In Vitro Analysis of DNA–Protein Interactions in Gene Transcription

Subscriber access provided by Uppsala universitetsbibliotek

Article

In Vitro Analysis of DNA-Protein Interactions in Gene Transcription Using DNAzyme-Based Electrochemical Assay Chao Li, Yaqin Tao, Yi Yang, Yang Xiang, and Genxi Li Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b00329 • Publication Date (Web): 03 Apr 2017 Downloaded from http://pubs.acs.org on April 4, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

In Vitro Analysis of DNA-Protein Interactions in Gene Transcription Using DNAzyme-Based Electrochemical Assay †

†

†

†

†‡

Chao Li, Yaqin Tao, Yi Yang, Yang Xiang, Genxi Li*, ,

State Key Laboratory of Pharmaceutical Biotechnology and Collaborative Innovation Center of Chemistry for Life Sciences, Department of Biochemistry, Nanjing University, Nanjing 210093, P. R. China †

‡

Center for Molecular Recognition and Biosensing, School of Life Sciences, Shanghai University, Shanghai 200444, P. R. China

ABSTRACT: The interaction between protein and DNA elements controls a variety of functions of genomes. The development of convenient and cost-effective method for investigating the sequence specificity of DNA-binding proteins represents an important challenge. In response, we have introduced an electrochemical assay in this work for specific and sensitive analysis of interaction between protein and nucleic acid in nucleic extracts, based on the protein-induced distinctive motion behavior of DNA deoxyribozyme (DNAzyme) on an electrode surface. As a proof of principle, we have also presented assays for the rapid, sensitive and selective detection of three transcription factors (NF-κB, SP6 RNA polymerase and HNF-4α), as well as the analysis of binding affinity of the mutated protein-binding sequence, and even screening of the binding sequence of HNF-4α protein in vitro. This work may open new opportunity for in-depth profiling of the sequence specificity of DNA-binding proteins and study of nucleotide polymorphisms in known protein binding sites.

electrophoretic mobility shift assay (EMSA) to some extent,11 it requires additional substrate preparation and expensive instruments, limiting its wide application in ordinary libraries. Moreover, most of the above methods require complex DNA probe design and extensive molecular modification. Therefore, the development of convenient and cost-effective method for the measurement of DNA-protein binding still remains challenging. Herein, a new strategy based on the DNA-protein interactions-induced distinctive motion behavior of DNA deoxyribozyme (DNAzyme) on an electrode surface has been introduced in this work to attain quantitative information of DNA-binding protein (at nanomolar level) and further to screen the specific binding sequence of these proteins. The DNAzyme strand used here can freely diffuse to the electrode surface and then efficiently cleave the substrate DNA modified on the surface.12

Sequence-specific DNA binding by proteins (e.g., promoters and transcription factors) plays crucial roles in controlling a variety of biological activities such as genome replication, transcription, recombination, and repair. Although tremendous importance not only in basic biology but also in clinical diagnosis and drug development, it still remains a challenging task to identify proteins binding to the specific nucleic acid sequence at a molecular level. For example, the degree of gene expression is not only controlled by the number of RNA polymerase but also is affected by the promoter sequence that tunes the binding and initiating efficiency of polymerase.1 Therefore, a method that can comprehensively analyze these information about DNA-protein interaction is rather important for understanding the underlying mechanisms of cellular activity. Traditional methods for detecting protein content (e.g., Western blotting assay, ELISA), binding activity (e.g., gelshift assays,2 DNase footprinting assay3,4), or transcript efficacy (e.g., quantitative PCR) are complicated and timeconsuming. So, alternative strategies such as gold nanoparticle-based colorimetric assay,5,6 DNA beacon-based fluorescence assay,7,8 and DNA switch-based electrochemical assay9,10 have been explored in recent years. These newer methods, however, have their own drawbacks. For example, most of them are only concerned with the detection of the DNAbinding protein concentration, few of them have attempted to identify genomic binding sites of proteins, which may lose many key information about the sequence specificity of DNAbinding proteins. Although dark-field microscopy has been developed to identify significant position of protein-binding DNA sequence that can make up the inherent shortcomings of

Scheme 1. Schematic illustration of the principle of the proposed assay for transcription factor (TF) detection. 1

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 5

saturated calomel electrode (SCE) as the reference electrode, a platinum wire as the counter electrode, and a gold electrode as the working electrode. Peak currents were fit using the manual fit mode in the CHI software. Experimental data were collected using square wave voltammetry from -0.05 to -0.45 V in increments of 0.001 V vs. Ag/AgCl, with an amplitude of 50 mV and a frequency of 15 Hz. 5 mM of Fe(CN)63−/4− with 1 M KNO3 was employed for the electrochemical impedance spectra (EIS) experiments. For EIS, spectra were recorded by applying a bias potential of 0.213 V vs SCE and 5 mV amplitude in the frequency range from 0.1 Hz to 100 kHz. Gain represents difference in peak currents obtained before and after target addition divided by initial peak current.

