Chinese Chemical Letters  2017, Vol. 28 Issue (10): 1916-1924   PDF    
Fluorescent probes for recognition of ATP
Ying Wu, Jia Wen, Hongjuan Li, Shiguo Sun, Yongqian Xu    
Shaanxi Key Laboratory of Natural Products & Chemical Biology, College of Chemistry & Pharmacy, Northwest A & F University, Yangling 712100, China
Abstract: Adenosine 5'-triphosphate (ATP) not only participates in various physiological activities as the universal energy currency but also implicates in various pathological processes in living cells. Consequently, sensitive and selective detection ATP in live cells, tissues, as well as environmental samples, are urgently demanded. Due to the simple and convenient operation, economy cost, high selectivity for analyte, well biocompatibility and low cytotoxicity, fluorescent sensors for monitoring ATP have aroused great attention of researchers. In recent years, a large number of fluorescent sensors for detecting ATP have developed. This manuscript summarized most of these sensors and the interaction-mechanism between ATP and sensors, mainly including electrostatic interaction, π-π interaction, covalent bonding or hydrogen bond, or combinations of them, and the advantages of each strategy were also generalized. Here, a viewpoint of classification was shown where the sensors were divided into five typed ones according to the structure of probes used.
Key words: Fluorescent probes     Adenosine 5'-triphosphate (ATP)     Cell imaging     Sensitive detection     Multi-interactions     Recognition mechanism    
1. Introduction

Adenosine 5'-triphosphate (ATP) is composed of adenine, ribose and three phosphate groups. ATP is not only the universal energy currency in living cells, but also a signaling mediator participating in a variety of biological processes, including triphosphoric acid cycle [1], ion channels [2], and neurotransmission [3], cell division [4], DNA synthesis [5]. Furthermore, the abnormal levels of ATP closely correlate with several pathological conditions, such as ischemia, hypoglycemia Parkinson's disease and cardiovascular disease [6-9]. Therefore, the selective detection and accurate quantification of ATP in biological and environmental samples is urgent.

Many analysis methods have been developed for the detection of ATP, such as high performance liquid chromatography, ion chromatography, mass spectrometry and electrochemistry [10-13]. These instrumentally intensive methods can only measure total nucleoside triphosphates content and often suffer from the need of specific equipment, complicated procedures for manipulation and poor accuracy for detection. Thus, a simple and highly precise strategy is essential to explore for not only detecting but also quantifying ATP in biological samples.

Among the diverse methods for detecting ATP, fluorescence detection [14] stands out due to its simplicity, high sensitivity, good selectivity, and real-time reaction monitoring. More importantly, fluorescence sensors could be used for sensitive detection of level fluctuations of intracellular biomolecules, which in turn may reflect the individual's health condition to a certain extent. Based on this, a wide variety of fluorescent sensors have been created and designed for detection of ATP in vitro and in vivo.

Due to the special forms of ATP, i.e., three negatively charged phosphoric acid groups, aromatic adenosine, ribose with polyhydroxy, a large number of fluorescence sensors possess positively charged, large conjugated structure of the aromatic ring, electronegativity of the atoms. These probes could sensor ATP by electrostatic interaction between negatively charged phosphates of ATP and positively charged recognition groups, and/or a covalent bond between electronegativity atom and ribose, and/or π-π interaction between large conjugated structure and adenine, which in turn could cause changes of the fluorescence signal of sensors [15, 16]. In this review, fluorescent molecules for ATP detection are classified to five types of ones according to the structure of the sensors. The five types of probes include chemosensors using metal ion complexes, sensors that can form excimers, organic small molecule based on multiple interactions, sensors bearing polythiophene and sensors containing biomolecules.

2. Five types of sensors for detection of ATP 2.1. Chemosensors using metal ion complexes

The use of dipicolylamine (DPA) unit for fluorescence-based sensing of ATP is a promising approach [17]. Fluorescence sensors based on metal coordination complexes possessing one or two coordination sites to afford stronger affinity for anions would be superior for highly selective detection of ATP [18]. The 2, 2'-dipicolylamine-Zn(Ⅱ) complex, a specific binding motif for phosphate anions, has become a common receptor in the design of phosphate sensors [19, 20]. To date, several Zn2+ complexes coupled with a variety of chromophores for the detection of ATP have been reported.

Artur J. Moro reported a dipicolylamine (DPA)-Zn(Ⅱ) complex (probe 1) coupled with a naphthalimide chromophore using an ethylamine spacer for ATP detection in HEPES buffer [21-23]. As shown in Fig. 1, the fluorescence of DPA is significantly quenched in alkaline media due to the deprotonation of the aromatic amine in the excited state [24], once coordinating with Zn2+, the deprotonation of the aromatic amine can occur at neutral pH due to the interaction between Zn2+ and the aromatic NH group. On the other hand, the addition of Zn2+ to a buffered solution of DPA leads to an increase of fluorescence due to the decrease in the photoinduced electron transfer effect (PET) from the lone electron pair of the tertiary amine on DPA group to naphthalimide fluorophore. Combined with the two hands, the fluorescence signal only showed a modest enhancement. While in the presence of ATP, binding of ATP to Zn2+ of Probe 1 decreased the strength of the Zn2+-NH interaction, which inhibited the deprotonation of the aromatic amine of probe 1 and resulting in an obvious fluorescence recovery of probe 1 [25]. Probe 1 can specifically monitor ATP against other anions with a large Stokes' shift (85 nm) and large magnitude of fluorescence enhancement in 100% water system.

Download:
Fig. 1. Structure of probe 1 and possible recognition mechanism for ATP. Inset: Structure of DPA-Zn.

