Real-time fluorescence assay of alkaline phosphatase in living cells using boron-doped graphene quantum dots as fluorophores
Author’s Accepted Manuscript Real-time fluorescence assay of alkaline phosphatase in living cells using boron-doped graphene quantum dots as fluoropho...
Author’s Accepted Manuscript Real-time fluorescence assay of alkaline phosphatase in living cells using boron-doped graphene quantum dots as fluorophores Li Chen, Guancao Yang, Ping Wu, Chenxin Cai www.elsevier.com/locate/bios
To appear in: Biosensors and Bioelectronic Received date: 6 February 2017 Revised date: 26 April 2017 Accepted date: 11 May 2017 Cite this article as: Li Chen, Guancao Yang, Ping Wu and Chenxin Cai, Realtime fluorescence assay of alkaline phosphatase in living cells using boron-doped graphene quantum dots as fluorophores, Biosensors and Bioelectronic, http://dx.doi.org/10.1016/j.bios.2017.05.022 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.
Real-time fluorescence assay of alkaline phosphatase in living cells using boron-doped graphene quantum dots as fluorophores Li Chen, Guancao Yang, Ping Wu* and Chenxin Cai* Jiangsu Key Laboratory of New Power Batteries, Jiangsu Collaborative Innovation Center of Biomedical Functional Materials, National and Local Joint Engineering Research Center of Biomedical Functional Materials, College of Chemistry and Materials Science, Nanjing Normal University, Nanjing 210097, P. R. China.
ABSTRACT: This work reports a convenient and real-time assay of alkaline phosphatase (ALP) in living cells based on a fluorescence quench-recovery process at a physiological pH using the boron-doped graphene quantum dots (BGQDs) as fluorophore. The fluorescence of BGQDs is found to be effectively quenched by Ce3+ ions because of the coordination of Ce3+ ions with the carboxyl group of BGQDs. Upon addition of adenosine triphosphate (ATP) into the system, the quenched fluorescence can be recovered by the ALP-positive expressed cells (such as MCF-7 cells) due to the removal of Ce3+ ions from
1
BGQDs surface by phosphate ions, which are generated from ATP under catalytic hydrolysis of ALP that expressed in cells. The extent of fluorescence signal recovery depends on the level of ALP in cells, which establishes the basis of ALP assay in living cells. This approach can also be used for specific discrimination of the ALP expression levels in different type of cells and thus sensitive detection of those ALP-positive expressed cells (for example MCF-7 cells) at a very low abundance (10 ± 5 cells mL–1). The advantages of this approach are that it has high sensitivity because of the significant suppression of the background due to the Ce3+ ion quenching the fluorescence of BGQDs, and has the ability of avoiding false signals arising from the nonspecific adsorption of non-target proteins because it operates via a fluorescence quench-recovery process. In addition, it can be extended to other enzyme systems, such as ATP-related kinases. Keywords: Fluorescence quenching; boron-doped graphene quantum dots; alkaline phosphatase; adenosine triphosphate; biomarker.
1. Introduction Alkaline phosphatase (ALP) is responsible for the catalytic removal of the phosphate groups from various substrates including nucleic acids, proteins, and carbohydrates (Coleman, 1992) etc. Its activity in living cells significantly affects the phosphorylation/dephosphorylation state, which plays important roles in signal transduction and regulation of intracellular processes (Choi et al., 2007). Moreover, ALP has been regarded as an important biomarker in medical diagnosis since abnormal levels of ALP in cells or serum are closely related to many diseases, such as breast and prostatic cancer, bone disease (osteoporosis and bone tumor), diabetes, hepatitis, and liver dysfunction (Al Mamari et al., 2013; Gyurcsányi et al., 2002; Ooi, et al., 2007) etc. Therefore, the development of convenient and reliable
2
