Introduction

Copper ions (Cu2+), an essential heavy metal element for living organisms, are widely involved in a series of processes in biological systems, such as electron transfer and redox reactions. They play very significant roles as a catalytic cofactor in different cellular physiological processes [1]. A lack of copper in organisms can lead to growth and metabolic disorders, while excessive copper can also produce toxic effects. For example, neurodegenerative diseases including Parkinson’s, Myankos, Wilson’s, Alzheimer’s, and amyotrophic lateral sclerosis are caused by the disruption of copper metabolism balance in vivo [2]. Pyrophosphate ions (PPi, P2O74−), as the major anions in biological systems, are produced during DNA polymerization, enzymatic reactions, and cellular metabolism [3]. Published reports suggest that abnormal concentrations of PPi in living cells can cause some diseases, such as arthritis and chondrocalcinosis [4]. Driven by the great significance of Cu2+ and PPi, substantial efforts have been made and various strategies including electrochemistry [5], colorimetry [6], surface-enhanced Raman spectroscopy (SERS) [7], surface plasmon resonance (SPR) [8], and fluorescence [9] have been adopted. These methods supply effective and diverse approaches for the determination of Cu2+ and PPi in a large variety of cases. For example, Wang et al. have reported several fluorescent biosensors for the determination of heavy ions and PPi with a low detection limit and wide linear range [10,11,12,13]. However, some inevitable limitations, such as expensive instruments and sophisticated operating procedures remain. To tackle this problem, it is of significance to develop new methods that are low cost and portable.

Point-of-care testing (POCT) involves clinical and bedside tests performed at the bedside of a patient. It has attracted increasing attention because it is portable, low cost, user friendly, and on site [14, 15]. The most comprehensive POCT involves the use of a personal glucose meter (PGM), which is widely used by diabetes patients to monitor blood glucose levels at home [16]. It allows the values of the glucose concentration in the patient’s body to be read off easily through a drop of blood. Consequently, it has raised the quality of life for millions of diabetic patients from all four corners of the world. However, it usually detects only one object, glucose. To make use of a PGM with many comparable advantages for other targets, Lu et al. reported the first example of utilizing a PGM to detect nonglucose targets [17]. To date, various targets, including proteins [18], nucleic acids [19, 20], enzyme activity [21], SARS-CoV-2 [22], foodborne pathogenic bacteria [23], ochratoxin A [24], pesticides and veterinary drug residues [25], and heavy metal ions [26], have been detected by PGM. We have also summarized the application of a blood glucose meter from three aspects: biomedical analysis, food analysis, and environmental analysis [27]. However, there are only two articles in the literature about Cu2+ detection based on a PGM. In 2014, Xiang et al. developed a Cu2+-sensing system based on an invertase-modified magnetic bead signal amplification probe and a PGM [28]. In 2017, aided by a ligation DNAzyme releasing strategy and a PGM, Lv et al. developed a portable and sensitive Cu2+-sensing system [29]. To our knowledge, PGM-based sensors for selective Cu2+ and PPi have not been reported to date. Usually, multiple analytes are often limited by tedious sampling and time-consuming pretreatments. Sequential detection with a single sensor has attracted increasing attention due to its low cost, fast speed, and effective analysis.

Inspired by the above considerations, we first report a novel sequential detection of Cu2+ and PPi with a single sensor based on the use of a PGM as a readout tool and a magnetic sphere as the separation tool. The designed strategy is shown in Scheme 1. Magnetic Fe3O4 spheres are prepared by the hydrothermal method, and then capped with SiO2. After functionalization with streptavidin, invertase was modified with a DNA-named Cu substrate and Cu DNA enzyme to form a sensing system. Upon targeting Cu2+, Cu2+ will recognize the Cu DNA substrate and form an “on-off” signal switch, thereby resulting in the separation of invertase from the Fe3O4 nanospheres. In the presence of PPi, PPi will recognize Cu2+ to form a Cu2+-PPi complex, resulting in an “off-on” signal switch. This signal change was collected by a portable PGM. Therefore, a cost-effectiveness and portable biosensor for the consecutive detection of Cu2+ and PPi was built on the basis of the above “on-off-on” system. This biosensor will provide potential applications in the fields of molecular diagnosis and environmental monitoring.