For biological and clinical applications, near infrared (NIR) fluorescent sensors are highly desirable because they can effectively avoid photodamage, scattering of light and serious interference from short wavelength emission of biological media [26-29]. Recently, two NIR ATP fluorescent sensors based on two self-assemblies of DPA-Zn(Ⅱ) derivatives (DPA-12@Zn(Ⅱ) (probe 2) and DPA-16@Zn(Ⅱ)) (probe 3) with a hydrophobic squaraine dye (SQ) were presented by our group [27, 28, 30-32]. As a hydrophobic dye, SQ was embedded into the spherical micelles of DPA-Zn(Ⅱ) derivatives to construct self-assembly, where hydrophilic DPA-Zn (Ⅱ) segments were distributed on the periphery of the micelles (Fig. 2) [33]. In the presence of ATP, the phosphate anion of ATP is attracted by DPA-Zn(Ⅱ) of the SQ-embedded micelles via the metal-anion coordination interaction. Meanwhile, ATP can interact with SQ located on the surface of micelles through electrostatic and π-π interactions. The multiple interactions enable the assembly to improve the selectivity for ATP. The two sensors could specifically track for ATP over other structurally similar nucleoside polyphosphates such as adenosine diphosphate (ADP), adenosine monophosphate (AMP), guanosine 50-diphosphate (GTP) and cytidine 50-triphosphoric acid (CTP). In addition, the platform has the potential application in monitoring the level fluctuation of ATP during the mitotic period. This is the first example of self-assembly of the DPA-Zn(Ⅱ) derivative with SQ that can be used for highly selective detection of ATP in living cells and for monitoring its level fluctuation during the mitotic period. In 2016, our group synthesized another NIR ATP sensor using silica nanoparticles functionalized with DPA-Zn(Ⅱ) recognized sites (SiNPs-DPA@Zn2+) (probe 4) for detection of ATP (Fig. 3) [34]. The DPA-Zn(Ⅱ) complex coupled with functional silica nanoparticles possessing many advantages provides proof-of-principle "seed crystals" for construction of supramolecular assemblies and platforms for ATP sensing with facile performance.

Download:
Fig. 2. The proposed ATP recognition mode of SQ-embedded self-assemblies of DPA-12@Zn(Ⅱ) (probe 2) and DPA-16@Zn(Ⅱ) (probe 3). Copied with permission [33]. Copyright 2016, Royal Society of Chemistry.

Download:
Fig. 3. Two known approaches (a and b) and the new strategy (c) of possible assembled assay between functionalized silica nanoparticle (SiNPs-DPA@Zn2+ or SiNPs-N+), SQ and ATP (probe 4). Copied with permission [34]. Copyright 2017, Nature Publishing Group.

Although zinc(Ⅱ) has good affinity to phosphate for ATP recognition, copper(Ⅱ) demonotrates more stronger interaction with phosphate [35]. Copper(Ⅱ)-phosphate interactions could provide a new approach for fluorescent sensing of ATP. Probe 5 (a 1:1:1 buffered mixture of ligand A, B and CuCl2) based on relatively simple fluorescent ligands that can form ternary complexes with copper and phosphates was designed by Kataev [36, 37] (Fig. 4). Probe 5 could chelate ATP through the copper(Ⅱ)-phosphate interaction and π-π interactions between adenine and anthracene, finally achieving ratiometric detection for ATP.

Download:
Fig. 4. The structure of Probe 5 (a 1:1:1 buffered mixture of ligand A, B and CuCl2). Reproduced with permission [37]. Copyright 2012, American Chemical Society.

Ga3+ ions were found to easily coordinate with oxygen atoms of hydroxyphenyl segments [38-40]. Combined with this discovery and previous progress at that time, a Ga3+ self-assembled fluorescent probe 6 was reported by our group, which can selectively recognize ATP with fluorescence enhancement from ADP, AMP and other structurally similar nucleoside triphosphates in vitro and in vivo [41] (Fig. 5). In addition, probe 6 (DHBO-Ga3+ (1:20)) can be applied to detect ATP-relevant enzyme activity. The tendencies of spectra changes upon addition of ATP are similar to DHBO with Ga3+, which suggested that ATP facilitates DHBO to interact with Ga3+ and stabilizes the self-assembly of metal-ligandanalyte. In addition, the selective π-π stacking interaction between the adenosine segment of ATP and DHBO ligand could also play an important role in the sensing process, which accounted for the selective recognition of ATP over other nucleoside triphosphates and pyrophosphate. Probe 6 was successfully applied to the detection of ATP in SKOV-3 cells with non-or very low cell cytotoxicity.

Download:
Fig. 5. Structure of DHBO and one of possible multi-membered arrays assembled between DHBO and Ga3+ ions (probe 6). Reproduced with permission [41]. Copyright 2015, Elsevier Ltd.

2.2. Sensors bearing pyrene groups to form excimers

The most interesting feature of pyrene derivatives is their capability to form excimers and to impart monomer and excimer fluorescence [42-44]. The π-π stacking interaction based on pyrene and adenine base of nucleoside triphosphates would affect the self-assembly of pyrene, resulting in the observed fluorescence change [35].

Based on this, Yoon et al. reported a new ATP selective fluorescent sensor (probe 7) bearing two pyrene moieties, which can display a unique ratiometric fluorescent change with ATP over other similar nucleoside triphosphates (Fig. 6) [1]. Ratiometric fluorescent sensors have an obvious advantage that they can be used to evaluate the analyte concentration and provide built-in correction from environmental effects [34, 45]. Moreover, probe 7 with a relatively simple imidazolium receptor can efficiently discriminate ATP over other nucleoside triphosphates through ionic hydrogen bonding between imidazolium (C-H)+ and triphosphate group, and π-π stacking interaction between pyrene and adenine base. The formed unique sandwich stacking of pyreneadenine-pyrene of probe 7 with ATP indicated that probe 7 is an excellent fluorescent sensor for investigations of ATP-relevant biological processes. Two pyrene-based zinc complexes (probes 8 and 9) were also synthesized and successfully applied to selectively detect and discriminate ATP from ADP by Yoon group in 2011 [46] (Fig. 7).

