- Email: [email protected]
Development of luminol-N-hydroxyphthalimide chemiluminescence system for highly selective and sensitive detection of superoxide dismutase, uric acid and Co2+
Author’s Accepted Manuscript Development of luminol-N-hydroxyphthalimide chemiluminescence system for highly selective and sensitive detection of superoxide dismutase, uric acid and Co2+ Muhammad Saqib, Liming Qi, Pan Hui, Anaclet Nsabimana, Mohamed Ibrahim Halawa, Wei Zhang, Guobao Xu
PII: DOI: Reference:
www.elsevier.com/locate/bios
S0956-5663(17)30563-8 http://dx.doi.org/10.1016/j.bios.2017.08.028 BIOS9938
To appear in: Biosensors and Bioelectronic Received date: 1 July 2017 Revised date: 8 August 2017 Accepted date: 11 August 2017 Cite this article as: Muhammad Saqib, Liming Qi, Pan Hui, Anaclet Nsabimana, Mohamed Ibrahim Halawa, Wei Zhang and Guobao Xu, Development of luminol-N-hydroxyphthalimide chemiluminescence system for highly selective and sensitive detection of superoxide dismutase, uric acid and Co2+, Biosensors and Bioelectronic, http://dx.doi.org/10.1016/j.bios.2017.08.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Development of luminol-N-hydroxyphthalimide chemiluminescence system for highly selective and sensitive detection of superoxide dismutase, uric acid and Co2+ Muhammad Saqiba,b, Liming Qia,b, Pan Huia, Anaclet Nsabimanaa,b, Mohamed Ibrahim Halawaa,b, Wei Zhang,a,* Guobao Xua,* a
State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun, Jilin 130022, PR China b
University of Chinese Academy of Sciences, No. 19A Yuquanlu, Beijing 100049, China
Abstract N-hydroxyphthalimide (NHPI), a well known reagent in organic synthesis and biochemical applications, has been developed as a stable and efficient chemiluminescence coreactant for the first time. It reacts with luminol much faster than N-hydroxysuccinimide, eliminating the need of a prereaction coil used in N-hydroxysuccinimide system. Without using prereaction coil, the chemiluminescence peak intensities of luminol-NHPI system are about 102 and 26 times greater than that of luminol-N-hydroxysuccinimide system and classical luminolhydrogen peroxide system, respectively. The luminol-NHPI system achieves the highly sensitive detection of luminol (LOD=70 pM) and NHPI (LOD=910 nM). Based on their excellent quenching efficiencies, superoxide dismutase and uric acid are sensitively detected with LODs of 3 ng/mL and 10 pM, respectively. Co2+ is also detected a LOD of 30 pM by its remarkable enhancing effect. Noteworthily, our method is at least 4 orders of magnitude more sensitive than previously reported uric acid detection methods, and can detect uric acid in human urine and Co2+ in tap and lake water real samples with excellent recoveries in the range of 96.35–102.70%. This luminol-NHPI system can be an important candidate for biochemical, clinical and environmental analysis.
Keywords: Chemiluminescence, Luminol, N-Hydroxyphthalimide, Superoxide dismutase, Uric acid, Cobalt ions
*Corresponding Authors: Prof. Guobao Xu, Wei Zhang E-mails: [email protected], [email protected] Tel/Fax: (+86) 431-85262747 1
1. Introduction The luminol chemiluminescence (CL) has remained one of the most efficient methods for biological compounds and metals ion detection (Khan et al. 2014). The luminol reaction utilize some oxidants and catalysts to generate CL (Li et al. 2010). Since 1928, it has been broadly employed in the presence of hydrogen peroxide (H2O2) and can be catalyzed by a range of materials (Albrecht 1928; Wang et al. 2012). The unstable property and easy reaction of H2O2 with numerous biological compounds, transition metal cations and complexes result in the poor selectivity of most luminol CL methods. Therefore, it would be beneficial to develop stable and efficient coreactants and new CL detection methods with high selectivity. However, both new coreactants for luminol CL and new CL detection methods with high selectivity have seldom been reported recently (Su et al. 2016; Zhang et al. 2014). N-hydroxyphthalimide (NHPI) is a well known reagent in synthetic chemistry and biochemical applications. It has been used for the synthesis of O-alkyl hydroxylamines, precursor for catalytic reaction to the phthalimido-N-oxyl radical (PINO), and have capacity of mild and selective hydrogen abstraction from a range of compounds with C–H bonds. It has been frequently employed for the oxidation of C–H bonds and generation of free radicals in synthetic schemes (Bauer and Miarka 1957; Gambarotti et al. 2001). The use of NHPI has become now more frequently in industry as well, due to its numerous advantages (Melone et al. 2014). In particular, NHPI is a versatile reagent for promoting hydrogen abstraction processes as well as behaves as a relatively good hydrogen donor (Melone and Punta 2013). However, it has never been used in CL. Oxygen radicals have been considered as highly toxic and major cause of aging, cancer, cellular injuries in extrahepatic and hepatic organs, and heart disease (Harman 1981; Yang et al. 2013). Interestingly, superoxide dismutase (SOD) is widely accepted as a vital enzyme for biological defense (Yang et al. 2013). SOD converts superoxide radicals (O2•−) to H2O2 and molecular oxygen (Yang et al. 2013). To date, several methods available for SOD determination, however, most of them require complicated and exhaustive steps (Table S1). Uric acid (UA) is a primary product of purine metabolism during breakdown processes. UA has been reported as a potential biomarker for several metabolic disorders, such as hyperuricaemia, gout, Lesch-Nyhan syndrome, renal failure (Kiran et al. 2012), and also for preeclampsia (Zhao et al. 2015). Currently, enzyme based methods are widely utilized for the determination of UA in urine and serum, which are costly, need to be stored, time-consuming, 2
and only be used under special conditions (Azmi et al. 2015; Chauhan and Pundir 2011; Lu et al. 2015; Lv et al. 2002; Zhao et al. 2015). Several other methods such as capillary electrophoresis and liquid chromatography have also been documented (Zhao 2013; Zhao et al. 2008; Zhou et al. 2013). However, these methods need organic solvents, laborious sample preparation/measurement procedures, and expensive instrumentation that make the analysis unsuitable for rapid and routine UA monitoring (Villa and Poppi 2016; Zhao 2013; Zhao et al. 2008; Zhou et al. 2013). Cobalt is an essential metal nutrient for the human body and plays significant roles in various biological processes. It is the main constituent of cobalamin (also known as vitamin B12) and some other metalloproteins (Lindsay and Kerr 2011). Its deficiency can cause severe diseases like pernicious anemia (Tvermoes et al. 2013). Exposure to cobalt ions contaminated environment may also cause pathogenic infections, allergic dermatitis, asthma, rhinitis, and even lung cancer (Au-Yeung et al. 2012; Selden et al. 2007). Therefore, it is highly desirable to develop a simple, rapid and efficient analytical method for the highly selective and sensitive determination of SOD, UA, and Co2+ in biofluids for disease diagnosis, health assessment, and environmental analysis. In this work, we demonstrated NHPI as an efficient and stable coreactant of luminol CL for the first time. NHPI can react with luminol to generate intense CL without additional catalysts. This new luminol-NHPI system enables the sensitive detection of luminol and NHPI. Also, we find that SOD and UA can effectively quench the CL of luminol-NHPI system. By taking advantage of the remarkable quenching effect, highly selective and sensitive methods were developed for the detection of SOD and UA. By the addition of Co2+, the CL intensity can be enhanced significantly, which was utilized for the highly sensitive detection of Co2+. Remarkably, this CL system shows excellent selectivity for both UA and Co2+ in the presence of several biological compounds and other metal ions. 2. Experimental section 2.1 Chemicals and materials N-hydroxyphthalimide (NHPI) was purchased from Accela Chem Bio Co. Ltd. (China). Luminol and H2O2 both were purchased from Beijing Chemical Reagent Company. CoCl2·6H2O and thiourea were purchased from Sinopharm Chemical Reagent Co. Ltd. (Beijing, China). UA was purchased from Bio Basic Inc. Superoxide dismutase (SOD) was purchased from Beijing HWRK Chem. Co. Ltd (China). Sodium azide (NaN3) was purchased 3
