English

• The accurate detection of biomarkers is essential for the diagnosis and prognosis of diseases and the development of precision medicine [1-3]. DNA glycosylases and structure-specific nucleases have been recognized as potential biomarkers due to their important role in DNA replication and repair to hold the genetic stability and integrity of the organism [4-7]. The DNA glycosylases such as UDG and hAAG could specifically recognize and excise the damaged sites from DNA backbone by base excision repair pathway followed by reinstalling the undamaged base, thereby repairing the damaged base site [8]. The structure-specific nuclease, FEN1 was more efficient at binding and catalyzing the excision of the 5’ overhanging DNA flap during lagging-strand DNA synthesis for DNA replication and repair [9]. However, the abnormal expression of DNA glycosylases and FEN1 are strongly correlated with the onset and progression of various cancers, such as lung cancer [10,11], breast cancer [12,13] and epithelial ovarian cancer [14]. Moreover, the occurrence and development of cancers are often associated with simultaneous aberrant expression of multiple DNA glycosylases and FEN1 [12,13,15]. Importantly, it is well known that the activity of enzymes extracted from living organisms gradually decreased with the increase of time [16]. Thus, it is of great interest to detect multiple DNA glycosylases and FEN1 simultaneously, providing sufficient evidence for early diagnosis and treatment monitoring of cancer.

  Nowadays, a variety of methods have been reported for simultaneous detection of multiple biomarkers based on the amplification strategy coupled with different detection method including high-performance liquid chromatography (HPLC) [17], cationic polymers [18], lanthanide labeling coupled with ICP-MS [19], fluorescent analysis [3] and single molecule analysis [20]. However, these methods suffer from the problems of the relatively large sample consumption (20 µL) [17], the cumbersome steps for the synthesis of broad-spectrum polymer quencher [18], the tedious lanthanide tags labeling process [19], the different sensitivity causing by different fluorescent labels [3], and the fluorescent labels with spectral overlap [3,20]. Thus, it is desirable to develop a simple and sensitive method for simultaneously detecting more enzymes with small amount of biological sample.

  CE is one of the most important separation techniques with many advantages such as high separation efficiency, quick analysis, cost-effective way, automation, less sample and reagent consumption [21-23], and it has been used as a powerful micro analytical platform for the detection of multiplex miRNAs. For example, Krylov's group demonstrated a method for accurate quantitation miRNAs in cell lysis solution with CE on the basis of hybridization between tagged DNA probes and miRNA targets [24]. Zhang's group demonstrated multiplex immuno-PCR coupled with CE-LIF method for the assay of multiple surface cluster of differentiations on exosomes [25]. However, they suffered from complicated synthesis of DNA probes with varying size peptide drag tags [24] and the non-baseline separation as well as long analysis time (1 h for a single separation) [25]. Although Li's group successfully separated the encoding DNA probes with different lengths for the assay of multiple methylated CpG sites by CE, expensive custom-made POP-4 polymer was used to dynamic coat on the surface of the wall of capillary to separate the different length DNA probe [26]. Moreover, capillary zone electrophoresis (CZE) is the simplest and most used CE separation mode, but it could not be used to separate the protein directly, because the inner wall of capillary could adsorb protein, resulting in poor detection reproducibility [27]. And it is desirable to develop a CZE method for the baseline separation DNAs to achieve the assay of multiple biomarker proteins.

  The specific DNA and RNA fragments can be amplified at a certain temperature by isothermal nucleic acid amplification technique which enables to transform biomarker protein into DNA with high sensitivity [28-31]. Recently, some nucleic acid amplification methods including strand displacement amplification [32], rolling circle amplification [33], exponential isothermal amplification [34] and Endonuclease Ⅳ (Endo Ⅳ)-assisted signal amplification [35] have been developed to improve the detection signal. Despite the enhanced sensitivity, they suffer from the troublesome circle-template synthesis [36], multistep reaction and diverse enzymes [36,37], inevitable nonspecific amplification and relatively high background [36,37]. Endo Ⅳ-assisted signal amplification could be used for the conversion and amplification of signals with the merits of simple design, high efficiency and selectivity [35]. And terminal deoxynucleotidyl transferase (TdT) enables the repetitive addition of various mononucleotides to the 3′-OH end of DNA sequences without the requirement of any DNA templates, being to fabricate versatile sensing methods [38]. Thus, we developed the simultaneous detection of multiple tumor-associated enzymes in cell extracts based on the integration of CZE with Endo Ⅳ-assisted template-free amplification. Notably, the different amplification products with the same FAM-fluorescence label could avoid the different sensitivity causing by different fluorescent labels. The method can be used to detect multiple enzymes in cell extracts, discriminate normal cells from cancer cells, evaluate kinetic parameters and identify potential inhibitors. It has good potential for simultaneously detecting other enzymes system.