It is hypothesized that sequence-specific binding of large protein to the DNAzyme strand significantly retards its activity, compared with random DNA-protein interactions. This study also aims to identify critical positions in the binding site by monitoring selective protein binding. As shown in Scheme 1, a bifunctional DNA probe that contains a protein binding element at 3’ terminus and a DNAzyme sequence at 5’ end has been designed to achieve both target recognition and signal transduction. In the absence of target protein, the DNAzyme can hybridize and then cleave substrate DNA in a multiple turnover manner. However, once in the presence of target protein, the formed large DNA-protein complex will slow the regeneration of DNAzyme and keep the redox tag (methylene blue, MB) on the electrode surface, leading a distinguished signal response eventually. In this fashion, complicated, inefficient and costly detection of DNA-binding protein molecules is replaced by a straightforward, sensitive, and specific detection. Moreover, without complex probe design, this method can be adapted into any other DNA-binding protein by just adjusting the DNA-binding sequence of the DNAzyme.

Antibody-modified gold nanoparticle (GNPs) preparation. GNPs are synthesized according to citrate-reduction method. First, the synthesized citrate-covered GNPs solution (96 µL) was adjusted by K2CO3 buffer (0.1 M) to be around pH 9.0. Then, Ab (0.5 mg/mL, 4 µL) in a PBS buffer was added to the solution. The sample was shaken at 4 °C for 6 h for sufficient immobilization. Remaining active sites of GNPs were passivated with 1% BSA solution for another 3 h. The Ab-GNPs were collected by centrifugation, washed for three times with deionized water to remove the excess protein molecules, and re-dispersed in 0.1 mL of PBST buffer (1x PBS with 0.05% Tween-20). Dynamic lighting scattering (DLS) measurements were performed through a Zetasizer Nano ZS (Malvern Instruments).

EXPERIMENTAL SECTION Materials and Reagents. NF-κB p65, rabbit anti-NF-κB antibody (Ab) and HNF-4α protein were purchased from Abcam. SP6 RNA polymerase (20 U/µL) was obtained from Beyotime. Bovine serum albumin (BSA), ZnCl2 and tris (2carboxyethyl) phosphine hydrochloride (TCEP) were obtained from Sigma. All oligonucleotides were synthesized and purified by Sangon Inc. (Shanghai, China). The DNA sequences and modifications are listed in Table S1.

Cell culture and cell extract preparation. HepG2 cells were grown in 5% CO2 DMEM medium supplemented with 10% FCS, L-glutamine, and antibiotics (Biological Industries). Cells were harvested into lysis buffer (10 mM Tris-HCl, pH 7.4, 1% NP-40, 150 mM NaCl, 1 mM EDTA, 1 mM benzamidine, 10 mg/mL leupeptin, 1 mM NaF, 1 mM Na3VO4). This lysate was centrifuged at 14000g for 20 min and used for electrochemical experiments. Malignant breast epithelial cells (MDA-MB-231) were grown in 5% CO2 RPMI medium 1640 supplemented with 10% FCS, L-glutamine, and antibiotics (Biological Industries).

The buffers employed in this work were as follows. DNA immobilization buffer: 10 mM Tris-HCl, 1 mM EDTA, and 0.1 M NaCl (pH 7.4). Reaction buffer: 10 mM phosphate buffered saline (PBS, pH 7.4) with 140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 20 µM ZnCl2 and 1 mM CaCl2, and 0.1% BSA. Buffers for electrode washing are 10 mM PBS solution with 0.1 M NaCl and 0.05% Tween-20 (pH 7.4). All solutions were prepared with NANOpure H2O (>18.0 MΩ) from a Millipore system.

