Anal. Chem. 2003, 75, 6236-6243
A Simple Quenching Method for Fluorescence Background Reduction and Its Application to the Direct, Quantitative Detection of Specific mRNA Rhiannon L. Nolan, Hong Cai,* John P. Nolan, and Peter M. Goodwin*
Bioscience Division, Los Alamos National Laboratory, Mail Stop M888, Los Alamos, New Mexico 87545
New genome sequence information is rapidly increasing the number of nucleic acid (NA) targets of use for characterizing and treating diseases. Detection of these targets by fluorescence-based assays is often limited by fluorescence background from unincorporated or unbound probes that are present in large excess over the target. To solve this problem, energy transfer-based probes have been developed and used to reduce the fluorescence from unbound probes. Although these probes have revolutionized NA target detection, their use requires scrupulous attention to design constraints, extensive probe quality control, and individually optimized experimental conditions. Here, we describe a simpler background reduction approach using singly labeled quencher oligomers to suppress excess unbound probe fluorescence following probe-target hybridization. A second limitation of most fluorescence-based NA target detection and quantification assays is the requirement for enzymatic amplification of target or signal for sensitivity. Amplification steps make quantification of original target copy number problematic because of variations in amplification efficiencies between the sequence targets and the experimental conditions. To avoid amplification, we coupled our quenching approach to a two-color NA assay with correlated, two-color, single-molecule fluorescence detection. We demonstrate a >100-fold background reduction and detection of targets present at concentrations as low as 100 fM using the two-color assay. The application of this technique to the detection and quantification of specific mRNA sequences enabled us to estimate β-actin copy numbers in cell-derived total RNA without an amplification step. The decoding of genome sequences is revealing additional nucleic acid (NA) markers for disease research, susceptibility, diagnosis, prognosis, and treatment. Specific NA markers are popularly detected using a variety of fluorescence-based assays.1 Because the sensitivity of fluorescence-based methods can be limited by high backgrounds from unincorporated or nonspecifically bound fluorescent probes, many fluorescence-based NA detection approaches have evolved to include background reduc* Corresponding authors. Fax: 505-665-3024. E-mails: (Goodwin)
[email protected]; (Cai)
[email protected]. (1) Epstein, J. R.; Biran, I.; Walt, D. R. Anal. Chim. Acta 2002, 469, 3-36.
6236 Analytical Chemistry, Vol. 75, No. 22, November 15, 2003
ing quencher moieties that suppress unbound probe fluorescence via energy transfer.2 Examples include molecular beacons,3 which fluoresce upon target hybridization by opening of a hairpin structure which keeps fluorophore and quencher in proximity to quench fluorescence in the absence of a target complementary to the loop region; and Taqman probes,4 which become fluorescent when a DNA polymerase cleaves off the quencher group upon target binding. Assays incorporating these probes call for stringent probe design and careful optimization of the temperature and salt conditions to prevent quenching of positive signals. In addition, the challenge of synthesizing, purifying and characterizing duallabeled probes is an issue, since fluorescent probes lacking the quencher group contribute to the background and can reduce the sensitivity of the assay. One important NA detection application is the analysis of specific mRNA transcript levels. Quantitative gene expression analysis is important for understanding cellular physiology. Tissue cell types are defined both by the unique complement of proteins that they express and the levels at which these proteins are expressed. Expression of proteins is dependent on mRNA transcript levels. In cancerous cells or cells that have been exposed to external stimuli, some mRNA transcript levels change with respect to the normal cell levels. Quantitative tracking of specific mRNA transcript levels is therefore useful for understanding the functions of cells as well as for identifying abnormal tissue. To determine mRNA levels, many fluorescence-based techniques require signal or target amplification5 or separation or wash steps, or both, prior to analysis to improve sensitivity. Most attempts to quantify mRNA levels have focused on the use of exponential target amplification by the RT-PCR (reverse transcription-polymerase chain reaction).6 Although exponential amplification methods such as the RT-PCR are useful for detecting RNA targets, they are not ideal for quantifying the amount of target present. Nonlinearity, variations in priming efficiency, site accessibility, and differing cycling requirements between primers prevent uniform amplification between different targets. An alternative to exponential target amplification methods is the use of linear signal amplification. One of these linear amplification (2) Morrison, L. J. Fluorescence 1999, 7, 187-196. (3) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303-308. (4) Heid, C. A.; Stevens, J.; Livak, K. J.; Williams, P. M. Genome Res. 1996, 6, 986-994. (5) Andras, S. C.; Power, J. B.; Cocking, E. C.; Davey, M. R. Mol. Biotechnol. 