Introduction

Alkaline phosphatase (ALP) is an essential enzyme in phosphate metabolism to catalyze the dephosphorylation process of nucleic acids, proteins, and some small molecules [1,2,3,4]. As the most common human alkaline phosphatase, ALP has been identified as an important biomarker in the diagnosis of many diseases. The abnormal level of ALP in serum is closely related to various diseases such as breast and prostatic cancer [5, 6], bone disease [7], liver dysfunction and diabetes [8, 9]. Therefore, significant efforts have been devoted to explore new method for the detection of ALP.A number of technologies have been employed for the detection of ALP, such as colorimetry [10,11,12], surface enhanced resonance Raman scattering [13, 14], fluorescence [15,16,17], electrochemistry [18]. Because of its fast analysis and high sensitivity, fluorescence method had become a hot topic for the detection of ALP owing to their high sensitivity, cost–effectiveness, simplicity, and convenience. Many fluorescent chemosensors have been successfully applied to detect ALP activity based on organic fluorescent dyes [19, 20], conjugated polyelectrolytes [21, 22], and inorganic semiconductor quantum dots (QDs) [23, 24]. However, organic dyes and QDs in practical applications can be limited by relative poor photostability (for organic fluorophores) or toxicity (e.g.QDs). There is a need to develop sensitive, simple and nontoxic methods for ALP activity monitoring.

Carbon dots (C-dots), one of the C-based materials, are superior to quantum dots and organic dyes in terms of excellent physicochemical and photochemical properties, good biocompatibility, low toxicity and easy to preparation [25,26,27,28,29]. B and N, the neighboring elements of carbon (C) in the periodic table, have atomic radii similar to C, which make it possible to efficiently modulate the properties of C-dots after doping. Most of previous methods introduce two or more compounds as the starting materials to synthesize B and N doped C-dots. Single starting material to synthesize B and N co-doped C-dots, containing B and N simultaneously, can simplify the synthesis process and reduce the byproducts [30]. β-Cyclodextrin (β-CD) can form host-guest complexes with a wide range of hydrophobic guest species through noncovalent interactions [31]. Previous works reported β-cyclodextrin serve as a modulator to greatly improve the catalytical activity of nanomaterial, because they can form complexes with hydrophobic compounds to offer a pocket for the substrate recognition [32,33,34]. However, there are only limited work focusing on the synthesis and application of β-cyclodextrin based C-dots [35,36,37,38,39,40].

Herein, we introduce 3-aminophenylboronic acid (APBA) as starting material to fabricate B and N co-doped carbon dots (N/B-C-dots) with high quantum yield through one-pot hydrothermal method without further passivation. Then, we synthesized the cyclodextrin functionalized B and N co-doped carbon dots (β-CD-N/B-C-dots) using mono-6-thio-β-CD as modifier and N/B-C-dots as precursors. With the aid of β-CD, the detection performance of N/B- C-dots is excellent because the cavity of β-CD offered a pocket for substrate recognition (Scheme 1). Mono-6-thio-β-CD was directly introduced on the surfaces of the N/B-C-dots for ALP activity through host-guest recognition. The p-nitrophenylphosphate (PNPP) as a common substrate in colorimetric determination of ALP was employed as the ALP substrate in this sensing system. The maximum absorption wavelength of its ALP reaction products (p-nitrophenol, PNP) can well overlap with excitation spectra of the N/B-C-dots, leading to the efficient quenching of N/B-C-dots because of the inner filter effect (IEF). The method shows many merits including rapidity, low cost, high sensitivity, and excellent selectivity, providing a new insight on the application of N/B-C-dots to develop the facile and sensitive biometric technology.

Scheme 1
scheme 1

Mechanism of the assay

Full size image

Materials and methods

Reagents and chemicals

APBA, alkaline phosphatase, p-nitrophenylphosphate (PNPP), p-nitrophenol (PNP), Thrombin, glucose oxidase (GOx) and p-nitrophenylphosphate were purchased from Aladdin Chemical Co., Ltd. (Shanghai, China, http://aladdin.company.lookchem.cn/). Mono-6-SH-β-CD was purchased from Shandong Binzhou Zhiyuan Bio-Technology Co., Ltd. (Shandong, China, http://www.bzzysw.com). Sodium hydroxide (NaOH), ascorbic acid (AA), cysteine (Cys), glutathione (GSH), lysine (Lys), tryptophan (Try) and glycine (Gly) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China, http://www.shreagent.lookchem.com).

