625 1.0 /FI 80 565

0.8 60

SI=0.881 0.6 SI=0.833 40

0.4

20 Immob. efficiency[%]

Ratio of Fl.intensities, FI SI=0.109

0.2 0 0.00.20.40.60.81.0 .5 5 .5 -2 D- -7 CD C CD Creatinine [mM] Immobilzed CD

Figure 6. (a) The response of the creatinine-sensing membrane immobilized with 2.5, 5, and 7.5 U Figurecreatinine 6. (a deiminase) The response (CD) of in the the creatinine-sensing presence of 0.1 to me 1.0mbrane mM creatinine. immobilized SI representswith 2.5, 5, slope and 7.5 value. U creatinine(b) Immobilization deiminase efficiency (CD) in the of various presence amounts of 0.1 to of 1.0 creatinine mM creatinine. deiminase. SI represents slope value. (b) Immobilization efficiency of various amounts of creatinine deiminase. 3.4. Characterization of the Creatinine-Sensing Membrane 3.4. Characterization of the Creatinine-Sensing Membrane As shown in Figure7, the linear detection range of creatinine was 0.1–1.0 mM with a high As shown in Figure 7, the linear detection range of creatinine was 0.1–1.0 mM with a high regression coefficient value of r2 = 0.96, while LOD (S/N = 3) was 0.0325 mM. Given the use of a regression coefficient value of r2 = 0.96, while LOD (S/N = 3) was 0.0325 mM. Given the use of a single enzyme, creatinine deiminase, the response of the creatinine-sensing membrane was similar to single enzyme, creatinine deiminase, the response of the creatinine-sensing membrane was similar that of O17-EC membrane to different ammonia concentrations reported in our previous study [32]. to that of O17-EC membrane to different ammonia concentrations reported in our previous study The fluorescence intensity of the creatinine-sensing membrane increased at λem = 565 nm and decreased [32]. The fluorescence intensity of the creatinine-sensing membrane increased at λem = 565 nm and at λem = 625 nm in response to an increase in creatinine concentrations from 0.1 to 10 mM. Thus, the decreased at λem = 625 nm in response to an increase in creatinine concentrations from 0.1 to 10 mM. sensing membrane with only one enzyme involved in the direct hydrolysis of the analytes to produce Thus, the sensing membrane with only one enzyme involved in the direct hydrolysis of the analytes ammonia seems to show higher sensitivity and faster response than that with two enzymes. to produce ammonia seems to show higher sensitivity and faster response than that with two The activity of creatinine deiminase immobilized in the matrix of EC and PU was evaluated enzymes. via Michaelis-Menten kinetics. Kinetic parameters were calculated from the ratio of two emission The activity of creatinine deiminase immobilized in the matrix of EC and PU was evaluated via fluorescence intensities at λem = 565 and 625 nm. Lineweaver-Burk plot revealed the maximal reaction Michaelis-Menten kinetics. Kinetic parameters were calculated from the ratio of two emission rate (Vmax) and Michaelis-Menten constant (Km) to be 1.1862 1/min and 22.05 mM, respectively. fluorescence intensities at λem = 565 and 625 nm. Lineweaver-Burk plot revealed the maximal Studies have shown Km values for free creatinine deiminase to be 1.27 [40] and 0.15 mM [41], while the reaction rate (Vmax) and Michaelis-Menten constant (Km) to be 1.1862 1/min and 22.05 mM, Km for the creatinine deiminase immobilized on the polyaniline-copper-nanocomposite was calculated respectively. Studies have shown Km values for free creatinine deiminase to be 1.27 [40] and 0.15 to be 0.163 mM [42]. The higher Km value observed in our study may be associated with the very mM [41], while the Km for the creatinine deiminase immobilized on the low affinitiy of creatinine deiminase to creatinine in EC and PU matrix as compared with EC-O17 polyaniline-copper-nanocomposite was calculated to be 0.163 mM [42]. The higher Km value membrane. The low affinity may result from the supporting matrix of EC and PU that was partially observed in our study may be associated with the very low affinitiy of creatinine deiminase to Sensors 2017, 17, x 12 of 17

Sensors 2017, 17, 2570 12 of 17 creatinine in EC and PU matrix as compared with EC-O17 membrane. The low affinity may result from the supporting matrix of EC and PU that was partially bound at the active site of the enzyme, boundowing atto the its active conformational site of the enzyme, change. owing The tomaximal its conformational reaction rate change. (Vmax The = 0.263 maximal 1/min) reaction for ratethe (Vimmobilizedmax = 0.263 urease1/min) reported for the immobilizedin our previous urease study reported [32] indicates in our previousthe strong study hydrolysis [32] indicates reaction the of strongcreatinine hydrolysis by creatinine reaction ofdeiminase, creatinine leading by creatinine to a deiminase,short response leading time to aof short the response creatinine-sensing time of the creatinine-sensingmembrane of about membrane t95 = 1–3 min of about at all creatinine t95 = 1–3 min concentrations. at all creatinine concentrations.