Scheme 1
scheme 1

Schematic illustration for portable and quantitative monitoring of pyrophosphatase activity using DNAzyme with a personal glucose meter

Full size image

Experimental

Reagents

We purchased FeCl3.6H2O and sucrose from Sangon Biotech Co., Ltd. (China). CH3COONa.3H2O, ammonia, ethylene glycol, tetraethylorthosilane (TEOS), and 3-aminopropyl) triethoxysilane (APTES) were provided by Aladdin Industrial Co., Ltd. (China). 4-(N-Maleimidomethyl) cyclohexane-1-carboxylic acid 3-sulfo-N-hydroxysuccinimide ester sodium salt (sulfo-SMCC), 1-(3-(dimethylamino)-propyl)-3-ethylcarbodiimide hydrochloride (EDC), N-hydroxysulfosuccinimide sodium salt (sulfo-NHS), invertase (β-fructose-dase, from Baker’s yeast), Tris (2-car-boxyethyl) phosphine hydrochloride (TCEP), and Tween-20 were obtained from Sigma-Aldrich Co., Ltd. (USA). Reagents used in the polyacrylamide gel electrophoresis experiment and DNA sequences were provided by Sangon Biotechnology Co., Ltd. (Shanghai, China). The laboratory water was deionized water (≥ 18 MΩ·cm, Millipore).

Other buffer solutions were also employed:

  • Buffer A: 0.1 M sodium phosphate buffer including 0.1 M NaCl (pH 7.4)

  • Buffer B: 0.1 M sodium phosphate buffer including 0.1 M NaCl and 0.05% Tween-20 (pH 7.4)

  • Buffer C: 0.05 M HEPES including 1.5 M NaCl (pH 7.0)

The DNA sequences used are shown below:

  • Cu-Enz (5′-GGTAAGCCTGGGCCTCTTTCTTTTTAAGAAAGAAC-3′)

  • Cu-Sub (5′-biotin-(A)9AGCTTCTTTCTAATACGGCTTACC-(A)9(CH2)6SH-3′

Instruments

Field emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM) images were collected using a JSM-7500 field emission electron microscope and a Tecai G2 F20 S-TWIN (FEI, America) electron microscope. Fourier-transform infrared (FTIR) spectra were measured using a Nicolet Nexus FTIR 670 spectrometer (Thermo, USA) with a measurement range of 400~4000 cm−1. X-ray diffraction (XRD) patterns were measured using a MinFlex 600 X-ray diffractometer. The ultraviolet visible absorption spectrum was measured using a UV-2600 ultraviolet spectrophotometer (Shimadzu, Kyoto, Japan) or a Nanodrop One (Thermo Fisher Scientific, America). A glucose meter (ACCU-CHEK Active, Germany) was used to determine the glucose concentration.

Synthesis of Fe3O4-NH2 nanoparticles

The synthesis method for Fe3O4 nanoparticles was performed according to the alcohol-thermal method reported in the literature and was appropriately modified [30]. Specifically, FeCl3·6H2O (1.35 g) was dissolved in ethylene glycol (40 ml) with constant stirring to form a transparent solution. Then, 3.6 g NaAc and 1.0 g polyethylene glycol were added to the above solution. After stirring vigorously for 1 h, the above mixture was placed into a Teflon-lined stainless-steel autoclave and heated for 8 h at 200 °C. A cooling method was used to cool the films from the furnace temperature to room temperature. The precipitate was washed three times with ultrapure ethanol and water. The product was dried for 6 h at a temperature of 60 °C.