  The principle of the proposed method for the simultaneous detection of three biomarkers enzymes is shown in Scheme 1: (1) The specific excision of double-branched substrate by FEN1, UDG and hAAG, (2) TdT-initiated extension coupled with Endo Ⅳ-assisted cyclic production of different length DNAs labeled with FAM-fluorophore and (3) the separation of the amplification products by CE. A double-branched substrate containing multiple target enzymes recognition sites is composed of three single strands (S_U, S_D, S_T), and the 3′ ends of which are modified with amino groups to prevent non-specific extension. Meanwhile, three different fluorescent probes were modified with apurinic/apyrimidinic (AP) sites in the middle position and labelled with a fluorophore (FAM) at their 5′ ends. FEN1 could recognize and cleave the 5′- flap ssDNA segment in the double-branched substrate, producing a 5′-flap strand with 3′-OH terminus (green color, Scheme 1). Similarly, hAAG could recognize and excise the hypoxanthine base (Ⅰ) in the double-branched substrate to produce an AP site. In the presence of UDG, the uracil (U) in the substrate could be specifically recognized and cleaved to form an AP site. And then both of the AP sites produced by hAAG and UDG could be excised by Endo Ⅳ to produce 3′-OH end single strands, respectively (purple and red color, Scheme 1). And then all of the 3′-OH end single strands could be extended by TdT polymerase to form poly-T sequences. Subsequently, different signal probes (19 nt, 22 nt and 26 nt) for FEN1, hAAG and UDG could hybridize with their respective extension product to form partially complementary double strand DNA (dsDNA), and the signal probes in the dsDNAs could be cleaved by Endo Ⅳ at the position of AP site, releasing the fluorophore labeled DNAs (6 nt, 11 nt and 15 nt) and the extension products. And then the extension products could bind to other free signal probes to form dsDNA, inducing the next cyclic cleavage of signal probes and releasing more different length DNAs labeled with fluorophores. At last, the released fluorophore labeled DNAs can be separated and detected by CE, enabling the simultaneous detection of the three enzymes in the homogeneous system.

  Scheme 1

  

  Scheme 1.  Schematic illustration of the CE-based fluorescence signal amplification strategy. (1) FEN1 product, (2) hAAG product, (3) UDG product, (4) FEN1 probe, (5) hAAG probe and (6) UDG probe.

  DownLoad: Full-Size Img PowerPoint

  In order to verify the feasibility of the method, Gel electrophoresis was carried out to investigate the cleavage and extension reaction under different conditions (Fig. S1 in Supporting information). As can be seen from Fig. S1A, the double-branched substrate (Fig. S1A, lane 7) migrated slower than the single strands (Fig. S1A, lanes 1-3) and the partially complementary double strand DNA (dsDNA) (Fig. S1A, lane 5 and lane 6). In the presence of UDG and the double-branched substrate, two new bands (14 nt and c) (Fig. S1B, lane 3) with faster mobility compared to the double-branched substrate band (Fig. S1B, lane 1a) can be observed, with 14 nt band corresponding to product of cleavage reaction generated by UDG and c band corresponding to the double-branched substrate intermediate. Likewise, two distinct new bands of 8 nt and d can be visualized in the presence of hAAG (Fig. S1B, lane 4), indicating that hAAG could cleave the substrate to form the single strand (8 nt) and partial complementary dsDNA (d). Moreover, the addition of FEN1 induces the appearance of the bands of 9 nt and e (Fig. S1B, lane 5), indicating that FEN1 could cleave the substrate to form the single strand (9 nt) and partial complementary dsDNA (e). In the presence of the three enzymes, four new bands of 8 nt, 9 nt, 14 nt and b (Fig. S1B, lane 2) were observed, with the three single strand bands corresponding to the products of cleavage reaction by UDG, hAAG and FEN1 and b band corresponding to the double-branched substrate intermediate. Moreover, the addition of TdT to the different cleavage products, respectively, resulted in the disappearance of multiple DNA strand bands, as well as the appearance of new extension products with well-defined bands (Fig. S1C, lanes 1-4). However, the band of double-branched substrate did not disappear in the absence of three enzymes (Fig. S1C, lane 5). These results indicated that the cleavage products could initiate extension reaction to form poly (T) sequences.