(a) (b) 18,000 1.4

0.1 mM 565 nm 625 nm 0.2 mM 15,000 0.4 mM 1.2 625 0.6 mM 0.8 mM /FI 1 mM 565 12,000 10 mM 2 mM 1.0 4 mM 6 mM 9000 8 mM 10 mM 0.8

6000 0.6 0.1 mM

Fluorescence intensity [a.u.] intensityFluorescence Creatinine 3000 RatioFl.intensities, of FI 0.4

0 525 550 575 600 625 650 675 700 0246810 Emission wavelength [nm] Creatinine [mM]

Figure 7. (a) Fluorescence emission spectra of the creatinine-sensing membrane in the presence of creatinine at 0.1 to 10 mM concentration monitored at an excitation wavelength of 460 nm. Figure 7. (a) Fluorescence emission spectra of the creatinine-sensing membrane in the presence of (b) Calibration curve for creatinine calculated by the ratio of two fluorescence intensities at emission creatinine at 0.1 to 10 mM concentration monitored at an excitation wavelength of 460 nm. (b) wavelengths of 565 and 625 nm. Calibration curve for creatinine calculated by the ratio of two fluorescence intensities at emission wavelengths of 565 and 625 nm. Ammonia was probably produced directly through the hydrolysis of creatinine. Therefore, the reproducibilityAmmonia ofwas the probably creatinine-sensing produced membranedirectly through was rapid, the hydrolysis as evident of from creatinine. the repeated Therefore, exposure the ofreproducibility the membrane of to 0.1the and creatinine-sensing 1.0 mM creatinine. membra Figure8ne shows was therapid, reversibility as evident of the from creatinine-sensing the repeated membraneexposure of based the membrane on the ratio to of 0.1 the and fluorescence 1.0 mM creatinine. intensities Figure at λem 8= shows 565 and the 625 reversibility nm. The sensing of the membranecreatinine-sensing exhibited membrane a very low based relative on the standard ratio of deviationthe fluorescence of 1.1% intensities and 1.9% at for λ 0.1em = and565 1.0and mM 625 creatinine,nm. The sensing respectively. membrane exhibited a very low relative standard deviation of 1.1% and 1.9% for 0.1 andThe 1.0 creatinine-sensing mM creatinine, respectively. membrane also preferred an alkaline (pH range of 8.0 to 9.0) to acidic (pH 5.0The to creatinine-sensing pH 7.0) medium (datamembrane not shown). also preferred At 0.4 mMan alkaline creatinine (pH concentration, range of 8.0 to the 9.0) ratio to ofacidic the fluorescence(pH 5.0 to pH intensities 7.0) medium of the (data creatinine-sensing not shown). membraneAt 0.4 mM atcreatinineλem = 565 concentration, and 625 nm increased the ratio with of the an increasefluorescence in the intensities pH to the of basic the creatinine-sensing range. The response membrane of the creatinine-sensing at λem = 565 and membrane625 nm increased at different with temperaturesan increase in was the also pH studied to the withbasic various range. creatinine The response concentrations. of the creatini The temperaturene-sensing membrane range of 30 toat 40different◦C failed temperatures to exert any significantwas also studied effect on with the sensitivity various creatinine of the membrane concentrations. at creatinine The concentration temperature ofrange 0.1 toof 1.0 30 mM to 40 (p =°C 0.901) failed (data to exert not shown). any significant effect on the sensitivity of the membrane at creatinineThe long-term concentration stability of 0.1 ofthe to 1.0 creatinine-sensing mM (p = 0.901) membrane(data not shown). was tested after 1.5 months of use and storage.The The long-term membrane stability maintained of the creatinine-sensing its high sensitivity membrane to variouscreatinine was tested concentrations after 1.5 months (Figure of use9), asand evident storage. from The the membrane increase inmaintained SI of the linearits high curve sensitivity at creatinine to various concentration creatinine of concentrations 0.1 to 1.0 mM (SI(Figure0.1–1.0 9), mM as) from evident 0.832 from (initial the use)increase to 0.936 in SI (after of th 1.5e linear months). curve The at creatinine increase in concentration SI after 1.5 months of 0.1 ofto use1.0 mM may (SI be0.1–1.0 related mM) tofrom the good0.832 maintenance(initial use) to of 0.936 the enzyme (after 1.5 activity months). in the The polymers increase EC in and SI after PU [ 391.5], whichmonths created of use amay versatile be related environment to the good for the maintena enzymence catalysis of the enzyme and response activity of in O17-EC the polymers membrane EC toand ammonia PU [39], produced. which created In addition, a versatile the increase environment in the backgroundfor the enzyme signal catalysis of the creatinine-sensing and response of membraneO17-EC membrane could be recognizedto ammonia after produced. testing theIn addition, sensing membrane the increase with in differentthe background pH solutions. signal Thisof the is attributedcreatinine-sensing to the slight membrane leakage of co O17uld dye be uponrecognized its exposure after to testing strong pHthe solutions, sensing therebymembrane increasing with thedifferent ratio ofpH the solutions. fluorescence This intensities is attributed at λem to= 565the andslight 625 leakage nm. The of long-term O17 dye soakingupon its of exposure the polymer to instrong the aqueous pH solutions, solution thereby may have increasing contributed the ratio to the of increasethe fluorescence in the reflection intensities of the at incidentλem = 565 light and and625 fluorescencenm. The long-term emission soaking during of creatinine the polymer measurements. in the aqueous solution may have contributed to the Sensors 2017, 17, x 13 of 17