methods for a sensitively and continuously assaying the ALP in living cells is important and valuable, not only for clinical diagnoses but also for biomedical research. ALP has been typically assayed by using several commercially available kits, which employ specially designed substrates such as p-nitrophenyl phosphate, 4-methyl-umbellyferyl phosphate, and paminophenyl phosphate, etc. These substrates are readily hydrolyzed by ALP to produce yellow colored (p-nitrophenol), highly fluorescent (4-methyl-umbelliferone), or electroactive (p-aminophenyl) products, that are quantitatively detected by spectroscopic or electrochemical methods (Park et al., 2014; Wang et al., 2009). Other methods, devised to determine ALP activity, include colorimetric (Choi et al, 2007), chemiluminescence (Freeman et al., 2010; Jiang and Wang, 2012), and surface enhanced Raman scattering (SERS) (Ruan et al., 2006), CEST MRI imaging (Daryaei et al., 2016), etc. Despite many achievements from these studies, the method with the potential for ALP assay in living cells is still rare and is highly desired. Fluorescent assay, with its intrinsic advantages of low background noise and high sensitivity, is regarded as a more desirable method in bioassay, especially in the visualization of targeting analytes in living cells (Lu et al., 2014; Sankara et al., 2017; Tu et al., 2012; Wu et al., 2013; Wu et al., 2014; Zhao et al., 2015). Many fluorophores such as organic dyes (Gong et al., 2011; Gu et al., 2013; Hou et al., 2015; Song et al., 2014; Zhang et al., 2015; Zhao, et al., 2017), conjugated polymers (Jornet-Martínez et al., 2017; Li et al., 2014; Liu and Schanze, 2008), inorganic semiconductor dots (Liu et al., 2014; Qian et al., 2015), and noble metal nanoclusters (Choi et al., 2007; Sun et al., 2014), etc. have been developed for assaying ALP. These fluorophores, however, have some drawbacks such as poor photostability and solubility in water for dyes, laborious synthesis procedure for fluorescent polymers, high toxicity for inorganic semiconductor dots, and high cost and poor stability in an aqueous system for noble metal nanoclusters. In addition, most of the reported assays were built by employing synthetic chemicals as a substrate rather than natural biomolecules, and performed in a medium with high pH value (~9.5) (Gong et al., 2011; Park et al., 2014; Song et al., 2014). The artificial substrates not only require complicated preparation procedures but also have potential toxicity to organisms. Furthermore, most assays are 3
based on the fluorescence turn-off mechanism (Freeman et al., 2010; Liu et al., 2014; Park et al., 2014; Sun et al., 2014), which makes the assay a relatively low sensitivity. Therefore, it is high desirable to develop a fluorescence assay of ALP based on turn-on mechanism, which makes the assay a high sensitivity. With comparison to those above fluorophores, graphene quantum dots (GQDs), in particular the heteroatom-doped GQDs have greater advantages such as stable light emitting, good photostability, and good biocompatibility, and more importantly easy modulation in their intrinsic properties, thus they have been used to design new fluorescence chemosensors and biosensors in vitro and in vivo (Cai et al., 2015; Chen et al., 2017a; Chen et al., 2017b; Ji et al., 2016; Yang et al., 2017; Zhang et al., 2016). This work reports a convenient and real-time assay with high sensitivity for monitoring ALP in living cells based on a fluorescence quench-recovery process (the net fluorescence turn-on process, which makes the assay a high sensitivity) at a physiological pH by using the boron-doped graphene quantum dots (BGQDs) as fluorophore and ATP (a physiological substrate of ALP) as substrate. As illustrated in Fig. 1A, BGQDs possess strong fluorescence (under excitation of an UV lamp with wavelength of 365 nm), and this fluorescence is found to be effectively quenched in the presence of Ce3+ ions because of the coordination of Ce3+ ions with the carboxyl group of BGQDs. However, upon addition of ATP into the system, the quenched fluorescence is recovered by the ALP-positive expressed cells (such as MCF-7 cells, human breast cancer cells) due to the removal of Ce3+ ions from the surface of BGQDs by phosphate ions, which are generated from ATP under catalytic hydrolysis of ALP that expressed in cells. The extent of fluorescence signal recovery depends on the level of ALP in cells, which establishes the basis of ALP assay in living cells and the capability of directly visualizing the ALP expression in ALP-positive expressed living cells.