from Tianjin Fuchen Chemical Reagents Factory (China). Luminol stock solution (10 mM) was prepared by dissolving 0.1772 g luminol in 0.10 M NaOH and then diluted with water up to 100 mL. Different working solutions with varying pH values were prepared by diluting the luminol stock solution with sodium hydroxide or carbonate buffer. Small volume of 1M NaOH was used to adjust the pH, where needed with 0.1M carbonate buffer. All chemicals were analytical grade reagents and were used as received. Double distilled water was utilized during all the experiments. 2.2 Apparatus CL emission–time curves were recorded with a flow injection analysis (FIA) based CL instrument. It consists of an IFIS-C mode intelligent flow injection sampler (ReMax Inc., Xi'an, China), a BPCL ultra-weak luminescence analyzer (Institute of Biophysics, Chinese Academic of Sciences), and a home-made flow cell. The home-made flow cell was kept in a light-tight (dark black) box of the luminescent analyzer to measure the CL signals.The photomultiplier tube (PMT) voltage was kept at 800V during all the experiments. The loop injector was equipped with an injection loop of 50 L to inject the analytes. 2.3 Procedure of luminol detection The general description of the used FIA setup is given in Scheme S1A. In a typical experiment, two flow channels were used. In flow channel A, 0.10 M carbonate buffer solution (pH 11.5), while in flow channel B, 0.5 mM aqueous solution of NHPI were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different standard solutions of luminol were prepared in water and were injected through the loop injector. 2.4 Procedure of NHPI detection The FIA system for NHPI detection is shown in Scheme S1A. 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5) in flow channel A and double distilled water in flow channel B were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different standard solutions of NHPI in water were injected through the loop injector. 2.5 Procedure of SOD detection Scheme S1A shows the FIA system for SOD detection. For SOD detection, carrier stream A used is 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5) and carrier stream B used is double distilled water. Both carrier streams were pumped into the flow cells at a flow 4
rate of 1.25 mL/min. Different SOD standard solutions in water were injected using the loop injector. 2.6 Procedure of UA detection In flow channel A, 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5), while in flow channel B, double distilled water were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different concentrations of UA were mixed with 0.5 mM NHPI first (in water), and then the mixture were injected through the loop injector immediately. Stock solution of UA (10 mM) was prepared by dissolving 0.0168 g solid UA into 10 ml double distilled water. Different standard concentrations of UA were prepared by simple dilution of stock solution with double distilled water and analyzed as described above. 2.7 Procedure of Co2+ detection In flow channel A, 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5), while in flow channel B, double distilled water were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different concentrations of Co2+ were mixed with 0.5 mM NHPI first (in water), and then immediately the mixture were injected through the loop injector. 2.8 Procedure of sample preparation For UA real samples application, three human urine samples were collected from healthy individuals. For Co2+ real samples application, tap and lake water samples were collected from the Changchun, Jilin province, China. The human urine samples were then diluted 1000 times with double distilled water to bring the concentrations of UA in calibration range, and to avoid the interference from potentially interfering components in urine samples. The percentage recoveries were evaluated with spiking UA and Co2+ at three different stated concentrations, 0.1 µM, 0.5 µM, 1.0 µM and 10 nM, 20 nM, and 30 nM, respectively. Each measurement was repeated three times. 3. Results and discussion 3.1 CL of the luminol-NHPI system It has been reported that N-hydroxysuccinimide (NHS) can react with luminol to generate CL. However, the reaction of NHS with luminol is very slow. Therefore, a prereaction coil of about 150 cm length was used before measurement (Scheme S1B) (Saqib et al. 2016). Both NHS and NHPI have been widely used in synthetic chemistry and biochemical applications. 5
According to literatures, NHPI can react with many compounds more rapidly than NHS (Melone and Punta 2013). Therefore, we test whether NHPI can react with luminol more rapidly and thus eliminate the use of a prereaction coil in this study. The reaction of NHPI with luminol is very fast. As shown in Fig. 1, intense CL can be measured immediately upon mixing without using pre-reaction coil (Scheme S1A). Without using pre-reaction coil, the CL peak intensity of luminol-NHPI is about 102 times higher than that of luminol-NHS CL system (Fig. 1) and is about 26 times greater than that of the luminol–H2O2 system (Fig. 2). Moreover, the reaction of NHPI with luminol needs lower concentrations of NHPI to achieve high CL intensity (Fig. 1-5 & S1-S9), while luminol-NHS requires higher concentrations of NHS to achieve high CL intensity (Saqib et al. 2016). It indicates that NHPI is superior to NHS for CL. Fig. S1 demonstrates that the maximum CL emission of luminol-NHPI system is noted at wavelength of 440 nm at pH 11.5, which is consistent with the typical spectrum of luminol (White et al. 1964). To further disclose the mechanism, we have investigated the dependence of CL intensities on SOD, sodium azide, and thiourea which are effective radical scavengers of O2•−, 1O2, and HO•, respectively. As shown in Fig. 3 (A), the CL intensities strongly quenched upon increasing the concentrations of SOD. In contrast, both NaN3 and thiourea showed no significant quenching effect even at higher concentration (Fig. 3 (C). These results indicated that O2•− was critical intermediate in the luminol-NHPI CL reaction. Moreover, NHPI has already been reported as an oxidant in alkaline solutions (Bauer and Miarka 1957; Gambarotti et al. 2001; Melone and Punta 2013), and it has been proposed that the bonds cleavage happens from the –NOX (in our report -NOH) position in alkaline medium in the case of N-chlorosuccinimide and N-bromosuccinimide (Safavi and Karimi 2002). Thus, NHPI may decompose in basic medium to generate O2•− radical species (eq. (1) in Scheme 1). O2•− then react efficiently with the deprotonated luminol to generate excited 3-aminophthalate and ultimately resulted in emission of light (eq. 3 in Scheme 1). 3.2 Effect of oxygen on CL According to literatures, luminol can react with dissolved oxygen to generate some CL signals under some conditions (Lin et al. 2001). Thus, the oxygen effect has been evaluated on the luminol-NHPI CL system. The dissolved oxygen was bubbled out by purging the testing solutions with nitrogen gas for 30 min. As shown in Fig. S2, the CL intensity decreases only about 17% after deaeration of the testing solutions. These results show that dissolved oxygen is not necessary for the generation of CL from luminol-NHPI system. 6
3.3 Effect of pH on CL Generally, luminol CL occurs in alkaline medium. Therefore, the pH effect on luminolNHPI CL system was also evaluated in the range of pH value 9.1–13.5. As shown in Fig. S3, the CL intensities increase sharply with increasing pH from 9.1 to 11.5, and then gradually decrease at above pH value 11.5. The sharp enhancement in CL emission with increase in pH from 9.1 to 11.5 is because of the rapid generation of O2•− radical species from NHPI and efficient luminol deprotonation at higher pH values. Above pH 11.5 range, NHPI decomposition reactions may result in less O2•− species and ultimately decrease in CL emission is observed. 3.4 Detection of luminol Luminol-NHPI system is utilized for the detection of luminol with better sensitivity. As shown in Fig. S4, the linear calibration curve is obtained in the range of 0.1 to 10 nM for luminol detection. The obtained linear regression equation is I = 16.15 + 36.75 c with a correlation coefficient (r) of 0.996, where I represent the CL intensity and c is the concentration of luminol. The detection limit was estimated to be 70 pM (S/N=3), which is much lower than that in the previous reports (Qi et al. 2016). By comparing with previously reported analytical methods for luminol detection, our method is simple and more sensitive. Since luminol is frequently used as label in bioassay, the higher sensitivity makes the present system promising for bioassay. The logarithmic linear calibration curve shows wider range from 0.1 to 3000 nM (Fig. S5). This method achieves reproducible results with relative standard deviation (RSD) of 2.80% at luminol concentration of 1 nM for nine consecutive measurements (n=9). 3.5 Detection of NHPI The proposed method was successfully applied for the detection of NHPI. The CL intensity increased linearly with the concentration of NHPI in the range of 1 to 300 µM with “r” of 0.995 (Fig. S6). The obtained linearity equation is I = 40.11+ 13.75 c. The detection limit was estimated to be 910 nM (S/N=3). The logarithmic linear calibration curve shows wider range from 1 to 1000 μM (Fig. S7). The RSD of nine repeated measurements is 3.21% at NHPI concentration of 10 µM, representing excellent reproducibility. To the best of our knowledge, it is the first report on the detection of NHPI. 3.6 Detection of SOD 7
