Letter pubs.acs.org/ac
Unimolecular Chemically Modified DNA Fluorescent Probe for OneStep Quantitative Measurement of the Activity of Human Apurinic/ Apyrimidinic Endonuclease 1 in Biological Samples Simin Fang, Lu Chen, and Meiping Zhao* Beijing National Laboratory for Molecular Sciences, MOE Key Laboratory of Bioorganic Chemistry and Molecular Engineering, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China S Supporting Information *
ABSTRACT: A novel DNA structure containing a 3′ internal-loop modified abasic site has been constructed which enables effective differentiation between apurinic/apyrimidinic endonuclease (APE1) and nonspecific endonuclease (DNase I). When this unique substrate structure is employed, a double-loop frayed-end chimeric fluorescent probe is successfully developed for quantitative measurement of the activity of APE1 in biological samples without the need of additional cleanup or preconcentration steps. The method is simple and rapid and has a single-step with a linear working range between 0.1 and 5.0 U/mL and a lower limit of detection of 0.1 U/mL. It holds great potential in real-time monitoring of the variation of intracellular and extracellular APE1, which will be very useful for further understanding of the DNA repair pathways in different organisms.
I
APE1 specifically incises the phosphodiester immediately 5′ to an AP site in the DNA duplex. In our preliminary experiments, we have tested a DNA probe in which only the phosphodiester bond immediately 5′ to the AP site is a natural one, while the rest of the internucleoside linkages in the sequence were all phosphorothioate modified. However, such modification did not effectively inhibit the degradation by DNase I. In contrast, the APE 1 incision activity notably declined, indicating that other phosphodiester bonds neighboring the AP site were also very important for APE1 to recognize and cleave its target phosphodiester bond immediately 5′ to the AP site. Thus, we next focused our efforts on modification of the bases near the AP site in order to construct a DNA structure that is more favorable for the incision by APE1 whereas relatively more resistant to the nonspecific degradation by DNases. Previous studies have demonstrated that the cleavage rate of AP site by APE1 was only slightly affected by the type of the opposite base.24 Among the four normal bases, AP:T and AP:A pairs showed a relatively higher cleavage rate by APE1 than the other two. Considering that apurinic sites are more commonly found than apyrimidinic sites in cells,25 we chose thymine (T) as the base opposite the AP site in this study. In addition to the opposing base, other bases neighboring the AP site (≥4 base pairs 5′ to and ≥3 base pairs 3′ to the AP site) were found to be required for APE1 to recognize and cleave the target bond.
ntracellular DNA repair is vital to the genome integrity and fidelity.1 As a multifunctional enzyme in human cells, human apurinic/apyrimidinic endonuclease 1 (APE1) plays crucial roles in the base excision repair (BER) pathway by recognizing AP sites in double-stranded DNA (dsDNA) and performing an incision at the 5′ side of the AP site, resulting in a nick in the phosphodiester backbone with 3′-OH and 5′-deoxyribose-5phosphate ends.2−5 APE1 can also cleave DNA 5′ to some base-containing damaged deoxynucleotides in the nucleotide incision repair pathway for the repair of oxidative DNA damage.6 In addition to its role in DNA repair, APE1 acts as an important regulator of the redox state of several transcription factors as well, including AP-1, HIF1R, and p53.7−9 More recent studies further reveal that the serum APE1 levels are closely correlated to certain tumors, implying the high potential of APE1 in cancer diagnosis and treatment.10−12 Despite a number of available approaches for APE1 detection, such as enzyme-linked immunosorbent assay,13 electrochemical immunoassay,14 gel electrophoresis,15,16 and fluorescent DNA probes,17,18 most of them were timeconsuming and required additional steps to separate the target enzyme from the complex matrices prior to detection. No single-step assay is available for direct measurement of this enzyme in biological samples because of the serious interference from other nonspecific endonucleases (typically represented by DNase I).19−23 In this work, we demonstrate a novel chemical structure, which can directly distinguish APE1 from DNase I and thus enable one-step detection of the activity of APE1 in human serum samples without the requirement of additional enzyme isolation steps. © XXXX American Chemical Society
Received: October 17, 2015 Accepted: November 24, 2015
A
