Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (96K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (23)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Walker, G.
Right arrow Articles by Nadeau, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Walker, G.
Right arrow Articles by Nadeau, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 1996 Oxford University Press 348-353

Footnote

DNA detection by strand displacement amplification and fluorescence polarization with signal enhancement using a DNA binding protein

DNA detection by strand displacement amplification and fluorescence polarization with signal enhancement using a DNA binding protein G. Terrance Walker* , C. Preston Linn and James G. Nadeau

Becton Dickinson Research Center, PO Box 12016, 21 Davis Drive, Research Triangle Park , NC 27709-2016, USA

Received August 22, 1995; Revised and Accepted November 7, 1995

ABSTRACT

Strand displacement amplification (SDA) is an isothermal in vitro method of amplifying a DNA sequence prior to its detection. We have combined SDA with fluorescence polarization detection. A 5 ' -fluorescein-labelled oligodeoxynucleotide detector probe hybridizes to the amplification product that rises in concentration during SDA and the single- to double-strand conversion is monitored through an increase in fluorescence polarization. Detection sensitivity can be enhanced by using a detector probe containing an Eco RI recognition sequence at its 5 ' -end that is not homologous to the target sequence. During SDA the probe is converted to a fully double-stranded form that specifically binds a genetically modified form of the endonuclease Eco RI which lacks cleavage activity but retains binding specificity. We have applied this SDA detection system to a target sequence specific for Mycobacterium tuberculosis .

INTRODUCTION

In its simplest sense nucleic acid probe technology is based upon hybridization of a labelled nucleic acid probe to a target sequence. Numerous methods have been devised for detecting this hybridization event, most involving physical separation of hybridized and unhybridized forms of the probe. Fluorescence polarization (FP) is a simple technique for monitoring hybridization of a fluorescently labelled oligonucleotide without separation of single- and double-stranded forms ( 1 ). A single-stranded oligonucleotide is relatively flexible and tumbles quickly in solution compared with its double-stranded counterpart. Consequently, a fluorescent dye attached to an oligonucleotide will experience slower tumbling (longer correlation time) upon hybridization, although the dye movement may not completely reflect that of the oligonucleotide, depending upon the nature of the linkage between the two bodies. The average correlation time for a dye population is reflected in the FP value of the sample, independent of the total dye concentration. Therefore, hybridization of an oligonucleotide probe can be detected without removal of excess probe as long as an appreciable percentage of the total probe is converted to the double-stranded form.

Although FP provides a simple and accurate means of DNA detection, the associated change in FP values is not extremely large. For example, hybridization of a fluorescein-labelled oligodeoxynucleotide is typically accompanied by an FP change from ~45 to ~65 millipolarization units (mP). Although this change can be adequate for detecting the presence of a target sequence, a 20 mP change does not allow one to estimate the concentration of hybridized target over a significant range. For example, if a 100 [mu]l sample contains 2 nM detector probe (~10 11 molecules), then a 20 mP change allows measurement of target levels between 10 11 and 10 10 molecules if FP values are accurate to within ~2 mP. On the other hand, if the associated change in FP is 200 mP, then the range of detectable target is extended to between 10 11 and 10 9 molecules. Expanding the dynamic range obviously improves detection sensitivity. The change in FP for hybridization of a detector probe can be enhanced by including a protein that binds specifically to the double-stranded form, thereby increasing its molecular weight and correlation time ( 2 - 4 ). We have enhanced the FP change by including the endonuclease Eco RI or a genetically modified form [ Eco RI(Gln111)], the latter of which binds specifically to the Eco RI recognition site but does not cleave it ( 10 ). Eco RI or Eco RI(Gln111) binding to a double-stranded, fluorescein-labelled detector probe containing an Eco RI recognition site increases the FP change for the single- to double-stranded conversion by ~6-fold, providing a convenient means of improving DNA detection.

We have combined FP detection with an in vitro nucleic acid amplification technique known as strand displacement amplification (SDA), which provides 10 8 -fold amplification of a target DNA sequence during constant temperature incubation ( 5 - 8 ). SDA is based upon the ability of a restriction enzyme to nick a hemi-modified recognition site and the ability of a polymerase to displace a downstream DNA strand during replication. The method consists of a target generation process (Fig. 1 A) that makes copies of the target sequence flanked by nickable restriction sites followed by exponential amplification of these modified target sequences (Fig. 1 B) through repeated nicking, strand displacement and priming of displaced strands. Despite the complicated appearance of Figure 1 , the experimental protocol of SDA is simple. Target DNA is heat denatured in the presence of all reagents except the restriction enzyme and polymerase, the sample is cooled to ~40oC, the enzymes are added and the sample is incubated at ~40oC.