Download:
Fig. 6. Proposed binding between probe 7 with ATP. Reproduced with permission [1]. Copyright 2009, American Chemical Society.

Download:
Fig. 7. Structure of probes 8 and 9. Reproduced with permission [46]. Copyright 2011, Wiley-VCH.

Another pyrene derivative (probe 10) was reported by Yang and coworkers. The probe 10 (DPTB-IMI-EG) contains three section: A triarylboron luminogen dipyren-1-yl (2, 4, 6-triisopropylphenyl) borane (DPTB), di(1H-imidazol-1-yl)methane dication (IMI) and 1-ethoxy-2-(2-methoxyethoxy)ethane (EG) moieties (Fig. 8) [47]. The probe provides a novel strategy to achieve analyte-induced finite aggregation and conquer the well-known aggregationcaused quenching (ACQ), a short-range interchromophoric energy migration and transfer process [48, 49]. The fluorescence intensity of probe 10 dramatically increases (12-fold) in the presence of ATP at the excitation wavelength of 440 nm. Probe 10 can be further successfully applied to detect cellular expression of ATP in NIH/3T3 fibroblast cells and MDA-MB 231 cells by using fluorescence microscopy and FLIM methods.

Download:
Fig. 8. The indicator system between ATP and probe 10. Copied with permission [47]. Copyright 2014, Wiley-VCH.

2.3. Organic small molecule based on multiple interactions

The current methods for detection of ATP are mainly based on positive and negative charge interactions between negatively charged phosphates of ATP and positively charged recognition groups. Compared with the probes with the single interaction point, multi-site ones could selectively and stably detect ATP because multiple interactions could produce cooperative effect to achieve the specific recognition of ATP.

The rhodamine-based probe 11 bearing three responsive sites was prepared by Lin et al. (the left section of Fig. 9) [50-52]. The three responsive sites of the probe with ATP are covalently interactions between boronic acid and ribose, π-π interactions between xanthene and adenine, and electrostatic interactions between amino and phosphate groups, respectively. The three kinds of interactions perform cooperatively to facilitate the ring opening structure of the Probe 11, thereby a strong fluorescence signal being generated. About 18.1-fold turn-on fluorescent response appeared after addition of ATP (the right section of Fig. 9B). Owing to the multisite-binding strategy, probe 11 demonstrated excellent selectivity to ATP over ADP and AMP (the right section of Fig. 9C). Taking the dynamics into account, the sensing process could be achieved within 20 seconds, thus enabling in-time monitoring of intracellular ATP (the right section of Fig. 9D). More importantly, probe 11 was perfectly applied to monitor mitochondrial ATP level fluctuation in live cells [53].

Download:
Fig. 9. Left: A) Structure of ATP and ATP probes (probes 11, 12, 13 and 14). B) Proposed mechanism for sensing ATP. Right: UV absorption A) and fluorescence B) spectra of probe 11 (10 mmol/L) responding to ATP at intracellular concentrations (0–10 mmol/L). C) Fluorescence titration curve of probe 11 upon the addition of ATP, ADP and AMP. [Analytes] = 0–10 mmol/L. D) Time-dependence of the fluorescence intensity change before and after treatment of ATP. [ATP] = 5 mmol/L. Time interval: 10 s. F/F0 represents the fluorescence intensity ratio (590 nm) and F0 is the initial fluorescence intensity (590 nm) of probe 11 in the absence of nucleoside polyphosphates. Conditions: λex = 510 nm, [probe 11] = 10 mmol/L, Glycerol/Krebs buffer solution (60/40, pH 7.8). Copied with permission [52]. Copyright 2016, Wiley-VCH.

A smart off-on molecular scaffold/fluorescent probe 15 has been designed and synthesized by Arvind Misra et al. [54] (Fig. 10). The H-bonding and electrostatic interactions between triphosphate unit of ATP and piperazine N atoms of probe 15 (the ratio between ATP and probe is 2:1), the CH-π and π-π stacking between anthracene and purine rings worked together to form a synergistic effect. This synergistic effect suppressed the process of PET. At the same time, a fluorescent "turn-on" signal and a nakedeye sensitive blue-green color in the medium were observed.

Download:
Fig. 10. The structure of probe 15 and plausible mode of interaction between probe 15 and ATP. Reproduced with permission [54]. Copyright 2015, Elsevier Ltd.

A novel amphiphilic imidazolium-based probe containing a dansyl fluorophore and a long cetyl chain for ATP recognition has been developed by Cao and Kim groups [54]. In comparison, probe probe 17 bearing a short methyl chain was also prepared [55] (the left section of Fig. 11). Cao and Kim groups studied the self-assemble ability of probe probe 16 by measuring the surface tension of the sensor. The result showed that hydrophobicity of cetyl chain of probe probe 16 is strong enough to form micelles and aggregate. However, probe probe 17 with a short methyl chain hardly self-assemble. Probe 16 shows a weak emission at 535 nm due to dansyl-based internal charge transfer (ICT). Upon addition of different equivalent of ATP, the fluorescence emission of probe probe 16 in PBS (1 mmol/L, pH 7.4) is dramatically enhanced with a blue shift from 535 nm to 508 nm owing to π-π stack, hydrogen bonding electrostatic interaction between ATP and probe probe 16 (the right section of Fig. 11). The above phenomenon proves that ICT from the dimethyl amino group to the sulfonyl moiety is inhibited. As a result, the maximal fluorescence wavelength experiences a blue-shift. This novel probe was successfully used to sense ATP with a turn-on fluorescence signal. Furthermore, the detection of ATP was performed in absolute aqueous solutions and in vivo.