4
At the same time, this approach can also be used for specific discrimination of the ALP expression in different type of cells and highly sensitive detection of those ALP-positive expressed cells (for example MCF-7 cells) at a very low abundance (~10 cells mL–1). 2. Materials and methods 2.1. Instruments The electrochemical experiments were conducted on a CHI 660B electrochemical workstation (CH Instruments) in a two-compartment two-electrode cell. The fluorescence spectra were collected with a Cary Eclipse fluorescence spectrophotometer (Varian) equipped with a Xenon lamp excitation source. Confocal laser scanning microscopic (CLSM) images were recorded with a confocal laser scanning microscope (MRC-1024, Bio-Rad) equipped with an oil immersion 100× objective. 2.2 Synthesis of boron-doped graphene quantum dots (BGQDs) BGQDs were synthesized by a constant potential electrolysis method using our previously published procedures (Chen et al., 2017a; Ji et al., 2016). Briefly, the electrolysis was performed in a twocompartment two-electrode cell in aqueous 0.1 M borax (99.5%, Sigma-Aldrich) solution (the pH was adjusted to ~7). The high purity graphite rod (99.9%, 3 mm in diameter, Shanghai Carbon Co. Ltd.) was used as an anode and a Pt sheet (1.5 cm × 1.5 cm) was used as a cathode. The voltage between the anode and cathode was controlled at 3 V. After 2-h electrolysis, BGQDs were collected by filtering the resulting solution using 0.22-μm microporous nylon membrane to remove the precipitated graphite oxide and graphite particles. Then the obtained pale-yellow solution was dialyzed over deionized water in a dialysis bag (retained molecular weight 3500 Da) for 48 h to remove the electrolyte of borax, the deionized water was changed every 12 h. 2.2. Cell culture
5
MCF-7 (human breast cancer cells), A549 (human lung cancer cells), HeLa (human cervical cancer cells), HEK-293T cells (human embryonic kidney cells), and HT-29 cells (human colon adenocarcinoma cells) cell lines were obtained from the cell bank of type culture collection of the Chinese Academy of Sciences (Shanghai, China) and cultured at 37 °C in RPMI 1640 medium supplemented with 10% fetal bovine serum (FEB), 2 mM L-glutamine, 100 U mL‒1 penicillin, and 100 μg mL‒1 streptomycin in a 5% CO2 environment. After growing to 90% confluence, the cells were washed with PBS (0.145 M NaCl, 1.9 mM NaH2PO4, 8.1 mM K2HPO4, pH 7.4) and replaced the culture medium by 1 mL PBS and the cell number was estimated by a hemocytometer. 2.3. Fluorescence assays MCF-7, A549, HeLa, HEK-293T, and HT-29 cells are adherent growth cells. For fluorescence measurements, these cells were dissociated by treating them with trypsin (0.25%) for 30 seconds. The suspended cells were transferred to a centrifuge tube and washed with PBS thrice. The measurements were performed in 10 mM Tris-HCl buffer at pH 7.4. In a typical assay, 20 μL of Ce3+ ion solution (20 mM) were first mixed with 1 mL BGQDs solution (15 μg mL‒1). After a stable background fluorescence was attained, 10.0 μL of ATP solution (20.0 mM) was introduced into this mixture. Then, MCF-7 cells (~1 × 106 cells mL‒1) were added into this system. The fluorescence spectrum at each step was recorded. The excitation wavelength is 380 nm. To evaluate the selectivity of the proposed assay, A549, HeLa, HEK-293T, and HT-29 cells, and several randomly selected biological proteins (enzymes) such as glucose oxidase (GOx), human serum albumin (HSA), phospholipase A2 (PLA2), horseradish peroxidase (HRP), and hemoglobin (Hb) were used to replace MCF-7 cells and record the fluorescence recovery of BGQDs. 2.4. MTT assay MCF-7 cells (~1 × 105 cells each well) were first incubated with BGQDs at a concentration of 0‒200 μg mL‒1 for 24 h in 96-well plates, and then repeatedly rinsed with PBS to remove the nonspecifically
6
bound BGQDs. The cytotoxicity of the BGQDs was estimated by an 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide (MTT) assay. MTT solution (10 μL, 5 mg mL–1, pH 7.4) was added to each well and incubated for 4 h. Then, dimethylsulfoxide (DMSO, 100 μL) was added to each well. Absorbance was recorded at 550 nm on a Synergy 2 microplate reader (Biotek). The viabilities of the cells incubated with the BGQDs were obtained by comparing to cells incubated without the BGQDs. 2.5. CLSM imaging Before recording the CLSM images, the cells were first transferred into a Petri dish and then incubated with BGQDs (containing Ce3+ ions and ATP) at 37 °C for 30 min. The CLSM images were recorded under excitation of 405 nm. Of note, although the highest intensity of the emission spectrum of the BGQDs is observed under the excitation of 380 nm, the excitation source of the confocal laser scanning microscope cannot be adjusted continuously, the excitation wavelengths that we can selected are 405, 457, 488, 514, 543, and 633 nm. Thus, we selected the 405 nm as excitation to record the CLSM images because the 405 nm is closed to 380 nm, under which the highest intensity of the emission spectrum of the BGQDs was observed. 