Next, 0.04 g Fe3O4 was placed in 20 ml ethanol and ultrasonically dispersed for 1 h. Then, 12 ml of ultrapure water and 2 ml of ammonia solution were poured into the solution as described above and transferred into a three-neck flask, and the flask was mechanically stirred in a water bath at 45 °C. Meanwhile, 400 μl TEOS was dissolved in 20 ml ethanol, with mechanical stirring for 10 min. Subsequently, the above mixture was poured into a three-neck flask and mechanically stirred for 6 h. The precipitate was washed three times with ultrapure ethanol and water. The Fe3O4-SiO2 product was dried for 6 h at a temperature of 60 °C.

Third, the obtained Fe3O4-SiO2 was added to 100 ml ethanol (98%, v/v) and mechanically stirred for 20 min. After 100 μl APTES was added, the solution as described above was mechanically stirred for 6 h. Then, the Fe3O4-NH2 nanoparticles were washed three times with ultrapure ethanol and water and dried for 6 h at a temperature of 60 °C.

Synthesis of Fe3O4-SA

Streptavidin was coupled to Fe3O4-NH2 nanoparticles by glutaraldehyde. Typically, 5 mg Fe3O4-NH2 nanoparticles were placed in buffer A (5 ml). After ultrasonication for 20 min, 200 μl glutaraldehyde (25%) was added and gently vortexed at 37 °C for 3 h. Subsequently, the solution as described above was washed with PBS four times and suspended in 4.5 ml PBS. After 1 mg/ml streptavidin (100 μg/ml) was added, the solution described above was gently vortexed at 37 °C for 12 h and then finally washed with buffer A.

Synthesis of the Cu sub-invertase

DNA-invertase was synthesized according to the literature procedures [17]. In brief, 0.5 mg sulfo-SMCC was mixed with 100 μl invertase (20 mg/ml) and then vortexed at room temperature for 2 h. Subsequently, excess insoluble sulfo-SMCC was removed by centrifugation (3000 rpm/min, 10 min). The supernatant was then subjected to ultrafiltration with a 100-kDa MW filter and termed solution A. Meanwhile, 60 μl Cu-Sub (20 μM) and 2 μl TCEP (3 mM) were added to 2 μl PBS (0.1 M, pH 5.5). After reacting at room temperature for 1 h, the solution described above was washed with a 30-kDa-MW filter by buffer A and termed solution B. Finally, solutions A and B were mixed and then reacted at room temperature for 48 h. After the solution was purified with a 100-kDa-MW filter using buffer A, the mixture was kept in 100 μl buffer A.

Synthesis of Cu-Sub-Fe3O4

Cu sub-invertase was added to the streptavidin-modified Fe3O4 solution (5 μM of the final concentration). After the solution was mixed gently at room temperature for 0.5 h, the Cu-Sub-Fe3O4 was washed with buffer A three times with a magnet.

Detection of Cu2+

First, Cu-Sub-Fe3O4 (1 μM), Cu-Enz (2 μM), 5 μl of 5 mM ascorbate, and various concentrations of Cu2+ were added to 200 μl buffer C for 2 h at 43 °C to form a sensing system. Then, the above system was washed with a magnet to remove the excess Cu-Enz. Subsequently, 5 μl sucrose (1 M) was added to the 10-μl sensing system. After reacting at room temperature for 40 min, a PGM was used to measure the glucose concentration.

Detection of PPi

First, 1 μM Cu-Sub-Fe3O4, 2 μM Cu-Enz, 5 μl of 5 mM ascorbate, Cu2+ (5 μM), and various concentrations of PPi were added to 200 μl buffer C for 3 h at 43 °C to form a sensing system. Then, we removed the excess Cu-Enz from the above system with a magnet. Afterwards, 5 μl sucrose (1 M) was added to the 10-μl sensing system. After reacting at room temperature for 40 min, a PGM was used to measure the glucose concentration.