Download:
Fig. 11. Left: Structure of probe probe 16 and probe probe 17 and a cartoon illustration for ATP sensing. Right: The fluorescence emission of probe probe 16 (2.0 × 10-5 mol/L, λex = 360 nm) upon addition of various amounts of ATP in PBS (1.0 mmol/L, pH 7.4) Inset: Photograph of the solution color change of probe probe 16 before and after addition of 2 equiv. of ATP. Copied with permission [55]. Copyright 2017, Royal Society of Chemistry.

2.4. Sensors bearing polythiophene

Recently, in view of the signal amplification effect of conjugated polymers, polythiophene derivatives have been widely used as fluorophores of ATP sensors [56, 57]. In addition, conjugated polythiophene derivatives have good water-solubility, which provides a unique platform for sensing of biologically relevant targets [58, 59]. Polythiophene is a readily available and representative polymer, and its structure is sensitive to external stimuli, which is useful for highly sensitive detection of analytes [60]. On the other hand, ATP has been demonstrated to be versatile building block for the construction of supramolecular aggregates to induce significant spectral changes of polymer [61-63].

A water-soluble cationic polythiophene derivative (probe 18) has been reported by Shinkai et al., which can colorimetric and fluorescent responses to ATP [64]. The sensing mechanism was based on electrostatic and hydrophobic cooperative interactions between probe 18 and ATP (Fig. 12). The probe based on a conjugated polymer for detection of ATP can operate in aqueous solution with physiological pH value.

Download:
Fig. 12. Structure of probe 18. Reproduced with permission [64]. Copyright 2005, Wiley-VCH.

Lu et al. developed a sensitive fluorescent and colorimetric dual-modal Probe 19 for the detection of ATP (Fig. 13) [65-67]. On account of the electrostatic interactions, probe 19 and ATP gathered together to form polymer aggregation. The aggregation was further promoted by the inter chain π-π stacking interaction of the large aromatic plane anthracene groups on the side chain of probe 19. Upon addition of ATP, the maximal absorption peak of probe 19 gradually shifted from 450 nm to 588 nm (the right section of Fig. 13). The solution color changed from yellow to deep violet, which can be used for the colorimetric detection of ATP [53]. Correspondingly, the fluorescence intensity at about 585 nm was significantly enhanced. The limit of detection (D) of probe 19 for ATP was estimated to be 2.3 ×10-9 mol/L, far below the intracellular ATP concentration (1-10 mmol/L).

Download:
Fig. 13. Left: Schematic illustration of the detection of ATP using the self-promoting aggregation of probe 19. Right: UV-vis titration spectra of probe 19 (100 mmol/L) in trisHCl buffer solution with increasing amounts of ATP. [ATP] = 0, 4, 8, 12, 16, 20, 24, 28, 32, 36, 44, 52, 60, 68 and 80 μmol/L from top to down. Copied with permission [67]. Copyright 2015, Royal Society of Chemistry.

Three cationic conjugated polyelectrolytes (CPEs) based on a common poly(π-πhenylene ethynylene terthiophene) backbone with side chains of different lengths, named as probe 20, probe 21, and probe 22, were designed and synthesized by Jiang et al. [68] (Fig. 14). The three polymers were successfully applied to monitor ATP in HeLa cells. The results showed that the longer the length of the side chain, the stronger the capacity of probe to stain cell is. In addition, probe 21 was selected to test ATP in vitro owing to its stronger light absorption capability and longer maximum absorption wavelength among the three sensors.

Download:
Fig. 14. Structures of probe 20, probe 21 and probe 22. Reproduced with permission [68]. Copyright 2016, American Chemical Society

A water-soluble polythiophenes (probe 23) were reported by Lu et al. [69]. The probe could realize ratiometric detection of ATP, which is built on binding-induced modulation of fluorescence resonance energy transfer (FRET) cooperated with the aggregation-caused quenching (ACQ) (Fig. 15). Owing to unique signal amplification effect of conjugated polyelectrolyte (CPE), the detection limit for ATP in Tris-HCl buffer solution (2 mmol/L, pH 7.4) was as low as 2.9 ×10-8 mol/L.

Download:
Fig. 15. Structure of probe 23 and nucleoside polyphosphates, and the plausible interaction between probe 23 and ATP. Reproduced with permission [69]. Copyright 2017, Royal Society of Chemistry.

A novel cationic water-soluble polythiophene derivative of poly [N, N, N-trimethyl-3-(thiophen-3-yl) propenaminium chloride] (probe 24) was synthesized by Zhang et al. [70] (Fig. 16). Probe 24 showed sensitive, colorimetric detection and discrimination for ATP over other structurally similar nucleoside triphosphates.

Download:
Fig. 16. Structure of probe 24. Reproduced with permission [70]. Copyright 2013, Taylor & Francis.

2.5. Sensors containing biomolecules

Lately, sensors based on biomolecules possessing various merits, such as unique selectivity and sensitivity to analytes, good water-soluble, low cytotoxicity, fine membrane permeability, captured the attention of the researcher. The interactions between biomolecules modified with fluorophores and ATP produce the variation of fluorescence signal to detect ATP. Moreover, ATP contains multiple sites to react with biomolecules on sensors. Consequently, various sensors containing biomolecule are designed and used for ATP detection.

Strano et al. reported a single-walled carbon nanotubes (SWNT)/Luciferase (Luc) enzyme conjugated probe 25 (SWNTLuc) for detection of ATP, which is based on the bioluminescent reaction between luciferase, D-luciferin and ATP which served as an energy currency [71]. As shown in Fig. 17, the SWNT of probe 25 showed near-infrared (NIR) fluorescence and in turn it was almost completely quenched by oxyluciferin (oxyLrin) generating from the Luc-mediated bioluminescent reaction that involves selective ATP consumption. The turn-off NIR fluorescence of probe 25 could be rationalized as photoinduced excitedstate electron transferred from the nanotube conduction band to the lowest unoccupied molecular orbital (LUMO) of an adsorbing molecule of oxyLrin [72, 73]. The strategy provided a new optical sensing mechanism that a fluorescence of probe 25 is quenched by oxyLrin. Additionally, SWNTLuc is the first SWNT-based optical sensor for the detection of ATP in living cells.

