binding assay on the basis of a modified enzyme-linked ... We describe a method based on fluorescence ..... National Institute on Aging, 6200 Seaforth. Street ...
Short Technical Reports Fluorescence resonance energy transfer-based method for detection of DNA binding activities of nuclear factor κB Hua-Jun He1,4, Rick Pires2, Tie-Nian Zhu3, Anhong Zhou4, Adolfas K. Gaigalas1, Sige Zou3, and Lili Wang1 1National
Institute of Standards and Technology, Gaithersburg, MD, 2Montgomery College, Germantown, MD, 3National Institute on Aging, Baltimore, MD, and 4Utah State University, Logan, UT, USA BioTechniques 43:93-98 (July 2007) doi 10.2144/000112475
The DNA binding protein nuclear factor κB (NF-κB) and the cellular signaling pathways in which it participates are the central coordinators of many biological processes, including innate and adaptive immune responses, oxidative stress response, and aging. NF-κB also plays a key role in diseases, for example, cancer. A simple, convenient, and high-throughput detection of NF-κB activation is therefore important for systematically studying signaling pathways and for screening therapeutic drug targets. We describe a method based on fluorescence resonance energy transfer (FRET) to directly measure the amount of activated NF-κB. More specifically, a double-stranded DNA (dsDNA) probe was designed to contain a pair of FRET fluorophores at the same end of the probe and an endonuclease binding site within the NF-κB consensus sequence. The activated NF-κB was detected by FRET following the restriction enzyme digestion. Using three different analyte materials—(i) purified recombinant NF-κB p50, (ii) nuclear extracts, and (iii) whole cell lysates—we demonstrated that this assay is as sensitive as the traditional, widely used electrophoretic mobility shift assay (EMSA), but much less labor-intensive for measuring NF-κB DNA binding activities. In addition, this FRET-based assay can be easily adapted for high-throughput screening of NF-κB activation.
INTRODUCTION Nuclear factor κB (NF-κB) plays an important role in the inducible transcriptional response to pathogenic signals, oxidative stresses, and proinflammatory cytokines (1). NF-κB binds directly to its target genes as a homo- or heterodimer formed with members of the Rel family proteins that include p50, p52, p65/Rel, Rel B, and cRel. In the resting cells, NF-κB is mostly localized at the cytosol in inactive forms bound by the inhibitory IκB proteins (2,3). After stimulation by cytokines, stressors, or pathogenic signals, signal transduction pathways are activated, which lead to the phosphorylation of IκB and result in ubiquitination and degradation of IκB by proteasomes (4). Consequently, NF-κB is released and then translocated to the nucleus, followed by binding to the κB consensus sequence located in the promoter regions of a variety of genes to induce gene expresVol. 43 ı No. 1 ı 2007
sions (1,5,6). Hence, DNA binding activities of NF-κB reflect cellular responses to various signals. Simple and flexible methods for the detection of DNA binding activities of NF-κB are highly desirable. A number of techniques have been developed for studying protein-DNA interactions. A commonly used method, electrophoretic mobility shift assay (EMSA) (7,8), detects slower migration of protein-DNA complexes relative to free DNA molecules using a nondenaturing gel electrophoresis. A reporter assay has also been used to detect DNA binding activity (9). The luciferase or β-galactosidase gene is placed under the control of a promoter that contains the consensus sequence recognized by the DNA binding protein. In addition, a chromatin immunoprecipitation (CHIP) assay has been developed for studying protein-DNA interactions (10). However, these assays are laborious, time-consuming, and difficult to apply for high-throughput screening.