  CE-LIF detection platform was also constructed to analyze the different length DNA amplification products. As shown in Fig. 1, only the FEN1 probe peak appeared in the absence of FEN1 (Fig. 1A, light purple curve). However, a new peak appeared in the presence of FEN1 accompanied by the decrease of the peak area of FEN1 probe, demonstrating that FEN1 induced amplification reaction to generate the new DNA strand with fluorophore (Fig. 1A, bluish violet curve). Similarly, only the hAAG probe peak appeared in the absence of hAAG (Fig. S2A in Supporting information, orange curve), as well as a new peak appeared in the presence of hAAG accompanied by the exhaustion of the hAAG probe (Fig. S2A, purple curve). Moreover, only the UDG probe peak appeared in the absence of UDG (Fig. S2B in Supporting information, cyan curve) as well as a new peak appeared in the presence of UDG (Fig. S2B, green curve). We further investigated the simultaneous detection of the three enzymes in the homogeneous system, as shown in Fig. 1B, only three peaks of the probes are present in the absence of target enzymes (Fig. 1B, orange curve). However, three new product peaks appeared when three enzymes existed simultaneously (Fig. 1B, pink curve). These results demonstrated that target enzymes could initiate the amplification reaction to produce different length DNAs which could be separated by CE.

  Figure 1

  

  Figure 1.  Feasibility analysis. Electropherograms of different samples: (A) The amplification product with (bluish violet curve) and without (light purple curve) FEN1, Electrophoresis separation condition: BGE, 1× TAE, pH 8.5; uncoated fused-silica capillary: 48.7 cm (38.5 cm to detector) × 50 µm ID. Applied voltage: 25 kV. Cartridge temperature: 25 ℃. (B) The amplification products with (pink curve) and without (orange curve) FEN1, hAAG and UDG. Electrophoresis separation condition for B: BGE, 4× TAE, other conditions were the same as the separation conditions in (A).

  DownLoad: Full-Size Img PowerPoint

  In order to obtain the best CE separation condition, FEN1, hAAG and UDG probes were used as the model analytes in Fig. 2A. In theory, CZE separation is based on the difference in electrophoretic mobility which determined by the difference in charge to mass ratio between analytes, and different length DNA strands could not be separated by CZE due to the same charge to mass ratio. However, the resolution of three probes increased gradually with the increase of borate buffer concentration in the range of 1-50 mmol/L with CZE separation, and the baseline separation could be achieved at 50 mmol/L borate buffer as shown in Fig. 2A. Similarly, the same results as that in borate buffer were also obtained in phosphate buffer system (Fig. S3 in Supporting information). Moreover, the resolution of the mixture of probes and amplification products also increased with increase of tris hydroxy methyl aminomethan, acetic acid, and ethylene diamine tetra-acetic acid (TAE) buffer concentration from 2 × to 4 × TAE buffer, and the baseline separation could be achieved in 4× TAE buffer (Fig. 2B). This behavior is most likely due to the opposition to electrophoresis alignment caused by Brownian motion [39], and Brownian motion induced the shorter analytes displaying a larger deviation from perfect electrophoresis alignment than being observed for longer analytes in the electric field force [40]. It is probably that the shorter the DNA strand length, the more obvious the Brownian motion, so the shorter DNA strand have a greater deviation from the direction of electrophoresis alignment. Moreover, the Joule heat gradually increased with the increase of buffer concentration in capillary, resulting in the increase of DNA strand Brownian motion. Under the positive voltage, the negatively charged DNA strand will migrate to the capillary inlet (anode), and the longer DNA strand migrated fast due to decreased hydrodynamic resistance resulting from their greater alignment along the electrophoresis alignment. The electroosmotic flow (EOF) caused by the inner wall of the capillary resulted in the net flow toward the outlet (cathode), and the DNA strands will still ultimately flow toward the cathode, but the longer DNA strands will migrate to the cathode late. Moreover, the electrophoretic migration time gradually increased with the increase of the buffer concentration, and the 5^th and 6^th peaks for hAAG and UDG probes with 4.5 × TAE buffer system did not appear at 40 min due to the dramatical decrease of EOF in Fig. 2B. Also, the peak shape gradually broadened with the increased electrophoretic migration time. Thus, compared with the separation effect of different buffer systems, 4× TAE buffer system was selected as the optimum electrophoresis separation condition.