Sensors 2017, 17, 2570 13 of 17 increase in the reflection of the incident light and fluorescence emission during creatinine Sensorsmeasurements.increase2017, 17, 2570 in the reflection of the incident light and fluorescence emission during13 creatinine of 17 measurements. 1.2 1.0 mM creatinine 1.2 625 1.0 1.0 mM creatinine /FI 625 565 1.0 /FI

0.8 565

0.8 0.6

0.6 0.4 atio of Fl.intensities, FI Fl.intensities, of atio R 0.40.1 mM creatinine

0.2FI Fl.intensities, of Ratio 00.1 10203040506070 mM creatinine 0.2 Time, [min] 0 10203040506070 Time, [min] Figure 8. Reversibility of the creatinine-sensing membrane at 0.1 and 1.0 mM creatinine. FigureFigure 8. Reversibility 8. Reversibility of the of creatinine-sensing the creatinine-sensing membrane membrane at 0.1 andat 0.1 1.0 and mM 1.0 creatinine. mM creatinine.

after 1.5 months 15,000 10 mM 1.6 12,000 after 1.5 months 15,000 1.6 10 mM 9000 1.4 12,000 625

/FI 6000 0.1 mM 9,000

565 1.4 1.2 625 creatinine 3000 /FI 6,000 0.1 mM 565 1.2 1.0 creatinine 0 3,000

initial use 15,000 1.0 0

0.8 10 mM 12,000 Initial use initial use 15,000 1.5 months 0.8 10 mM 9000 0.6 1.5 months Initial (+background) use 12,000 1.5 months 6000 Fluorescence intensity [a.u.] Ratioof Fl.intensities, FI 9,000 0.6 0.1 mM 0.4 1.5 months (+background) 3000 Fluorescence intensity [a.u.] creatinine 6,000 Ratioof Fl.intensities, FI 0.1 mM 02468100.4 0 550 600creatinine 650 700 3,000 Creatinine [mM] 0246810Emission wavelength [nm] 0 550 600 650 700 Creatinine [mM] Figure 9. Long-term stability of the creatinine-sensing membraneEmission wavelength at 0.1 to[nm] 10 mM creatinine Figure 9. Long-term stability of the creatinine-sensing membrane at 0.1 to 10 mM creatinine concentration upon initial use and after 1.5 months of use (left). Two fluorescence emission spectra of concentration upon initial use and after 1.5 months of use (left). Two fluorescence emission spectra the creatinine-sensingFigure 9. Long-term membrane stability with of respect the creatinine-sensing to creatinine concentrations membrane at λatex 0.1= 460 to nm.10 mM creatinine of theconcentration creatinine-sensing upon membrane initial use withand afterrespect 1.5 tomonths creatinine of use concentrations (left). Two fluorescence at λex = 460 nm. emission spectra