Download:
Fig. 17. Illustration of the SWNTLuc (probe 25) for detection of ATP. Copied with permission [71]. Copyright 2010, Wiley-VCH.

Schmuck et al. reported an artificial nucleotide-binding peptide as a probe (probe 26), which contained a naphthalimide fluorophore and two symmetric peptidic arms equipped with a tailor-made anion-binding motif of the guanidiniocarbonyl pyrrole moiety for the detection of ATP (Fig. 18) [74]. A metal-free but biomolecule-based probe 26 showing turn-on fluorescence response and was applied to detect for ATP in vitro with high sensitivity and selectivity. Additionally, probe 26 was successfully used for the imaging of ATP in cells with low cytotoxicity and background fluorescence.

Download:
Fig. 18. Top: Molecular structure of probe 26. Bottom: Schematic representation for detection of ATP. Copied with permission [74]. Copyright 2017, Royal Society of Chemistry.

Two fluorometric sensors based on a tri-serine and tri-lactone scaffold with thiourea or sulfonamide moieties were reported by Anzenbacher Jr. et al. [75] (Fig. 19). Upon addition of ATP, probe 27 with thiourea showed an enhancement of fluorescence owing to the interaction between thiourea and ATP. However, probe 28 with sulfonamide showed quenching upon addition of ATP. The two sensors achieved quantitative analysis of ATP in human blood serum with high accuracy.

Download:
Fig. 19. Structures of probe 27 and probe 28 (A) and analytes (B). Reproduced with permission [75]. Copyright 2016, Royal Society of Chemistry.

Chan et al. developed a novel fluorescent chemosensor (probe 29) based on an anthracene moiety and ditopic cholic acid with C17 side chain. Probe 29 could selectively detected ATP through interaction between hosts and guests molecule (Fig. 20) [76]. Versatile hosts formed complexes with ATP via multiple hydrogen bonding interactions. The binding interactions can adjust the fluorescence of probe 29 via a photo-induced electron transfer (PET) process. Consequently, this sensor could selectively quantify ATP over other nucleoside triphosphates with turn-off fluorescence response.

Download:
Fig. 20. Structure of probe 29 and the proposed binding mode of it and ATP through multiple hydrogen bonding. Reproduced with permission [76]. Copyright 2008, Royal Society of Chemistry.

Versatile dumbbell molecule DM (probe 30) with 50-end phosphorylation was designed for sensitive monitoring of ATP by He et al. [77] (Fig. 21). Probe 30 was protected from the degradation of exo-nucleases after addition of ATP, which could modulate the fluorescence of probe 30. Obvious turn-on fluorescence appeared after that the protected probe binds to intercalation dye. The probe with 50-end phosphorylation was designed through cooperation with the ATP-dependent ligation reaction for selective detection of ATP. The detection limit (D) was found to be 4.8 pmol/L. Probe 30 was used to detect ATP in human serum and cell. In addition, the visual detection of ATP in droplet-based microfluidic platform was successfully applied.

Download:
Fig. 21. Illustration of the mechanism of DM (probe 30) for detection of ATP. Copied with permission [77]. Copyright 2016, Elsevier Ltd.

Unmodified cysteamine capped nanocrystalline cadmium sulfide quantum dots (Cys-CdS QDs) (probe 31) were designed by Ngeontae et al. (Fig. 22) [65, 78, 79]. Probe 31 sensitively recognized ATP with a turn-on fluorescence response and wide detection range (20-80 mmol/L) through positive and negative charge interactions. In addition, probe 31 was successfully designed and applied to determine ATP in urine samples.

Download:
Fig. 22. Illustration of the mechanism of probe 31 for detection of ATP. Copied with permission [79]. Copyright 2013, Elsevier Ltd.

Aptamer is an artifical single-stranded oligonucleotide (DNA or RNA), which could bound target analytes with high affinity and specificity. The affinity of ATP aptamer to ATP is stronger than that of its complementary nucleic acid. Combined with the highly specific interaction between aptamers and ATP, probes with ATP aptamer can detect ATP with high sensitivity and selectivity. The fluorescence response was based on the binding of ATP with the corresponding aptamer sequence and re-conformation of a new structure. A large number of aptamer probes were developed to detect ATP. Most of ATP probes with aptamer depend on the general detection mechanisms of the strong interaction between ATP and ATP aptamers. Therefore, only two representative ATP probes based on aptamer were discussed here.

Ju and Jun reported ATP aptamer coated on MoS2 nanoplates of a chlorine e6 (Ce6) to form a nanoprobe (probe 32, Fig. 23) for fluorescent imaging of intracellular ATP and ATP-controllable photodynamic therapy (PDT) [80]. Probe 32 alone nearly has no fluorescence due to the high quenching ability of MoS2 nanoplates via energy transfer [81]. In the presence of ATP, aptamer bound with ATP to form a folded and rigid structure, resulting in the release of the Ce6 from the MoS2 and in turn the recovery of fluorescence. Compared with widely used carbon nanomaterials such as graphene oxide, MoS2 nanoplates with smaller volume and much less hazardous show good permeability to cells and definite potential of application in biomedicine [82, 83]. ATP-mediated release of Ce6 led to the generation of 1O2 under laser irradiation at 660 nm to induce the death of tumor cells. A smart MoS2 nanoplate-based nanoprobe with low dark toxicity was successfully applied to fluorescent imaging of intracellular ATP and ATPcontrollable PDT, which presented an intelligent multifunctional theranostic platform for cellular ATP imaging and ATP-controllable cancer therapy with excellent efficiency.