  Figure 2

  

  Figure 2.  Optimization of electrophoresis separation conditions. The electropherogram obtained under different buffer system: (A) Borate buffer system, pH 8.8; (1) FEN1 probe, (2) hAAG probe, (3) UDG probe. (B) TAE buffer system, pH 8.5; (1) FEN1 product, (2) hAAG product, (3) UDG product, (4) FEN1 probe, (5) hAAG probe, (6) UDG probe. Other CE separation conditions were the same as those in Fig. 1A.

  DownLoad: Full-Size Img PowerPoint

  The dependency of the amplification products peak area on the concentration of the corresponding enzymes was investigated at the optimal experimental conditions (Figs. S4 and S5 in Supporting information). The electropherograms obtained in different enzymes concentration are shown in Fig. 3A. As can be seen from Fig. 3B, the peak area of amplification product increased gradually with the increasing FEN1 enzyme concentration. To examine the linear relationship for quantitative detection, we further plotted the FEN1 enzyme concentration against the peak area (A) of its product and obtained the linear equation between the logarithmic (log) value of FEN1 enzyme concentration and peak area. As shown in the inset in Fig. 3B, the corresponding calibration plot of A versus the logarithm of FEN1 concentration is well linear. Moreover, we plotted the enzyme concentration for hAAG and UDG against their product peak area to obtain the linear equations between the log value of both of the enzyme concentration and their peak area (Fig. S6 in Supporting information). As can be seen in Table S2 (Supporting information), the correlation coefficients (R) calculated were between 0.9950 and 0.9970 over the concentration ranges (0.08-160 U/mL for FEN1, 2.5-250 U/mL for hAAG, and 0.0004-2.5 U/mL for UDG). The limit of detection (LOD) was 0.07, 2.40 and 2.1 × 10⁻⁴ U/mL for FEN1, hAAG and UDG, following the rules of signal-to-noise ratio of 3. The results demonstrated that the method can be used to simultaneously quantify the concentration of different enzymes.

  Figure 3

  

  Figure 3.  The sensitivity and selectivity for the strategy. (A) Electropherograms of FEN1, hAAG and UDG in a defined range. (B) The linear relationship between the peak area and concentration of FEN1. (C) The selectivity of the method. T7 Exo (250 U/mL), λ Exo (250 U/mL), BSA (5 mg/mL), hOGG1 (250 U/mL), TDG (250 U/mL), FEN1 (160 U/mL) + hAAG (250 U/mL) + UDG (2.5 U/mL).

  DownLoad: Full-Size Img PowerPoint

  The selectivity of this method was also investigated in the presence of some interferent proteins. Exonucleases (T7 exo, λ exo), bovine serum albumin (BSA) and human 8-oxoguanine-DNA glycosylase 1 (hOGG1) and thymine DNA glycosylase (TDG) were selected as interfering substances. As illustrated in Fig. 3C, the significant product peak area appeared in the presence of the three target enzymes. However, the peak area of products was nearly the same as that of the control in the presence of the interfering substances. These results indicated that our method has good selectivity in distinguishing three target enzymes from other interfering substances.

  In addition, we investigated the intra-/inter-day deviations of peak time and area to evaluate the repeatability of this method. The amplification products and signal probes were used as the subjects for analysis in the presence of FEN1(80 U/mL), hAAG (200 U/mL) and UDG (2.0 U/mL). As shown in Tables S3 (Supporting information), the RSDs of peak time and peak area for different analytes were as follows: 2.50%-4.37% and 3.24%-7.18% (inter-day); 1.37% -2.71% and 1.43%-3.02% (intra-day); 4.28%-6.08% and 4.16%-7.57% (column to column), respectively. The results of RSDs indicated that the method had good repeatability for quantifying three target enzymes activity.