Results and discussion

The principle of the fabricated biosensor

The framework of the fabricated biosensor for the determination of Cu2+ and PPi is schematically described in Scheme 1. It involves six principal processes: (1) synthesis of magnetic Fe3O4@SiO2 nanoparticles, (2) preparation of Cu-sub-invertase, (3) connection of Cu-sub-invertase with magnetic Fe3O4@SiO2 nanoparticles, (4) formation of the magnetic sensing probe, (5) fabrication of an “off-on” sensing system and PGM detection, and (6) fabrication of an “on-off” sensing system and PGM detection. In step 1, magnetic Fe3O4@SiO2 nanoparticles are first synthesized by the hydrothermal method and then modified with a SiO2 shell. In step 2, the above magnetic Fe3O4 nanoparticles are modified with streptavidin by glutaraldehyde. In step 3, magnetic Fe3O4@SiO2 nanoparticles are connected with DNA through the specific reaction between streptavidin and biotin. In step 4, magnetic sensing probes are fabricated by the hybridization reaction between Cu-EnZ and Cu-sub. After magnetic separation, sucrose is added to the above system to prompt sucrose into glucose, and then the glucose signal is detected by a PGM. At this time, the glucose signal is very low. In step 5, with the existence of Cu2+, it can cut DNA Cu-sub. After magnetic separation, invertase left the magnetic nanoparticles and stayed in the sensing system to prompt sucrose into glucose, and then the glucose signal was detected by a PGM. This will catalyze the conversion of sucrose into glucose and thus obtain a high signal. This is an “off-on” sensing system. Subsequently, with the existence of PPi, a Cu2+-PPi complex was formed by the binding reaction between Cu2+ and PPi. After magnetic separation, invertase remained in the system, and thus, a low signal was obtained. This is an “on-off” sensing system. Through the use of this “off-on-off” sensing system, Cu2+ and PPi were detected consecutively with one sensor.

Characterization of Fe3O4-NH2 nanoparticles

To realize our design, the most basic precondition for the successful fabrication of DNAzyme-Fe3O4 nanosystems is to gain a highly efficient magnetic probe. TEM, XRD, and FTIR were carried out to characterize the morphology of Fe3O4-NH2 nanoparticles. As shown in Fig. 1A, Fe3O4 nanoparticles have a better dispersion, particle size uniformity, and homogeneous spherical structure. Compared with bare Fe3O4 nanoparticles, a white silicon shell was discovered outside the black Fe3O4 nanosphere, indicating the successful preparation of Fe3O4@SiO2 nanospheres. Figure 2B gives the XRD pattern for Fe3O4 nanoparticles. Eight sharp diffraction peaks are observed located at 2θ = 30.1°, 35.6°, 37.1°, 43.1°, 53.4°, 56.99°, and 62.6°, which belong to the (220), (311), (222), (400), (422), (511), (440), and (533) crystal planes of the cubic spinel structures for Fe3O4 nanoparticles, respectively, from a comparison with JCPDS file Card No. 99-0073. This is consistent with the previous reported literature [31], indicating that Fe3O4 nanoparticles were successfully synthesized using this method. Finally, FTIR spectra for Fe3O4, Fe3O4-SiO2, and Fe3O4-NH2 were studied and are shown in Fig. 1C. Compared with line a, line b and line c show a new absorbance at 789 cm−1 and 1110 cm−1, respectively, corresponding to the plane stretching vibration peaks and bending vibration absorption kurtosis of Si–O–Si, indicating successful preparation of the SiO2 shell. The stretching vibration peak for N–H at 1612 cm−1 was discovered at line c, indicating that the amino groups were successfully connected to the surface of SiO2-coated magnetic particles [32].