Download:
Fig. 23. (a) Schematic illustration of the Ce6-aptamer loaded MoS2 nanoprobe (probe 32) for 1O2 production upon target binding. (b) Endocytosis of the nanoprobe for ATP imaging and ATP-mediated release of 1O2 in lysosomes under 660 nm laser irradiation, which leads to the lysosomal enzyme-induced cell death. The 1O2 induces the lysosomal membrane permeabilization and consequent release of lysosomal enzymes into the cytosol to trigger apoptosis. Copied with permission [80]. Copyright 2015, Royal Society of Chemistry.

Xiang developed a new label-free and fluorescent probe with hairpin aptamer (probe 33) for detection of ATP, which is based on exonuclease Ⅲ (Exo Ⅲ)-catalyzed target recycling (ECTR) amplification and indication of SYBR Green Ⅰ (SGI) [84]. Exo Ⅲ was employed to recycle to generate amplified fluorescent signal (Fig. 24). In the absence of ATP, probe 33 with sequence for ATP binding and five protruding mononucleotides at the 3' terminus was resistant to Exo Ⅲ digestion. In this case SGI generated a significant fluorescent signal at 520 nm due to its effective impaction into the stem of probe 33. However, the binding of ATP to the corresponding aptamer sequence induces probe 33 to be cleaved off by Exo Ⅲ to release ATP. The released ATP triggers the recycling process. The autonomous cyclic process led to the cleavage of a large number of probe 33 and in turn the release of SGI from the stems of probe 33, resulting in the turn-off fluorescent response for sensitive detection of ATP.

Download:
Fig. 24. Exo Ⅲ-catalyzed target recycling amplification for label-free and sensitive fluorescent detection of ATP. The association of ATP with the hairpin aptamer probe (probe 33) changed the structure of the probe and created catalytic sites for Exo Ⅲ to initiate the target ATP recycling process, which led to the digestion of a large number of probes and significantly suppressed fluorescent emission for sensitive ATP detection. Copied with permission [84]. Copyright 2014, Elsevier Ltd.

3. Conclusion

Among the various nucleoside triphosphates, ATP plays important role in energy supply. The change of its concentration could reflect the health of the living body to a certain extent. Consequently, a large number of fluorescent sensors for detecting ATP have developed in recent years. The majority of these sensors have been successfully applied to imaging in live cells and biological sample. In this review, we summarized the fluorescent sensors for detecting ATP from a viewpoint of classification where the sensors were divided into five types of ones according to the structures of the probes. Meanwhile, the design intention and advantages of each kind of probe for detection of ATP were reasonably generalized, which could inspire researchers to develop new creative probes for detecting ATP. We believe that the discovery of highly efficient and advanced probes for ATP sensing would be in very near further.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 21676218, 21476185, 21472016, 21272030), the Fundamental Research Funds for the Central Universities (Nos. 2014YB027, 2452015447, 2452013py014), Shaanxi Province Science and Technology.