  The three enzymes are often considered as potential tumor biomarkers because they are generally present at higher level in tumor cells than in normal cells [41,42]. Evaluation of the inhibitory effect of inhibitors on these enzymes in vitro can effectively screen some chemicals used as the antitumor drugs [42]. Therefore, aurintricarboxylic acid (ATA) and uracil glycosylase inhibitor (UGI) were selected as the model inhibitors of FEN1 and UDG, respectively. As shown in Fig. 4A, FEN1 activity decreased significantly with the raising amount of ATA in the range of 0-10 µmol/L, indicating that the FEN1 activity could be effectively inhibited. According to the relationship between the inhibitor concentration and the peak area, the half-maximum inhibition concentration (IC₅₀) of ATA was 0.80 µmol/L, which was consistent with that acquired by six-fold cascade amplification method for FEN1 activity detection (0.83 µmol/L) [37]. Moreover, the inhibitory effect of UGI on UDG activity was also performed (Fig. S7 in Supporting information), and the IC₅₀ value of UGI was 0.065 U/mL, consistent with that in the CRISPR/ Cas12a system coupled with enhanced SDA (0.079 U/mL) [43]. In addition, the Z-factor in the inhibitor screening system established with FEN1 as the target was 0.902, and in the inhibitor screening system established with UDG as the target was 0.921. And the Z-factor in both of the above systems was >0.5. These results indicated our method could be used to estimate the inhibitory effect of drugs on different enzymes and it can be considered as a highly efficient tool for drug screening.

  Figure 4

  

  Figure 4.  Inhibition and enzyme kinetic analysis. (A) Relationship between the relative activity of FEN1 and the concentration of ATA. (B) Enzyme kinetic curve for FEN1 (80 U/mL) assay in the substrate concentration range of 0-400 nmol/L.

  DownLoad: Full-Size Img PowerPoint

  Enzyme kinetics analysis could be used as a method to assess the magnitude of the affinity between enzymes and substrates. Fig. 4B showed the kinetic plot corresponding to FEN1 enzyme obtained by adding different concentrations of substrate (0-400 nmol/L) to the system. According to Michaelis−Menten equation v = v_max[S]/(K_m + [S]) (where v_max is the maximum initial velocity, and [S] is the concentration of DNA substrate, and K_m is the Michaelis−Menten constant.), the K_m value of 85.55 ± 15.95 nmol/L for FEN1. As shown in Fig. S8 (Supporting information), the K_m value of 98.81 ± 9.80 nmol/L for hAAG and 80.06 ± 10.31 nmol/L for UDG, consistent with the value of 39.06 nmol/L for FEN1 obtained by dual exponential amplification-based fluorescent method [37], 47.85 nmol/L for hAAG obtained by the electrochemical assay [31] and 63.2 ± 7.4 nmol/L for UDG obtained by target enzyme-initiated Endo Ⅳ-assisted bicyclic cascade signal amplification method [44].

  The enzymes activity in cells gradually decreased once they leave the culture environment, and it is important to simultaneously detect three enzymes activity, ensuring uniformity in the determination time and operation, making the determination results more informative. In order to investigate the practical application capability of the method, Hela cells were selected as the model biological sample for detection. As shown in Fig. 5A, the peak area of amplification products corresponding to three enzymes increased as the concentration of the cell extract increased, and the peak area of the amplification product related to FEN1 enzyme versus the logarithm of the extracted protein concentration was plotted to obtain the relevant linear equation (Fig. 5B). The linear equations between the peak area of amplification products for hAAG and UDG enzymes versus the logarithm of the extracted protein concentration were also shown in Fig. S9 (Supporting information). In a certain concentration range, the three enzymes showed good linearity, and their linear equations and limits of detection are listed in Table S4 (Supporting information). In an attempt to verify whether there is a large difference in the level of the three enzymes in tumor cells and normal cells, we selected HeLa cell (tumor cell) and 293T cell (normal cell) as experimental subjects to determine the concentration of three enzymes in different intercellular and nuclei, comparing them with control and inactivated Hela cells. As shown in Figs. 5C and D, the concentration of three enzymes were significantly higher in tumor cell compared to normal cell, as the same results in the previously reports [37,45]. It suggested the capability of our method could perform simultaneous assay of multiple enzymes in complex biological samples.

  Figure 5

  

  Figure 5.  Practical application of the method. (A) Electropherograms of three enzymes in different concentration of Hela cell extract. Numbers 1-6 as labeled in Fig. 2B. The concentration of cell extract (µg/mL), A: 0.0082; B: 0.082; C: 0.82; D: 8.23; E: 82.26; F: 822.6. (B) The relationship between the peak area and logarithmic value of Hela nuclear extract concentration. (C) Electropherograms and (D) histogram for comparison of enzyme level in different cell extract, respectively. The total amount of both 293T cell and Hela cell extract is 2 × 10⁶.