Renard et al. (11) established a DNA binding assay on the basis of a modified enzyme-linked immunosorbent assay (ELISA). Recently, methods in relative high-throughput platforms have been developed for studying protein-DNA interactions based on fluorescence resonance energy transfer (FRET) (12– 15). For instance, Lu and his colleagues have described an assay that combines exonuclease III (ExoIII) protection strategy and FRET detection or SYBR® Green I staining method (14,15). In their assay, a double-stranded DNA (dsDNA) probe is designed to contain a pair of FRET fluorophores in the middle and two identical protein binding sites on each side of the FRET fluorophores. The NF-κB protein can protect the probe from ExoIII digestion, which results in a high FRET signal. This binding configuration, however, does not necessarily reflect endogenous DNA binding. Further, protein binding on one side may interfere the binding on the other side, because of the steric effect from complexation with other cotranscription factors likely present in cell extracts. In this study, we developed a method employing both restriction endonuclease digestion and FRET detection strategy to study NF-κB-DNA interaction. The DNA-FRET probe used is a dsDNA that contains a pair of FRET fluorophores at the same end of the probe and an endonuclease recognition site of the κB DNA consensus sequence. When the transcription factor binds to the dsDNA, it prevents the DNA probe from binding and subsequently being cleaved by the endonuclease, which then results in a high FRET signal. We compared this assay with the commonly used EMSA using purified recombinant NF-κB p50, nuclear extracts, and whole cell lysates. We examined the suitability of this FRET-based assay for high-throughput screening of NFκB activation. MATERIALS AND METHODS FRET Probes Oligonucleotides were synthesized and high-performance liquid chromatography (HPLC)-purified from Invitrogen (Carlsbad, CA, www.biotechniques.com ı BioTechniques ı 93
Short Technical Reports
USA). The sequences of two pairs of NF-κB FRET probes are: (probe 1) 5′(FAM)AAGTGGGAAATTCCTCT G-3′ 5′-CAGAGGAATTTCCCAC TT(TAMRA)-3′; and (probe 2, a mutant probe) 5′-(FAM)AAGTGTTAAATTC CTCTG-3′, 5′-CAGAGGAATTT AACACTT(TAMRA)-3′ (16). The donor [carboxyfluorescein (FAM)] and acceptor [tetramethyl-6-carboxyrhodamine (TAMRA)] fluorophores are attached to 5′ (dA) and 3′ (dT) via a C6 linker, respectively. The bold sequences represent the NF-κB binding sites, and the underlined sequences are the recognition sites for restriction enzyme ApoI, respectively. To obtain dsDNA-FRET probes, complementary oligonucleotide pairs were mixed at the same molar concentrations of 20 μM in a 100-μL Tris buffer solution (10 mM Tris-HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA). The mixture was heated for 4 min at 92°C and cooled down slowly to 25°C. The formed dsDNA probes were then purified by using native polyacrylamide gel electrophoresis (PAGE). DNA Binding Proteins, Cell Culture, and Preparation of Cell Extracts Purified recombinant NF-κB p50 (rhNF-κB p50) was purchased from Promega (Madison, WI, USA). The restriction endonuclease, ApoI, was obtained from New England Biolabs (Ipswich, MA, USA). A control protein, glutathione S-transferase, was purified from Escherichia coli in house. To obtain NF-κB protein mixtures, HeLa cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS), 50 U/mL penicillin, and 50 mg/mL streptomycin in a humidified chamber containing 5% CO2 at 37°C, and were then treated with or without 20 ng/mL tumor necrosis factor-α (TNF-α) from R&D Systems (Minneapolis, MN, USA) for 30 min. Nuclear and cytoplasmic fractions were extracted with the NE-PER Nuclear and Cytoplasmic Extraction kits from Pierce Biotechnology (Rockford, IL, USA). The nuclear extracts were transferred to the binding reaction buffer using Microcon ® YM-10 centrifugal filters (Millipore, 94 ı BioTechniques ı www.biotechniques.com
A �
�
�
B
Figure 1. The fluorescence resonance energy transfer (FRET) and restriction enzyme-based assay. (A) Representation of the assay. FRET-double-stranded DNA (dsDNA) probe 1 is used as an example. The italicized sequence refers to the nuclear factor κB (NF-κB) p50 binding site, and the underlined sequence represents the restriction enzyme recognition site, respectively. (B) Emission spectra of the FRET-dsDNA probe alone (curve 1), the FRET probe digested for 60 min by a 10 U ApoI (curve 2), and the FRET probe digested for 60 min by 10 U ApoI in the presence of 20 pmol recombinant NF-κB p50 (curve 3). Fam, carboxyfluorescein; TAMRA, tetramethyl-6-carboxyrhodamine; A.U., arbitrary units.