of the creatinine-sensing membrane with respect to creatinine concentrations at λex = 460 nm. TheThe limitation limitation of of the the spectrometric spectrometric method method is the is interference the interference from froma few afactors few factorsin the samples in the samples [17,18] that have similar absorption and emission wavelengths or the change in the refractive [17,18] thatThe have limitation similar of absorptionthe spectrometric and emission method iswavelengths the interference or the from change a few infactors the refractivein the samples medium. The influence of some components in urine samples on the creatinine-sensing membrane is medium.[17,18] The that influence have similarof some absorption components and in uremissionine samples wavelengths on the creati or thenine-sensing change in membrane the refractive shown in Figure 10. According to the results of one-way ANOVA, the presence of 0.2 mM histidine is shownmedium. in Figure The influence 10. According of some to components the results inof ur one-wayine samples ANOVA, on the thecreati presencenine-sensing of 0.2 membrane mM in 1.0 mM creatinine sample resulted in a significant difference in the measurement as compared to histidineis shown in 1.0 inmM Figure creatinine 10. According sample resulted to the resultin a significants of one-way difference ANOVA, in thethe measurementpresence of 0.2as mM samples without histidine (p = 0.003). However, the presence of other components such as glycine, comparedhistidine to samplesin 1.0 mM without creatinine histidine sample (p =resulted 0.003). However,in a significant the presence difference of inother the componentsmeasurement as phenylalanine, tryptophan, urea, and creatine failed to exert any effect on the measurement of such comparedas glycine, to phenylalanine, samples without tryptophan, histidine urea,(p = 0.003).and creatine However, failed the topresence exert any of othereffect componentson the 1.0 mM creatinine (p > 0.05); however, the mixture of these components containing histidine also measurementsuch as glycine,of 1.0 mM phenylalanine, creatinine (p >tryptophan, 0.05); however, urea, the and mixture creatine of thesefailed components to exert any containing effect on the hadhistidine a significant also had effect a significant on the creatinine effect on measurement the creatinine using measurement the creatinine-sensing using the creatinine-sensing membrane (t-test measurement of 1.0 mM creatinine (p > 0.05); however, the mixture of these components containing2+ withmembranep = 0.005). (t-test No with significant p = 0.005). interference No significant was interference observed with was aobserved mixture with of cations a mixture such of as cations Ca , + histidine+ also2+ had a significant effect on the creatinine measurement using the creatinine-sensing Na ,K , and2+ Mg+ (data+ not shown).2+ Aside from their influence on the pH of the solution, these such membraneas Ca , Na (t-test, K , withand pMg = 0.005). (data No not significant shown). interferenceAside from wastheir observed influence with on athe mixture pH of of the cations cations seemed to have less effect on the response of the creatinine-sensing membrane. Maintaining solution,such theseas Ca 2+cations, Na+, Kseemed+, and Mgto 2+have (data less not effeshown).ct on Asidethe response from their of influencethe creatinine-sensing on the pH of the an alkaline medium during creatinine measurements may guarantee the high sensitivity of the membrane.solution, Maintaining these cations an alkaline seemed medium to have duri lessng effe creatininect on the measurements response of may the guaranteecreatinine-sensing the creatinine-sensing membrane. high membrane.sensitivity of Maintaining the creatinine-sensing an alkaline membrane. medium duri ng creatinine measurements may guarantee the A few studies have systematically reviewed the techniques of electrochemical high sensitivity of the creatinine-sensing membrane. enzymic/non-enzymic and immuno-sensors for creatinine detection [17]. Aside from electrochemical Sensors 2017, 17, 2570 14 of 17 Sensors 2017, 17, 2570 14 of 17 A few studies have systematically reviewed the techniques of electrochemical enzymic/non-enzymic and immuno-sensors for creatinine detection [17]. Aside from transducers, some optical methods based on Jaffe’s reaction or enzyme-catalyzed reactions were also electrochemical transducers, some optical methods based on Jaffe’s reaction or enzyme-catalyzed used for the measurement of creatinine. In Table2, the analytical performance of some optical methods reactions were also used for the measurement of creatinine. In Table 2, the analytical performance for creatinine detection is compared with that of the ratiometric fluorescence creatinine biosensor of some optical methods for creatinine detection is compared with that of the ratiometric developed in this study. The ratiometric fluorescent creatinine biosensor exhibited good analytical fluorescence creatinine biosensor developed in this study. The ratiometric fluorescent creatinine performance in terms of high sensitivity, good selectivity, low-cost, and rapidity. In particular, biosensor exhibited good analytical performance in terms of high sensitivity, good selectivity, many samples can be analyzed simultaneously using a 96-well microtiter plate with immobilized low-cost, and rapidity. In particular, many samples can be analyzed simultaneously using a 96-well creatinine-sensing membranes. microtiter plate with immobilized creatinine-sensing membranes.

