Sensitive Detection of DNA Methyltransferase Activity Based on

Publication Date (Web): October 17, 2014 ... Fax: +86-27-68755595. .... Assay of DNA methyltransferase 1 activity based on uracil-specific excision re...
0 downloads 0 Views 365KB Size
Download PDF
Article pubs.acs.org/ac

Sensitive Detection of DNA Methyltransferase Activity Based on Exonuclease-Mediated Target Recycling Xi-Wen Xing, Feng Tang, Jun Wu, Jie-Mei Chu, Yu-Qi Feng, Xiang Zhou,* and Bi-Feng Yuan* Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), Department of Chemistry, Wuhan University, Wuhan, Hubei 430072, P.R. China S Supporting Information *

ABSTRACT: DNA methylation plays vital roles in various biological processes in both prokaryotes and eukaryotes. In bacteria, modification of adenine at N6 can protect bacterial DNA against cleavage by restriction enzymes, and bacterial DNA adenine methyltransferases are essential for bacterial virulence and viability. DNA adenine methyltransferase (DAM) targets the sequence of 5′-GATC-3′ and can convert adenine into N6methyladenine (m6A). Because mammals do not methylate DNA at adenine, bacterial DAM represents an excellent candidate for antibiotic development. Here, we developed an exonuclease III-aided target recycling strategy to sensitively assay activity of DAM. In this method, a hairpin probe labeled with a donor fluorophore (FAM) at the 5′ end and a quencher (BHQ) close to the 3′ end (FQ probe) was employed as reporter. Another hairpin substrate containing sequence of GATC was used as the methylation substrate of DAM. Once the hairpin substrate was methylated by DAM, it could be recognized and cleaved by Dpn I, which allows the release of a single-stranded oligodeoxynucleotide (ssODN). The ssODN can then hybridize to the 3′ protruding terminus of FQ probe, which subsequently triggers the exonuclease III-mediated target recycling reaction and therefore can significantly improve the detection sensitivity of DAM. The exonuclease-mediated target recycling strategy is extremely sensitive and as low as 0.01 U/mL DAM can be distinctly determined. Using this developed method, we evaluated DAM activity in different growth stages of E. coli cells, and we also demonstrated that the assay has the potential to screen suitable inhibitor drugs for DAM for disease(s) treatment.

D

of them either are labor-intensive or involved the use of radioactive materials. In this respect, some new methods have been developed for real-time and nonradioactive detection of DAM to overcome the above drawbacks. For example, using methylation-responsive DNA gold nanoparticles (AuNPs) assembly17,18 and methylation-triggered DNAzyme-based strand displacement amplification,19 the colorimetric approaches have been developed to assay DNA methyltransferases activity, whereas the sensing platforms often suffer from tedious nanoparticle preparation and functionalization or relative low detection limits. Recently, the electrochemical and electrogenerated chemiluminescence (ECL) biosensors have been described for the DNA methyltransferases activity assay with improved detection limits down to 0.04 U/mL level.13,20 In addition, the bioluminescence assay based on in vitro protein expression has been reported for monitoring DNA methyltransferases activity.21 Despite the attractive sensitivity, their performances have been compromised by cumbersome multistep processes and long assay times. Fluorescence sensing systems also have been developed, which enables the assay of

NA methylation, which is catalyzed by DNA methyltransferases using S-adenosylmethionine (SAM) as the methyl donor, plays critical roles in various biological processes in both prokaryotes and eukaryotes.1,2 Aberrant DNA methylation is a well-recognized hallmark of many diseases.3 In bacteria, modification of cytosine at C5 or N4 as well as adenine at N6 can protect bacterial DNA against cleavage by restriction enzymes.4 Therefore, bacterial DNA methyltransferases are essential for bacterial virulence and viability.5,6 DNA adenine methyltransferase (DAM) ubiquitously exists in prokaryotic organisms and targets the sequence 5′-GATC-3′. DAM can convert adenine into N6-methyladenine (m6A) and is responsible for the regulation of various cellular processes in E. coli, including DNA mismatch repair, DNA replication, and regulation of gene expression.7−9 In bacterial species, inhibition or overexpression of DAM could result in virulence alterations.10 Because mammals do not methylate DNA at adenine, the bacterial DAM represents an excellent candidate target for developing new therapeutics for diseases.11 Hence, sensitive detection of the activity of DAM and fast screening of its inhibitors have attracted intense interest in the past decades for both biochemical and clinical research.12−15 Traditional methods for the DAM activity assay frequently employed the radioactive labeling strategy and immune-based detection.16 Although these methods are well established, most © XXXX American Chemical Society