Inspired by the above phenomenon, luminol-NHPI CL system was utilized for the sensitive detection of SOD. As shown in Fig. 3 (A, B), the CL intensities strongly decrease with increasing concentrations of SOD due to the scavenging effect of SOD on superoxide anions. The CL intensity decreased linearly with the concentration of SOD in the range of 0.01 to 0.25 µg/mL with “r” of 0.998 (Fig. 3B) according to stern-volmer equation (He et al. 2013): I0/I = 1 + KSV [SOD] where I0 and I refers to the CL intensities of the control sample (absence) and after the addition of SOD, respectively. The obtained linearity equation is I0/I = 0.99 + 4.48 c. The quenching constant (KSV) is 4.48. The detection limit (S/N=3) was estimated to be 3.0 ng/mL. Table S1 shows that our method is simple and sensitive in comparison with other methods. 3.7 Detection of UA The luminol–NHPI CL can be remarkably quenched by UA and the CL intensity is quenched up to 99.9% in the presence of 100 µM UA, which is then employed for the detection of UA. Fig. 4A shows the linear calibration curve for the detection of UA. The relationship between the CL quenching efficiencies and the concentrations of UA was evaluated with the help of Stern–Volmer equation (He et al. 2013): I0/I = 1 + KSV [UA] where I0 and I refers to the CL intensities of the control sample (absence) and after the addition of UA (presence), respectively. KSV is the Stern−Volmer quenching constant, and [UA] is the concentration of UA. The CL quenching efficiency of I0/I has a linear relationship over concentrations of UA from 50 to 8000 pM with a linear equation of I0/I = 0.99 + 1.13 c (r = 0.994). The quenching constant (KSV) is 1.13. The detection limit was estimated to be 10 pM (S/N=3). The present method demonstrated at least 4 orders of magnitude more sensitive than previously reported analytical methods for UA detection and is much simpler without adding any catalysts (Table S2). This method achieves reproducible results with RSD of 3.15% at UA concentration of 1 nM for nine consecutive measurements (n=9). 3.8 Selectivity for UA The selectivity of luminol-NHPI CL system toward UA was investigated in the presence of typical interfering compounds such as ascorbic acid, creatinine, tryptophan, cysteine, glutathione, glucose, sucrose, Cl, PO43, NO3, SO42 and other metal ions were added separately into the control sample solution. The much higher concentrations of other 8
biological compounds (100 µM) were chosen as compared to UA (30 µM) for selectivity study. As shown in Fig. 4C, the CL intensity quenched efficiently only in the presence of UA. Noteworthily, no significant quenching in CL emission was noted by adding above mentioned biological compounds and metal ions. The obtained results show that this CL system has excellent selectivity for UA among several other potential biological interfering compounds. As shown in Scheme 1, the reaction of luminol with NHPI can generate O2•− radical species. It has been reported that UA can more effectively quench the luminol CL in the presence of these oxygen radicals than other radical scavengers (Becker et al. 1989). Moreover, one previous luminol CL study showed that UA can be oxidized to allantoin after reaction with these O2•− radical species (Wang et al. 2013). Thus the decreased amount of reactive oxygen species resulted in the inhibition of CL emission. It has been reported that ascorbic acid is a versatile scavenger for O2•− radical species and can decrease CL in some papers (Tu et al. 2017) but behaves as a good enhancer of CL in the presence of O2•− radical species in other papers (Li et al. 2016). It seems that the effect of ascorbic acid on CL depends on CL system studied. In our CL system, UA can quench CL stronger as compared to ascorbic acid. The higher selectivity of this novel CL system highlights its distinction from previous luminol systems which show low selectivity in the presence of potential interfering compounds such as ascorbic acid, cysteine, glucose, creatinine, and glutathione. 3.9 Detection of Co2+ The CL behavior of luminol–NHPI system was investigated in the presence of Co2+. Upon addition of Co2+, the CL emission is significantly enhanced, which is efficiently applied for the detection of Co2+. The excellent linear response has achieved between concentration of Co2+ from 0.1 to 10 nM and CL enhancement ((I-I0)/I0) (Fig. 5 A, B). A linear equation of (I-I0)/I0 = 0.13 + 0.09 c (r = 0.998) has obtained. The detection limit was estimated to be 30 pM (S/N=3). The logarithmic linear calibration curve showed wide range from 0.1 to 300 nM (Fig. S8). The present method demonstrated more sensitivity than previously reported analytical methods for Co2+ detection. In some previous reports, it is reported that only Co2+ and luminol can also produce ultra-weak CL signals. Therefore, the effect of NHPI on detection of Co2+ was also investigated. Without the addition of NHPI, the detection limit for Co2+ is estimated to be 30.0 nM under same experimental conditions. The CL behaviour shows that NHPI can remarkably improve both the sensitivity and CL intensity. The CL signals with the addition of NHPI are about 39 times higher than in the absence of NHPI (Fig. S9).These intense CL signals (by using NHPI) are also advantageous for the 9
future implementation of portable and low cost detectors in the detection protocols. The RSD of nine repeated measurements is 3.91% at Co2+ concentration of 1 nM, representing excellent reproducibility. By comparison, the present method is simple, cost-effective, and more sensitive than other reported methods by two orders of magnitude (Table S3). 3.10 Selectivity for Co2+ To investigate the selectivity of luminol/NHPI CL system toward Co2+ ions, typical metal ions such as K+, Na+, Mg2+, Ni2+, Co2+, Mn2+, Cu2+, Zn2+, Pb2+, Hg2+, Cd2+, Fe3+, and Cr3+ ions (10 µΜ) were added separately into the control sample solution. Fig. 5C depicts that the CL signals enhanced dramatically only with the addition of Co2+. In contrast, negligible enhancement in CL intensity was observed with the addition of above mentioned metal ions at the same concentrations (10µM). It indicates that this CL method has excellent selectivity for Co2+ detection. This remarkable selectivity supports the distinction of this novel CL system from previous luminol systems which commonly show poor selectivity in the presence of other metal ions. The exquisite selectivity of this CL system for Co2+ is attributed to the following reasons. On the one hand, it has already reported that Co2+ can catalyze the decomposition of NHPI much more efficiently to generate reactive oxygen species than other metal cations (Gambarotti et al. 2001). On the other hand, it has also been reported that Co2+ is most efficient catalyst of luminol reaction in the presence of reactive oxygen species than other metal cations (Burdo and Seitz 1975). These multiple catalytic properties of Co2+ may contribute to the exquisite selectivity than other metal ions. 3.11 Determination of UA and Co2+ in real samples In order to evaluate the practical applicability of this method, the determinations of UA and Co2+ ions in human urine and tap and lake water samples have been carried out, respectively. For UA determination in real samples, three urine samples were collected from healthy individuals. These urine samples were diluted 1000 times to bring the concentration of UA in the linear range and avoid the interfering species. The UA concentrations were calculated to be 2.09 µM, 1.87 µM, and 1.65 µM in three different urine samples by standard addition method, respectively. Tap and lake water samples were collected from the Changchun, Jilin province, China. The Co2+ ions concentration in the tap water sample solution was calculated to be 0.091 ± 0.02 nM by standard addition method. The Co2 + ions concentration is not detected in lake water samples. To check the recoveries, urine and water samples were spiked with standard UA and Co2+ solutions of 0.10 µM, 0.50 µM, 1.00 µM, 10