1.2 100%

625 1.0

/FI 80% 75% 565 0.8

0.6

0.4

0.2 RatioFl.intensirties, of FI

0.0 Crn +Gly +His Phe Trp +Ur +Cr Mix Crn Crn Crn+ Crn+ Crn Crn Crn+ Components

FigureFigure 10.10. ResponseResponse of of the the creatine-sensing creatine-sensing membrane membrane to aqueousto aqueous solutions solutions of 1.0 of mM 1.0 creatininemM creatinine (Crn) and(Crn) 0.2 and mM of0.2 each mM of of the each following of the components: followingGly: components: glycine, His: Gly: histidine, glycine, Phe: His: phenylalanine, histidine, Phe: Trp: tryptophan,phenylalanine, Ur: Trp: urea, tryptophan, Cr: creatine, Ur: and urea, Mix: Cr: mixture creatine, of Gly, and His, Mix: Phe, mixture Trp, Ur, of andGly, Cr. His, Phe, Trp, Ur, and Cr. Table 2. Summary of the analytical performance of some optical methods for creatinine detection. Table 2. Summary of the analytical performance of some optical methods for creatinine detection Linear Detection Limit of Response Detection Method (Enzyme or Assay Used) Ref. Detection Method LinearRange LimitDetection of ResponseTime Ref. Light(Enzyme diffraction or Assay (CD) Used) Detection 0.01–0.7 Range mM Detection 6 µM Time 14 min [43] Light diffraction (CD)2+ 0.01–0.7 mM 6 µM 14 min [43] Chemiluminescence (H2O2-Co ) 0.1–30 µM 0.07 µM 1 min [44] ColorimetryChemiluminescence (uric acid-Hg 2+(H-AuNPs)2O2-Co2+) 0.1–301–12 µMµM 0.07 1.6µM nM 1 min 5 min [44] [28] ColorimetryColorimetry (citrate-AuNPs) 0.1–20 mM 0.72 mM 24 min [45] 1–12 µM 1.6 nM 5 min [28] Colorimetry-enzPAD(uric acid-Hg (CNN2+-AuNPs) + CTN + SOD) 0.22–2.2 mM 0.17 mM 11 min [46] Fluorescence (QDs-CD conj.) 0.3–5 mM n.a. 3–4 min [47] Colorimetry (citrate-AuNPs) 0.1–20 mM 0.72 mM 24 min [45] Ratiometric fluorescence (CD) 0.1–1.0 mM 0.0325 mM 1–3 min This work Colorimetry-enzPAD 0.22–2.2 mM 0.17 mM 11 min [46] CD: creatinine deiminase,(CNN + CTN AuNP: + SOD) gold nanoparticles, CNN: creatininase, CTN: creatinase, SOD: sarcosine oxidase, QD: quantum dots, enz-PAD: enzyme paper-based analytical device. Fluorescence (QDs-CD conj.) 0.3–5 mM n.a. 3–4 min [47] Ratiometric fluorescence (CD) 0.1–1.0 mM 0.0325 mM 1–3 min This work 3.5. DeterminationCD: creatinine of deiminase, Creatine and AuNP: Creatinine gold nanoparticles, in Artificial Urine CNN: Solution creatininase, (AUS) CTN: creatinase, SOD: Thesarcosine creatine- oxidase, and QD: creatinine-sensing quantum dots, enz-PAD: biosensors enzyme developed paper-based for analytical the detection device. of creatine and creatinine were used to evaluate their recovery capabilities. The measurement results of creatine and3.5. Determination creatinine in theof Creatine standard and solution Creatinine prepared in Artificial in 10Urine mM Solution phosphate (AUS) buffer saline (pH 7.2) and AUSThe are showncreatine- in Figureand creati 11.nine-sensing The difference biosensors plot in Figure developed 11 also for shows the detection the recovery of creatine percentage and ofcreatinine the concentrations were used ofto creatine evaluate and their creatinine recovery in capabilities. the AUS. The The recovery measurement percentage results of two of sensingcreatine membranesand creatinine was in quite the standard high (85–115%). solution The prepared large difference in 10 mM in phosphate the two measurement buffer saline results(pH 7.2) at lowand concentrationsAUS are shown of in creatine Figure 11. and The creatinine difference (0–0.2 plot mM) in Figure may be11 associatedalso shows with the recovery the presence percentage of anions of 2− suchthe concentrations as SO4 in the of AUS. creatine and creatinine in the AUS. The recovery percentage of two sensing membranes was quite high (85–115%). The large difference in the two measurement results at low concentrations of creatine and creatinine (0–0.2 mM) may be associated with the presence of anions such as SO42− in the AUS. Sensors 2017, 17, 2570 15 of 17 Sensors 2017, 17, 2570 15 of 17

Creatine-sensing membrane Creatinine-sensing membrane 30 30

15 15

0 0

Difference [%] -15 -15

625 -30 -30

/FI 1.0 1.0 565

0.8 0.8

0.6 0.6

0.4 STD-Cr STD-Crn 0.4 AUS-Cr AUS-Crn 0.2 0.2

0.0 0.0 0246810 0246810 Ratio of Fl.intensities, FI Fl.intensities, of Ratio Creatine, [mM] Creatinine, [mM]