DOI: 10.1021/acs.analchem.5b03939 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry Therefore, we tried to modify these bases, aiming to find the difference between APE1 and DNase I in the specific requirement of the substrate structure. In our initial experiments, we found that methylation of the bases adjacent to the AP site could enhance the incision rate by APE1. Unfortunately, a similar increase of cleavage rate by DNase I was also observed.26 Next, we tried to vary the types of bases around the AP site to obtain different minor groove widths at the two sides of AP site, which has been found to be closely correlated to the cleavage rates of DNA duplex by both APE1 and DNase I.27 Though DNase I is commonly regarded as a nonspecific endonuclease, recent studies have shown that it actually has some sequence preferences.26 It exclusively interacts with DNA via the minor groove, where it recognizes about six consecutive base pairs. In a hexamer-based model study to quantify the relative rate of cleavage by DNase I in terms of local DNA sequence context, it was found that the phosphodiester bond in a GC-rich sequence (GACpGCG, p indicates the position of the investigated phosphodiester bond) could hardly be cut by DNase I. We adopted this hexamer sequence and inserted an AP:T basepair at position p, aiming to generate a relatively poor substrate to DNase I. Using a dually labeled DNA duplex probe containing the hexamer structure (see probe P-1, Figure 1B) as a control probe, we systematically investigated the influences of adjacent
flanking bases of the AP site on the cleavage efficiency of both APE 1 and DNase I. According to previous study,24 a mismatch (C:C) positioned immediately 5′ to the AP site significantly reduced the rate of APE1 cleavage, while a 3′ mismatch (T:T) adjacent to the AP site slightly enhanced the APE1 incision rates. Under the sequence context of P-1, we compared the effects of different types of 3′-mismatched base pairs on the cleavage rate by APE1. As shown in Figure 1A, the presence of a 3′-C:C mismatch (probe P-2) leads to a much higher increase of the cleavage rate by APE1 (2 U/mL, about 2.5-fold increase) in comparison with the other three types of 3′ mismatched base pairs. Similar results were obtained at lower concentration levels of APE1 (1 U/mL; see Figure 1C, P-2). It was also confirmed that the presence of a 5′-mismatched base pair to the AP site had significant inhibitory effects on the APE1 incision activity (Figure 1C, P-3). Then, we measured the cleavage rates of the above three probes by DNase I (Figure 1C). Interestingly, DNase I exhibited an opposite preference to the mismatches at the two sides of AP site, though the differences between the absolute signals of P-1 and P-2 or P-1 and P-3 were not so distinct because DNase I could also cleave other phosphodiester sites in the probes, and thus, the background fluorescence signals were much higher than APE1. The above results gave us a valuable hint that the presence of a mismatched base pair at the 3′ side of the AP site was favorable for the incision by APE1, whereas it was disadvantageous for the degradation by DNase I. Extending this line of thought, we further investigated the effect of the introduction of a base pair at the 3′ + 1 position to the AP site (probe P-4) on the cleavage rates by the two enzymes. To our delight, the 3′ + 1 mismatch showed a similar increasing effect on the incision rate by APE1, while the cleavage rate by DNase I was even lower than that of P-2. These results substantially suggested that, in comparison with APE1, DNase I has a stronger requirement for duplex DNA structure 3′ to the AP site, most likely because incorporation of mismatches at the 3′ side would widen the minor groove, thus inhibiting the cleavage efficiency of DNase I. In contrast, the double-helical structure on the 5′ side of the AP site is more important for APE1 than that on the 3′ side.24 Furthermore, the results of P-2 and P-4 in Figure 1C indicated that placement of unstable mismatched base pairs on the 3′ side of the AP site could even facilitate the incision by APE1, most likely because destabilization of the duplex at the 3′ side of the AP site was favorable for the bending of the strand and intercalation of an aromatic residue side-chain on APE1 into the base stack at the AP site.27 For further confirmation and exaggeration of such a discrimination effect between APE1 and DNase I, we synthesized probe P-5 in which two consecutive mismatched base pairs were incorporated at the 3′ side of the AP site. Amazingly, about 2.5-fold enhancement of the APE1 incision rate was observed, while the cleavage rate by DNase I remained as low as that of P-4. Since the abasic site itself may also be regarded as a type of mismatch, the double mismatched base pairs and adjacent AP site actually formed a small internal loop in the DNA duplex. This unique loop structure containing an AP site offers an ideal active region for highly sensitive detection of target enzyme APE1 and effective discrimination against DNase I. Taking advantage of the unique property of the above APsite containing internal loop structure, we designed Probe-APE1 which has a self-hybridized hairpin structure with an internal loop containing the AP site in the stem part and the AP-site located close to the 5′ end (see Figure 2A and Table S1). The