Protein detection. Unless otherwise indicated, assays were performed by incubating samples (diluted in 10 mM PBS, pH 7.4, 5 mM KCl, 1 mM MgCl2, and 1% BSA) with 5 nM DNAzyme (diluted in 10 mM Tris-HCl, pH 7.4, 1 mM MgCl2, 0.1% BSA) in 2 µL incubations at room temperature for 15 min, before addition of a 3 µL reaction buffer required for probe binding and cleavage. Then, the 5 µL drop of the mixed solution was placed on the top of the substrate DNA modified electrode and left under the lid at 37 °C for incubation, followed by a washing buffer rinse (∼10 s). For nanoparticleassistant assay, samples were firstly incubated with 1 µL synthesized nanoparticle for 10 min and then the same procedures are performed. Experiments performed in cell extract were realized by adding 1 µl of DNAzyme probe (0.5 µM) to 19 µl of cell extract. The volume percentage of the cell extract in the samples is thus 95%.

Electrode cleaning and sensor preparation. The gold electrode (diameter 3.0 mm) was immersed in piranha solution (H2SO4: 30% H2O2 = 3:1) for 5 min to remove the adsorbed organic matter, and then rinsed with pure water. After that, the electrode was successively polished with alumina powder (particle sizes: 1.0 and 0.3 µm) on microcloth. Lastly, it was sonicated for 5 min in both ethanol and water, and electrochemically activated in 0.5 M H2SO4 until a stable cyclic voltammogram was obtained. After incubating substrate DNA with TCEP (1 mM) for 40 min to allow reduction of disulfide bonds, electrodes with substrate DNA self-assembly monolayers (SAMs) were obtained by incubation with 0.5 µM substrate strands for 1 h at room temperature, followed by a 1 h treatment with an aqueous solution of 1 mM MCH to get well-aligned DNA monolayers. The electrode was then further rinsed with pure water and dried again with nitrogen.

RESULTS AND DISCUSSIONS To verify this new sensing architecture, we have firstly designed a sensor against the common eukaryotic transcription factor (TF) NF-κB whose aberrant expression is related with

Electrochemical measurements. Electrochemical measurements were carried out on 660D Electrochemical Analyzers (CH Instruments) with a conventional three-electrode cell at room temperature. The three-electrode system consisted of a 2

ACS Paragon Plus Environment

Page 3 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 1. (A) Square wave voltammogram obtained upon analyzing 200 nM NF-κB with (curve c) or without (curve b) the help of Ab-GNPs for a fixed time interval of 45 min. Curve a is for the control experiment which is in the absence of NF-κB. Curves d and e are obtained when (d) 1 µM heated NF-κB and (e) 1 mg/mL BSA are used instead of normal NF-κB. (B) Electrochemistry profile responded to different protein targets with the help of antibody-GNPs for a fixed time interval of 45 min (from bottom to top: 0, 0.5, 3, 15, 75, 200 nM). Inset is a plot of current intensity as a function of target concentrations.

Figure 2. The DNAzyme-based assay is versatile, sensitive and rapid through changing the DNA-binding motif. (A, B) Sensitive detection of SP6 RNA polymerase (Kd = 87 ± 8 nM) and HNF-4α (squares, Kd = 69 ± 12 nM) by plotting current intensities versus various concentrations of protein targets. (C, D) Fast and selective detection of SP6 RNA polymerase (triangles) and HNF-4α (squares) and little cross reactivity with the other target (circles, 1 µM NF-κB).