Received: July 30, 2014 Accepted: October 17, 2014

A

dx.doi.org/10.1021/ac502845b | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

cleaved by Dpn I, which allows the release of a single-stranded oligodeoxynucleotide (ssODN). The ssODN can hybridize to the 3′ protruding terminus of the FQ probe, which subsequently triggers the exonuclease III-mediated target recycling reaction and therefore can significantly improve the detection sensitivity of DAM. Using the simple, rapid, and sensitive method, we examined DAM activity in different growth stages of E. coli cells and the inhibition of DAM by certain drugs. Given the attractive analytical characteristics, the sensing strategy could be used to evaluate DAM activity and screen suitable inhibitor drugs of DAM for disease(s) treatment in biomedical research.

DNA methyltransferases activity with detection limits being 0.4−0.8 U/mL.22,23 Therefore, it is still a demand to develop a simple, rapid, and highly sensitive sensing strategy for the assay of DNA methyltransferases activity. Recently, isothermal amplification techniques, such as strand displacement amplification24 and nicking enzyme signal amplification25 were developed for the DNA methyltransferase assay due to the signal amplification. The strand displacement amplification-based colorimetric method allows simple visualization of DNA methyltransferase activity, but the detection sensitivity of the colorimetric method is limited. The nicking enzyme signal amplification-based method exhibits improved sensitivity, while it requires the initial heating to denature the molecule beacon. Inspired by the signal amplification strategy, here, we developed an exonuclease III-aided target recycling to sensitively detect the activity of DAM. In this strategy, a hairpin probe labeled with a donor fluorophore (FAM) at the 5′ end and a quencher (BHQ) close to the 3′ end (FQ probe) was employed as the reporter. The FQ probe contains an exonuclease III-resistant 3′ protruding terminus (Figure 1).

■

MATERIAL AND METHODS

Reagents. DAM, Dpn I, M.Sss I methyltransferase, and HhaI methyltransferase were purchased from New England Biolabs (Ipswich, MA, USA). Exonuclease III and FQ probe DNA 5′-FAM-AGGAAGACGTACGTATCTTCCT-(BHQ)CTAATGA-3′ were purchased from Takara Biotechnology Co., Ltd. (Dalian, China). Hairpin substrate oligonucleotides were purchased from Invitrogen Technology (Shanghai, China), and their sequences are listed in Table 1. Yeast extract, peptone, and agar were purchased from Becton Dickinson Medical Devices Co., Ltd. (Shanghai, China). The cell lysis buffer RIPA and Bradford Protein Assay Kit were bought from Beyotime Institute of Biotechnology (Shanghai, China). The water used throughout the study was purified by using a Milli-Q apparatus (Millipore, Bedford, MA). Other chemicals with analytical grade were purchased from Sigma-Aldrich (Beijing, China). Culture of Bacterial Cells. The GW5100 (DAM positive) and JM110 (DAM negative) E. coli cells were cultured according to our previous method.26 Briefly, a colony was inoculated into 3 mL of liquid medium (5 g/L yeast extract, 10 g/L Trypton, 10 g/L NaCl) and incubated at 37 °C in a shaker (250 rpm) for 12 h. Then, 30 μL of the cell suspension was subsequently added into 3 mL of medium and incubated for 2.5 h or 12 h. Subsequently, the cell suspension was centrifuged at 5000 rpm to obtain a cell pellet followed by washing twice with Milli-Q water. The resulting E. coli cells were lysed using RIPA lysis buffer. The protein contents were determined using the

Figure 1. Schematic diagram of the assay for DAM activity based on exonuclease-mediated target recycling.