Fig. 1
figure 1

A TEM of Fe3O4@SiO2 and B XRD of Fe3O4, and C FTIR of Fe3O4, Fe3O4-SiO2, and Fe3O4-NH2

Full size image
Fig. 2
figure 2

A UV-Vis absorbance spectroscopy of Fe3O4@SiO2 (a) and Fe3O4-DNA (b). B The PGM signal intensity under different conditions: (a) buffer solution, (b) Fe3O4-DNA, (c) magnetic sensing probe, (d) magnetic sensing probe in the presence of Cu2+, (e) magnetic sensing probe in the presence of Cu2+ and PPi

Full size image

Characterization of the Fe3O4-DNA nanocomposite

To achieve our design, the key process for the successful fabrication of DNAzyme-Fe3O4 nanosystems is to connect DNA Cu-Enz with Fe3O4 nanoparticles. To certify the successful preparation of the magnetic sensing probe, UV-Vis absorbance spectroscopy was conducted for Fe3O4@SiO2 and Fe3O4-DNA, as demonstrated in Fig. 2A. A stark absorption peak located at 260 nm is equivalent to that for DNA that appeared for Fe3O4@SiO2 after modification of DNA. In addition, these results also demonstrate the successful synthesis of Fe3O4@SiO2-DNA.

The feasibility of the fabricated biosensor

A series of control experiments were performed and the PGM readings were collected to demonstrate the feasibility of the fabricated biosensor. As demonstrated in Fig. 2B, the PGM signals for buffer solution (histogram a), Fe3O4-DNA (histogram b), and the magnetic sensing probe (histogram c) were low due to no obvious glucose production, indicating that this biosensor has a low background signal. However, in the presence of Cu2+, a much higher signal was generated (histogram d). This indicated that the invertase was introduced into the magnetic sensing probe with the addition of Cu2+ because Cu2+ could cut Cu-Enz DNA and release the invertase into the sensing system to form an “off-on” sensing system. The generated invertase will catalyze sucrose conversion into glucose, and thus, signals can be captured by a PGM. Subsequently, with the addition of PPi and Cu2+, a lower signal was detected (histogram f). This is because Cu2+ is captured by PPi and then the magnetic sensing probe is finally restored shortly to form an “on-off” system. Invertase was removed with the magnetic sensing probe. No amount of glucose was generated, and the signal was very small. This result indicated the feasibility of the designed magnetic sensing probe for sequential detection of Cu2+ and PPi.

Optimization of the experimental conditions

For the purpose of realizing the best sensing performance, several experimental parameters for the designed magnetic sensing probe and the detection procedures were optimized. First, the preparation of a magnetic sensing probe was the key step for sequential detection of Cu2+ and PPi. Thus, the Cu-sub to Cu-Enz ratio was investigated and is demonstrated in Fig. 3A. As the ratio increased, the PGM signal increased and then reached a plateau after 2:1. This indicated that a ratio of 2:1 is sufficient for the formation of the sensing system, and this value was thus chosen for the optimized condition. Subsequently, the cleavage time for Cu2+ with DNAzyme influences the sensitivity of our sensing system. Thus, the cleavage time of Cu2+ with DNAzyme between 20 and 180 min was studied. As shown in Fig. 3B, the signal was increased with the cleavage time until 120 min. A longer cleavage time did not result in any obvious change. To save the assay, 120 min was selected as the cleavage time for Cu2+ with DNAzyme. Furthermore, the enzyme catalysis time also affects the system signal. Different catalysis times ranging from 5 to 50 min were investigated. As shown in Fig. 3C, the signal increased at a steady rate upon increasing the enzyme catalysis time until 40 min. Thus, a reaction time of 40 min was used in further steps.

Fig. 3
figure 3

The effect of the ratio of Cu-sub and Cu-Enz (A), the enzymatic cleaving time of Cu2+ with DNAzyme (B), and enzymatic catalysis time (C) on the PGM signal