Figure 1. (A) Effects of different types of 3′-mismatched base pairs on the cleavage rate of APE1. (B) Sequences of probes used in the experiments of (A) and (C). “X” as shown in blue represents the tetrahydrofuran (THF) AP site. Dashes represent the identical sequences as for P-1 (sequences P-2 to P-5). The mismatched base pairs are shown in red. F is FAM (fluorescein). Dabcyl is 4-((4(dimethylamino)phenyl) azo)-benzoyl. Q represents BHQ1 (Black Hole Quencher 1). (C) Comparison of the effects of the introduction of mismatched basepair(s) at different positions near the AP site on the cleavage rates of APE1 and DNase I. B
DOI: 10.1021/acs.analchem.5b03939 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
exonucleases. On the other hand, APE1 requires at least four natural nucleotides at the 5′ side and three at the 3′ side of the AP site; phosphorothioate modification of these nucleotides will significantly inhibit the incision activity of APE1. Taken together, there should be at least 10 bases between the AP site and the 5′ end, which seems to be too long for the cleavage product to quickly dissociate from the probe and generate fluorescent signals. To address this issue, we figured out a frayed end for the probe, leaving the last three bases near the 3′ end and five bases near the 5′ ends unpaired. Due to the phosphorothioate modification of the phosphodiester bonds between these bases, the nonspecific degradation by exonucleases could be effectively blocked. In addition to the above modification, the stem part at the 3′ side to the quencher was also protected by phosphorothioated phosphodiester bonds to avoid being degraded by DNase I, as destruction of this part would destabilize the rest of the stem and affect the recognition and incision of the substrate structure by APE1. Figure 2B summarizes the selectivity of Probe-APE1 for APE1 against DNase I and other possibly coexisting nucleases. The probe shows an outstanding discrimination effect against DNase I. Under the tested concentration levels (1.0 U/mL), the fluorescence signal of DNase I was negligible in comparison to that of APE1. To the best of our knowledge, this is the highest selectivity so far reported for the detection of APE1 against DNase I by using a fluorescent DNA probe.17,18 To find out the contribution of the AP site in the loop structure on the above discrimination effect, we further synthesized a control probe for Probe-APE1 (referred to as Probe-APE1-C) which had the same sequence as Probe-APE1 but with the AP site replaced by a normal A base (see Table S1). The fluorescence response signals of Probe-APE1-C to DNase I were measured and compared with that of Probe-APE1 in Figure S1. As can be seen, the AP site itself also plays critical roles in discriminating the two enzymes. The exact mechanisms for such difference merit further detailed investigation. It is worth mentioning that Exonuclease III (Exo III) is both a 3′−5′ exonuclease and an AP endonuclease. It is actually the major constituting expressed AP endonuclease in E. coli.31 Figure 2B shows that the fluorescence signal of 0.25 U/mL of Exo III was comparable to about 60% of 1.0 U/mL of APE1, indicating that Probe-APE1 is also suitable for measurement of the AP endonuclease activity of Exo III with high sensitivity. To further prove this, we used Probe-APE1-C to react with APE1 and Exo III, respectively. As shown in the inset of Figure 2B, the fluorescence responses were both negligible, which proved that the 3′ exonuclease activity of the two enzymes were completely prohibited by the phosphorothioate modified frayed ends. Thus, the fluorescence signals observed by Probe-APE1 were exclusively generated by the AP endonuclease activities of APE1 and Exo III. No significant fluorescent signals were observed when Probe-APE1 was mixed with other possibly coexisting exonucleases (including Exo I, T7 exo, and S1), suggesting that the probe can be applied to measure the AP endonuclease activities without suffering from the interferences from other nucleases. Then, we examined the linear response range of Probe-APE1 to different concentrations of APE1. The reaction buffer was optimized in detail to obtain stable and sensitive signals under different concentration levels. The results were compared and discussed in the Supporting Information. Figure 3A shows the time curves for measurement of the activity of APE1 at different concentrations in the optimum buffer (10 mM Bis Tris
Figure 2. (A) Schematic representation of the double-loop frayed-end Probe-APE1 for measurement of the activity of APE1. The blue strand represents nucleotides with native phosphodiester; the red strand represents phosphorothioated nucleotides. (B) Selectivity of ProbeAPE1 (100 nM) to APE1 (1.0 U/mL) over other nucleases (DNase I: 1.0 U/mL; Exo III 0.25 U/mL; Exo I 5.0 U/mL; T7 10 U/mL; S1 10 U/mL). Inset: Fluorescence responses of Probe-APE1-C (a control probe which has the same sequence as Probe-APE1 but with the AP site replaced by a normal A base; see Table S1) to APE1 and Exo III.