2001, 19, 29-44. (6) Bustin, S. A. J. Mol. Endocrinol. 2002, 29, 23-39. 10.1021/ac034803r CCC: $25.00
© 2003 American Chemical Society Published on Web 09/30/2003
methods, termed Invader, capitalizes on the recognition and cleavage by a flap endonuclease of a specific structure formed by a probe-target complex containing a 3′ flap sequence.7 Although quantification of target based on linear amplification is more straightforward than with exponential amplification methods, amplification efficiency is still strongly dependent on the target sequences, enzyme processivity, and other subtle reaction differences. A method sensitive enough to detect specific mRNA or other NA sequences without amplification is preferable for quantitative analysis and will be a valuable addition to many cellular and analytical methods. The advent of single-molecule fluorescence detection methods beginning with the seminal work of Hirschfeld8 enables the identification, sorting, and quantitative comparison of individual (fluorescing) members of heterogeneous molecular populations.9-12 The ability to detect individual fluorescently labeled molecules is in essence a background reduction challenge. The background arises from Raman and Rayleigh solvent scattering, fluorescent impurities, emission from optical components, and detector dark counts. Ways of mitigating some these include band-pass filtering, small measurement volumes, pulsed excitation with time-gated detection, and prephotobleaching of solvent components.9 In welloptimized instruments, fluorescence from single fluorophores is detected with high signal-to-background ratios. However, the use of single-molecule fluorescence methods for ultrasensitive NA target detection has been limited by fluorescence background introduced by the high probe concentrations required for specific labeling at low target concentrations. Nevertheless, some success has been reported for single-molecule detection of specific NA sequences.13-18 In one technique, first reported by Castro and Williams,13 unamplified genomic DNA target sequences were labeled with a pair of hybridization probes bearing spectrally distinct dyes. An instrument able to monitor time-resolved singlemolecule fluorescence at both probe emission wavelengths was used to identify colabeled targets. Detection of a single color was disregarded as unbound probe. Simultaneous detection of singlemolecule fluorescence at both wavelengths, referred to here as two-color single-molecule detection, signaled the presence of target labeled with both probes. Here, we describe a simple quenching procedure for the suppression of background due to unbound fluorescent NA (7) de Arruda, M.; Lyamichev, V. I.; Eis, P. S.; Iszczyszyn, W.; Kwiatkowski, R. W.; Law, S. M.; Olson, M. C.; Rasmussen, E. B. Expert Rev. Mol. Diagn. 2002, 2, 487-496. (8) Hirschfeld, T. Appl. Opt. 1976, 15, 2965-2966. (9) Keller, R. A.; Ambrose, W. P.; Goodwin, P. M.; Jett, J. H.; Martin, J. C.; Wu, M. Appl. Spectrosc. 1996, 50, A12-A32. (10) Xie, X. S.; Trautman, J. K. Annu. Rev. Phys. Chem. 1998, 49, 441-480. (11) Ambrose, W. P.; Goodwin, P. M.; Jett, J. H.; Van Orden, A.; Werner, J. H.; Keller, R. A. Chem. Rev. 1999, 99, 2929-2956. (12) Keller, R. A.; Ambrose, W. P.; Arias, A. A.; Gai, H.; Emory, S. R.; Goodwin, P. M.; Jett, J. H. Anal. Chem. 2002, 74, 316a-324a. (13) Castro, A.; Williams, J. G. K. Anal. Chem. 1997, 69, 3915-3920. (14) Knemeyer, J.-P.; Marme´, N.; Sauer, M. Anal. Chem. 2000, 72, 3717-3724. (15) Anazawa, T.; Matsunaga, H.; Yeung, E. S. Anal. Chem. 2002, 74, 50335038. (16) Goodwin, P. M.; Nolan, R. L.; Cai, H. Proc. SPIE 2003, 4962, 78-88. (17) Li, H. T.; Ying, L. M.; Green, J. J.; Balasubramanian, S.; Klenerman, D. Anal. Chem. 2003, 75, 1664-1670. (18) Korn, K.; Gardellin, P.; Liao, B.; Amacker, M.; Bergstro ¨m, A° .; Bjo ¨rkman, H.; Camacho, A.; Do¨rho ¨fer, S.; Do¨rre, K.; Enstro¨m, J.; Ericson, T.; Favez, T.; Go¨sch, M.; Honegger, A.; Jaccoud, S.; Lapczyna, M.; Litborn, E.; Thyberg, P.; Winter, H.; Rigler, R. Nucleic Acids Res. 2003, 31, e89.