Another hairpin substrate containing sequence of GATC was used as the methylation substrate of DAM. Once the hairpin substrate was methylated by DAM, it could be recognized and

Table 1. Sequences of Hairpin Substrates Used in the Experimenta

a

The bases underlined can form the stem of the hairpin substrate. Those highlighted in red are the recognition sites (GATC) of DAM and Dpn I, and those highlighted in italic are the liberated ssODN after Dpn I digestion. B

dx.doi.org/10.1021/ac502845b | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

results showed that the methylated S1 can be efficiently cleaved by Dpn I (Figure S1, lane 4, Supporting Information), while in the reaction mixture that lacks either DAM or Dpn I, S1 cannot be cleaved by Dpn I (Figure S1, lanes 2 and 3, Supporting Information). We then also examined the digestion of doublestranded DNA from the blunt 3′ terminus by exonuclease III. The results suggested that double-stranded DNA was efficiently digested by exonuclease III (Figure S2, lane 3, Supporting Information). Subsequently, we used the proposed strategy to evaluate the feasibility of the assay. As shown in Figure S3, Supporting Information, a significant fluorescent signal was observed when all the three enzymes of DAM, Dpn I, and Exo III were present. However, in the absence of any of the enzymes, there was no fluorescent response, which demonstrated the feasibility of this method for the assay of DAM activity. Optimization of Fluorescence Monitoring. The ssODN released from the hairpin substrate can affect the hybridization with the FQ probe, which eventually determines the detection sensitivity. In this respect, we evaluated the different hairpin substrates that can release different lengths of the ssODN upon methylation by DAM and digestion by Dpn I. Our results showed that the hairpin substrate of S1 can offer the highest relative fluorescent intensity (Figure 2A). Since the duplex part

Bradford Protein Assay Kit according to the manufacturer’s recommended protocol. Fluorescence Monitoring. All fluorescence measurements were performed on a PerkinElmer LS-55 spectrofluorometer at 25 °C. The emission spectra were collected from 505 to 600 nm with an excitation wavelength of 494 nm. Slit widths for the excitation and emission were 5 and 10 nm, respectively. Assay of DAM Activity. All of these standard solutions were prepared under 4 °C and stored at −20 °C. The reaction mixture (200 μL) consisted of 100 nM FQ probe, 50 nM hairpin substrate, 160 μM SAM, 5 U of Dpn I endonuclease, 10 U of exonuclease III, and various amounts of DAM. The reaction buffer contained 50 mM Tis-HCl (pH 7.5), 10 mM MgCl2, 50 mM NaCl, and 1 mM DTT. The experiment was performed at 37 °C. To achieve the best performance, the incubation time of DAM was optimized. Selectivity of the DAM Assay. To investigate the selectivity of the proposed DAM assay, two other methyltransferases, M.Sss I and HhaI, were selected as the potential interfering enzymes. The selectivity experiments were conducted with 25 and 125 U/mL of interfering enzymes in the same way as the DAM activity detection procedure. Influence of Inhibitors on DAM Activity. To further study the inhibitors screening ability of the proposed assay, the inhibition effects of gentamycin and benzylpenicillin (broadspectrum antibiotic) on the DAM activity were carried out with various concentrations. After the DAM methylation and Dpn I digestion, the fluorescent intensity was recorded using the aforementioned procedure.

■

RESULTS AND DISCUSSION Design Strategy of DAM Activity Assay. As restriction endonuclease Dpn I is a DNA methylation dependent enzyme and can quickly cleave the methylated DNA, it is used to evaluate the methylation level of the hairpin DNA substrate. The candidate sequences of the hairpin substrates are listed in Table 1, and the strategy of the assay is shown in Figure 1. Upon methylation by DAM and digestion by Dpn I, the stem of the hairpin substrates will detach and release a singlestranded ODN (ssODN), which can hybridize to the 3′ protruding terminus of the FQ probe. Then, the stem-loop structure of the FQ probe was opened and can form a doublestranded DNA with ssODN with a blunt 3′ terminus, which initiated the digestion by exonuclease III to degrade the FQ probe from the 3′ terminus and liberate the BHQ quencher. Thus, the FAM fluorescence can be observed. It should be noted that exonuclease III can recognize and cleave from the 3′ end of a duplex DNA with a blunt or receded 3′ terminus, and a duplex DNA with a protruded 3′ terminus is not a substrate for exonuclease III. Since the 3′ terminus of the ssODN in the double-stranded DNA is not fully hybridized to its complementary stand, the ssODN will not be digested by exonuclease III. The ssODN then can hybridize with another FQ probe and enter a new cycle, which therefore can amplify the signal and dramatically increase the detection sensitivity with an extremely low amount of ssODN. Thus, the exonuclease III-aided signal amplification allows the sensitive assay of DAM. The FQ probe was designed to form a hairpin configuration with FAM fluorophore labeled at the 5′ end and BHQ quencher labeled close to the 3′ end to reduce the background signal. We first used polyacrylamide gel electrophoresis to evaluate the feasibility of this assay. In this respect, we examined the cleavage of the methylated substrate 1 (S1) by Dpn I. The