many diseases such as asthma, cancer, and diabetes.13 As shown in Figure 1A, the electrochemical signals under different conditions have been investigated. In the absence of NFκB protein, the current intensity of MB is relatively small (~ 0.988 µA, curve a), resulting from the efficient cleavage of DNAzyme (Figure S1). Upon addition of 200 nM NF-κB to the system, the increase of current intensity is 64.9% (curve b), consistent with the inhibited activity of DNAzyme to some extent. Considering the above results, it seems that the binding of large protein molecule (Mw = 65 kDa) onto the designed probe indeed prevents the cleavage of substrate DNA, so it is reasonable to assume that better performance may be obtained through taking advantage of size-tunable metal nanoparticle. As expected, when NF-κB antibody modified gold nanoparticles (d = ~ 54 nm) have been employed (Figure S2), addition of 200 nM NF-κB results in a 212.4% increase in current intensity within 30 min (curve c), which implies the activity of DNAzyme is almost completely inhibited. Thus, the proposed assay strategy can be used for amplified detection of NF-κB. In the meantime, the control groups fail to generate any responsive signals (curve d, e), indicating the high selectivity of the proposed assay. With the help of nanoparticle-based amplification, the current intensity is proportional to the NF-κB concentration and the detection limit is near 0.08 nM (Figure. 1B) after optimization of reaction time (Figure S3). In contrast, the detection limit is about 1 nM when we have used the 1:1 binding strategy without amplification (Figure S4). More importantly, this DNAzyme-based method is highly compatible with the complex detecting media such as serum and cell nucleic extract, which displays no compromise of sensitivity (Figure. S4). This improved performance may be due to two aspects: one is related to the high activity of DNAzyme and direct molecular recognition between DNA sequence and target protein; the other is related to the inherent advantages of electrochemical assay which is sensitive to the change of electrode surface. Motivated by our success in the detection of NF-κB, we have just replaced the recognition domain of DNA probe to target SP6 RNA polymerase and HNF-4α to test the versatility of the proposed assay. SP6 RNA polymerase is a DNAdependent RNA polymerase that is highly specific for the

SP6 phage promoter (from −17 to −1, ATTTAGGTGACACTATA)14 and HNF-4αbelongs to a highly-conserved member of the nuclear receptor superfamily (classical binding site: GGTCAAAGGTCAA) and is considered as a major regulator of the hepatocyte phenotype.15 Under the same condition, both achieve similar sensitivity and rapid detection of the DNA- binding proteins and show little cross reactivity with other TFs (Figure 2). After demonstrating the quantitative ability of the proposed assay, we next have made use of this method to analyze the interaction between DNA and these DNA-binding proteins. To

Figure 3. Current intensities according to the binding of RNA polymerase to each promoter variant. (A) positive control (consensus binding site); mutated binding sites at (B) −16; (C) −10; (D) −6; (E) negative control (random sequence). Inset: schematic of different protein binding affinity of the mutated sequences.

3

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 5

ware (Figure 4), we have found that the oligonucleotides with higher S/N value are also those most significantly similar to the HNF-4α consensus binding sites. As the S/N value decreases, the similarity to the consensus binding sequences reduces as well, but the most important base sites are constant (positions 4, 5, 6, 8, 11, and 12). So, through analysis of these information, we may screen the potential binding sequences of HNF-4α in the whole genome.

CONCLUSION In this study, we have demonstrated that the attachment of protein molecule onto the DNAzyme strand may result in a reduced enzymatic activity. Based on this finding, we have then proposed a novel method for the analysis of DNA-protein interaction. This DNAzyme-based technique has several important merits over existing methods. First, this assay allows one-step detection of DNA-binding protein in the nuclear extracts without dangerous radioactive label and overmuch sample. Second, this assay generates quantitative results, making affinity to be calculated. Third, this assay can be further used to analyze the nucleotide polymorphisms in known TF binding sites and screen the potential binding sites of TF, which largely broadens the application range of the proposed method. Finally, the concept proposed in this report may also be further developed for sensitive, on-chip, and simultaneous detection of multiple molecular interactions, making this method economical and easily adaptable into currently available microfluidic platforms.

Figure 4. HNF-4α binding sequences identified by the proposed assay. The 32 sequences are divided in three groups based on their S/N values. Then, each of the group of sequences is illustrated with the program WebLogo. The previously known classic binding sequence of HNF-4 is indicated at the bottom. conduct the research, we have firstly randomly mutated several binding sites of SP6 RNA polymerase (T to G at −16, T to C at −10, and A to G at −6) to evaluate sequence-specific binding efficiency. Typically, the activity of DNAzyme is preserved when the polymerase fails to interact the binding sequence, which results in a reduced signal response. So, in Figure 3, not surprisingly, the maximum electrochemical signal is observed upon the presence of a SP6 polymerase consensus sequence. In contrast, current intensities reduce more or less when the protein binds to the point-mutated oligonucleotides, which also indicates differential effects of the base mutations on the interactions between the polymerase to the promoters. Moreover, the mutation at − 9 position most significantly disturbs the binding of the SP6 to the promoter sequence, implying its more important role in the specific protein binding. Taken together, these results demonstrate that the proposed method can be used to analyze the influence of even a single-base mutation on the interaction between DNAbinding protein and corresponding oligonucleotide. We have also explored to use this method for screening the sequence of DNA-binding protein. An additional 32 binding oligonucleotides are randomly chosen from ChIP-chip analysis in HepG2 cells that may have potential HNF-4α binding sites predicted by the software.16 The signal-to-noise ratio (S/N) determined by the electrochemical detection of each sequences is contrasted with the previously obtained prediction score that suggests the possibility that HNF-4α binds to the oligonucleotide motifs (Figure S5). The relationship between S/N and prediction score is satisfied in that oligonucleotides with the higher score generate high S/N, whereas most of oligonucleotides with the lower score are only accompanied by low S/N (< 1). When these selected oligonucleotides are grouped according to their S/N values and analyzed by WebLogo soft-