fluorophore and quencher were labeled at the fourth base from the 3′-side of the AP site and the fifth base from the 5′-side of the AP site, respectively. Thus, the initial fluorescence could be effectively quenched via fluorescence resonance energy transfer (FRET), and the interactions between APE1 and the active region of the probe would not be hindered by the two labels. In the presence of APE1, the deoxyribose-phosphate backbone 5′ to the AP site will be rapidly cleaved and the resultant short strand bearing the fluorophore will quickly dissociate from the rest of the probe, leading to strong fluorescence emission. It is worth mentioning that, in P-1 to P-5, FAM was used as the fluorophore. In Probe-APE1, we replaced the fluorescent label with 6-carboxy-X-rhodamine (ROX) because FAM can be nonspecifically bound and quenched by human serum albumin (HSA), the most abundant protein in human serum samples.28 The quencher was accordingly changed to BHQ2 in ProbeAPE1. To avoid the attack from exonucleases, we modified the internucleoside linkages close to the two ends of Probe-APE1 with phosphorothioated phosphodiester bonds (as indicated with an asterisk after the nucleotides in the sequence; see Table S1).29,30 It was observed that at least three phosphorothioated phosphodiester bonds at the 3′ end and five at the 5′ end were needed for efficient resistance to the degradation by C
DOI: 10.1021/acs.analchem.5b03939 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
preconcentration steps. It provides a useful tool for highthroughput measurement of the APE1 level in a large number of biological samples. Furthermore, it holds great potential in real-time monitoring of the variation of intracellular and extracellular APE1, which will be very useful for further understanding of the DNA repair pathways in different organisms.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b03939. Experimental section and supplementary Tables S1−S2 and Figures S1−S3 (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS This work was financially supported by the NSFC (21375004, 21575008) and the Beijing NSF (2152014). REFERENCES
(1) Browner, W. S.; Kahn, A. J.; Ziv, E.; Reiner, A. P.; Oshima, J.; Cawthon, R. M.; Hsueh, W. C.; Cummings, S. R. Am. J. Med. 2004, 117, 851−860. (2) Evans, A. R.; Limp-Foster, M.; Kelley, M. R. Mutat. Res., DNA Repair 2000, 461, 83−108. (3) Mol, C. D.; Izumi, T.; Mitra, S.; Tainer, J. A. Nature 2000, 403, 451−456. (4) Wilson, D. M., 3rd; Barsky, D. Mutat. Res., DNA Repair 2001, 485, 283−307. (5) Demple, B.; Sung, J. S. DNA Repair 2005, 4, 1442−1449. (6) Ischenko, A. A.; Saparbaev, M. K. Nature 2002, 415, 183−187. (7) Abate, C.; Patel, L.; Rauscher, F. J., 3rd; Curran, T. Science 1990, 249, 1157−1161. (8) Gaiddon, C.; Moorthy, N. C.; Prives, C. EMBO J. 1999, 18, 5609−5621. (9) Xanthoudakis, S.; Miao, G.; Wang, F.; Pan, Y. C.; Curran, T. EMBO J. 1992, 11, 3323−3335. (10) Liu, L.; Yan, L.; Donze, J. R.; Gerson, S. L. Mol. Cancer Ther. 2003, 2, 1061−1066. (11) Wang, D.; Luo, M.; Kelley, M. R. Mol. Cancer Ther. 2004, 3, 679−686. (12) Silber, J. R.; Bobola, M. S.; Blank, A.; Schoeler, K. D.; Haroldson, P. D.; Huynh, M. B.; Kolstoe, D. D. Clin. Cancer Res. 2002, 8, 3008−3018. (13) Cesaratto, L.; Codarin, E.; Vascotto, C.; Leonardi, A.; Kelley, M. R.; Tiribelli, C.; Tell, G. PLoS One 2013, 8, e70909. (14) Han, J.; Zhuo, Y.; Chai, Y.; Xiang, Y.; Yuan, R.; Yuan, Y.; Liao, N. Biosens. Bioelectron. 2013, 41, 116−122. (15) Weinfeld, M.; Liuzzi, M.; Paterson, M. C. Biochemistry 1990, 29, 1737−1743. (16) Kreklau, E. L.; Limp-Foster, M.; Liu, N.; Xu, Y.; Kelley, M. R.; Erickson, L. C. Nucleic Acids Res. 2001, 29, 2558−2566. (17) Maksimenko, A.; Ishchenko, A. A.; Sanz, G.; Laval, J.; Elder, R. H.; Saparbaev, M. K. Biochem. Biophys. Res. Commun. 2004, 319, 240− 246. (18) Leung, K. H.; He, H. Z.; Ma, V. P.; Zhong, H. J.; Chan, D. S.; Zhou, J.; Mergny, J. L.; Leung, C. H.; Ma, D. L. Chem. Commun. (Cambridge, U. K.) 2013, 49, 5630−5632.