Figure 2. (A) Optimization of hairpin substrate. (B) Optimization of the incubation time of DAM.

of the double-stranded DNA formed by the FQ probe and ssDNA released from substrate S1 is the longest, it will generate the most stable double-stranded DNA; therefore, the performance of the signal amplification strategy by substrate S1 is better than the other substrates of S2, S3, and S4. However, if the substrates DNA is too long, the loop part of the hairpin substrate DNA could partially form double-stranded DNA with the FQ probe even without DAM methylation and Dpn I digestion, which will cause high background signal. Thus, we chose S1 as the substrate for further experiments. We then further optimized the DAM incubation time. The result showed that the increased folds of fluorescent intensity reached a plateau when the incubation time was 40 min (Figure C

dx.doi.org/10.1021/ac502845b | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

2B). Therefore, 40 min of incubation time was used for the following experiments. Assay of DAM Activity. The advantage of this exonuclease III-aided assay of DAM is that fluorescent signal can be amplified by exonuclease-mediated target recycling. Therefore, it benefits the detection of low amounts of target. In this respect, we compared the relative fluorescent intensity of the assay with and without exonuclease III-aided signal amplification. The results showed that exonuclease III-aided signal amplification strategy can dramatically increase the detection sensitivity and even as low as 0.01 U/mL DAM can be readily observed (Figure 3), which is much lower than the previously

Figure 3. Effect of exonuclease-mediated target recycling on the detection sensitivity of the assay for DAM activity. Figure 4. (A) Fluorescent signal response with different concentrations of DAM. (B) Linear curve of the assay for DAM activity based on exonuclease-mediated target recycling.

reported assay for DAM activity.23,25 However, only 1.5 U/mL DAM was detected without exonuclease III-aided signal amplification (Figure 3). Therefore, the detection sensitivity increased by around 150-fold due to the exonuclease III-aided signal amplification. The linear range of the assay for DAM activity was evaluated using DAM with the range from 0 to 50 U/mL (Figure 4A). As shown in Figure 4B, a relatively good linearity was obtained across 3 orders of magnitude (R2 = 0.9812). We further evaluated the selectivity of this assay for DAM activity. In this respect, M.Sss I and HhaI that function as methyltranferases with the recognition sequence of 5′-CG-3′ and 5′-GC-3′, respectively, were selected as interference methyltranferases to assess the selectivity of this method. The results showed that only DAM can induce a remarkable fluorescence enhancement, while M.Sss I and HhaI have no obvious fluorescence change, even when adding 5-fold more of the methyltranferases (Figure 5). These results demonstrated that the amplification strategy for DAM activity exhibited a good selectivity originating from the high specific sequence recognition between DAM and the hairpin substrate. Generally, the strategy we developed here is simple and sensitive and can be accomplished in approximately 60 min. As far as we know, the assay time of this method is relatively short compared with most existing approaches which ranged from 120 to 420 min, and the detection limit of our assay is 0.01 U/ mL, which is better than most of the other developed approaches that have the detection limits of 0.04−0.8 U/mL. The detailed comparison of these DAM activity assays is listed in Table S1, Supporting Information. Evaluation of DAM Activity Inhibition. DAM inhibition is attractive because humans lack this particular enzymatic activity; therefore, DAM represents an excellent candidate for antibiotic development. Here, we further examined the inhibition effects of gentamycin and benzylpenicillin (broad-

Figure 5. Selectivity of the assay for DAM.

spectrum antibiotic) on DAM activity using our developed assay. A previous report suggested that gentamycin had no influence on the activity of Dpn I when the concentrations of gentamycin was lower than 100 μM;22 therefore, the effect of gentamycin should go through the inhibition of DAM. Here, we also examined the effects of gentamycin and benzylpenicillin on the activity of Dpn I and exonuclease III. The results showed that both gentamycin and benzylpenicillin exhibited no obvious effects on the activities of Dpn I and exonuclease III (Figures S1 and S2, Supporting Information), whereas gentamycin and benzylpenicillin displayed dose dependent inhibition for DAM (Figure 6). The calculated IC50 for gentamycin and benzylpenicillin were 10.0 and 28.4 μM, respectively, which were at the same level as a previous report.25 These results demonstrated that the proposed DAM activity assay can be successfully applied in DAM inhibitor D

dx.doi.org/10.1021/ac502845b | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 6. Inhibition of DAM activity by gentamycin (A) and benzylpenicillin (B).