References
[1]
Z. Xu, N.J. Singh, J. Lim, et al., J. Am. Chem. Soc. 131(2009) 15528-15533. DOI:10.1021/ja906855a
[2]
F.M. Ashcroft, F.M. Gribble, Diabetologia 42(1999) 903-919. DOI:10.1007/s001250051247
[3]
F.M. Impellizzeri, S.M. Marcora, C. Castagna, et al., Int. J. Sports Med. 27(2006) 483-492. DOI:10.1055/s-2005-865839
[4]
X.T. Shen, G. Mizuguchi, A. Hamiche, C. Wu, Nature 406(2000) 541-544. DOI:10.1038/35020123
[5]
P.B. Dennis, A. Jaeschke, M. Saitoh, et al., Science 294(2001) 1102-1105. DOI:10.1126/science.1063518
[6]
K.T. Bush, S.H. Keller, S.K. Nigam, J. Clin. Invest. 106(2000) 621-626. DOI:10.1172/JCI10968
[7]
S. Przedborski, M. Vila, Clin. Neurosci. Res. 1(2001) 407-418. DOI:10.1016/S1566-2772(01)00019-6
[8]
A. Makela, T. Kuusi, T. Schroder, Scand. J. Clin. Lab. Invest. 57(1997) 401-407. DOI:10.3109/00365519709084587
[9]
J.A. Cruz-Aguado, Y. Chen, Z. Zhang, et al., J. Am. Chem. Soc. 126(2004) 6878-6879. DOI:10.1021/ja0490424
[10]
L. Mora, A.S. Hernandez-Cazares, M.C. Aristoy, F. Toldra, Food Chem. 123(2010) 1282-1288. DOI:10.1016/j.foodchem.2010.05.072
[11]
P.J. Xie, M.L. Ye, Z.Y. Hu, et al., Chin. Chem. Lett. 22(2011) 1485-1488. DOI:10.1016/j.cclet.2011.09.008
[12]
Y.F. Huang, H.T. Chang, Anal. Chem. 79(2007) 4852-4859. DOI:10.1021/ac070023x
[13]
P. Yu, X. He, L. Zhang, L. Mao, Anal. Chem. 87(2015) 1373-1380. DOI:10.1021/ac504249k
[14]
S. Leng, Q.L. Qiao, Y. Gao, et al., Chin. Chem. Lett.(2017). DOI:10.1016/j.cclet.2017.03.034
[15]
K.D. Bhatt, D.J. Vyas, B.A. Makwana, et al., Chin. Chem. Lett. 27(2016) 731-737. DOI:10.1016/j.cclet.2016.01.012
[16]
Y.L. Xu, X.Y. Niu, H.L. Chen, S.G. Zhao, X.G. Chen, Chin. Chem. Lett. 28(2017) 338-344. DOI:10.1016/j.cclet.2016.10.003
[17]
A.P. de Silva, T.S. Moody, G.D. Wright, Analyst 134(2009) 2385-2393. DOI:10.1039/b912527m
[18]
H.T. Ngo, X. Liu, K.A. Jolliffe, Chem. Soc. Rev. 41(2012) 4928-4965. DOI:10.1039/c2cs35087d
[19]
A.J. Moro, P.J. Cywinski, S. Koersten, G.J. Mohr, Chem. Commun. 46(2010) 1085-1087. DOI:10.1039/B919661G
[20]
S.Y. Jiao, K. Li, X. Wang, et al., Analyst 140(2015) 174-181. DOI:10.1039/C4AN01615G
[21]
Z.Y. Wu, J.N. Cui, X.H. Qian, T.Y. Liu, Chin. Chem. Lett. 24(2013) 359-361. DOI:10.1016/j.cclet.2013.03.027
[22]
W. Feng, Q.L. Qiao, S. Leng, et al., Chin. Chem. Lett. 27(2016) 1554-1558. DOI:10.1016/j.cclet.2016.06.016
[23]
A.J. Moro, P.J. Cywinski, S. Korsten, G.J. Mohr, Chem. Commun. 46(2010) 1085-1087. DOI:10.1039/B919661G
[24]
M.S. Alexiou, V. Tychopoulos, S. Ghorbanian, et al., J. Chem Soc. Perkin Trans. 21(1990) 837-842.
[25]
M. Crego-Calama, D.N. Reinhoudt, Adv. Mater. 13(2001) 1171-1174. DOI:10.1002/(ISSN)1521-4095
[26]
J.R. Hou, D. Jin, B. Chen, et al., Chin. Chem. Lett.(2017). DOI:10.1016/j.cclet.2017.03.037
[27]
S.Y. Lin, H.J. Zhu, W.J. Xu, G.M. Wang, N.Y. Fu, Chin. Chem. Lett. 25(2014) 1291-1295. DOI:10.1016/j.cclet.2014.04.027
[28]
X. Fan, D. Zhang, H. Li, S. Sun, Y. Xu, Sensor. Actuators B:Chem. 245(2017) 290-296. DOI:10.1016/j.snb.2017.01.122
[29]
Y. Xu, B. Li, W. Li, et al., Chem. Commun. 49(2013) 4764-4766. DOI:10.1039/c3cc41994k
[30]
B. Li, W. Li, Y. Xu, et al., Chem. Commun. 51(2015) 14652-14655. DOI:10.1039/C5CC06086A
[31]
Y. Xu, B. Li, L. Xiao, et al., Chem. Commun. 49(2013) 7732-7734. DOI:10.1039/c3cc43223h
[32]
Q. Zhang, X. Dong, K.P. Wang, et al., Chin. Chem. Lett. 28(2017) 777-781. DOI:10.1016/j.cclet.2017.03.001
[33]
R. Feng, Y. Xu, H. Zhao, X. Duan, S. Sun, Analyst 141(2016) 3219-3223. DOI:10.1039/C6AN00646A
[34]
R. Feng, W. Shi, D. Wang, et al., Sci. Rep. 7(2017) 43491. DOI:10.1038/srep43491
[35]
V. Amendola, G. Bergamaschi, A. Buttafava, L. Fabbrizzi, E. Monzani, J. Am. Chem. Soc. 132(2010) 147-156. DOI:10.1021/ja9046262
[36]
A. Ojida, S. Park, Y. Mito-oka, I. Hamachi, Tetrahedron Lett. 43(2002) 6193-6195. DOI:10.1016/S0040-4039(02)01317-5
[37]
E. Kataev, R. Arnold, T. Ruffer, H. Lang, Inorg. Chem. 51(2012) 7948-7950. DOI:10.1021/ic300805q
[38]
C.J. Hastings, R.G. Bergman, K.N. Raymond, J. Chem. Eur. 20(2014) 3966-3973. DOI:10.1002/chem.201303885
[39]
R.M. McKinlay, P.K. Thallapally, J.L. Atwood, Chem. Commun.(2006), 2956-2958.
[40]
R.M. McKinlay, P.K. Thallapally, G.W.V. Cave, J.L. Atwood, Angew. Chem. Int. Ed. 44(2005) 5733-5736. DOI:10.1002/(ISSN)1521-3773
[41]
L. Xiao, S. Sun, Z. Pei, et al., Biosens. Bioelectron. 65(2015) 166-170. DOI:10.1016/j.bios.2014.10.038
[42]
J. Huang, Y. Wu, Y. Chen, et al., Angew. Chem. Int. Ed. 50(2011) 401-404. DOI:10.1002/anie.201005375