Preparation of N/B-dots

The N/B-dots were synthesized by a hydrothermal method [28]. In a typically experimental procedure, 0.1 g of APBA was dissolved in 10 mL ultrapure water, followed by adding 0.2 mL of 0.1 M NaOH under stirring. Then, the clear and homogeneous solution was transferred into the Teflon-lined autoclave chamber and heated to 180 °C for 4 h. After cooling down to room temperature, the solution was centrifuged at 10000 rpm (5595 rcf) for 10 min to remove large precipitates. The final purified N/B-dots were stored at 4 °C for further use.

Preparation of β-CD-N/B-C-dots

Typically, 0.5 mL of prepared N/B-C-dots was diluted to 5 mL, and then 50 μL of 2 M NaOH and 11.35 mg 6-SH-β-CD were added and homogenized by stirring. The mixture was heated at 70 °C for 4 h. The final solution was stored at 4 °C in refrigerator for use in the next step.

Characterizations of β-CD-N/B-C-dots

Transmission electron microscopy (TEM) images were obtained on a Tecnai G20 microscope (FEI, America). Fourier-transform infrared (FT-IR) spectra were recorded on FT-IR spectrophotometer (Perkin Elmer, America). UV–vis absorption spectra (UV–vis) were performed on Lamber35 UV spectrometer (Perkin Elmer, America). The fluorescence measurements were recorded on LS55 fluorescence spectrometer (Perkin Elmer, America).

IFE-based fluorometric determination of ALP

The IFE based fluorescent ALP activity assay was performed under the following procedures. A total of 10 μL of ALP with activities from 0 to 114 U·L−1 (or 10 μL of serum sample) was added into the reaction system (Tris-HCl, pH =8.0), which consisted of 80 μM PNPP and 0.1 μM MgSO4. The reaction solution was incubated at 37 °C for 20 min, after that the mixture was transferred to 1 mL of C-dots solution and then subjected to fluorescence spectral measurements at the excitation wavelength of 400 nm.

Results and discussions

Synthesis and characterization of β-CD-N/B-C-dots

The N/B-C-dots were synthesized by a hydrothermal method [28]. We selected the APBA as the nitrogen and boron-containing starting material for the synthesis of N/B-C-dots. Then, we synthesized the cyclodextrin functionalized B and N co-doped carbon dots (β-CD-N/B-C-dots) using mono-6-thio-β-CD as modifier and N/B-C-dots as precursors. To obtain the optimal synthetic conditions, the influence of reactants concentrations including NaOH and CD, reaction time, and reaction temperature on the fluorescence intensity of mono-6-thio-β-CD modificated N/B-C-dots (β-CD-N/B-C-dots) were investigated, respectively (Fig. S1). Typically, 0.5 mL of prepared β-CD-N/B-C-dots was diluted to 5 mL, and then 50 μL of 2 M NaOH and 11.35 mg 6-SH-β-CD were added and homogenized by stirring. The mixture was heated at 70 °C for 4 h. The final solution was stored at 4 °C in refrigerator for use in the next step.

The morphological structure of β-CD-N/B-C-dots characterized by TEM is displayed at Fig. 1a. The TEM image demonstrates that the β-CD-N/B-C-dots shows well-dispersed without obvious aggregation with the average size is 5 nm. The fluorescence spectra indicate that the optimal excitation and emission wavelengths of β-CD-N/B-C-dots are 400 nm and 500 nm, respectively (Fig. 1b). As shown in Fig. S2 a, the strong peak of 235 nm is ascribed to n → π* transition of C = C-N, and the peak of 290 nm is attributed to π → π* transition of C = C bond. The surface composition of the β-CD-N/B-C-dots is investigated in Fig. S2 b, the peak at 3375 cm−1, 1621 cm−1 and 1372 cm−1 in the FTIR curve of N/B-C-dots are shifted to 3391 cm−1, 1653 cm−1 and 1405 cm−1 after the modification of β-CD, demonstrating the formation of β-CD-N/B-C-dots.

Fig. 1
figure 1

a TEM images of the β-CD-N/B-C-dots, b Fluorescence excitation and emission spectra of the β-CD-N/B-C-dots