  DownLoad: Full-Size Img PowerPoint

  In this work, we constructed a CE detection platform combined with single FAM-fluorescent labeled and template-free amplification system for the simultaneously quantifying multiple enzymes with the detection limit of 0.07 U/mL for FEN1, 2.40 U/mL for hAAG and 2.1 × 10⁻⁴ U/mL for UDG. In addition to showing good analytical performance, the method was also applied to evaluate kinetic parameters and identify potential inhibitors. Meanwhile, the method is valuable to achieve the successful detection of multiple enzymes in complicated cell extract simultaneously, avoiding the incongruity in the determination time and operation resulting in the different enzymes activity. Compared with the reported methods for simultaneous detection of multiple enzymes in Table S5 (Supporting information), this method has the obvious superiorities: (1) The use of signal amplification strategy coupled with CE detection platform could simultaneously detect three repair enzymes in complicated cell extracts. (2) This amplification reaction could be rapidly performed under isothermal conditions without template, quencher, tedious surface modification and complicated material synthesis, greatly simplifying operation steps and reducing costs. (3) The amplification products could be separated effectively by CE without the complicated modification of the capillary inner wall or labeling different tails on signal probes for separation. (4) The different amplification products with the same FAM-fluorescence label could avoid the different sensitivity causing by different fluorescent labels. (5) The amplification strategy was simply designed and easily incorporated into the CE detection platform, facilitating the construction of a versatile platform for simultaneously detecting multiple DNA repair enzymes, with extensive applications in the clinical diagnosis and therapy of disease.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Huige Zhang: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Methodology, Investigation, Data curation, Conceptualization. Wei Chen: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Yuyan Huang: Validation, Data curation. Mingfang Wu: Software, Data curation. Hongli Chen: Writing – review & editing, Project administration, Funding acquisition. Cuiling Ren: Software, Funding acquisition. Xiaoyan Liu: Software, Funding acquisition. Haixia Zhang: Writing – review & editing, Supervision, Project administration, Conceptualization.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 21874060 and 22174058, U21A20282) and the Science and Technology program of Gansu Province (No. 22JR5RA476).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110721.

  
  
• 1. [1]

     Y.C. Zhao, J.Z. Xiang, H. Cheng, X.L. Liu, F. Li, Biosens. Bioelectron. 194 (2021) 113581. doi: 10.1016/j.bios.2021.113581

  2. [2]

     L.M. Yang, X.H. Yin, B. An, F. Li, Anal. Chem. 93 (2021) 1709–1716. doi: 10.1021/acs.analchem.0c04308

  3. [3]

     Y. Zhang, C.C. Li, B. Tang, C.Y. Zhang, Anal. Chem. 89 (2017) 7684–7692. doi: 10.1021/acs.analchem.7b01655

  4. [4]

     M.H. Liu, C.R. Wang, W.J. Liu, et al., J. Mater. Chem. B 10 (2022) 3277–3284. doi: 10.1039/d2tb00491g

  5. [5]

     S.E. Tsutakawa, S. Classen, B.R. Chapados, et al., Cell 145 (2011) 198–211. doi: 10.1016/j.cell.2011.03.004

  6. [6]

     Q. Zhang, J. Hu, D.L Li, et al., Anal. Chim. Acta 1298 (2024) 342395. doi: 10.1016/j.aca.2024.342395

  7. [7]

     H.G. Zhang, L.L Wang, Y. Xie, et al., Analyst 144 (2019) 3064–3071. doi: 10.1039/c9an00200f

  8. [8]

     L.L. Wang, H.G. Zhang, W. Chen, et al., Chin. Chem. Lett. 33 (2022) 3603–3612. doi: 10.1016/j.cclet.2021.10.053

  9. [9]

     M.S. Zaher, F. Rashid, B. Song, et al., Nucleic Acids Res. 46 (2018) 2956–2974. doi: 10.1093/nar/gky082

  10. [10]

      T. Paz-Elizur, J. Natl. Cancer Inst. 95 (2003) 1312–1319. doi: 10.1093/jnci/djg033

  11. [11]

      M. Sato, L. Girard, I. Sekine, et al., Oncogene 22 (2003) 7243–7246. doi: 10.1038/sj.onc.1206977

  12. [12]

      L. Xu, J.L. Qu, N. Song, et al., Oncol. Rep. 44 (2020) 2443–2454. doi: 10.3892/or.2020.7812

  13. [13]

      Z.Y. Wang, S.Q. Teng, N.N. Zhao, et al., Chem. Commun. 60 (2024) 3075–3078. doi: 10.1039/d4cc00408f

  14. [14]