concentration using the Bradford Protein Assay. Then, a certain volume of E. coli cell lysate was added into the reaction buffer; so, DAM activity (U/mL) was determined on the basis of the calibration curve in Figure 4B. Therefore, DAM activity (U/mg) can be obtained with knowing the DAM activity (U) and the total protein amount (mg) in a certain volume of E. coli cell lysate. The exponential growth stage of E. coli occurred between 1.5 and 3 h, and the growth rate slowed down significantly and then entered into the stationary growth stage after 12 h. The result showed that the activity of DAM is slightly higher in the exponential growth stage than in the stationary growth stage (Figure 7B), with the DAM activity being 0.54 U/mg at exponential growth stage and 0.45 U/mg at stationary growth stage. This phenomenon may be attributed to the higher level of DAM in E. coli cells at the exponential growth stage, which is required for the fast growth of E. coli cells.27,28 In addition, a previous report indicated that polyphosphate, an endogenous inhibitor of DAM, was low in the exponential growth stage, but it increased in the stationary stage, which may also contribute to the relative higher activity of DAM in exponential growth stage than in stationary stage.29 In addition, we further examined the inhibition of DAM from E. coli cell lysate by gentamycin and benzylpenicillin. The results showed that DAM activity can be efficiently inhibited by gentamycin and benzylpenicillin (Figure 7B). Furthermore, no obvious DAM activity was observed in DAM-negative E. coli of JM110 (Figure 7B). These results suggested that the obtained signal is derived from DAM activity, not from the nonspecific cleavage of probe. Taken together, the developed assay can be readily applied to evaluate DAM activity in complex biological samples.

screening and is a potentially useful tool for antibiotic drug discovery. Assay of DAM Activity in E. coli. We further investigated the endogenous DAM activity of E. coli cells using our developed method. As shown in Figure 7A, the fluorescence

■

CONCLUSIONS We developed a fluorescence sensing method for simple, rapid, and highly sensitive detection of the activity and inhibition of DAM by an exonuclease III-mediated target recycling strategy. By taking advantage of the unique strategy of signal amplification, the detection limit can be as low as 0.01 U/ mL. Owing to the specific site recognition of DAM, this new approach could discriminate DAM from other methyltranferases with high selectivity. The application in E. coli cell extract further demonstrated that the assay has a great potential for sensitive detection of DAM activity in complex biological samples. In addition, this assay can be applied to screen inhibitors of DAM. Given the attractive analytical characteristics, the sensing strategy might find many important applications in biomedical research.

Figure 7. (A) Measurement of the activity of DAM in E. coli cells. (B) The activity of DAM at different growth stages of E. coli cells and in DAM negative E. coli cells, and the inhibition of DAM activity by gentamycin (30 μM) and benzylpenicillin (30 μM).

intensity increased with the increased amounts of total protein of E. coli cells, and we can readily observe the fluorescence even when 9 ng of total protein was used, suggesting the high sensitivity of this assay. We then compared the DAM activity in different growth stage of E. coli cells. As for the calculation of DAM activity in E. coli cell lysate, we first determined the total protein E

dx.doi.org/10.1021/ac502845b | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

■

Article

(24) Zhu, C.; Wen, Y.; Peng, H.; Long, Y.; He, Y.; Huang, Q.; Li, D.; Fan, C. Anal. Bioanal. Chem. 2011, 399, 3459−3464. (25) Chen, F.; Zhao, Y. Analyst 2013, 138, 284−289. (26) Xing, X. W.; Liu, Y. L.; Vargas, M.; Wang, Y.; Feng, Y. Q.; Zhou, X.; Yuan, B. F. PLoS One 2013, 8, No. e72993. (27) Lobner-Olesen, A.; Skovgaard, O.; Marinus, M. G. Curr. Opin. Microbiol. 2005, 8, 154−160. (28) Wion, D.; Casadesus, J. Nat. Rev. Microbiol. 2006, 4, 183−192. (29) Elsawy, H.; Jeltsch, A. World Acad. Sci., Eng. Technol. 2013, 7, 643−645.