and 10, 20, 30 nM, respectively. Table S4 shows that the recoveries of all samples are achieved in the range of 96.35-102.70%. The obtained results show that luminol-NHPI CL system is appropriate for real sample applications. 4. Conclusions We have developed NHPI as an efficient coreactant for luminol CL. Co2+ can dramatically enhance CL signals of the luminol-NHPI system, while SOD and UA can effectively decrease CL signals of the luminol-NHPI system. This CL system achieves more sensitive detection of UA, SOD, Co2+, luminol, and NHPI. Furthermore, this method displays excellent selectivity for the detection of UA and Co2+ and can detect UA in urine samples and Co2+ in tap and lake water samples, which makes the luminol-NHPI CL system attractive. The popularity and versatile applications of NHPI in synthetic chemistry makes NHPI promising coreactant for other CL systems. Because of the favourable features of NHPI and the broad applications of CL, NHPI-based CL will find lots of applications, such as biochemical applications, clinical analysis, and environmental detection. Acknowledgments This project was kindly supported by the National Key Research and Development Program of China [No. 2016YFA0201300], the National Natural Science Foundation of China [Nos. 21675148 and 21475123], and the Chinese Academy of Sciences (CAS)–the Academy of Sciences for the Developing World (TWAS) President’s Fellowship Programme [2013-053]. References Albrecht, H.O., 1928. Z. Phys. Chem. 136, 321-330. Au-Yeung, H.Y., New, E.J., Chang, C.J., 2012. Chem. Commun. 48(43), 5268-5270. Azmi, N.E., Ramli, N.I., Abdullah, J., Abdul Hamid, M.A., Sidek, H., Abd Rahman, S., Ariffin, N., Yusof, N.A., 2015. Biosens. Bioelectron. 67, 129-133. Bauer, L., Miarka, S.V., 1957. J. Am. Chem. Soc. 79(8), 1983-1985. Becker, B.F., Reinholz, N., Özçelik, T., Leipert, B., Gerlach, E., 1989. Pflügers Archiv. 415(2), 127-135. Burdo, T.G., Seitz, W.R., 1975. Anal. Chem. 47(9), 1639-1643. Chauhan, N., Pundir, C.S., 2011. Anal. Biochem. 413(2), 97-103. Gambarotti, C., Punta, C., Recupero, F., Zlotorzynska, M., Sammis, G., 2001. Encyclopedia of Reagents for Organic Synthesis. John Wiley & Sons, Ltd. 11
Girotti, S., Fini, F., Ferri, E., Budini, R., Piazzi, S., Cantagalli, D., 2000. Talanta 51(4), 685692. Gobi, K.V., Mizutani, F., 2001. Anal. Sci. 17(1), 11-15. Harman, D., 1981. Proc. Natl. Acad. Sci. U. S. A. 78(11), 7124-7128. He, D., Zhang, Z., Huang, Y., Hu, Y., Zhou, H., Chen, D., 2005. Luminescence 20(4-5), 271275. He, Y., Huang, G.M., Cui, H., 2013. ACS Appl. Mater. Interfaces 5(21), 11336-11340. Jin, D., Seo, M.-H., Huy, B.T., Pham, Q.-T., Conte, M.L., Thangadurai, D., Lee, Y.-I., 2016. Biosens. Bioelectron. 77, 359-365. Khan, P., Idrees, D., Moxley, M.A., Corbett, J.A., Ahmad, F., von Figura, G., Sly, W.S., Waheed, A., Hassan, M.I., 2014. Appl. Biochem. Biotechnol. 173(2), 333-355. Kiran, R., Scorsone, E., Mailley, P., Bergonzo, P., 2012. Anal. Chem. 84(23), 10207-10213. Li, N., Wang, W., Tian, D., Cui, H., 2010. Chem. Commun. 46(9), 1520-1522. Li, Q., Li, F., Shen, W., Liu, X., Cui, H., 2016. J. Mater. Chem. C 4(16), 3477-3484. Li, Z., Feng, M., Lu, J., 1998. Microchem. J. 59(2), 278-283. Lin, J.M., Shan, X., Hanaoka, S., Yamada, M., 2001. Anal. Chem. 73(21), 5043-5051. Lin, Y.Y., Dai, H., Xu, G.F., Yang, T., Yang, C.P., Tong, Y.J., Yang, Y.S., Chen, G.N., 2013. Microchim. Acta 180(7-8), 563-572. Lin, Z.Y., Huang, L.Z., Liu, Y., Lin, J.M., Chi, Y.W., Chen, G.N., 2008. Electrochem. Commun. 10(11), 1708-1711. Lindsay, D., Kerr, W., 2011. Nat. Chem. 3(6), 494. Lu, J., Xiong, Y., Liao, C., Ye, F., 2015. Anal. Methods 7(23), 9894-9899. Lv, Y., Zhang, Z., Chen, F., 2002. Analyst 127(9), 1176-1179. Melone, L., Prosperini, S., Ercole, G., Pastori, N., Punta, C., 2014. J. Chem. Technol. Biotechnol. 89(9), 1370-1378. Melone, L., Punta, C., 2013. Beilstein J. Org. Chem. 9, 1296-1310. Noroozifar, M., Khorasani-Motlagh, M., Jahromi, F.Z., Rostami, S., 2013. Sens. Actuator BChem. 188, 65-72. Qi, L., Xia, Y., Qi, W., Gao, W., Wu, F., Xu, G., 2016. Anal. Chem. 88(2), 1123-1127. Reddy, K.K., Gobi, K.V., 2012. Electrochim. Acta 78, 109-114. Safavi, A., Karimi, M.A., 2002. Talanta 57(3), 491-500. Saqib, M., Li, S., Gao, W., Majeed, S., Qi, L., Liu, Z., Xu, G., 2016. Anal. Bioanal. Chem. 408(30), 8851-8857. 12
Selden, A.I., Norberg, C., Karlson-Stiber, C., Hellstrom-Lindberg, E., 2007.
Environ.
Toxicol. Pharmacol. 23(1), 129-131. Su, Y., Deng, D., Zhang, L., Song, H., Lv, Y., 2016. S TrAC-Trend. Anal. Chem. 82, 394411. Tu, Y.-J., Njus, D., Schlegel, H.B., 2017. Org. Biomol. Chem. 15(20), 4417-4431. Tvermoes, B.E., Finley, B.L., Unice, K.M., Otani, J.M., Paustenbach, D.J., Galbraith, D.A., 2013. Food Chem. Toxicol. 53, 432-439. Villa, J.E.L., Poppi, R.J., 2016. Analyst 141(6), 1966-1972. Wang, D.M., Zhang, Y., Zheng, L.L., Yang, X.X., Wang, Y., Huang, C.Z., 2012. J. Phys. Chem. C 116(40), 21622-21628. Wang, J., Tan, X., Song, Z., 2013. J. Anal. Methods Chem. 2013, 7. Wei, W., Wang, H., Jiang, C., 2008. Spectrochim. Acta Mol. Biomol. Spectrosc. 70(2), 389393. White, E.H., Zafiriou, O., Kagi, H.H., Hill, J.H.M., 1964. J. Am. Chem. Soc. 86(5), 940-941. Yang, C., Zhang, Z., 2010. Talanta 81(1-2), 477-481. Yang, X., Dou, Y., Zhu, S., 2013. Analyst 138(11), 3246-3252. Zhang, R., Hu, Y., Li, G., 2014. Anal. Chem. 86(12), 6080-6087. Zhao, J., 2013. Anal. Methods 5(23), 6781. Zhao, L., Blackburn, J., Brosseau, C.L., 2015. Anal. Chem. 87(1), 441-447. Zhao, S., Wang, J., Ye, F., Liu, Y.-M., 2008. Anal. Biochem. 378(2), 127-131. Zhou, S., Zuo, R., Zhu, Z., Wu, D., Vasa, K., Deng, Y., Zuo, Y., 2013. Anal. Methods 5(5), 1307-1311.
13
Figure and Scheme captions Fig. 1. Comparison of CL profiles of luminol-NHS and luminol-NHPI system under same experimental conditions. Inset: the enlarged version of CL profiles of luminol-NHS system. The comparison was performed with 10 μM luminol at pH 11.5 in the presence of 0.5 mM NHS or 0.5 mM NHPI. Fig. 2. A comparison of CL behavior for the luminol–H2O2 (red color) and luminol–NHPI systems (violet color). Inset: enlarged version of chemiluminescent kinetic profile of luminol–H2O2 system. The comparison was conducted in the presence of 10 µM luminol, 2 mM NHPI, and 2 mM H 2O2 at pH 11.5. Fig. 3. (A) CL kinetic curves in the absence and the presence of different concentrations of SOD: 0.01–2.0 μg/mL. (B) Shows the relationship between CL quenching and the concentration of SOD from 0.01 to 2.0 μg/mL. Inset shows linear calibration curve for SOD in the range of 0.01–0.25 μg/mL. (C) CL kinetic curves in the absence and the presence of thiourea and NaN3: 5 mM each. The CL measurement was carried out in the presence of 0.3 mM NHPI and 10 µM luminol at pH 11.5. Scheme 1. Proposed mechanism of luminol-NHPI CL detection system. Fig. 4. (A) CL emission-time curves in different concentrations of UA, (B) Linear calibration curve of UA in the range of 50-8000 pM. Inset: the enlarged version of UA concentrations from 0 to 500 pM and (C) selectivity for the detection of UA. I symbolize the CL intensity at a specific UA concentration and other interfering compounds; I0 is the CL intensity of the control sample. I0/I represent the CL quenching efficiency after the addition of UA and different biological compounds. The concentration of UA was 30 µM and for all other compounds were 100 µM for selectivity study. The measurements were performed in the presence of 10 µM luminol, and 0.5 mM NHPI at pH11.5. Fig. 5. (A) CL emission-time curves in different concentrations of Co2+, (B) shows the relationship between CL enhancement and the concentration of Co2+ from 0.1 to 300 nM. Inset shows the linear calibration curve for the detection of Co2+ in the range of 0.1-10 nM. (C) Selectivity for the detection of Co2+. I symbolizes the CL intensity of the system after Co2+ and other metal cations addition, I0 indicates the CL intensity of the control sample and (I-I0)/I0 is the expression of the CL enhancement efficiency after the addition of Co2+ and other metal ions. The concentrations of all the metal ions were 10 µM for selectivity study. The measurements were performed in the presence of 10 µM luminol, and 0.5 mM NHPI at pH11.5.
14
Highlights Chemiluminescence of luminol-NHPI has been reported for the first time. Superoxide dismutase and uric acid can remarkably quench the chemiluminescence. Co2+ can dramatically enhance the chemiluminescence of luminol-NHPI. Highly sensitive detection of superoxide dismutase, uric acid and Co2+ is achieved. It can successfully detect uric acid in human urine and Co2+ in water real samples after dilution.