Figure 1 . Schematic representation of SDA. SDA is performed using an excess of four primers (B 1 , B 2 , S 1 and S 2 ). S 1 and S 2 contain target binding regions at their 3'-ends and a recognition site (5'-GTTGAC) for the restriction enzyme Hin cII located immediately 5' of target binding regions ( Hin cII recognition sites are designated by the raised boxes). S 1 and S 2 bind to opposite strands of the target sequence, flanking the region to be amplified. B 1 and B 2 are simply target binding sequences (containing no Hin cII recognition sites) and bind at positions 5' to S 1 and S 2 . ( A ) Starting at the top of the figure, the target DNA is heat denatured. S 1 and B 1 then hybridize to one strand of the target upon lowering the temperature to 41oC. (Only one of the two target strands is shown. A corresponding series of reactions orginate on the other strand.) Hin cII and exo - Klenow (an exonuclease-deficient form of DNA polymerase I from E.coli ) are then added to the sample. At this point the remaining steps proceed as a single cascade. Exo - Klenow, which is present in large molar excess over the number of target strands, simultaneously extends S 1 and B 1 using dGTP, dCTP, dUTP and dATP[alpha]S. As S 1 is extended the extension product (S 1 -ext) is displaced through extension of B 1 . S 1 -ext serves as the target for binding of S 2 and B 2 . Simultaneous extension of S 2 and B 2 results in displacement of an S 2 extension product (S 2 -ext). An S 1 primer binds to S 2 -ext and is extended, forming a double-stranded structure with a hemiphosphorothioate Hin cII site at each end. Hin cII nicks the unmodified strand of the hemiphosphorothioate site on S 1 , leaving intact the thio-modified complementary strand of the Hin cII site. [ Hin cII nicking can also occur at the hemiphosphorothioate site on the opposite end of the fragment (not shown).] Exo - Klenow then extends the 3'-end at the nick and displaces the downstream strand. An S 2 primer binds to the displaced stand and is extended, forming an intermediate in the SDA cycle, shown in (B) by the dashed arrow. ( B ) The SDA cycle is where exponential amplification occurs. During each round of the cycle the 3'-end of S 1 binds to the 3'-end of the displaced target strand T 2 , forming a duplex with 5'-overhangs. Likewise, S 2 binds to T 1 , the complement of T 2 . Exo - Klenow extends the recessed 3'-ends of the duplexes, producing hemiphosphorothioate recognition sites that are nicked by Hin cII. These nicking and extension/displacement steps cycle continuously (short upturned arrows), because extension at a nick regenerates a nickable Hin cII recognition site. The strand displaced from the S 1 [middot]T 2 duplex is identical to T 1 . Likewise, the displaced strand from the S 2 [middot]T 1 duplex is identical to T 2 . Consequently, target amplification is exponential, because each displaced T 2 binds a new S 1 primer while each displaced T 1 binds a new S 2 (long upturned arrows). Sense and antisense strands are differentiated by thin and thick lines. Intact and nicked Hin cII recognition sequences are depicted by _[squf][squf]_ and _[squf][squf]_. The partial Hin cII recognition sequence 5'-GAC and its complement 5'-GTC are present at the 5'- and 3'-ends of displaced strands, as represented by [squf]_ and _[squf]. Additional details are reported elsewhere (5,6).We have modified the original SDA protocol to include a 5'-fluorescein-labelled detector probe that is converted from a single- to double-stranded form in a target-dependent manner during SDA. FP values are then recorded to detect target-dependent conversion of the probe using the sequence-specific protein Eco RI(Gln111) to enhance the signal.

MATERIALS AND METHODS

All oligodeoxynucleotides were synthesized on an Applied Biosystems Inc. instrument (Model 380B) and purified by denaturing gel electrophoresis. 5'-Fluorescein-labelled oligodeoxynucleotides were prepared by standard procedures using the reagent 6-FAM Amidite from Applied Biosystems Inc. (P/N 401527) according to the product insert protocols. Two different 5'-fluorescein detector probes were used. Probe D 1 (5'-dATCCGTATGGTGGATAACGTCTTTCA) binds nucleotide positions 985-1010 of the IS6110 element of Mycobacterium tuberculosis ( 9 ), which is contained within the sequence being amplified by SDA (IS6110 nucleotide positions 972-1023). The other detector probe, D 2 , (5'-dGGAATTCATCCGTATGGTGGATAACGTCTTTCA) is identical to D 1 but has an additional sequence (5'-dGGAATTC) at its 5'-end which contains the Eco RI recognition sequence.