Figure 11. Comparison of creatine (Cr) and creatinine (Crn) concentrations in standard solution Figure 11. Comparison of creatine (Cr) and creatinine (Crn) concentrations in standard solution (STD) and artificial urine solution (AUS) using creatine- and creatinine-sensing membranes, and their difference(STD) and plots. artificial The differenceurine solution represents (AUS) 100using× (creatinecreatine- orand creatinine creatinine-s in AUSensing− standard membranes, creatine and ortheir creatinine)/(standard difference plots. The creatine difference or creatinine). represents The 100 dashed × (creatine lines indicate or creatinine the recovery in AUS percentage − standard of 85–115%creatine foror creatinecreatinine)/(standard or creatinine increatine AUS samples. or creatinine). The dashed lines indicate the recovery percentage of 85–115% for creatine or creatinine in AUS samples. 4. Conclusions 4. Conclusions Enzymatic fluorescence assay techniques are highly specific and sensitive, but their use is restricted Enzymatic fluorescence assay techniques are highly specific and sensitive, but their use is due to enzyme instability and assay complexity. In this study, ratiometric fluorescent biosensors restricted due to enzyme instability and assay complexity. In this study, ratiometric fluorescent for creatine and creatinine detection were successfully fabricated. These biosensors showed good biosensors for creatine and creatinine detection were successfully fabricated. These biosensors sensitivity in the linear concentration range of 0.1–1.0 mM for creatine and 0.1–1.0 mM for creatinine, showed good sensitivity in the linear concentration range of 0.1–1.0 mM and 1.0–10 mM for creatine while their LOD was 0.015 and 0.0325 mM for creatine and creatinine, respectively. The sensing and 0.1–1.0 mM for creatinine, while their LOD was 0.015 and 0.0325 mM for creatine and membranes displayed negligible interference from amino acids and salts, with the exception of creatinine, respectively. The sensing membranes displayed negligible interference from amino acids 0.2 mM histidine. The reproducibility of the sensing membranes for creatine and creatinine was and salts, with the exception of 0.2 mM histidine. The reproducibility of the sensing membranes for excellent, with a very low relative standard deviation (<2.67%), and their sensitivity to creatine and creatine and creatinine was excellent, with a very low relative standard deviation (<2.67%), and creatinine was retained for at least 2 months. The high recovery percentage of two sensing membranes their sensitivity to creatine and creatinine was retained for at least 2 months. The high recovery in artificial urine samples highlights their potential application for the determination of creatine percentage of two sensing membranes in artificial urine samples highlights their potential and creatinine concentrations in clinical chemistry. In addition, the successful development of the application for the determination of creatine and creatinine concentrations in clinical chemistry. In ratiometric fluorescent biosensors would be of a great significance in high-throughput screening addition, the successful development of the ratiometric fluorescent biosensors would be of a great techniques in analytical biochemistry. significance in high-throughput screening techniques in analytical biochemistry. Acknowledgments: This work was partly supported by the National Research Foundation (NRF), Republic of KoreaAcknowledgments: (Grant Number: This 2013R1A1A2058628), work was partly supported and by Koreaby the MinistryNational ofResearch Environment Foundation as “Global (NRF), Top Republic Project” of (ProjectKorea (Grant No.: 2016002210007). Number: 2013R1A1A2058628), and by Korea Ministry of Environment as “Global Top Project” Author(Project Contributions: No.: 2016002210007).H.D. and J.R. conceived and designed the experiments; H.D. performed the experiments and analyzed the data. Author Contributions: H.D. and J.R. conceived and designed the experiments; H.D. performed the Conflicts of Interest: The authors declare no conflict of interest. experiments and analyzed the data.

ReferencesConflicts of Interest: The authors declare no conflict of interest. 1. Wyes, M.; Kaddurah-Daouk, R. Creatine and creatinine metabolism. Physiol. Rev. 2000, 80, 1107–1213. References 2. Greenhaff, P.L.; Casey, A.; Short, A.H.; Harris, R.; Soderlund, K.; Hultman, E. Influence of oral creatine 1. supplementationWyes, M.; Kaddurah-Daouk, of muscle torque R. Creatine during and repeated creatinine bouts-of metabolism. maximal Physiol. voluntary Rev. exercise 2000, 80 in, 1107–1213. man. Clin. Sci. 2. 1993Greenhaff,, 84, 565–571. P.L.; Casey, [CrossRef A.; ][Short,PubMed A.H.;] Harris, R.; Soderlund, K.; Hultman, E. Influence of oral creatine 3. Williams,supplementation M.H.; Branch, of muscle J.D. torque Creatine during supplementation repeated bouts-of and maximal exercise voluntary performance: exercise An in update. man. Clin.J. Am. Sci. Coll.1993 Nutr., 84, 565–571.1998, 17 , 216. [CrossRef][PubMed] 3. Williams, M.H.; Branch, J.D. Creatine supplementation and exercise performance: An update. J. Am. Coll. Sensors 2017, 17, 2570 16 of 17

4. Mora, L.; Sentandreu, M.A.; Toldra, F. Contents of creatine, creatinine and carnosine in porcine muscles of different metabolic types. Meat Sci. 2008, 79, 709–715. [CrossRef][PubMed] 5. Koncki, R.; Walcerz, I.; Ruckruh, F.; Glab, S. Bienzymatic potentiometric electrodes for creatine and L-arginine Determination. Anal. Chim. Acta 1996, 333, 215–222. [CrossRef] 6. Karakus, E.; Pekyardımcı, S.; Kılıc, E. Potentiometric bienzymatic biosensor based on PVC membrane containing palmitic acid for determination of creatine. Proc. Biochem. 2006, 41, 1371–1377. [CrossRef] 7. Madarag, M.B.; Popescu, I.C.; Ufer, S.; Buck, R.P. Microfabricated amperometric creatine and creatinine biosensors. Anal. Chim. Acta 1996, 319, 335–345. [CrossRef] 8. Ramanavicius, A. Amperometric biosensor for the determination of creatine. Anal. Bioanal. Chem. 2007, 387, 1899–1906. [CrossRef][PubMed]