Billerica, MA, USA) prior to the analysis. To collect the whole cell lysates, cell pellets were suspended in the lysis buffer [20 mM HEPES, pH 7.5, 0.35 M NaCl, 20% glycerol, 1% Nonidet® P40 (NP40), 0.5 mM EDTA] with Protease Inhibitor Cocktail Set I from Calbiochem (La Jolla, CA, USA). After setting on ice for 10 min, the lysates were centrifuged at 14,000× g for 20 min at 4°C, and the supernatant was collected and stored at -70°C until analysis. The protein concentration was determined using a BCA™ kit from Pierce Biotechnology. Electrophoretic Mobility Shift Assay EMSA was performed at room temperature as follows: recombinant NF-κB p50 or cell extracts were incubated with a DNA probe in the DNA binding buffer [10 mM HEPES, pH 7.9, 50 mM KCl, 0.1 mM EDTA,
2.5 mM dithiothreitol (DTT), 0.05% NP40, 0.01 U poly(dI-dC) (Pierce Biotechnology), 0.5 mg/mL bovine serum albumin (BSA), and 10% (v/v) glycerol] for 10 min. To measure the dissociation constants in a competitive EMSA, restriction endonuclease ApoI was mixed with NF-κB p50 in the binding buffer. After addition of dsDNA-FRET probes, the mixture was incubated for additional 20 min. Two microliters of the gel-loading buffer (250 mM Tris-HCl, pH 7.5, 40% glycerol) were then added to the reaction mixture, and the mixture was loaded on a pre-run 5% polyacrylamide gel and electrophoresed at 100 V in 0.5× Tris-borate-EDTA (TBE) buffer. DNA detection was performed using an FMBIO III MultiView scanner. The band intensity was analyzed using FMBIO Image Analysis 3.0 software (both from MiraiBio, Alameda, CA, USA). Vol. 43 ı No. 1 ı 2007
Short Technical Reports
Fluorescence Measurement and FRET Analysis For FRET analysis, the dsDNAFRET probe and the DNA binding protein were incubated in the binding buffer for 20 min, followed by addition of MgCl2 and restriction endonuclease ApoI to a final concentration of 2 mM. The reaction mixture was kept at 25°C for different time periods. EDTA was then added to a final concentration of 5 mM in a total volume of 20 μL to terminate the restriction reaction. The mixture was further diluted with 480 μL buffer solution containing 10 mM Tris-HCl, pH 8.0, 100 mM NaCl, and 1 mM EDTA and was immediately subjected to fluorescence measure-
A NF-κB p50 Probe
ApoI Probe
Free Probe
B NF-κB p50 Probe
ApoI Probe
Free Probe Figure 2. Competitive electrophoretic mobility shift assay (EMSA) for evaluating the feasibility of the experimental design. In the assays, the amount of the fluorescence resonance energy transfer (FRET)-double-stranded DNA (dsDNA) probe was fixed at 0.28 pmol in a total volume of 20 μL. (A) Competitive binding of the DNA probe by various amount nuclear factor κB (NFκB) p50 in presence of the fixed amount of 1.2 pmol ApoI. Lane 1, 10 pmol; lane 2, 5 pmol; lane 3, 2 pmol; lane 4, 1 pmol; lane 5, 0.5 pmol; and lane 6, 0.2 pmol. The only FRET probe is presented in lane 7. (B) Competitive binding of the DNA probe by various amount of ApoI at the fixed amount of 3 pmol NF-κB p50: lane 2, 0 pmol; 3, 0.12 pmol; 4, 0.24 pmol; 5, 0.6 pmol; 6, 1.2 pmol; 7, 2.4 pmol; 8, 4.8 pmol. The only FRET probe is presented in lane 1. Arrows indicate the positions for free FRET-dsDNA probe (lower), ApoI-DNA probe complex (middle), and NF-κB p50-DNA probe complex (upper), respectively. Vol. 43 ı No. 1 ı 2007
ments at 30°–32°C. Emission spectra were collected by using a Starna® Semi-Micro cuvette, type 9F, and a spectrofluorimeter with a 488-nm excitation source (17). Approximately 0.5 mL reaction mixture was used for the measurements while stirred with a small magnetic bar. The FRET efficiency, R, is calculated as R = 1 - [(F(t) - F(0)]/[F(max) - F(0)], where F(t) refers to the fluorescence intensity of a reaction mixture under the endonuclease treatment for a given time t at 518 nm, and F(0) represents the fluorescence intensity at 518 nm at zero-time point following endonuclease addition in the absence of DNA binding proteins (14). The parameter F(max) is the fluorescence emission intensity at 518 nm under endonuclease treatment in the absence of DNA binding proteins for a maximal time point of an experiment. A relative FRET efficiency, ΔR = R(t) R(c), where R(t) is the transfer efficiency of an analyte protein sample and R(c) is the value of the corresponding control sample (purified GST protein, nuclear extracts, or whole cell lysates without TNF-α treatment), respectively.
the stability of dsDNA. Within the conserved GG cores is the recognition site for the restriction enzyme ApoI. In the absence of NF-κB and a endonuclease, the free probe contains two fluorophores close to each other, which results in very low donor fluorescence at 518 nm and high FRET signal at 588 nm as shown in Figure 1B (curve 1). With the presence of the endouclease, the free probe was cleaved into small fragments. Since the melting temperatures (Tm) of these fragments were