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org/.

■

AUTHOR INFORMATION

Corresponding Authors

*Tel.:+86-27-68755595. Fax: +86-27-68755595. E-mail: [email protected]. *Tel.:+86-27-68755595. Fax: +86-27-68755595. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the financial support from the National Basic Research Program of China (973 Program) (2012CB720600, 2012CB720601, 2012CB720603) and the National Natural Science Foundation of China (21205091).

■

REFERENCES

(1) Smith, Z. D.; Meissner, A. Nat. Rev. Genet. 2013, 14, 204−220. (2) Vanyushin, B. F. Mol. Biol. 2005, 39, 473−481. (3) Robertson, K. D. Nat. Rev. Genet. 2005, 6, 597−610. (4) Jeltsch, A. ChemBioChem 2002, 3, 274−293. (5) Kossykh, V. G.; Lloyd, R. S. J. Bacteriol. 2004, 186, 2061−2067. (6) Heithoff, D. M.; Sinsheimer, R. L.; Low, D. A.; Mahan, M. J. Science 1999, 284, 967−970. (7) Coffin, S. R.; Reich, N. O. Biochemistry 2009, 48, 7399−7410. (8) Marinus, M. G.; Casadesus, J. FEMS Microbiol. Rev. 2009, 33, 488−503. (9) Low, D. A.; Weyand, N. J.; Mahan, M. J. Infect. Immun. 2001, 69, 7197−7204. (10) Heusipp, G.; Falker, S.; Schmidt, M. A. Int. J. Med. Microbiol. 2007, 297, 1−7. (11) Mashhoon, N.; Pruss, C.; Carroll, M.; Johnson, P. H.; Reich, N. O. J. Biomol. Screening 2006, 11, 497−510. (12) Zaleski, P.; Wojciechowski, M.; Piekarowicz, A. Microbiology 2005, 151, 3361−3369. (13) He, X.; Su, J.; Wang, Y.; Wang, K.; Ni, X.; Chen, Z. Biosens. Bioelectron. 2011, 28, 298−303. (14) Fang, G.; Munera, D.; Friedman, D. I.; Mandlik, A.; Chao, M. C.; Banerjee, O.; Feng, Z.; Losic, B.; Mahajan, M. C.; Jabado, O. J.; Deikus, G.; Clark, T. A.; Luong, K.; Murray, I. A.; Davis, B. M.; KerenPaz, A.; Chess, A.; Roberts, R. J.; Korlach, J.; Turner, S. W.; Kumar, V.; Waldor, M. K.; Schadt, E. E. Nat. Biotechnol. 2012, 30, 1232−1239. (15) Lobner-Olesen, A.; Marinus, M. G.; Hansen, F. G. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 4672−4677. (16) Boye, E.; Marinus, M. G.; Lobner-Olesen, A. J. Bacteriol. 1992, 174, 1682−1685. (17) Song, G.; Chen, C.; Ren, J.; Qu, X. ACS Nano 2009, 3, 1183− 1189. (18) Liu, T.; Zhao, J.; Zhang, D.; Li, G. Anal. Chem. 2010, 82, 229− 233. (19) Li, W.; Liu, Z.; Lin, H.; Nie, Z.; Chen, J.; Xu, X.; Yao, S. Anal. Chem. 2010, 82, 1935−1941. (20) Liu, S.; Wu, P.; Li, W.; Zhang, H.; Cai, C. Chem. Commun. (Cambridge, U. K.) 2011, 47, 2844−2846. (21) Jiang, C.; Yan, C. Y.; Huang, C.; Jiang, J. H.; Yu, R. Q. Anal. Biochem. 2012, 423, 224−228. (22) Li, J.; Yan, H.; Wang, K.; Tan, W.; Zhou, X. Anal. Chem. 2007, 79, 1050−1056. (23) Tian, T.; Xiao, H.; Long, Y.; Zhang, X.; Wang, S.; Zhou, X.; Liu, S.; Zhou, X. Chem. Commun. (Cambridge, U. K.) 2012, 48, 10031− 10033. F

dx.doi.org/10.1021/ac502845b | Anal. Chem. XXXX, XXX, XXX−XXX