3. Results and discussion The synthesized BGQDs exhibit a strong fluorescence signal at ~530 nm (~2.34 eV), which is believed to result from the radiative π*→n transition of electrons in the B–C bond (Chen et al., 2017a), in Tris-HCl buffer (10 mM, pH 7.4) under excitation of 380 nm (curve a, Fig. 1B). This fluorescence is very stable because no time-dependent fluorescence changes can be observed for several months (at least a span of four months in our experiments) at the ambient temperature (Fig. S1). This feature is crucial because the BGQDs must be water-soluble and stable in the ambient environment for their practical use in sensing applications. Addition of the Ce3+ ions (Ce(NO3)3, 0.4 mM) into solution causes the complete quenching of the fluorescence (curve b, Fig. 1B. The Ce3+ ion concentration-dependent quenching characteristic is presented in Fig. S2A). Kinetic studies show that the fluorescence quenching
7
of Ce3+ ions is fairly fast because the fluorescence intensity of BGQDs decreased rapidly to ∼20% of the initial intensity within ~1 min and was completely quenched within 3 min (Fig. S2B), implying high quenching efficiency of Ce3+ ions to BGQDs. The high quenching efficiency results from the nonradiative electron/hole recombination processes between the Ce3+ ions and BGQDs triggered by the rapid formation of the complexation of Ce‒BGQD through the high binding ability of the carboxylate groups on BGQD surface to Ce3+ ions (the equilibrium constant of Ce3+ ion bound to carboxylate group is ~1×1019) (Zhou, 1997). To further elucidate the mechanism of the Ce3+ ion-induced fluorescence quenching of BGQDs, we performed DFT (Density Functional Theory) calculations to microscopically understand the interaction of the Ce3+ ions with BGQDs (the calculation details are depicted in Supplemental Material). The calculation indicates that BGQDs have HOMO (the highest occupied molecular orbital) and LUMO (the lowest unoccupied molecular orbital) energy level of –5.05 and –1.83 eV, respectively. The HOMO and LUMO levels of Ce3+ ion were –4.90 and –2.14 eV, respectively, with a HOMO–LUMO energy gap of 2.76 eV. The binding of Ce3+ ions onto the surface of BGQDs can induce the partial transfer of excited electrons from BGQDs to LUMO orbitals of Ce3+ ions, thereby facilitating nonradiative electron/hole recombination (Fig. 1C); thus, the electron transition in radiative forms (fluorescence emission) was restrained, causing the fluorescence quenching of BGQDs. After complete quenching of the fluorescence of BGQDs, ATP (0.2 mM. This concentration is high enough to ensure the ALP-catalyzed hydrolysis is under substrate-saturated condition, Fig. S3) was introduced into the system. When MCF-7 cells are absent in the system, ATP can barely interact with Ce3+ ions and the alteration of the fluorescence signal of BGQDs is negligible (curve c, Fig. 1B), implying that ATP cannot cause the dissociation of Ce3+ ions from BGQD surface because of its weak binding ability to Ce3+ ions (the equilibrium constant of ATP binding to Ce3+ ions is only ~4) (Orabi et al., 2010). Under our conditions, ATP may be slowly hydrolyzed to adenosine diphosphate (ADP) and adenosine monophosphate (AMP), thus, we also evaluated the binding ability of the Ce3+ ions to ADP and AMP. The results depicted in Fig. S4 suggest that the binding ability of Ce3+ ions to ADP and AMP 8
is weak because fluorescence recovery of the Ce3+-quenched BGQDs is negligible after introducing the ADP (0.2 mM), and AMP (0.2 mM) into the system. However, the introduction of MCF-7 cells (~1 × 106 cells mL‒1) can result in great enhancement of the fluorescence signal, leading to the fluorescence recovery occurrence (curve d, Fig. 1B). Of note, the cytotoxicity of the BGQDs is very low because MTT assay indicates that the cellular viability of the MCF-7 cells still remains at a high level (more than 92%) even after the cells were incubated with BGQDs, the concentration ranges from 0 to 200 μg mL‒1, for 24 h (Fig. S5). Addition of MCF-7 cells to the solution initiates the hydrolysis of ATP to produce phosphate ions owing to the specific catalytic ability of ALP expressed in MCF-7 cells. Because of stronger affinity of phosphate to Ce3+ ions than carboxylate (the equilibrium constant of Ce3+ ions bind to phosphate ions is ~1 × 1023) (Zhou, 1997), the generated phosphate ions will bind with