Full size image

Principle of ALP activity assay based on IFE

Introduction of heteroatoms into carbon nanomaterials has been explored to tune the conduction/valence band position of doped carbon material resulting in altered functions. For this purpose, nitrogen was widely used as electron donors, whereas boron was used as an electron acceptor. It is confirmed through various reports that excitons of carbon, emissive traps on C-dots, the quantum confinement effect, aromatic moieties, oxygen contacting groups, free zigzag sites, and edge defects contribute to the fluorescence. In addition to this, we also believe that N-doping as well as B doping also remarkably enhance the fluorescence of C-dots. In this study, PNPP was employed as the substrate of ALP to indirectly determine the activity of ALP based on IFE. IFE phenomenon is due to the absorption of the excitation or emission by absorbers in the detection system when the absorption spectra of the absorbers overlap with the excitation or emission spectra of fluorophores. As shown in Fig. 2a, the maximum UV-vis absorbance of PNPP (300 nm) has no effect on fluorescence intensity of β-CD-N/B-C-dots (500 nm) and the PNPP hydrolysis product (PNP) by ALP has a great molar absorptivity at 400 nm, which has a good overlap with the excitation spectrum of β-CD-N/B-C-dots. To further confirm the feasibility of our design for ALP detection, we recorded the fluorescence spectra of β-CD-N/B-C-dots mixed with PNPP and PNP (excitation peaks at 400 nm). As we expected, the addition of PNP (0.5 mM) quenches the fluorescence intensity of β-CD-N/B-C-dots at 500 nm significantly, while the same addition of PNPP quenches a little (Fig. 2b).

Fig. 2
figure 2

a UV-vis absorption spectra of PNPP and PNP solutions under alkaline conditions, b Fluorescence spectra of the β-CD-N/B-C-dots in the presence of PNPP and PNP (excitation peaks at 400 nm)

Full size image

IFE-based fluorescence assay for ALP activity

To obtain the optimum reaction conditions for ALP activity detection, the incubation time was chosen for investigation. As shown in Fig. 3a, the incubation time for quenching efficiency reaching the maximum are both about 20 min though in different concentration of ALP (25, 50 and 100 U·L−1). Then we measured ALP activity at the best incubation time. In Fig. 3b, with the increasing of ALP activity, the fluorescence intensity of β-CD-N/B-C-dots probe decreases continuously. A good fitted regression line between the relative fluorescence intensity (F0 - F) and ALP concentrations is obtained in the range of 0.003 to 5.5 U·L−1 (Fig. 3c–3d), with a linear regression equation as Y = 15.13X + 216.58 (R2 = 0.997). The detecting platform provides an ultralow detection limit of 0.0003 U·L−1. The performance of the β-CD-N/B-C-dots based assay is compared with literature reported fluorescence assay towards ALP. From Table S1, we can see the performance of our assay was better than literature report. In addition, compared with previous work by You’s group [41], because of the introduction of β-CD, the wider and more polar β-CD secondary faces are exposed to bind suitable guest species such as PNPP in the β-CD annulus and in close proximity to the catalytic surface [42] and thereby enhance the probability of catalytic oxidation of PNPP by β-CD-N/B-C-dots.

Fig. 3
figure 3

a The relationship between incubation time and quenching efficiency in different concentration of ALP, b Fluorescence spectra of β-CD-N/B-C-dots with ALP (0–114 U·L−1), c The plot of fluorescence responses of β-CD-N/B-C-dots (excitation/emission peaks at 400/500 nm), d Fluorescence response of β-CD-N/B-C-dots against the concentration of ALP from 0.003 to 5.5 U·L−1 (excitation/emission peaks at 400/500 nm)

Full size image

Selectivity of the method

To evaluate the specificity of the IFE-based assay for ALP, several potential interfering compounds were investigated under the same conditions. 10 μL of various small molecules including 50 mg·L−1 BSA and HSA, 0.1 mM Try, Lys, Cys, Gly, AA and GSH, 10 U·L−1 Gox, thrombin and ALP was added into β-CD-N/B-C-dots (dilution with Tris-HCl) with Mg2+ (0.1 μM) ion and PNPP (80 μM). After incubated for 20 min at 37 °C, the fluorescence spectrum was recorded under 400 nm excitation. As shown in Fig. 4, at the same class of concentration, all the other small molecules can not significantly affect the fluorescence signals of the β-CD-N/B-C-dots while ALP has an almost 100% quenching effect on the fluorescence intensity of β-CD-N/B-C-dots, indicating the excellent selectivity and applicability of the detecting strategy.

Fig. 4
figure 4

Selectivity of β-CD-N/B-C-dots to different substances: 50 mg·L−1 BSA and HSA, 0.1 mM Try, Lys, Cys, Gly, AA and GSH, 10 U·L−1 Gox, thrombin and ALP

Full size image

Conclusion

We demonstrate a simple approach to synthesize N/B-C-dots by one-pot hydrothermal method and further obtain β-CD-N/B-C-dots modified fluorescent probe by using mono-6-thio-β-CD as modifier. This modified fluorescent probe was developed for the specific recognition and quantitative detection of ALP in a cost-effective and time-saving way based on the inner filter effect (IFE) of N/B-C-dots. Besides, mono-6-thio-β-CD was directly introduced on the surfaces of the N/B-C-dots for ALP activity through host-guest recognition to improve the sensitivity of the detection system. We believe that this method paves the way for the further exploration and practical application in bioanalysis fields.