Full size image

Detection of Cu2+

According to the above scheme, the concentration of Cu2+ was determined by the final concentration of released probes. Under the optimized conditions, the analytical performance of the proposed “off-on-off” sensor for Cu2+ was investigated. As shown in Fig. 4A, the PGM signals increased with the Cu2+ concentration. A linear relationship between the PGM signal and the logarithmic Cu2+ concentration was achieved in the range of 0.01–5 μM, as demonstrated in Fig. 4B. The linear equation for the PGM signal = 42.3811 + 5.1396lgC[Cu2+] (R = 0.9917). The detection limit was 10 nM. Compared with other reported references, our sensing system exhibits comparative sensitivity, as demonstrated in Table 1. Furthermore, the selectivity of this biosensor was investigated by replacing Cu2+ with other ions, such as Zn2+, Pb2+, Fe2+, Mg2+, Ca2+, Ba2+, and Cd2+. No obvious changes in the PGM signal were achieved compared to that of Cu2+ at the same concentration, indicating the excellent selectivity of our biosensor. Subsequently, the stability of the sensing probe was evaluated by detecting the PGM signal over 14 days. The probe was prepared and stored in a refrigerator at 4 °C. After 2 weeks, the PGM signal of the biosensor kept about 91.2% of its initial value. The slight decrease in the signal response may be due to the gradual denature of SA and invertase. This is sufficient for the practical application of Cu2+ biosensors.

Fig. 4
figure 4

A PGM signals of the sensor system with various concentrations of Cu2+, B the corresponding linear relationship between the signal and the concentration of Cu2+ from 10 nM to 5 μM. C Selectivity of the sensor system for Cu2+ against Zn2+, Pb2+, Mg2+, Ca2+, Ba2+, and Ca2+. D Stability of the sensor system for Cu2+ determined by 14 days

Full size image
Table 1 Comparison of our “off-on-off” sensor with reports from other references
Full size table

Detection of PPi

Furthermore, the feasibility of the sensing system for PPi detection with the support of Cu2+ was investigated. Figure 5A demonstrates the PGM signal of the sensor system for different concentrations of PPi in the presence of 5.0 μM Cu2+. The PGM signals decrease with increasing PPi concentration. A linear relationship between the PGM signal and the PPi concentration is achieved in the range from 0.5 to 10 μM. The linear equation for the PGM signal = 15.0454 − 1.3163C[PPi] (R = 0.9920). The detection limit was determined to be 500 nM. Compared with other reported references, our sensing system exhibits comparative sensitivity, as shown in Table 2. Furthermore, the selectivity of this biosensor was investigated by replacing PPi with other potential interferences including PO43−, HPO42−, H2PO4−, HCO3−, CO32−, SO42−, and F−. No obvious changes in the PGM signal were observed compared to that for PPi at the same concentration, indicating the excellent selectivity of our biosensor. This is sufficient for the practical application of PPi biosensors.

Fig. 5
figure 5

A The corresponding linear relationship between the PGM signal and concentration of PPi from 0.5 to 10 μM. B Selectivity of the sensing system for PPi against other potential interferences

Full size image
Table 2 Comparison of our “off-on-off” sensor with reports from other references
Full size table

Detection of Cu2+ and PPi in real samples

To certify the potential application of our fabricated “off-on-off” biosensor for the determination of Cu2+, tap water samples from Hangzhou Dianzi University with Cu2+ added were tested using the standard addition method. As demonstrated in Table 3, recoveries were obtained in the range of 96.8% and 116%. Moreover, human urine from Hangzhou Cancer Hospital was selected as an example for PPi detection. As demonstrated in Table 4, recoveries were obtained in the range of 96.6% and 109.6%, indicating excellent practical application ability in real analytical samples.

Table 3 Sensing system results obtained for real tap water samples
Full size table
Table 4 Sensing system results obtained for human urine samples
Full size table

Conclusion

In summary, a portable, quantitative, and sequential monitoring system for copper ions and pyrophosphate based on a DNAzyme-Fe3O4 nanosystem and glucometer readout was fabricated. This “off-on-off” sensing system has the advantages of low cost, portability, and user friendliness. Satisfactory results were achieved for this biosensor for the detection of Cu2+ in real serum samples and PPi in human urine samples. Therefore, we anticipate that this biosensor has potential for application in clinical biological analysis and may lead to online, real-time sensing, and can be even employed for personal use in families.