15
16
Scheme 1
17
PII: DOI: Reference:
www.elsevier.com/locate/bios
S0956-5663(17)30563-8 http://dx.doi.org/10.1016/j.bios.2017.08.028 BIOS9938
To appear in: Biosensors and Bioelectronic Received date: 1 July 2017 Revised date: 8 August 2017 Accepted date: 11 August 2017 Cite this article as: Muhammad Saqib, Liming Qi, Pan Hui, Anaclet Nsabimana, Mohamed Ibrahim Halawa, Wei Zhang and Guobao Xu, Development of luminol-N-hydroxyphthalimide chemiluminescence system for highly selective and sensitive detection of superoxide dismutase, uric acid and Co2+, Biosensors and Bioelectronic, http://dx.doi.org/10.1016/j.bios.2017.08.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Development of luminol-N-hydroxyphthalimide chemiluminescence system for highly selective and sensitive detection of superoxide dismutase, uric acid and Co2+ Muhammad Saqiba,b, Liming Qia,b, Pan Huia, Anaclet Nsabimanaa,b, Mohamed Ibrahim Halawaa,b, Wei Zhang,a,* Guobao Xua,* a
State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun, Jilin 130022, PR China b
University of Chinese Academy of Sciences, No. 19A Yuquanlu, Beijing 100049, China
Abstract N-hydroxyphthalimide (NHPI), a well known reagent in organic synthesis and biochemical applications, has been developed as a stable and efficient chemiluminescence coreactant for the first time. It reacts with luminol much faster than N-hydroxysuccinimide, eliminating the need of a prereaction coil used in N-hydroxysuccinimide system. Without using prereaction coil, the chemiluminescence peak intensities of luminol-NHPI system are about 102 and 26 times greater than that of luminol-N-hydroxysuccinimide system and classical luminolhydrogen peroxide system, respectively. The luminol-NHPI system achieves the highly sensitive detection of luminol (LOD=70 pM) and NHPI (LOD=910 nM). Based on their excellent quenching efficiencies, superoxide dismutase and uric acid are sensitively detected with LODs of 3 ng/mL and 10 pM, respectively. Co2+ is also detected a LOD of 30 pM by its remarkable enhancing effect. Noteworthily, our method is at least 4 orders of magnitude more sensitive than previously reported uric acid detection methods, and can detect uric acid in human urine and Co2+ in tap and lake water real samples with excellent recoveries in the range of 96.35–102.70%. This luminol-NHPI system can be an important candidate for biochemical, clinical and environmental analysis.
Keywords: Chemiluminescence, Luminol, N-Hydroxyphthalimide, Superoxide dismutase, Uric acid, Cobalt ions
*Corresponding Authors: Prof. Guobao Xu, Wei Zhang E-mails: [email protected], [email protected] Tel/Fax: (+86) 431-85262747 1
1. Introduction The luminol chemiluminescence (CL) has remained one of the most efficient methods for biological compounds and metals ion detection (Khan et al. 2014). The luminol reaction utilize some oxidants and catalysts to generate CL (Li et al. 2010). Since 1928, it has been broadly employed in the presence of hydrogen peroxide (H2O2) and can be catalyzed by a range of materials (Albrecht 1928; Wang et al. 2012). The unstable property and easy reaction of H2O2 with numerous biological compounds, transition metal cations and complexes result in the poor selectivity of most luminol CL methods. Therefore, it would be beneficial to develop stable and efficient coreactants and new CL detection methods with high selectivity. However, both new coreactants for luminol CL and new CL detection methods with high selectivity have seldom been reported recently (Su et al. 2016; Zhang et al. 2014). N-hydroxyphthalimide (NHPI) is a well known reagent in synthetic chemistry and biochemical applications. It has been used for the synthesis of O-alkyl hydroxylamines, precursor for catalytic reaction to the phthalimido-N-oxyl radical (PINO), and have capacity of mild and selective hydrogen abstraction from a range of compounds with C–H bonds. It has been frequently employed for the oxidation of C–H bonds and generation of free radicals in synthetic schemes (Bauer and Miarka 1957; Gambarotti et al. 2001). The use of NHPI has become now more frequently in industry as well, due to its numerous advantages (Melone et al. 2014). In particular, NHPI is a versatile reagent for promoting hydrogen abstraction processes as well as behaves as a relatively good hydrogen donor (Melone and Punta 2013). However, it has never been used in CL. Oxygen radicals have been considered as highly toxic and major cause of aging, cancer, cellular injuries in extrahepatic and hepatic organs, and heart disease (Harman 1981; Yang et al. 2013). Interestingly, superoxide dismutase (SOD) is widely accepted as a vital enzyme for biological defense (Yang et al. 2013). SOD converts superoxide radicals (O2•−) to H2O2 and molecular oxygen (Yang et al. 2013). To date, several methods available for SOD determination, however, most of them require complicated and exhaustive steps (Table S1). Uric acid (UA) is a primary product of purine metabolism during breakdown processes. UA has been reported as a potential biomarker for several metabolic disorders, such as hyperuricaemia, gout, Lesch-Nyhan syndrome, renal failure (Kiran et al. 2012), and also for preeclampsia (Zhao et al. 2015). Currently, enzyme based methods are widely utilized for the determination of UA in urine and serum, which are costly, need to be stored, time-consuming, 2
and only be used under special conditions (Azmi et al. 2015; Chauhan and Pundir 2011; Lu et al. 2015; Lv et al. 2002; Zhao et al. 2015). Several other methods such as capillary electrophoresis and liquid chromatography have also been documented (Zhao 2013; Zhao et al. 2008; Zhou et al. 2013). However, these methods need organic solvents, laborious sample preparation/measurement procedures, and expensive instrumentation that make the analysis unsuitable for rapid and routine UA monitoring (Villa and Poppi 2016; Zhao 2013; Zhao et al. 2008; Zhou et al. 2013). Cobalt is an essential metal nutrient for the human body and plays significant roles in various biological processes. It is the main constituent of cobalamin (also known as vitamin B12) and some other metalloproteins (Lindsay and Kerr 2011). Its deficiency can cause severe diseases like pernicious anemia (Tvermoes et al. 2013). Exposure to cobalt ions contaminated environment may also cause pathogenic infections, allergic dermatitis, asthma, rhinitis, and even lung cancer (Au-Yeung et al. 2012; Selden et al. 2007). Therefore, it is highly desirable to develop a simple, rapid and efficient analytical method for the highly selective and sensitive determination of SOD, UA, and Co2+ in biofluids for disease diagnosis, health assessment, and environmental analysis. In this work, we demonstrated NHPI as an efficient and stable coreactant of luminol CL for the first time. NHPI can react with luminol to generate intense CL without additional catalysts. This new luminol-NHPI system enables the sensitive detection of luminol and NHPI. Also, we find that SOD and UA can effectively quench the CL of luminol-NHPI system. By taking advantage of the remarkable quenching effect, highly selective and sensitive methods were developed for the detection of SOD and UA. By the addition of Co2+, the CL intensity can be enhanced significantly, which was utilized for the highly sensitive detection of Co2+. Remarkably, this CL system shows excellent selectivity for both UA and Co2+ in the presence of several biological compounds and other metal ions. 2. Experimental section 2.1 Chemicals and materials N-hydroxyphthalimide (NHPI) was purchased from Accela Chem Bio Co. Ltd. (China). Luminol and H2O2 both were purchased from Beijing Chemical Reagent Company. CoCl2·6H2O and thiourea were purchased from Sinopharm Chemical Reagent Co. Ltd. (Beijing, China). UA was purchased from Bio Basic Inc. Superoxide dismutase (SOD) was purchased from Beijing HWRK Chem. Co. Ltd (China). Sodium azide (NaN3) was purchased 3