Ce3+ ions and remove them from BGQD surface, thus leading to fluorescence recovery of BGQDs. As ALP-catalyzed hydrolysis reaction proceeds, the amount of phosphate ions available for association with Ce3+ ions increases, and therefore, the BGQD fluorescence intensity increases with time. BGQDs respond to the MCF-7 cells rapidly because the fluorescence signal recovers to be more than 90% of the initial signal within 5 min, and levels off to a relatively stable value after ~30 min (Fig. S6). To verify the phosphate ions can remove the Ce3+ ions from the surface of BGQDs and lead to the fluorescence recovery, the recovered fluorescence signal of the Ce3+-quenched BGQDs was recorded after directly adding phosphate ions (0.2 mM Na3PO4) into the system. The results depicted in the Fig. S7 indicate that a big recovered fluorescence signal is obtained upon the addition of the phosphate ions, demonstrating the phosphate ions can remove the bound Ce3+ ions from the BGQD surface and result to the fluorescence recovery. However, the intensity of the fluorescence recovery induced by pyrophosphate ions (P2O74‒) ions (0.2 mM) is much lower than that induced by phosphate ions (Fig. S7), implying the binding ability of the P2O74‒ ions to Ce3+ ions is weaker than that of phosphate ions. There is no fluorescence response of MCF-7 cells themselves in the range of 800‒450 nm (Zhao et al., 2015). Moreover, no fluorescence recovery is observed if MCF-7 cells were pre-incubated with 9
levamisole hydrochloride (2 mM, an ALP inhibitor) (Fig. S8). These results further demonstrate that the observed fluorescence signal depicted in Fig. 1B is certainly ascribed to the removal of Ce3+ ions from BGQD surface by ALP-catalyzed generation of phosphate ions. It should also be noted that the ratio of signal to background obtained here is high (> 9) because of the significant suppression of the background due to the Ce3+ ion quenching to the fluorescence of BGQDs. This unique feature will make the proposed method a high assay sensitivity. The role of ATP to the recovery of the fluorescence signal of Ce3+-quenched BGQDs was evaluated. After the fluorescence of the BGQDs was quenched by Ce3+ ions, MCF-7 cells (~1 × 106 cells mL‒1) were introduced into the system. After 30-min incubation, the fluorescence recovery was recorded at the absence of ATP substrate. The results in Fig. S9 show that the recovery of fluorescence signal is negligible. Although the ALP-positive expressed cells were presented in the system, phosphate ions could not be generated due to the lack of the substrate (ATP). Thus, the Ce3+ ions bound on the surface of BGQDs cannot be removed and the quenched fluorescence cannot be recovered. This result implies the MCF-7 cell themselves in absence of ATP cannot induce the recovery of the fluorescence signal of Ce3+-quenched BGQDs, probably due to that the concentrations of phosphate ions in the cells are too low to effectively remove the Ce3+ ions from the surface of BGQDs. The effects of the ATP on the fluorescence characteristics of MCF-7 cell themselves were also studied by recording the CLSM images of the cells in the presence and absence of ATP (0.2 mM). The results depicted in Fig. S10 indicate that the fluorescence signal of the cells is negligible even in the presence of ATP in the system, implying ATP does not alter the fluorescence characteristics of the MCF-7 cells. These results further verify that the observed fluorescence signal in Fig. 1B comes from BGQD in cells. The effects of pH of the Tris-HCl buffer on the fluorescence recovery of the Ce3+-quenched BGQDs by the MCF-7 cells were studied. Three pH values (7, 8, and 9) were selected (the buffered range of Tris-HCl is 7‒9). The results depicted in Fig. S11 indicated that the recovered fluorescence signal
10
decreases with the increasing of the pH value of the buffer from 7 to 9. Therefore, a physiological pH (pH 7.4) is selected for this work.
The specificity of this assay was first evaluated by using ALP-negative expressed cells, such as HEK293T cells (human embryonic kidney cells), and HT-29 cells (human colon adenocarcinoma cells). The recovered fluorescence signal of Ce3+-quenched BGQDs is negligible after 30-min incubation with these cells in the presence of ATP (Fig. 2A), implying that these cells cannot catalyze the hydrolysis of ATP to generate phosphate ions because of their ALP-negative expressions. The CLSM images shown in Fig. 2B indicate that there is no fluorescence observed from the HEK-293T and HT-29 cells even these cells were incubated with Ce3+-quenched BGQDs for 30 min in the presence of ATP, further demonstrating these cells cannot catalyze the hydrolysis of ATP to generate phosphate ions.