FP values were recorded on an FPM 1 instrument from Jolley Consulting and Research, Inc. (Round Lake, IL), an instrument specifically designed for fluorescein. Excitation and emission wavelengths are fixed through interference filters appropriate for fluorescein. Samples were contained in disposable borosilicate glass test tubes (catalog no.14-962-10B; Fisher) and maintained at 37oC during FP measurement. FP is a dimensionless quantity that is expressed as

FP = ( I par - I perp )/( I par + I perp ),

where I par and I perp represent emission intensity when the emission polarizer is in the parallel and perpendicular position respectively in relation to the excitation polarizer. FP values are usually expressed in terms of millipolarization units (mP)

FP (mP) = 1000( I par - I perp )/( I par + I perp ).

Initial experiments designed to measure the FP change associated with hybridization of the 5'-fluorescein-labelled oligodeoxynucleotides (Fig. 3 ) were performed as follows. Samples (100 [mu]l) containing 10 nM 5'-fluorescein-labelled detector probe with or without an equivalent amount of complement oligodeoxynucleotide were prepared in 4 mM TAE, 50 mM NaCl, pH 7.8. Hybridization occurred over 30 min at 37oC. The samples were then diluted to 1 ml using 55 mM NaCl, 111 mM Tris-HCl, pH 7.5, 0.7 mM K 2 HPO 4 , pH 7.4, 1.1 mM EDTA, 0.7 mM [beta]-mercaptoethanol, 27 [mu]g/ml bovine serum albumin (BSA), 0.02% Triton X-100, 7% (v/v) glycerol. The 1 ml samples were equilibrated for 15 min at 37oC and FP values were recorded at 37oC. Then 5 [mu]l 100 000 U/ml Eco RI (New England BioLabs) or 5 [mu]l 1.6 [mu]M (as protein dimer) Eco RI(Gln111) ( 10 ) were added to the samples and FP readings were recorded at 37oC after the samples were incubated for 2 h at 37oC.

Experiments involving SDA and subsequent addition of Eco RI or Eco RI(Gln111) (Fig. 4 ) were performed as follows. SDA reactions were performed on samples containing M.tuberculosis target DNA generally as previously described ( 5 - 7 ). Each 100 [mu]l sample contained 50 mM K 2 HPO 4 , pH 7.6, 7 mM MgCl 2 , 0.5 mM dUTP, 0.2 mM each dGTP, dCTP and dATP[alpha]S (Pharmacia), 16% (v/v) glycerol, 0.1 mg/ml BSA, 100 ng human placental DNA, 50 nM primer S 1 (5'-dGCATTATAGTACCTGTCT GTTGAC ACTGAGATCCCCT; Hin cII recognition sequence italicized), 300 nM primer S 2 (5'-dTTGAATAGTCGGTTACTT GTTGAC GGCGTACTCGACC; Hin cII II recognition sequence italicized), 25 nM each primers B 1 (5'-dCGCTGAACCGGAT) and B 2 (5'-dTGGACCCGCCAAC), 300 U Hin cII (New England Biolabs), 1 U exo - Klenow (United States Biochemical), 10 nM 5'-fluorescein-labelled detector probe D 2 and the indicated amounts of M.tuberculosis target DNA. For each sample all reagents except Hin cII and exo - Klenow were assembled in a microcentrifuge tube and the sample was heated in a boiling water bath for 2 min and then equilibrated at 41oC in a water bath. Then 4 [mu]l 75 U/[mu]l Hin cII and 2 [mu]l 0.5 U/[mu]l exo - Klenow were added in a single aliquot. SDA proceeded for 3 h at 41oC and was terminated by addition of EDTA to 10 mM. Samples were then diluted with 0.9 ml 55 mM NaCl, 111 mM Tris-HCl, pH 7.5, 0.7 mM K 2 HPO 4 , pH 7.4, 1.1 mM EDTA, 0.7 mM [beta]-mercaptoethanol, 27 [mu]g/ml BSA, 0.02% Triton X-100, 7% glycerol. The 1 ml samples were equilibrated at 37oC and FP values were recorded at 37oC. Then 5 [mu]l 100 000 U/ml Eco RI (New England BioLabs) or 5 [mu]l 1.6 [mu]M Eco RI(Gln111) were added, the samples were incubated for 3.5 h at 37oC and FP readings were again recorded.