from Tianjin Fuchen Chemical Reagents Factory (China). Luminol stock solution (10 mM) was prepared by dissolving 0.1772 g luminol in 0.10 M NaOH and then diluted with water up to 100 mL. Different working solutions with varying pH values were prepared by diluting the luminol stock solution with sodium hydroxide or carbonate buffer. Small volume of 1M NaOH was used to adjust the pH, where needed with 0.1M carbonate buffer. All chemicals were analytical grade reagents and were used as received. Double distilled water was utilized during all the experiments. 2.2 Apparatus CL emission–time curves were recorded with a flow injection analysis (FIA) based CL instrument. It consists of an IFIS-C mode intelligent flow injection sampler (ReMax Inc., Xi'an, China), a BPCL ultra-weak luminescence analyzer (Institute of Biophysics, Chinese Academic of Sciences), and a home-made flow cell. The home-made flow cell was kept in a light-tight (dark black) box of the luminescent analyzer to measure the CL signals.The photomultiplier tube (PMT) voltage was kept at 800V during all the experiments. The loop injector was equipped with an injection loop of 50 L to inject the analytes. 2.3 Procedure of luminol detection The general description of the used FIA setup is given in Scheme S1A. In a typical experiment, two flow channels were used. In flow channel A, 0.10 M carbonate buffer solution (pH 11.5), while in flow channel B, 0.5 mM aqueous solution of NHPI were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different standard solutions of luminol were prepared in water and were injected through the loop injector. 2.4 Procedure of NHPI detection The FIA system for NHPI detection is shown in Scheme S1A. 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5) in flow channel A and double distilled water in flow channel B were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different standard solutions of NHPI in water were injected through the loop injector. 2.5 Procedure of SOD detection Scheme S1A shows the FIA system for SOD detection. For SOD detection, carrier stream A used is 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5) and carrier stream B used is double distilled water. Both carrier streams were pumped into the flow cells at a flow 4
rate of 1.25 mL/min. Different SOD standard solutions in water were injected using the loop injector. 2.6 Procedure of UA detection In flow channel A, 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5), while in flow channel B, double distilled water were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different concentrations of UA were mixed with 0.5 mM NHPI first (in water), and then the mixture were injected through the loop injector immediately. Stock solution of UA (10 mM) was prepared by dissolving 0.0168 g solid UA into 10 ml double distilled water. Different standard concentrations of UA were prepared by simple dilution of stock solution with double distilled water and analyzed as described above. 2.7 Procedure of Co2+ detection In flow channel A, 10 M luminol in 0.10 M carbonate buffer solution (pH 11.5), while in flow channel B, double distilled water were pumped into the flow cells at a flow rate of 1.25 mL/min, respectively. Different concentrations of Co2+ were mixed with 0.5 mM NHPI first (in water), and then immediately the mixture were injected through the loop injector. 2.8 Procedure of sample preparation For UA real samples application, three human urine samples were collected from healthy individuals. For Co2+ real samples application, tap and lake water samples were collected from the Changchun, Jilin province, China. The human urine samples were then diluted 1000 times with double distilled water to bring the concentrations of UA in calibration range, and to avoid the interference from potentially interfering components in urine samples. The percentage recoveries were evaluated with spiking UA and Co2+ at three different stated concentrations, 0.1 µM, 0.5 µM, 1.0 µM and 10 nM, 20 nM, and 30 nM, respectively. Each measurement was repeated three times. 3. Results and discussion 3.1 CL of the luminol-NHPI system It has been reported that N-hydroxysuccinimide (NHS) can react with luminol to generate CL. However, the reaction of NHS with luminol is very slow. Therefore, a prereaction coil of about 150 cm length was used before measurement (Scheme S1B) (Saqib et al. 2016). Both NHS and NHPI have been widely used in synthetic chemistry and biochemical applications. 5
According to literatures, NHPI can react with many compounds more rapidly than NHS (Melone and Punta 2013). Therefore, we test whether NHPI can react with luminol more rapidly and thus eliminate the use of a prereaction coil in this study. The reaction of NHPI with luminol is very fast. As shown in Fig. 1, intense CL can be measured immediately upon mixing without using pre-reaction coil (Scheme S1A). Without using pre-reaction coil, the CL peak intensity of luminol-NHPI is about 102 times higher than that of luminol-NHS CL system (Fig. 1) and is about 26 times greater than that of the luminol–H2O2 system (Fig. 2). Moreover, the reaction of NHPI with luminol needs lower concentrations of NHPI to achieve high CL intensity (Fig. 1-5 & S1-S9), while luminol-NHS requires higher concentrations of NHS to achieve high CL intensity (Saqib et al. 2016). It indicates that NHPI is superior to NHS for CL. Fig. S1 demonstrates that the maximum CL emission of luminol-NHPI system is noted at wavelength of 440 nm at pH 11.5, which is consistent with the typical spectrum of luminol (White et al. 1964). To further disclose the mechanism, we have investigated the dependence of CL intensities on SOD, sodium azide, and thiourea which are effective radical scavengers of O2•−, 1O2, and HO•, respectively. As shown in Fig. 3 (A), the CL intensities strongly quenched upon increasing the concentrations of SOD. In contrast, both NaN3 and thiourea showed no significant quenching effect even at higher concentration (Fig. 3 (C). These results indicated that O2•− was critical intermediate in the luminol-NHPI CL reaction. Moreover, NHPI has already been reported as an oxidant in alkaline solutions (Bauer and Miarka 1957; Gambarotti et al. 2001; Melone and Punta 2013), and it has been proposed that the bonds cleavage happens from the –NOX (in our report -NOH) position in alkaline medium in the case of N-chlorosuccinimide and N-bromosuccinimide (Safavi and Karimi 2002). Thus, NHPI may decompose in basic medium to generate O2•− radical species (eq. (1) in Scheme 1). O2•− then react efficiently with the deprotonated luminol to generate excited 3-aminophthalate and ultimately resulted in emission of light (eq. 3 in Scheme 1). 3.2 Effect of oxygen on CL According to literatures, luminol can react with dissolved oxygen to generate some CL signals under some conditions (Lin et al. 2001). Thus, the oxygen effect has been evaluated on the luminol-NHPI CL system. The dissolved oxygen was bubbled out by purging the testing solutions with nitrogen gas for 30 min. As shown in Fig. S2, the CL intensity decreases only about 17% after deaeration of the testing solutions. These results show that dissolved oxygen is not necessary for the generation of CL from luminol-NHPI system. 6
3.3 Effect of pH on CL Generally, luminol CL occurs in alkaline medium. Therefore, the pH effect on luminolNHPI CL system was also evaluated in the range of pH value 9.1–13.5. As shown in Fig. S3, the CL intensities increase sharply with increasing pH from 9.1 to 11.5, and then gradually decrease at above pH value 11.5. The sharp enhancement in CL emission with increase in pH from 9.1 to 11.5 is because of the rapid generation of O2•− radical species from NHPI and efficient luminol deprotonation at higher pH values. Above pH 11.5 range, NHPI decomposition reactions may result in less O2•− species and ultimately decrease in CL emission is observed. 3.4 Detection of luminol Luminol-NHPI system is utilized for the detection of luminol with better sensitivity. As shown in Fig. S4, the linear calibration curve is obtained in the range of 0.1 to 10 nM for luminol detection. The obtained linear regression equation is I = 16.15 + 36.75 c with a correlation coefficient (r) of 0.996, where I represent the CL intensity and c is the concentration of luminol. The detection limit was estimated to be 70 pM (S/N=3), which is much lower than that in the previous reports (Qi et al. 2016). By comparing with previously reported analytical methods for luminol detection, our method is simple and more sensitive. Since luminol is frequently used as label in bioassay, the higher sensitivity makes the present system promising for bioassay. The logarithmic linear calibration curve shows wider range from 0.1 to 3000 nM (Fig. S5). This method achieves reproducible results with relative standard deviation (RSD) of 2.80% at luminol concentration of 1 nM for nine consecutive measurements (n=9). 3.5 Detection of NHPI The proposed method was successfully applied for the detection of NHPI. The CL intensity increased linearly with the concentration of NHPI in the range of 1 to 300 µM with “r” of 0.995 (Fig. S6). The obtained linearity equation is I = 40.11+ 13.75 c. The detection limit was estimated to be 910 nM (S/N=3). The logarithmic linear calibration curve shows wider range from 1 to 1000 μM (Fig. S7). The RSD of nine repeated measurements is 3.21% at NHPI concentration of 10 µM, representing excellent reproducibility. To the best of our knowledge, it is the first report on the detection of NHPI. 3.6 Detection of SOD 7