The specificity of this assay was further evaluated by randomly selecting several biological proteins (enzymes) such as glucose oxidase (GOx), human serum albumin (HSA), phospholipase A2 (PLA2), horseradish peroxidase (HRP), and hemoglobin (Hb), each at 20 μg mL‒1, to replace MCF-7 cells and record the fluorescence recovery of BGQDs. None of the control proteins has a specific interaction with ATP. After 30-min incubation with these control proteins, we did not observe any recovery in the fluorescent signal of the Ce3+-quenched BGQDs (Fig. 3). Therefore, it is safe to conclude that the fluorescence recovery of noted in the assay arises from the specific enzymatic activity of ALP. This feature ensures the assay to be impervious to false signals caused by the nonspecific adsorption of nontarget proteins.
11
The interference of the other cells (such as ALP-negative expressed cells) on the fluorescence recovery ability of MCF-7 cells to the Ce3+-quenched BGQDs were also studied. We selected HEK293T cells as the interference cells. The results in Fig. S12 show that the presence of the HEK-293T cells in the assay system does not affect the fluorescence recovery characteristic of MCF-7 cells toward the Ce3+-quenched BGQDs because the recovered fluorescence intensity is almost identical for the system with and without HEK-293T cells.
Sensitivity is another key factor we studied for evaluating the feasibility of our method for ALP assay in cells. The recovered fluorescence signals increase with the number of cells (curves a‒g, Fig. 4A). This increase can also be directly visualized from the photographs depicted in the inset. The fluorescence signal linearly increases with the concentration of MCF-7 cells at least up to ~1 × 106 cells mL‒1. The detection limit can be estimated to lower than ~(10 ± 5) cells mL‒1 (Fig. 4B), demonstrating that BGQDs fluorophore can response to the very low concentrations of ALP and our strategy can also be used to detect the ALP-positive expressed target cells in a low abundance. The high sensitivity originates from the specific catalytic hydrolysis of ATP by ALP and high affinity binding of the generated phosphate to Ce3+ ions as well as ultralow background, which is due to the complete quenching of the fluorescence of BGQDs by Ce3+ ions.
This assay can also be applied to distinguish the ALP expression levels in different type of cells. We studied the fluorescence recovery of the Ce3+-quenched BGQDs by three different types of cells, MCF7, HeLa (human cervical cancer cells), and A549 cells (human lung cancer cells). The results show that the extents of the fluorescence recovery induced by A549 (curve a, Fig. 5A) and HeLa (curve b) is 12
lower than that induced by MCF-7 cells (curve c), implying that the expression levels of ALP in A549 and HeLa cells is lower than that in MCF-7 cells. The CLSM images shown in Fig. 5B further verify this conclusion. After incubated with BGQDs, moderate green fluorescence signals are observed in HeLa cells (Fig. 5B), which is weaker than that observed in MCF-7 cells, while a weak green fluorescent intensity is observed in A549 cells, implying that the expression levels of ALP increases in the order of MCF-7 > HeLa > A549 cells (of note, this conclusion is obtained by assuming that response characteristics of the MCF-7, HeLa, and A549 cells to BGQDs are similar). These results indicate that the proposed assay method can be used for estimating the ALP level in cells with good reliability. 4. Conclusions In conclusion, we have described a new assay for real-time monitoring the level of ALP in living cells. This assay is based on a fluorescence quench-recovery process at a physiological pH using the boron-doped graphene quantum dots (BGQDs) as fluorophore and ATP as substrate. This assay can also be used for specific discrimination of the ALP expression levels in different type of cells and thus sensitive detection of those ALP-positive expressed cells (for example MCF-7 cells) at a very low abundance (10 ±5 cells mL–1). In response to these previously reported methods that use the synthetic substrates, this method with