      T.M.A. Abdel-Fatah, R. Russell, N. Albarakati, et al., Mol. Oncol. 8 (2014) 1326–1338. doi: 10.1016/j.molonc.2014.04.009

  15. [15]

      Y. Leitner-Dagan, Z. Sevilya, M. Pinchev, et al., J. Natl. Cancer. Inst. 104 (2012) 1765–1769. doi: 10.1093/jnci/djs445

  16. [16]

      J. Sabotiű, J. Kos, Appl. Microbiol. Biotechnol. 93 (2012) 1351–1375. doi: 10.1007/s00253-011-3834-x

  17. [17]

      T. Qi, C. Song, J. He, et al., Anal. Chem. 92 (2020) 5033–5040. doi: 10.1021/acs.analchem.9b05301

  18. [18]

      J.Y. Chen, T. Jin, J.F. Li, et al., ACS Appl. Bio. Mater. 4 (2021) 820–828. doi: 10.1021/acsabm.0c01362

  19. [19]

      Q. Kang, M. He, B.B. Chen, G.Y. Xiao, B. Hu, Anal. Chem. 93 (2020) 737–744. doi: 10.3390/polym12040737

  20. [20]

      L.J. Wang, L.P. Pan, X. Zou, J.G. Qiu, C.Y. Zhang, Anal. Chem. 94 (2022) 17700–17708. doi: 10.1021/acs.analchem.2c04624

  21. [21]

      D. Zhou, L.J. Bao, H.Q. Long, et al., Chin. Chem. Lett. 36 (2025) 110093. doi: 10.1016/j.cclet.2024.110093

  22. [22]

      W.J. Zhang, J. Hong, L. Yang, et al., Chin. Chem. Lett. 35 (2024) 108695. doi: 10.1016/j.cclet.2023.108695

  23. [23]

      H.G. Zhang, J.H. Zhu, S.D. Qi, N. Yan, X.G. Chen, Anal. Chem. 81 (2009) 8886–8891. doi: 10.1021/ac901474j

  24. [24]

      L. Hu, A.S. Stasheuski, D.W. Wegman, et al., Anal. Chem. 89 (2017) 4743–4748. doi: 10.1021/acs.analchem.7b00707

  25. [25]

      N. Singh, L. Huang, D.B. Wang, N. Shao, X.E. Zhang, Anal. Chem. 92 (2020) 10569–10577. doi: 10.1021/acs.analchem.0c01464

  26. [26]

      F.X. Su, L.M. Wang, Y.Y. Sun, et al., Chem. Sci. 6 (2015) 1866–1872. doi: 10.1039/C4SC03135K

  27. [27]

      S. Li, Q.Y. Jiang, Y.H. Liu, et al., Anal. Chem. 93 (2021) 11275–11283. doi: 10.1021/acs.analchem.1c02264

  28. [28]

      X.J. Liu, Y.C. Zhao, F. Li, Biosens. Bioelectron. 173 (2021) 112832. doi: 10.1016/j.bios.2020.112832

  29. [29]

      C.C. Li, H.Y. Chen, J. Hu, C.Y. Zhang, Chem. Sci. 11 (2020) 5724–5734. doi: 10.1039/d0sc01652g

  30. [30]

      L.J. Wang, L. Liang, B.J. Liu, B. Jiang, C.Y. Zhang, Chem. Sci. 12 (2021) 5544–5554. doi: 10.1039/d1sc00189b

  31. [31]

      L.M. Yang, H. Guo, Q. Gao, et al., Anal. Chem. 95 (2023) 17834–17842. doi: 10.1021/acs.analchem.3c03914

  32. [32]

      J.B. Long, Y.X. Liu, Q.F. Cao, et al., Chin. Chem. Lett. 26 (2015) 1031–1035. doi: 10.1016/j.cclet.2015.05.036

  33. [33]

      J.Y. Liu, L.M. Zhang, W.T. Zeng, et al., Chin. Chem. Lett. 34 (2023) 108141. doi: 10.1016/j.cclet.2023.108141

  34. [34]

      Y. Han, D.L. Li, Q. Han, F. Ma, C.Y. Zhang, Anal. Chem. 96 (2024) 4487–4494. doi: 10.1021/acs.analchem.3c05090

  35. [35]

      C.C. Li, Y. Zhang, B. Tang, C.Y. Zhang, Chem. Commun. 54 (2018) 5839–5842. doi: 10.1039/c8cc01695j

  36. [36]