Inspired by the above phenomenon, luminol-NHPI CL system was utilized for the sensitive detection of SOD. As shown in Fig. 3 (A, B), the CL intensities strongly decrease with increasing concentrations of SOD due to the scavenging effect of SOD on superoxide anions. The CL intensity decreased linearly with the concentration of SOD in the range of 0.01 to 0.25 µg/mL with “r” of 0.998 (Fig. 3B) according to stern-volmer equation (He et al. 2013): I0/I = 1 + KSV [SOD] where I0 and I refers to the CL intensities of the control sample (absence) and after the addition of SOD, respectively. The obtained linearity equation is I0/I = 0.99 + 4.48 c. The quenching constant (KSV) is 4.48. The detection limit (S/N=3) was estimated to be 3.0 ng/mL. Table S1 shows that our method is simple and sensitive in comparison with other methods. 3.7 Detection of UA The luminol–NHPI CL can be remarkably quenched by UA and the CL intensity is quenched up to 99.9% in the presence of 100 µM UA, which is then employed for the detection of UA. Fig. 4A shows the linear calibration curve for the detection of UA. The relationship between the CL quenching efficiencies and the concentrations of UA was evaluated with the help of Stern–Volmer equation (He et al. 2013): I0/I = 1 + KSV [UA] where I0 and I refers to the CL intensities of the control sample (absence) and after the addition of UA (presence), respectively. KSV is the Stern−Volmer quenching constant, and [UA] is the concentration of UA. The CL quenching efficiency of I0/I has a linear relationship over concentrations of UA from 50 to 8000 pM with a linear equation of I0/I = 0.99 + 1.13 c (r = 0.994). The quenching constant (KSV) is 1.13. The detection limit was estimated to be 10 pM (S/N=3). The present method demonstrated at least 4 orders of magnitude more sensitive than previously reported analytical methods for UA detection and is much simpler without adding any catalysts (Table S2). This method achieves reproducible results with RSD of 3.15% at UA concentration of 1 nM for nine consecutive measurements (n=9). 3.8 Selectivity for UA The selectivity of luminol-NHPI CL system toward UA was investigated in the presence of typical interfering compounds such as ascorbic acid, creatinine, tryptophan, cysteine, glutathione, glucose, sucrose, Cl, PO43, NO3, SO42 and other metal ions were added separately into the control sample solution. The much higher concentrations of other 8
biological compounds (100 µM) were chosen as compared to UA (30 µM) for selectivity study. As shown in Fig. 4C, the CL intensity quenched efficiently only in the presence of UA. Noteworthily, no significant quenching in CL emission was noted by adding above mentioned biological compounds and metal ions. The obtained results show that this CL system has excellent selectivity for UA among several other potential biological interfering compounds. As shown in Scheme 1, the reaction of luminol with NHPI can generate O2•− radical species. It has been reported that UA can more effectively quench the luminol CL in the presence of these oxygen radicals than other radical scavengers (Becker et al. 1989). Moreover, one previous luminol CL study showed that UA can be oxidized to allantoin after reaction with these O2•− radical species (Wang et al. 2013). Thus the decreased amount of reactive oxygen species resulted in the inhibition of CL emission. It has been reported that ascorbic acid is a versatile scavenger for O2•− radical species and can decrease CL in some papers (Tu et al. 2017) but behaves as a good enhancer of CL in the presence of O2•− radical species in other papers (Li et al. 2016). It seems that the effect of ascorbic acid on CL depends on CL system studied. In our CL system, UA can quench CL stronger as compared to ascorbic acid. The higher selectivity of this novel CL system highlights its distinction from previous luminol systems which show low selectivity in the presence of potential interfering compounds such as ascorbic acid, cysteine, glucose, creatinine, and glutathione. 3.9 Detection of Co2+ The CL behavior of luminol–NHPI system was investigated in the presence of Co2+. Upon addition of Co2+, the CL emission is significantly enhanced, which is efficiently applied for the detection of Co2+. The excellent linear response has achieved between concentration of Co2+ from 0.1 to 10 nM and CL enhancement ((I-I0)/I0) (Fig. 5 A, B). A linear equation of (I-I0)/I0 = 0.13 + 0.09 c (r = 0.998) has obtained. The detection limit was estimated to be 30 pM (S/N=3). The logarithmic linear calibration curve showed wide range from 0.1 to 300 nM (Fig. S8). The present method demonstrated more sensitivity than previously reported analytical methods for Co2+ detection. In some previous reports, it is reported that only Co2+ and luminol can also produce ultra-weak CL signals. Therefore, the effect of NHPI on detection of Co2+ was also investigated. Without the addition of NHPI, the detection limit for Co2+ is estimated to be 30.0 nM under same experimental conditions. The CL behaviour shows that NHPI can remarkably improve both the sensitivity and CL intensity. The CL signals with the addition of NHPI are about 39 times higher than in the absence of NHPI (Fig. S9).These intense CL signals (by using NHPI) are also advantageous for the 9
future implementation of portable and low cost detectors in the detection protocols. The RSD of nine repeated measurements is 3.91% at Co2+ concentration of 1 nM, representing excellent reproducibility. By comparison, the present method is simple, cost-effective, and more sensitive than other reported methods by two orders of magnitude (Table S3). 3.10 Selectivity for Co2+ To investigate the selectivity of luminol/NHPI CL system toward Co2+ ions, typical metal ions such as K+, Na+, Mg2+, Ni2+, Co2+, Mn2+, Cu2+, Zn2+, Pb2+, Hg2+, Cd2+, Fe3+, and Cr3+ ions (10 µΜ) were added separately into the control sample solution. Fig. 5C depicts that the CL signals enhanced dramatically only with the addition of Co2+. In contrast, negligible enhancement in CL intensity was observed with the addition of above mentioned metal ions at the same concentrations (10µM). It indicates that this CL method has excellent selectivity for Co2+ detection. This remarkable selectivity supports the distinction of this novel CL system from previous luminol systems which commonly show poor selectivity in the presence of other metal ions. The exquisite selectivity of this CL system for Co2+ is attributed to the following reasons. On the one hand, it has already reported that Co2+ can catalyze the decomposition of NHPI much more efficiently to generate reactive oxygen species than other metal cations (Gambarotti et al. 2001). On the other hand, it has also been reported that Co2+ is most efficient catalyst of luminol reaction in the presence of reactive oxygen species than other metal cations (Burdo and Seitz 1975). These multiple catalytic properties of Co2+ may contribute to the exquisite selectivity than other metal ions. 3.11 Determination of UA and Co2+ in real samples In order to evaluate the practical applicability of this method, the determinations of UA and Co2+ ions in human urine and tap and lake water samples have been carried out, respectively. For UA determination in real samples, three urine samples were collected from healthy individuals. These urine samples were diluted 1000 times to bring the concentration of UA in the linear range and avoid the interfering species. The UA concentrations were calculated to be 2.09 µM, 1.87 µM, and 1.65 µM in three different urine samples by standard addition method, respectively. Tap and lake water samples were collected from the Changchun, Jilin province, China. The Co2+ ions concentration in the tap water sample solution was calculated to be 0.091 ± 0.02 nM by standard addition method. The Co2 + ions concentration is not detected in lake water samples. To check the recoveries, urine and water samples were spiked with standard UA and Co2+ solutions of 0.10 µM, 0.50 µM, 1.00 µM, 10