Figure 1. Two-color detection of specific DNA sequences with and without quenching. Simple hybridization of a pair of target-specific NA probes was used to label a NA sequence of interest (panel A). One of the probe sequences (red lines) was 5′-labeled with a redemitting fluorophore, A647, and the other probe sequence (green lines) 5′ was covalently labeled with a green-emitting fluorophore, RG. When both probes are mixed in excess with the target NA, the probes cohybridize to the target sequence (shown in black), labeling it fluorescent red and green. However, unbound probes are also fluorescent and contribute significantly to the fluorescence background. In addition, probes can bind nonspecifically to nontarget sequences present in the mix. In a second hybridization step (panel B), nontarget bound probe fluorescence was reduced by excess addition of a pair of quencher oligomers, NAs complementary to the probe sequences and 3′-labeled with spectrally matched quencher moieties. These quencher oligomers (black short lines) specifically bind to and quench the fluorescence of the free or weakly bound probes (shown in gray).
probes. To permit detection of specific NAs without amplification of signal or target, we have applied our quenching method to the two-color single-molecule detection approach demonstrated by Castro and Williams.13 The first step is to tag the target sequence of interest by hybridizing excess concentrations of two DNA probe sequences, one 5′ labeled with a red dye and the other 5′ labeled with a green dye (Figure 1A). In a second step, 3′ quencherlabeled oligomers with sequences perfectly complementary to the probe sequences are added to hybridize to remaining free probe (Figure 1B). Upon binding of a quencher oligomer to a probe with a complementary sequence, the fluorescent probe label is brought into close proximity to a quencher moiety that efficiently quenches probe emission by energy transfer. We show the advantages and sensitivity improvements possible using this quenching method by analysis of synthetic NA targets. To illustrate the potential of the quenching background reduction method for gene expression applications, we demonstrate quantitative detection of β-actin mRNA and estimate the number of β-actin transcripts per cell. EXPERIMENTAL SECTION Reagents. THP-1 and RS4[11] cell lines were purchased from American Type Culture Collection (Manassas, VA). Trizol reagent was from Invitrogen (Carlsbad, CA). Alexa Fluor 647 (A647) carboxylic acid, succinimidyl ester came from Molecular Probes (Eugene, OR). Salmon sperm DNA, SUPERase‚In, and total Escherichia coli RNA were procured from Ambion (Austin, TX). All other reagents were purchased from Sigma Chemical Co. (St. Louis, MO). Sources of the synthetic oligonucleotides are given in Table 1. Probe, Quencher Oligomer, and Target Preparation and Characterization. We designed two sets of linear single-stranded Analytical Chemistry, Vol. 75, No. 22, November 15, 2003
6237
Table 1. Synthetic Oligonucleotide Probe, Quencher Oligomer, and Target Characteristics target
function
sourcea
5′ label
5′ f 3′ sequence
oligonucleotide
red probe red quencher green probe green quencher synthetic target
ILT BT S BT IDT
A647 RG
red probe
M
A647
red quencher
BT
green probe
S
green quencher
BT
synthetic target
M
AAAAATTTCTTGGGCTCACTAGGAG CTCCTAGTGAGCCCAAGAAATTTTT AAAAAAATTTGAGTGAGTTTTTGAAGATG CATCTTCAAAAACTCACTCAAATTTTTTT CTCCTAGTGAGCCCAAGAAATTTTTCATCTTCAAAAACTCACTCAAATTTTTTT AGTCCGCCTAGAAGCATTTGCGGTGGACGATGGAG CTCCATCGTCCACCGCAAATGCTTCTAGGCGGACT TTTACACGAAAGCAATGCTATCACCTCCCCTGTGTGGACTTGG CCAAGTCCACACAGGGGAGGTGATAGCATTGCTTTCGTGTAAA CTCCATCGTCCACCGCAAATGCTTCTAGGCGGACTATGACCAAGTCCACACAGGGGAGGTGATAGCATTGCTTTCGTGTAAATTATG
β-actin mRNA
RG
3′ label BHQ3 BHQ1
labeling efficiency 0.50 1.0 0.71 1.0 0.95
BHQ3
0.99 0.80
BHQ1
0.91
a ILT, Invitrogen/Life Technologies, Carlsbad, CA; BT, BioSearch Technologies, Novato, CA; S, Synthegen, Houston, TX; IDT, Integrated DNA Technologies, Coralville, IA; M, Midland Certified Reagent Co., Midland, TX.