      J. Song, F. Yin, X. Li, et al., Analyst 143 (2018) 1593–1598. doi: 10.1039/c7an02032e

  37. [37]

      W. Chen, H.G. Zhang, Y.N. Zhang, et al., Anal. Chim. Acta 1263 (2023) 341275. doi: 10.1016/j.aca.2023.341275

  38. [38]

      L.L. Wang, H.G. Zhang, Y. Xie, et al., Talanta 194 (2019) 846–851. doi: 10.1016/j.talanta.2018.10.083

  39. [39]

      B.H. Fishbine, Fullerene Sci. Technol. 4 (1996) 87–100. doi: 10.1080/10641229608001539

  40. [40]

      S.K. Doorn, R.E. Fields, H. Hu, et al., J. Am. Chem. Soc. 124 (2002) 3169–3174. doi: 10.1021/ja012364c

  41. [41]

      S. Ba, H. Zhang, J.Y. Lee, et al., Bioorg. Med. Chem. 24 (2016) 1988–1992. doi: 10.1016/j.bmc.2016.03.025

  42. [42]

      E. Guo, Y. Ishii, J. Mueller, et al., Proc. Natl. Acad. Sci. U. S. A. 117 (2020) 19415–19424. doi: 10.1073/pnas.2009237117

  43. [43]

      X.L. Chen, Y.W. Wu, G.H. Cao, et al., ACS Sens. 5 (2020) 1615–1623. doi: 10.1021/acssensors.0c00081

  44. [44]

      L.J. Wang, M. Ren, Q. Zhang, B. Tang, C.Y. Zhang, Anal. Chem. 89 (2017) 4488–4494. doi: 10.1021/acs.analchem.6b04673

  45. [45]

      H.G. Zhang, Z.X. Gao, F. He, et al., Anal. Bioanal. Chem. 414 (2022) 6989–7000. doi: 10.1007/s00216-022-04267-1

• 

  Scheme 1  Schematic illustration of the CE-based fluorescence signal amplification strategy. (1) FEN1 product, (2) hAAG product, (3) UDG product, (4) FEN1 probe, (5) hAAG probe and (6) UDG probe.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Feasibility analysis. Electropherograms of different samples: (A) The amplification product with (bluish violet curve) and without (light purple curve) FEN1, Electrophoresis separation condition: BGE, 1× TAE, pH 8.5; uncoated fused-silica capillary: 48.7 cm (38.5 cm to detector) × 50 µm ID. Applied voltage: 25 kV. Cartridge temperature: 25 ℃. (B) The amplification products with (pink curve) and without (orange curve) FEN1, hAAG and UDG. Electrophoresis separation condition for B: BGE, 4× TAE, other conditions were the same as the separation conditions in (A).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Optimization of electrophoresis separation conditions. The electropherogram obtained under different buffer system: (A) Borate buffer system, pH 8.8; (1) FEN1 probe, (2) hAAG probe, (3) UDG probe. (B) TAE buffer system, pH 8.5; (1) FEN1 product, (2) hAAG product, (3) UDG product, (4) FEN1 probe, (5) hAAG probe, (6) UDG probe. Other CE separation conditions were the same as those in Fig. 1A.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  The sensitivity and selectivity for the strategy. (A) Electropherograms of FEN1, hAAG and UDG in a defined range. (B) The linear relationship between the peak area and concentration of FEN1. (C) The selectivity of the method. T7 Exo (250 U/mL), λ Exo (250 U/mL), BSA (5 mg/mL), hOGG1 (250 U/mL), TDG (250 U/mL), FEN1 (160 U/mL) + hAAG (250 U/mL) + UDG (2.5 U/mL).

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Inhibition and enzyme kinetic analysis. (A) Relationship between the relative activity of FEN1 and the concentration of ATA. (B) Enzyme kinetic curve for FEN1 (80 U/mL) assay in the substrate concentration range of 0-400 nmol/L.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Practical application of the method. (A) Electropherograms of three enzymes in different concentration of Hela cell extract. Numbers 1-6 as labeled in Fig. 2B. The concentration of cell extract (µg/mL), A: 0.0082; B: 0.082; C: 0.82; D: 8.23; E: 82.26; F: 822.6. (B) The relationship between the peak area and logarithmic value of Hela nuclear extract concentration. (C) Electropherograms and (D) histogram for comparison of enzyme level in different cell extract, respectively. The total amount of both 293T cell and Hela cell extract is 2 × 10⁶.

  下载: 全尺寸图片 幻灯片
•