and 10, 20, 30 nM, respectively. Table S4 shows that the recoveries of all samples are achieved in the range of 96.35-102.70%. The obtained results show that luminol-NHPI CL system is appropriate for real sample applications. 4. Conclusions We have developed NHPI as an efficient coreactant for luminol CL. Co2+ can dramatically enhance CL signals of the luminol-NHPI system, while SOD and UA can effectively decrease CL signals of the luminol-NHPI system. This CL system achieves more sensitive detection of UA, SOD, Co2+, luminol, and NHPI. Furthermore, this method displays excellent selectivity for the detection of UA and Co2+ and can detect UA in urine samples and Co2+ in tap and lake water samples, which makes the luminol-NHPI CL system attractive. The popularity and versatile applications of NHPI in synthetic chemistry makes NHPI promising coreactant for other CL systems. Because of the favourable features of NHPI and the broad applications of CL, NHPI-based CL will find lots of applications, such as biochemical applications, clinical analysis, and environmental detection. Acknowledgments This project was kindly supported by the National Key Research and Development Program of China [No. 2016YFA0201300], the National Natural Science Foundation of China [Nos. 21675148 and 21475123], and the Chinese Academy of Sciences (CAS)–the Academy of Sciences for the Developing World (TWAS) President’s Fellowship Programme [2013-053]. References Albrecht, H.O., 1928. Z. Phys. Chem. 136, 321-330. Au-Yeung, H.Y., New, E.J., Chang, C.J., 2012. Chem. Commun. 48(43), 5268-5270. Azmi, N.E., Ramli, N.I., Abdullah, J., Abdul Hamid, M.A., Sidek, H., Abd Rahman, S., Ariffin, N., Yusof, N.A., 2015. Biosens. Bioelectron. 67, 129-133. Bauer, L., Miarka, S.V., 1957. J. Am. Chem. Soc. 79(8), 1983-1985. Becker, B.F., Reinholz, N., Özçelik, T., Leipert, B., Gerlach, E., 1989. Pflügers Archiv. 415(2), 127-135. Burdo, T.G., Seitz, W.R., 1975. Anal. Chem. 47(9), 1639-1643. Chauhan, N., Pundir, C.S., 2011. Anal. Biochem. 413(2), 97-103. Gambarotti, C., Punta, C., Recupero, F., Zlotorzynska, M., Sammis, G., 2001. Encyclopedia of Reagents for Organic Synthesis. John Wiley & Sons, Ltd. 11
Girotti, S., Fini, F., Ferri, E., Budini, R., Piazzi, S., Cantagalli, D., 2000. Talanta 51(4), 685692. Gobi, K.V., Mizutani, F., 2001. Anal. Sci. 17(1), 11-15. Harman, D., 1981. Proc. Natl. Acad. Sci. U. S. A. 78(11), 7124-7128. He, D., Zhang, Z., Huang, Y., Hu, Y., Zhou, H., Chen, D., 2005. Luminescence 20(4-5), 271275. He, Y., Huang, G.M., Cui, H., 2013. ACS Appl. Mater. Interfaces 5(21), 11336-11340. Jin, D., Seo, M.-H., Huy, B.T., Pham, Q.-T., Conte, M.L., Thangadurai, D., Lee, Y.-I., 2016. Biosens. Bioelectron. 77, 359-365. Khan, P., Idrees, D., Moxley, M.A., Corbett, J.A., Ahmad, F., von Figura, G., Sly, W.S., Waheed, A., Hassan, M.I., 2014. Appl. Biochem. Biotechnol. 173(2), 333-355. Kiran, R., Scorsone, E., Mailley, P., Bergonzo, P., 2012. Anal. Chem. 84(23), 10207-10213. Li, N., Wang, W., Tian, D., Cui, H., 2010. Chem. Commun. 46(9), 1520-1522. Li, Q., Li, F., Shen, W., Liu, X., Cui, H., 2016. J. Mater. Chem. C 4(16), 3477-3484. Li, Z., Feng, M., Lu, J., 1998. Microchem. J. 59(2), 278-283. Lin, J.M., Shan, X., Hanaoka, S., Yamada, M., 2001. Anal. Chem. 73(21), 5043-5051. Lin, Y.Y., Dai, H., Xu, G.F., Yang, T., Yang, C.P., Tong, Y.J., Yang, Y.S., Chen, G.N., 2013. Microchim. Acta 180(7-8), 563-572. Lin, Z.Y., Huang, L.Z., Liu, Y., Lin, J.M., Chi, Y.W., Chen, G.N., 2008. Electrochem. Commun. 10(11), 1708-1711. Lindsay, D., Kerr, W., 2011. Nat. Chem. 3(6), 494. Lu, J., Xiong, Y., Liao, C., Ye, F., 2015. Anal. Methods 7(23), 9894-9899. Lv, Y., Zhang, Z., Chen, F., 2002. Analyst 127(9), 1176-1179. Melone, L., Prosperini, S., Ercole, G., Pastori, N., Punta, C., 2014. J. Chem. Technol. Biotechnol. 89(9), 1370-1378. Melone, L., Punta, C., 2013. Beilstein J. Org. Chem. 9, 1296-1310. Noroozifar, M., Khorasani-Motlagh, M., Jahromi, F.Z., Rostami, S., 2013. Sens. Actuator BChem. 188, 65-72. Qi, L., Xia, Y., Qi, W., Gao, W., Wu, F., Xu, G., 2016. Anal. Chem. 88(2), 1123-1127. Reddy, K.K., Gobi, K.V., 2012. Electrochim. Acta 78, 109-114. Safavi, A., Karimi, M.A., 2002. Talanta 57(3), 491-500. Saqib, M., Li, S., Gao, W., Majeed, S., Qi, L., Liu, Z., Xu, G., 2016. Anal. Bioanal. Chem. 408(30), 8851-8857. 12
Selden, A.I., Norberg, C., Karlson-Stiber, C., Hellstrom-Lindberg, E., 2007.
Environ.
Toxicol. Pharmacol. 23(1), 129-131. Su, Y., Deng, D., Zhang, L., Song, H., Lv, Y., 2016. S TrAC-Trend. Anal. Chem. 82, 394411. Tu, Y.-J., Njus, D., Schlegel, H.B., 2017. Org. Biomol. Chem. 15(20), 4417-4431. Tvermoes, B.E., Finley, B.L., Unice, K.M., Otani, J.M., Paustenbach, D.J., Galbraith, D.A., 2013. Food Chem. Toxicol. 53, 432-439. Villa, J.E.L., Poppi, R.J., 2016. Analyst 141(6), 1966-1972. Wang, D.M., Zhang, Y., Zheng, L.L., Yang, X.X., Wang, Y., Huang, C.Z., 2012. J. Phys. Chem. C 116(40), 21622-21628. Wang, J., Tan, X., Song, Z., 2013. J. Anal. Methods Chem. 2013, 7. Wei, W., Wang, H., Jiang, C., 2008. Spectrochim. Acta Mol. Biomol. Spectrosc. 70(2), 389393. White, E.H., Zafiriou, O., Kagi, H.H., Hill, J.H.M., 1964. J. Am. Chem. Soc. 86(5), 940-941. Yang, C., Zhang, Z., 2010. Talanta 81(1-2), 477-481. Yang, X., Dou, Y., Zhu, S., 2013. Analyst 138(11), 3246-3252. Zhang, R., Hu, Y., Li, G., 2014. Anal. Chem. 86(12), 6080-6087. Zhao, J., 2013. Anal. Methods 5(23), 6781. Zhao, L., Blackburn, J., Brosseau, C.L., 2015. Anal. Chem. 87(1), 441-447. Zhao, S., Wang, J., Ye, F., Liu, Y.-M., 2008. Anal. Biochem. 378(2), 127-131. Zhou, S., Zuo, R., Zhu, Z., Wu, D., Vasa, K., Deng, Y., Zuo, Y., 2013. Anal. Methods 5(5), 1307-1311.
13
Figure and Scheme captions Fig. 1. Comparison of CL profiles of luminol-NHS and luminol-NHPI system under same experimental conditions. Inset: the enlarged version of CL profiles of luminol-NHS system. The comparison was performed with 10 μM luminol at pH 11.5 in the presence of 0.5 mM NHS or 0.5 mM NHPI. Fig. 2. A comparison of CL behavior for the luminol–H2O2 (red color) and luminol–NHPI systems (violet color). Inset: enlarged version of chemiluminescent kinetic profile of luminol–H2O2 system. The comparison was conducted in the presence of 10 µM luminol, 2 mM NHPI, and 2 mM H 2O2 at pH 11.5. Fig. 3. (A) CL kinetic curves in the absence and the presence of different concentrations of SOD: 0.01–2.0 μg/mL. (B) Shows the relationship between CL quenching and the concentration of SOD from 0.01 to 2.0 μg/mL. Inset shows linear calibration curve for SOD in the range of 0.01–0.25 μg/mL. (C) CL kinetic curves in the absence and the presence of thiourea and NaN3: 5 mM each. The CL measurement was carried out in the presence of 0.3 mM NHPI and 10 µM luminol at pH 11.5. Scheme 1. Proposed mechanism of luminol-NHPI CL detection system. Fig. 4. (A) CL emission-time curves in different concentrations of UA, (B) Linear calibration curve of UA in the range of 50-8000 pM. Inset: the enlarged version of UA concentrations from 0 to 500 pM and (C) selectivity for the detection of UA. I symbolize the CL intensity at a specific UA concentration and other interfering compounds; I0 is the CL intensity of the control sample. I0/I represent the CL quenching efficiency after the addition of UA and different biological compounds. The concentration of UA was 30 µM and for all other compounds were 100 µM for selectivity study. The measurements were performed in the presence of 10 µM luminol, and 0.5 mM NHPI at pH11.5. Fig. 5. (A) CL emission-time curves in different concentrations of Co2+, (B) shows the relationship between CL enhancement and the concentration of Co2+ from 0.1 to 300 nM. Inset shows the linear calibration curve for the detection of Co2+ in the range of 0.1-10 nM. (C) Selectivity for the detection of Co2+. I symbolizes the CL intensity of the system after Co2+ and other metal cations addition, I0 indicates the CL intensity of the control sample and (I-I0)/I0 is the expression of the CL enhancement efficiency after the addition of Co2+ and other metal ions. The concentrations of all the metal ions were 10 µM for selectivity study. The measurements were performed in the presence of 10 µM luminol, and 0.5 mM NHPI at pH11.5.
14
Highlights Chemiluminescence of luminol-NHPI has been reported for the first time. Superoxide dismutase and uric acid can remarkably quench the chemiluminescence. Co2+ can dramatically enhance the chemiluminescence of luminol-NHPI. Highly sensitive detection of superoxide dismutase, uric acid and Co2+ is achieved. It can successfully detect uric acid in human urine and Co2+ in water real samples after dilution.
15
16
Scheme 1
17