ALP’s physiological substrate, ATP, as the substrate and conducting at physiological pH offers the possibility to develop a diagnostic ALP assay for pathological conditions (the assay of the ALP level in MCF-7 cells bearing BALB/c nude mice using the method reported in this work is under way), and can be helpful in the treatment of diseases related to ALP function. The method can be directly used in a high-throughput screening format using a multiwell plate reader because it is based on fluorescence intensity. Moreover, this assay can be extended to other enzyme systems, such as ATP-related kinases. Acknowledgements
13
This work is supported by NSFC (21335004, 21375063, 21405083, and 21675088), the Program of Outstanding Innovation Research Team of Universities in Jiangsu Province, and the Priority Academic Program Development of Jiangsu Higher Education Institutions. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios. References Al Mamari, S., Djordjevic, J., Halliday, J. S., Chapman, R. W., 2013. J. Hepatol. 58, 329‒334. Cai, Z., Fumin Li, F., Wu, P., Ji, L., Zhang, H., Cai, C., Gervasio, D. F., 2015. Anal. Chem. 87, 11803‒ 11811. Chen, L., Wu, C., Du, P., Feng, X., Wu, P., Cai, C.; 2017a. Talanta 164, 100‒109. Chen, Z., Liang, R., Guo, X., Liang, J., Deng, Q., Li, M., An, T., Liu, T., Wu, Y., 2017b. Biosens. Bioelectron. 91, 60–65. Choi, Y., Ho, N.-H., Tung, C.-H., 2007. Angew. Chem. 119, 721–723. Coleman, J. E., 1992. Annu. Rev. Biophys. Biomol. Struct. 21, 441–483. Daryaei, I., Ghaffari, M. M., Jones, H. M., Pagel, M. D., 2016. ACS Sens. 1, 857‒861. Freeman, R., Finder, T., Gill, R., Willner, I., 2010. Nano Lett. 10, 2192‒2196. Gong, H., Little, G., Cradduck, M., Draney, D. R., Padhye, N., Olive, D. M., 2011. Talanta 84, 941‒946. Gu, X., Zhang, G., Wang, Z., Liu, W., Xiao, L., Zhang, D., 2013. Analyst 138, 2427‒2431. Gyurcsányi, R. E., Bereczki, A., Nagy, G., Neuman, M. R., Lindner, E., 2002. Analyst 127, 235–240. Hou, X., Yu, Q., Zeng, F., Ye, J., Wu, S., 2015. J. Mater. Chem. B 3, 1042‒1048. Ji, L., Chen, L., Wu, P., Gervasio, D. F., Cai, C., 2016. Anal. Chem. 88, 3935−3944. Jiang, H., Wang, X., 2012. Anal. Chem. 84, 6986‒6993. Jornet-Martínez, N., Campíns-Falcó, P., Hall, E. A. H., 2016. Biosens. Bioelectron. 86, 14–19.
14
Li, Y., Li, Y., Wang, X., Su, X., 2014. New J. Chem. 38, 4574‒4579. Liu, S., Pang, S., Na, W., Su, X., 2014. Biosens. Bioelectron. 55, 249‒254. Liu, Y., Schanze, K. S., 2008. Anal. Chem. 80, 8605‒8612. Lu, Y., Wu, P., Yin, Y., Zhang, H., Cai, C., 2014. J. Mater. Chem. B 2, 3849–3859. Ooi, K., Shiraki, K., Morishita, Y., Nobori, T., 2007. J. Clin. Lab. Anal. 21, 133–139. Orabi, A. S., Azab, H. A., Sadd, F., Said, H., 2010. J. Solution Chem. 39, 319‒334. Park, K. S., Lee, C. Y., Park, H. G., 2014. Analyst 139, 4691‒4695. Qian, Z., Chai, L., Tang, C., Huang, Y., Chen, J., Feng, H., 2015. Anal. Chem. 87, 2966‒2973. Ruan, C., Wang, W., Gu, B., 2006. Anal. Chem. 78, 3379‒3384. Sankara, V. S. P. K., Jayanthi, A., Das, A. B., Saxena, U., 2017. Biosens. Bioelectron. 91, 15–23. Song, Z., Kwok, R. T. K., Zhao, E., He, Z., Hong, Y., Lam, J. W., Liu, B., Tang, B. Z., 2014. ACS Appl. Mater. Interfaces 6, 17245‒17254. Sun, J., Yang, F., Zhao, D., Yang, X., 2014. Anal. Chem. 86, 7883‒7889. Tu, Y., Wu, P., Zhang, H., Cai, C., 2012. Chem. Commun. 48, 10718–10720. Wang, J. H., Wang, K., Bartling, B., Liu, C.-C., 2009. Sensors 9, 8709‒8721. Wu, P., Gao, Y., Lu, Y., Zhang, H., Cai, C., 2013. Analyst 138, 6501–6510. Wu, P., Tu, Y., Qian, Y., Zhang, H., Cai, C., 2014. Chem. Commun. 50, 1012‒1014. Yang, Y., Liu, Q., Liu, Y., Cui, J., Liu, H., Wang, P., Li, Y., Chen, L., Zhao, Z., Dong, Y., 2017. Biosens. Bioelectron. 90, 31–38. Zhang, H., Xu, C., Liu, J., Li, X., Lin, L., Li, X., 2015. Chem. Commun. 51, 7031‒7034. Zhang, H., Kang, S., Wang, G., Zhang, Y., Zhao, H., 2016. ACS Sens. 1, 875‒881. Zhao, B., Wu, P., Zhang, H., Cai, C., 2015. Biosens. Bioelectron. 68, 763–770. Zhao, L., Xie, S., Song, X., Wei, J., Zhang, Z., Li, X., 2017. Biosens. Bioelectron. 91, 217–224. Zhou, T. H., 1997. Handbook of Analytical Chemistry, 2nd ed.; Chemical Industry Press: Beijing.