RESULTS

The associated change in exclusion volume that accompanies a single- to double-strand conversion of a fluorescently labelled oligodeoxynucleotide probe results in an increase in correlation time (slower tumbling in solution) that is detectable by FP. Correlation times can be increased further through protein binding to the double-stranded probe. We have designed an FP detection system in which a fluorescently labelled, single-stranded oligodeoxynucleotide is converted to a double-stranded form in a target-dependent manner during SDA (Fig. 2 ). We have used two types of detector probes (D 1 and D 2 ). D 1 consists simply of a target binding sequence. D 2 is identical to D 1 except that it contains an Eco RI recognition sequence at its 5'-end. Single- to double-strand conversion of a detector probe during SDA is first described for D 1 . As shown in Figure 2 A, D 1 binds to one of the strands displaced during the SDA cycle at a location immediately downstream of SDA primer S 1 (structure I). S 1 and D 1 are then extended by polymerase, resulting in displacement of the probe extension product (structure II) in a manner analogous to the strand displacement reaction intrinsic to SDA (Fig. 1 ). The displaced single-stranded probe extension product (structure III) binds the other SDA primer (S 2 ), forming a complex (structure IV) which becomes fully double-stranded through polymerase extension (structure V). This double-stranded complex (structure V) provides a template for linear SDA in which the Hin cII site on S 2 is nicked and polymerase extension/displacement at the nick produces single-stranded strands to which additional D 1 probes bind (structure VI) and upon which they are extended (structure VII). Structures I, II, V, VI and VII all account for double-stranded forms of the D 1 detector probe that are detectable by an increase in FP value.


Figure 2 . Single- to double-stranded conversion of the fluorescently labelled detector probe during SDA. This series of hybridization, extension and displacement steps occurs concurrently with SDA. Approximate FP values are indicated for various forms of the detector probes in the absence and presence of the protein enhancer Eco RI (Gln111) respectively. ( A ) The 5'-fluorescein-labelled detector probe (D 1 ) binds a displaced strand downstream from the SDA primer S 1 (structure I). This complex is identical to the complex shown at the top left side of the SDA cycle (Fig. 1B), where it is depicted without the detector probe. Exo - Klenow simultaneously extends D 1 and S 1 along the displaced strand (structure II). Extension of S 1 displaces the extension product of D 1 (structure III), which then binds the other SDA primer (S 2 ), forming structure IV. This complex is then extended by exo - Klenow, forming a complex (structure V) which undergoes linear SDA upon nicking of the primer sequence on S 2 by Hin cII and strand displacement by exo - Klenow. The strands displaced during this linear SDA bind additional detector probes (structure VI), which are extended by exo - Klenow (structure VII). Structures I, II, V, VI and VII all account for forms of double-stranded detector probe upon target-specific SDA. ( B ) An analogous series of steps occurs with the 5'-fluorescein detector probe D 2 , which is identical to D 1 except that it also contains an Eco RI recognition sequence at its 5'-end that is not homologous to the target sequence being amplified. In the case of D 2 , structures V-VII account for double-stranded sources of D 2 that are detectable through an increase in FP values and can bind Eco RI or Eco RI(Gln111). Sense and antisense strands are differentiated by thin and thick lines. Hin cII recognition sequences are depicted by _[squf][squf]_. The partial Hin cII recognition sequence 5'-GAC and its complement 5'-GTC are present at the 5'- and 3'-ends of displaced strands, as represented by [squf] _ and _ [squf]. The Eco RI recognition sequence is depicted by ___.


Figure 3 . FP values as a function of probe conformation and protein binding. Polarization values were recorded for single- and double-stranded forms of the two detector probes in the presence or absence of either Eco RI or Eco RI(Gln111). Detector probe D 1 lacks an Eco RI recognition site, while detector D 2 contains an Eco RI site at its 5'-end.


D 2 likewise binds to a displaced target strand immediately downstream of S 1 (structure I, Fig. 2 B) and undergoes a subsequent cascade of extension and displacement steps that are completely analogous to those of D 1 . In the case of D 2 , however, only structures V-VII account for double-stranded sources that can be detected by FP (Fig. 2 B). Likewise, these are the double-stranded structures that bind Eco RI and Eco RI(Gln111). Structures I-IV for D 2 have single-stranded 5'-ends that do not bind the proteins or exhibit higher FP values.