Figure 3. (A) Time courses of the hydrolytic reaction of Probe-APE1 (100 nM) by APE1 at different concentrations. (B) Linear working range of the assay: 0.1 to 5.0 U/mL, R2 = 0.998.
Propane-HCl, 10 mM MgCl2, 100 μg/mL BSA, 3.5 mg/mL HSA, pH 7.0) by using Probe-APE1 (200 nM). The initial reaction rates obtained from the linear section of the time curves showed a good linear relationship with the APE1 concentration in the range from 0.1 to 5.0 U/mL (Figure 3B). The detection limit was found to be 0.1 U/mL. The probe was then applied to measure the APE1 activity in healthy human serum samples, which was found to be 2.1 ± 0.2 U/mL (1.2 ± 0.1 ng/mL), with recoveries of 93 ± 10% (n = 3). The detailed data were provided in the Supporting Information. The results were in good agreement with those obtained by previous studies (1.2 ± 0.7 ng/mL) by using conventional immunoassay methods,32 which proved that Probe-APE1 could be used to directly measure APE activity in the diluted serum without the need of additional purification or concentration steps. Since the APE1 levels have been found to be elevated in coronary artery disease32 and a number of cancers such as ovarian, cervical, and prostate,2 the developed Probe-APE1 holds great potential in high-throughput detection of APE1 in clinical samples. In conclusion, we have constructed a novel double-loop frayed-end modified AP site-containing DNA fluorescent probe which enabled highly selective and sensitive detection of the activity of APE1. The method was simple and rapid and had a single-step, with a linear working range between 0.1 and 5.0 U/ mL and a lower limit of detection of 0.1 U/mL. It can be used for quantitative measurement of the activity of APE1 in biological samples without the need of additional cleanup or D
DOI: 10.1021/acs.analchem.5b03939 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry (19) Zheng, Y. C.; Zheng, M. L.; Chen, S.; Zhao, Z. S.; Duan, X. M. J. Mater. Chem. B 2014, 2, 2301−2310. (20) Shen, Q.; Zhou, L.; Yuan, Y.; Huang, Y.; Xiang, B.; Chen, C.; Nie, Z.; Yao, S. Biosens. Bioelectron. 2014, 55, 187−194. (21) Lu, L. H.; Chan, D. S. H.; Kwong, D. W. J.; He, H. Z.; Leung, C. H.; Ma, D. L. Chem. Sci. 2014, 5, 4561−4568. (22) Leung, K. H.; He, H. Z.; He, B. Y.; Zhong, H. J.; Lin, S.; Wang, Y. T.; Ma, D. L.; Leung, C. H. Chem. Sci. 2015, 6, 2166−2171. (23) Lin, S.; Gao, W.; Tian, Z. R.; Yang, C.; Lu, L. H.; Mergny, J. L.; Leung, C. H.; Ma, D. L. Chem. Sci. 2015, 6, 4284−4290. (24) Wilson, D. M., 3rd; Takeshita, M.; Grollman, A. P.; Demple, B. J. Biol. Chem. 1995, 270, 16002−16007. (25) Lindahl, T. Nature 1993, 362, 709−715. (26) Lazarovici, A.; Zhou, T.; Shafer, A.; Dantas Machado, A. C.; Riley, T. R.; Sandstrom, R.; Sabo, P. J.; Lu, Y.; Rohs, R.; Stamatoyannopoulos, J. A.; Bussemaker, H. J. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 6376−6381. (27) Mol, C. D.; Hosfield, D. J.; Tainer, J. A. Mutat. Res., DNA Repair 2000, 460, 211−229. (28) Vlasova, I. M.; Saletsky, A. M. Laser Phys. Lett. 2008, 5, 384− 389. (29) Su, X.; Zhu, X.; Zhang, C.; Xiao, X.; Zhao, M. Anal. Chem. 2012, 84, 5059−5065. (30) Su, X.; Zhang, C.; Zhu, X.; Fang, S.; Weng, R.; Xiao, X.; Zhao, M. Anal. Chem. 2013, 85, 9939−9946. (31) Mol, C. D.; Kuo, C. F.; Thayer, M. M.; Cunningham, R. P.; Tainer, J. A. Nature 1995, 374, 381−386. (32) Jin, S. A.; Seo, H. J.; Kim, S. K.; Lee, Y. R.; Choi, S.; Ahn, K. T.; Kim, J. H.; Park, J. H.; Lee, J. H.; Choi, S. W.; Seong, I. W.; Jeon, B. H.; Jeong, J. O. Korean Circ. J. 2015, 45, 364−371.
E
DOI: 10.1021/acs.analchem.5b03939 Anal. Chem. XXXX, XXX, XXX−XXX