15
Figure captions Fig. 1. (A) Illustration of the processes of assay of ALP in living cells based on a fluorescence quenchrecovery process using BGQDs as fluorophore and ATP as substrate. (B) Fluorescence spectra of BGQDs observed at different conditions. (a) BGQDs in 10 mM Tris-HCl buffer (15 μg mL‒1), (b) BGQDs in the presence of Ce3+ ions (0.4 mM), (c) BGQDs in the presence of Ce3+ ions (0.4 mM) and ATP (0.2 mM), (d) BGQDs in the presence of Ce3+ ions (0.4 mM), ATP (0.2 mM), and MCF-7 cells (~1 × 106 cells mL‒1). The excitation wavelength is 380 nm. (C) Diagram illustration of the fluorescence quenching of BGQDs after interaction with Ce3+ ions. Fig. 2. (A) Recovered fluorescence signals of BGQDs in Tris-HCl containing 0.4 mM Ce3+ ions and 0.2 mM ATP in presence of (a) HEK-293T (~1 × 106 cells mL‒1), and (b) HT-29 cells (~1 × 106 cells mL‒1). (B) CLSM images of HEK-293T, and HT-29 cells in different conditions. The CLSM images were recorded under excitation of 405 nm. The scale bar in the images is 20 μm. Fig. 3. The background-subtracted fluorescence signals of BGQDs in Tris-HCl buffer containing 0.4 mM Ce3+ ions and 0.2 mM ATP after incubated with MCF-7 cells (~1 × 106 cells mL‒1), GOx, HSA, PLA2, HRP, and Hb, respectively. The concentration of each protein is each at 20 μg mL‒1. Fig. 4. (A) Fluorescence spectra of BGQDs in Tris-HCl containing 0.4 mM Ce3+ ions and 0.2 mM ATP after incubated with MCF-7 cells at cell concentration of (a) 0 (without MCF-7 cells), (b) ~1 × 101, (c) 1 × 102, (d) 1 × 103, (e) 1 × 104, (f) 1 × 105, and (g) 1 × 106 cells mL‒1. Inset: photographic illustration of the fluorescence recovery of the system at different concentrations of MCF-7 cells. (B) Plot of background-subtracted fluorescence signals (ΔF) against the concentrations of MCF-7 cells. Error bars were calculated based on three measurements. Fig. 5. (A) Recovered fluorescence signals of BGQDs in Tris-HCl containing 0.4 mM Ce3+ ions and 0.2 mM ATP in presence of (a) A549 (~1 × 106 cells mL‒1), (b) HeLa (~1 × 106 cells mL‒1), and (c) MCF-7 cells (~1 × 106 cells mL‒1). (B) CLSM images of MCF-7, HeLa, and A549 cells in different conditions. The CLSM images were recorded under excitation of 405 nm. The scale bar in the images is 20 μm. 16
17
Highlights
18
► A method for real-time assay of alkaline phosphatase (ALP) level in living cells is reported. ►This assay is based on a fluorescence quench-recovery process using the boron-doped graphene quantum dots as fluorophore. ► It has high detection sensitivity and specific discrimination of the ALP expression levels in different type of cells. ► It has the ability of avoiding false signals, and extending to other enzyme systems.