DNA probes and quencher oligomers. One set was used to detect a synthetic DNA target sequence, and the other to detect β-actin mRNA. DNA probes consisted of two functional parts: a targetbinding sequence complementary to a target subsequence and a fluorescent dye moiety conjugated to an amine group spaced six carbons from the 5′ end. A647 and Rhodamine Green X (RG; “X” refers to a seven-atom aminohexanoyl spacer) fluorescent dyes were used to label probes red and green, respectively. β-Actin probe sequences were designed with the aid of OLIGO Primer Analysis Software Version 6.4 (Molecular Biology Insights, Inc., Cascade, CO) using a GenBank β-actin sequence (accession no. BC002409). Salt-adjusted melting temperatures (Tm) of the probe and quencher oligomers bound to their targets under our hybridization conditions (0.2 M NaCl) were estimated using standard methods.19,20 The probes and quenchers for the synthetic DNA target had Tm ∼ 70 °C; the probes and quenchers for the β-actin mRNA had Tm ∼ 90 °C. A647 carboxylic acid, succinimidyl ester was used to covalently label probe sequences red at an amine group coupled to a six carbon linker on the 5′ end. The labeling and PAGE purification procedures were carried out in our laboratory according to Molecular Probes “Amine-Reactive Probe” product information sheet MP 00143. Quencher oligomer sequences used were the reverse complements of the corresponding probe sequences above and were synthesized 3′-labeled with either Black Hole Quencher 1 (BHQ1) to quench emission from the green-labeled probes or Black Hole Quencher 3 (BHQ3) to quench red probe emission. All synthetic oligonucleotides were either PAGE, RP-HPLC, or RP cartridge-purified. See Table 1 for probe, quencher oligomer, and synthetic target sequences, labeling and sources. To obtain total RNA for β-actin detection, THP-1 and RS4[11] cell lines were cultured according to recommendations from American Type Culture Collection (Manassas, VA). Total RNA was extracted from the cells using Trizol reagent, and the absence of degradation products was verified by gel electrophoresis. (19) Breslauer, K. J.; Frank, R.; Blocker, H.; Marky, L. A. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 3746-3750. (20) Sugimoto, N.; Nakano, S.; Yoneyama, M.; Honda, K. Nucleic Acids Res. 1996, 24, 4501-4505.
6238
Analytical Chemistry, Vol. 75, No. 22, November 15, 2003
All NA and dye concentrations were determined from 1-cm path length absorbance values collected on a Beckman DU-640 spectrophotometer (Beckman Coulter, Fullerton, CA), published or vendor-supplied extinction coefficients, and Beer’s law. DNA molar concentrations were calculated using the sample optical density (OD) at 260 nm. Dye concentrations were determined using an OD measurement at the appropriate dye absorbance maximum (506 nm for RG, 649 nm for A647, 530 nm for BHQ1, and 653 nm for BHQ3). The absorbance contribution from the dyes at 260 nm was unknown for the quenchers and zero for the A647 and resulted in a