9. Kacar, C.; Erden, P.E.; Pekyardimci, S.; Kilic, E. An Fe3O4-nanoparticles-based amperometric biosensor for creatine determination. Artif. Cells Nanomed. Biotechnol. 2013, 41, 2–7. [CrossRef][PubMed] 10. Yao, T.; Kotegawa, K. Simultaneous flow-injection assay of creatinine and creatine in serum by the combined use of a 16-way switching valve, some specific enzyme reactors and a highly selective hydrogen peroxide electrode. Anal. Chim. Acta 2002, 462, 283–291. [CrossRef] 11. Dash, A.K.; Sawhney, A. A simple LC method with UV detection for the analysis of creatine and creatinine and its application to several creatine Formulations. J. Pharm. Biomed. Anal. 2002, 29, 939–945. [CrossRef] 12. Smith-Palmer, T. Separation methods applicable to urinary creatine and creatinine. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2002, 781, 93–106. [CrossRef] 13. Fernandez-Fernanadez, M.; Rodríguez-Gonzalez, P.; Alvarez, M.E.A.; Rodríguez, F.; Menendez, F.V.A.; Alonso, J.I.G. Simultaneous determination of creatinine and creatine in human serum by double-spike isotope dilution liquid chromatography–tandem mass spectrometry (LC-MS/MS) and gas chromatography−mass spectrometry (GC-MS). Anal. Chem. 2015, 87, 3735–3763. [CrossRef][PubMed] 14. Liotta, E.; Gottardo, R.; Bonizzato, L.; Pascali, J.P.; Bertaso, A.; Tagliaro, F. Rapid and direct determination of creatinine in urine using capillary zone electrophoresis. Clin. Chim. Acta 2009, 409, 52–55. [CrossRef] [PubMed] 15. Miura, C.; Funaya, N.; Matsunaga, H.; Haginaka, J. Monodisperse, molecularly imprinted polymers for creatinine by modified precipitation polymerization and their applications to creatinine assays for human serum and urine. J. Pharm. Biomed. Anal. 2013, 85, 288–294. [CrossRef][PubMed] 16. Mohabbati-Kalejahi, E.; Azimirad, V.; Bahrami, M.; Ganbari, A. A review on creatinine measurement techniques. Talanta 2012, 97, 1–8. [CrossRef][PubMed] 17. Pundir, C.S.; Yadav, S.; Kumar, A. Creatinine sensors. Trends Anal. Chem. 2013, 50, 42–52. [CrossRef] 18. Randvir, E.P.; Banks, C.E. Analytical methods for quantifying creatinine within biological media. Sens. Actuators B 2013, 183, 239–252. [CrossRef] 19. Lad, U.; Khokhar, S.; Kale, G.M. Electrochemical creatinine biosensors. Anal. Chem. 2008, 80, 7910–7917. [CrossRef][PubMed] 20. Killard, A.J.; Smyth, M.R. Creatinine biosensors: Principles and designs. Trends Biotechnol. 2000, 18, 433–437. [CrossRef] 21. Shih, Y.T.; Huang, H.J. A creatinine deiminase modified polyaniline electrode for creatinine analysis. Anal. Chim. Acta 1999, 392, 143–150. [CrossRef] 22. Yadav, S.; Kumar, A.; Pundir, C.S. Amperometric creatinine biosensor based on covalently coimmobilized enzymes onto carboxylated multiwalled carbon nanotubes/polyaniline composite film. Anal. Biochem. 2011, 419, 277–283. [CrossRef][PubMed] 23. Radomska, A.; Bodenszac, E.; Glab, S.; Koncki, R. Creatinine biosensor based on ammonium ion selective electrode and its application in flow-injection analysis. Talanta 2004, 64, 603–608. [CrossRef][PubMed] 24. Zinchenkoa, O.A.; Marchenkoa, S.V.; Sergeyevaa, T.A.; Kuklab, A.L.; Pavlyuchenkob, A.S.; Krasyukc, E.K.; Soldatkina, A.P.; El’skayaa, A.V. Application of creatinine-sensitive biosensor for hemodialysis control. Biosens. Bioelectron. 2012, 35, 466–469. [CrossRef][PubMed] 25. Soldatkin, A.P.; Montoriol, J.; Sant, W.; Martelet, C.; Jaffrezic-Renault, N. Creatinine sensitive biosensor based on ISFETs and creatinine deiminase immobilised in BSA membrane. Talanta 2002, 58, 351–357. [CrossRef] 26. Sant, W.; Pourciel-Gouzy, M.L.; Launay, J.; Conto, T.D.; Colin, R.; Martinez, A.; Temple-Boyer, P. Development of a creatinine-sensitive sensor for medical analysis. Sens. Actuators B 2004, 103, 260–264. [CrossRef] Sensors 2017, 17, 2570 17 of 17