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Oligonucleotide sequences, EIS results of electrode at different stages, DLS results, electrochemistry responded to different protein targets without antibody-GNPs, electrochemical responses to NF-κB in different biological fluids and table of protein-binding DNA sequences as well as additional figures and tables (PDF)

AUTHOR INFORMATION Corresponding Author * E-mail: [email protected]. Tel: +86-25-83593596.

ORCID Genxi Li: 0000-0001-9663-9914 Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work is supported by the National Natural Science Foundation of China (Grant Nos. 21235003, 21327902, 81672570), and and the Science Foundation of Jiangsu Province, China (Grant No. BM2015023).

REFERENCES (1) Babu, M. M.; Luscombe, N. M.; Aravind, L.; Gerstein, M.; Teichmann, S. A. Curr. Opin. Struct. Biol., 2004, 14, 283-291. (2) Garner, M. M.; Revzin, A. Nucleic Acids Res., 1981, 9, 30473060. (3) Diamond, M. I.; Miner, J. N.; Yoshinaga, S. K.; Yamamoto, 4

ACS Paragon Plus Environment

Page 5 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

K. R. Science 1990, 249, 1266-1272. (4) Ji, Z.; Song, R.; Huang, H.; Regev, A.; Struhl, K. Nat. Biotechnol. 2016, 34, 410-413. (5) Zhang, Y.; Hu, J.; Zhang, C.-y. Anal. Chem. 2012, 84, 95449549. (6) Tan, Y. N.; Su, X.; Zhu, Y.; Lee, J. Y. ACS nano 2010, 4, 5101-5110. (7) Vallée-Bélisle, A.; Bonham, A. J.; Reich, N. O.; Ricci, F.; Plaxco, K. W. J. Am. Chem. Soc. 2011, 133, 13836-13839. (8) Wang, K.; Tang, Z.; Yang, C. J.; Kim, Y.; Fang, X.; Li, W.; Wu, Y.; Medley, C. D.; Cao, Z.; Li, J. Angew. Chem., Int. Ed. 2009, 48, 856-870. (9) Bonham, A. J.; Hsieh, K.; Ferguson, B. S.; Vallée-Bélisle, A.; Ricci, F.; Soh, H. T.; Plaxco, K. W. J. Am. Chem. Soc. 2012, 134, 3346-3348.

(10) Ma, J.; Li, C.; Tao, Y.; Feng, C.; Li, G. Biosens. Bioelectron. 2016, 86, 933-938. (11) Song, M. S.; Choi, S. P.; Lee, J.; Kwon, Y. J.; Sim, S. J. Adv. Mater. 2013, 25, 1265-1269. (12) Ranallo, S.; Rossetti, M.; Plaxco, K. W.; Vallee-Belisle, A.; Ricci, F. Angew. Chem., Int. Ed. 2015, 54, 13214-13218. (13) Hernandez, N. Genes Dev., 1993, 7, 1291-1308. (14) Brown, J. E.; Klement, J. F.; McAllister, W. T. Nucleic Acids Res., 1986, 14, 3521-3526. (15) Parviz, F.; Matullo, C.; Garrison, W. D.; Savatski, L.; Adamson, J. W.; Ning, G.; Kaestner, K. H.; Rossi, J. M.; Zaret, K. S.; Duncan, S. A. Nat. Genet. 2003, 34, 292-296. (16) Rada-Iglesias, A.; Wallerman, O.; Koch, C.; Ameur, A.; Enroth, S.; Clelland, G.; Wester, K.; Wilcox, S.; Dovey, O. M.; Ellis, P. D. Hum. Mol. Genet. 2005, 14, 3435-3447.

For TOC Only

5

ACS Paragon Plus Environment