[43]
P. Thirupathi, J.Y. Park, L.N. Neupane, M.Y.L.N. Kishore, K.H. Lee, ACS Appl. Mater. Interfaces 7(2015) 14243-14253. DOI:10.1021/acsami.5b01932
[44]
Y. Zhang, Y. Fang, N.Z. Xu, et al., Chin. Chem. Lett. 27(2016) 1673-1678. DOI:10.1016/j.cclet.2016.04.011
[45]
J. Wu, W. Liu, J. Ge, H. Zhang, P. Wang, Chem. Soc. Rev. 40(2011) 3483-3495. DOI:10.1039/c0cs00224k
[46]
Z. Xu, D.R. Spring, J. Yoon, J. Chem Asian 6(2011) 2114-2122. DOI:10.1002/asia.201100120
[47]
X. Li, X. Guo, L. Cao, et al., Angew. Chem. Int. Ed. 53(2014) 7809-7813. DOI:10.1002/anie.201403918
[48]
W.Z. Yuan, P. Lu, S. Chen, et al., Adv. Mater. 22(2010) 2159-2163. DOI:10.1002/adma.v22:19
[49]
C. Zhang, Y. Yuan, S. Zhang, Y. Wang, Z. Liu, Angew. Chem. Int. Ed. 50(2011) 6851-6854. DOI:10.1002/anie.201100769
[50]
Q. Wang, Y. Feng, J. Jiang, et al., Chin. Chem. Lett. 27(2016) 1563-1566. DOI:10.1016/j.cclet.2016.02.021
[51]
H.R. Cheng, Y. Qian, Chin. Chem. Lett. 27(2016) 879-886. DOI:10.1016/j.cclet.2016.01.039
[52]
L. Wang, L. Yuan, X. Zeng, et al., Angew. Chem. Int. Ed. 55(2016) 1773-1776. DOI:10.1002/anie.201510003
[53]
Y. Chen, T. Wei, Z. Zhang, et al., Chin. Chem. Lett.(2017). DOI:10.1016/j.cclet.2017.05.010
[54]
P. Srivastava, S.S. Razi, R. Ali, et al., Biosens. Bioelectron. 69(2015) 179-185. DOI:10.1016/j.bios.2015.02.028
[55]
J.H. Zhu, C. Yu, Y. Chen, et al., Chem. Commun. 53(2017) 4342-4345. DOI:10.1039/C7CC01346A
[56]
L.P. Wang, G. Li, W.Z. Li, et al., Chin. Chem. Lett. 25(2014) 1620-1624. DOI:10.1016/j.cclet.2014.06.028
[57]
Q. Zhang, X. Dong, K.P. Wang, et al., Chin. Chem. Lett. 28(2017) 777-781. DOI:10.1016/j.cclet.2017.03.001
[58]
B. Liu, G.C. Bazan, Chem. Mater. 16(2004) 4467-4476. DOI:10.1021/cm049587x
[59]
B. Liu, G.C. Bazan, Proc. Natl. Acad. Sci. U.S.A 102(2005) 589-593. DOI:10.1073/pnas.0408561102
[60]
D.T. McQuade, A.E. Pullen, T.M. Swager, Chem. Rev. 100(2000) 2537-2574. DOI:10.1021/cr9801014
[61]
R. Iwaura, K. Yoshida, M. Masuda, et al., Angew. Chem. Int. Ed. 42(2003) 1009-1012. DOI:10.1002/anie.200390257
[62]
S.J. George, A. Ajayaghosh, P. Jonkheijm, Angew. Chem. Int. Ed. 43(2004) 3422-3425. DOI:10.1002/(ISSN)1521-3773
[63]
Z. Kejik, K. Zaruba, D. Michalik, et al., Chem. Commun(2006), 1533-1535.
[64]
C. Li, M. Numata, M. Takeuchi, S. Shinkai, Angew. Chem. Int. Ed. 44(2005) 6371-6374. DOI:10.1002/(ISSN)1521-3773
[65]
X.D. Xia, H.W. Huang, Chin. Chem. Lett. 25(2014) 1271-1274. DOI:10.1016/j.cclet.2014.03.008
[66]
F. Keshvari, M. Bahram, K. Farhadi, Chin. Chem. Lett. 27(2016) 847-851. DOI:10.1016/j.cclet.2016.01.022
[67]
D. Cheng, Y. Li, J. Wang, et al., Chem. Commun. 51(2015) 8544-8546. DOI:10.1039/C5CC01713K
[68]
Z. Chen, P. Wu, R. Cong, et al., ACS Appl. Mater. Interfaces 8(2016) 3567-3574. DOI:10.1021/acsami.5b06935
[69]
N. An, Q. Zhang, J. Wang, et al., Polym. Chem. 8(2017) 1138-1145. DOI:10.1039/C6PY02001A
[70]
X. Wang, Q. Feng, L. Wang, et al., Des. Monomers Polym. 17(2013) 26-32.
[71]
J.H. Kim, J.H. Ahn, P.W. Barone, et al., Angew. Chem. Int. Ed. 49(2010) 1456-1459. DOI:10.1002/anie.200906251
[72]
J.H. Kim, D.A. Heller, H. Jin, et al., Nat. Chem. 1(2009) 473-481. DOI:10.1038/nchem.332
[73]
B. Satishkumar, L.O. Brown, Y. Gao, et al., Nat. Nanotechnol. 2(2007) 560-564. DOI:10.1038/nnano.2007.261
[74]
D. Maity, M. Li, M. Ehlers, C. Schmuck, Chem. Commun. 53(2017) 208-211. DOI:10.1039/C6CC08386B
[75]
A. Akdeniz, M.G. Caglayan, I. Polivina, Anzenbacher P.Jr, Org. Biomol. Chem. 14(2016) 7459-7462. DOI:10.1039/C6OB01378C
[76]
H. Wang, W.H. Chan, Org. Biomol. Chem. 6(2008) 162-168. DOI:10.1039/B715086E
[77]
J. Chen, Y. Liu, X. Ji, Z. He, Biosens. Bioelectron. 83(2016) 221-228. DOI:10.1016/j.bios.2016.04.055
[78]
X.B. Gao, J. Yu, N. Li, H.Y. Yin, J.H. Yang, Chin. Chem. Lett. 18(2007) 1289-1292. DOI:10.1016/j.cclet.2007.07.010
[79]
W. Tedsana, T. Tuntulani, W. Ngeontae, Anal. Chim. Acta 783(2013) 65-73. DOI:10.1016/j.aca.2013.04.037
[80]
L. Jia, L. Ding, J. Tian, et al., Nanoscale 7(2015) 15953-15961. DOI:10.1039/C5NR02224J
[81]
C. Zhu, Z. Zeng, H. Li, et al., J. Am. Chem. Soc. 135(2013) 5998-6001. DOI:10.1021/ja4019572
[82]
W.Z. Teo, E.L.K. Chng, Z. Sofer, M. Pumera, J. Chem. Eur. 20(2014) 9627-9632. DOI:10.1002/chem.201402680
[83]
P. Huang, Z. Li, J. Lin, et al., Biomaterials 32(2011) 3447-3458. DOI:10.1016/j.biomaterials.2011.01.032
[84]
Y. Xu, J. Xu, Y. Xiang, R. Yuan, Y. Chai, Biosens. Bioelectron. 51(2014) 293-296. DOI:10.1016/j.bios.2013.08.002