27. Marchenko, S.V.; Soldatkin, O.O.; Kasap, B.O.; Kurc, B.A.; Soldatkin, A.P.; Dzyadevych, S.V. Creatinine deiminase adsorption onto silicalite-modified pH-FET for creation of new creatinine-sensitive biosensor. Nanoscale Res. Lett. 2016, 11, 173–180. [CrossRef][PubMed] 28. Du, J.; Zhu, B.; Leow, W.R.; Chen, S.; Sum, T.C.; Peng, X.; Chen, X. Colorimetric detection of creatinine based on plasmonic nanoparticles via synergistic coordination chemistry. Small 2015, 11, 4104–4110. [CrossRef] [PubMed] 29. Taussky, H.H. A procedure increasing the specificity of the Jaffe reaction for the determination of creatine and creatinine in urine and plasma. Clin. Chim. Acta 1956, 1, 210–224. [CrossRef] 30. Stefan, R.-L.; Bokretsion, R.G. Determination of creatine and creatinine using a diamond paste based electrode. Instum. Sci. Technol. 2003, 31, 183–188. [CrossRef] 31. Duong, H.D.; Rhee, J.I. A ratiometric fluorescence sensor for the detection of ammonia. Sens. Actuators B 2014, 190, 768–774. [CrossRef] 32. Duong, H.D.; Rhee, J.I. Development of a ratiometric fluorescent urea biosensor based on the urease immobilized onto the oxazine 170 perchlorate-ethyl cellulose membrane. Talanta 2015, 134, 333–339. [CrossRef][PubMed] 33. Lakowicz, J.R. Topics in Fluorescence Spectroscopy; Plenum Press: New York, NY, USA, 1994; Volume 4, pp. 3–6. 34. Nakata, E.; Wang, H.; Hamachi, I. Ratiometric fluorescent biosensor for real-time and label-free monitoring of fine saccharide metabolic pathways. ChemBioChem 2008, 9, 25–28. [CrossRef][PubMed] 35. Xie, H.; Zeng, F.; Wu, S. Ratiometric fluorescent biosensor for hyaluronidase with hyaluronan as both nanoparticle scaffold and substrate for enzymatic reaction. Biomacromolecules 2014, 15, 3383–3389. [CrossRef] [PubMed] 36. Liang, S.S.; Qi, L.; Zhang, R.L.; Jin, M.; Zhang, Z.Q. Ratiometric fluorescence biosensor based on CdTe quantum and carbon dots for double strand DNA detection. Sens. Actuators B 2017, 244, 585–590. [CrossRef] 37. Barker, S.L.R.; Clark, H.A.; Swallen, S.F.; Kopelman, R.; Tsang, A.W.; Swanson, J.A. Ratiometric and fluorescence-lifetime-based biosensors incorporating cytochrome c and the detection of extra- and intracellular macrophage nitric oxide. Anal. Chem. 1999, 71, 1767–1772. [CrossRef][PubMed]

38. Tiwari, A.; Dhakate, S.R. Chitosan–SiO2–multiwall carbon nanotubes nanocomposite: A novel matrix for the immobilization of creatine amidinohydrolase. Int. J. Biol. Macromol. 2009, 44, 408–412. [CrossRef][PubMed] 39. Weidgans, B.M.; Krause, C.; Klimant, I.; Wolfbeis, O.S. Fluorescent pH sensors with negligible sensitivity to ionic strength. Analyst 2004, 129, 645–650. [CrossRef][PubMed] 40. Uwajima, T.; Terada, O. Properties of crystalline creatinine deiminase from Corynebacterium lilium. Agric. Biol. Chem. 1980, 44, 1787–1792. [CrossRef] 41. Gottschalk, E.M.; Hippe, H.; Patzke, F. Creatinine deiminase (EC 3.5.4.21) from bacterium BN11: Purification, properties and applicability in a serum/urine creatinine assay. Clin. Chim. Acta 1991, 204, 223–238. [CrossRef] 42. Zhybak, M.; Beni, V.; Vagin, M.Y.; Dempsey, E.; Turner, A.P.F.; Korpan, Y. Creatinine and urea biosensors based on a novel ammonium ion-selective copper-polyaniline nano-composite. Biosens. Bioelectron. 2016, 77, 505–511. [CrossRef][PubMed] 43. Sharma, A.C.; Jana, T.; Kesavamoorthy, R.; Shi, L.; Virji, M.A.; Finegold, D.N.; Asher, S.A. A general photonic crystal sensing motif: Creatinine in bodily fluids. J. Am. Chem. Soc. 2004, 126, 2971–2977. [CrossRef] [PubMed] 2+ 44. Hanif, S.; John, P.; Gao, W.; Saqib, M.; Qi, L.; Xu, G. Chemiluminescence of creatinine/H2O2/Co and its application for selective creatinine detection. Biosens. Bioelect. 2016, 75, 347–351. [CrossRef][PubMed] 45. He, Y.; Zhang, X.; Yu, H. Gold nanoparticles-based colorimetric and visual creatinine-assay. Microchim. Acta 2015, 182, 2037–2043. [CrossRef] 46. Talalak, K.; Noiphung, J.; Songjaroen, T.; Chailapakul, O.; Laiwattanapaisal, W. A facile low-cost enzymatic paper-based assay for the determination of urine creatinine. Talanta 2015, 144, 915–921. [CrossRef][PubMed] 47. Ruedas-Rama, M.J.; Hall, E.A.H. Analytical nanosphere sensors using quantum dot-enzyme conjugates for urea and creatinine. Anal. Chem. 2010, 82, 9043–9049. [CrossRef][PubMed]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).