The entire process described in Figure 2 occurs simultaneously with the SDA cycle depicted in Figure 1 B. Although inclusion of additional primers tends to increase background reactions in some amplification systems (e.g. PCR), the fluorescent probe does not increase SDA background reactions, because any undesired mispriming between the probe and an SDA primer (B 1 , B 2 , S 1 or S 2 ) does not generate a product that can be exponentially amplified, since the fluorescent probe does not contain a Hin cII site. SDA requires participation of two Hin cII site-containing primers to achieve exponential amplification ( 8 ). This is in contrast to PCR, where any oligonucleotide containing an extendable 3'-end can serve as an amplification primer.

Before combining FP detection with SDA we performed a preliminary experiment where we measured FP changes for D 1 and D 2 hybridizing to their complementary oligodeoxynucleotides in the absence and presence of either Eco RI or Eco RI(Gln111). Oligodeoxynucleotides were hybridized in a Mg 2+ -free solution designed to support Eco RI and Eco RI(Gln111) binding but not cleavage by Eco RI. Preliminary FP values were then recorded at 37oC before addition of Eco RI or Eco RI(Gln111). Subsequently Eco RI or Eco RI(Gln111) was added to a final protein dimer concentration that was four times that of the oligodeoxynucleotides, to ensure stoichiometric binding, and FP values were again recorded at 37oC.

FP values increase from ~50 to ~65 mP upon hybridization of the detector probes to their complements in the absence of protein (Fig. 3 ). Eco RI and Eco RI(Gln111) increase the FP values for double-stranded D 2 , but not its single-stranded form. In contrast, protein binding does not affect FP values for either D 1 conformation, because it lacks an Eco RI site. Eco RI(Gln111) produces a larger increase in FP for double-stranded D 2 compared with Eco RI, perhaps due to its higher binding affinity ( 10 ).

Next we applied FP detection to an SDA system previously developed for M.tuberculosis DNA ( 5 , 7 ). Samples containing different amounts of M.tuberculosis DNA underwent SDA in the presence of D 2 . Following SDA FP values were determined before and after addition of Eco RI(Gln111) (Fig. 4 ). Before addition of Eco RI(Gln111) D 2 exhibits a target-dependent increase in FP value. Samples containing M.tuberculosis target DNA exhibit higher FP values than the samples lacking M.tuberculosis DNA, even down to a level of 10 M.tuberculosis genomes. Over a range of higher M.tuberculosis levels (10 3 -10 5 genomes) FP values reach a maximum value (~62 mP) consistent with 100% double-stranded detector probe (Fig. 3 ), due to complete conversion (probe saturation). Probe saturation occurs between 10 2 and 10 3 genomes, which corresponds to 10 3 and 10 4 copies of the IS6110 target sequence (~10 copies per genome). Since there are ~10 12 detector probes per SDA reaction, the observed amplification factor is between 10 8 - and 10 9 -fold, which is consistent with previous reports for this SDA system using 32 P detector probes with gel electrophoresis analysis ( 5 ). Nearly identical results are obtained with detector D 1 in the absence of Eco RI(Gln111) (data not shown).


Figure 4 . SDA and FP detection of M.tuberculosis DNA. SDA was performed in the presence of detector probe D 2 on samples containing the indicated amounts of target M.tuberculosis DNA. After SDA FP was measured before and after addition of Eco RI(Gln111) as indicated. The negative control sample contained dATP instead of dATP[alpha]S, which disables SDA but allows background hybridization or extension of the detector probe by exo - Klenow.

The sample lacking M.tuberculosis DNA should exhibit FP values identical to unhybridized detector probe. Higher FP values for the no M.tuberculosis DNA can arise primarily from two sources. First, the sample may be accidently contaminated with minute quantities of M.tuberculosis DNA or SDA products (amplicons) carried over from previous reactions. Low level contamination of negative samples with a few target molecules is a critical concern with amplification techniques like SDA, since they provide detection of a few target molecules. Alternatively, the detector probe in the no M.tuberculosis DNA sample could be converted to a double-stranded form either directly through background hybridization, perhaps to the 100 ng human DNA that was present in all the samples, or through some polymerase-mediated activity, such as extension of a transiently formed hairpin conformation by the detector probe. A control sample was included to elucidate possible sources of background signal in no M.tuberculosis DNA sample. The control sample contained dATP instead of dATP[alpha]S during SDA. Substitution with dATP still enables any non-specific extension of the detector probe by exo - Klenow, but it disables the SDA mechanism by allowing double-stranded cleavage of Hin cII sites. The FP value for this control sample indicates that non-specific hybridization of the detector probe is extremely low, since it is very close to the single-stranded value in Figure 3 . Likewise, the SDA sample lacking M.tuberculosis target DNA exhibits a signal that is also representative of single-stranded detector probe in the absence of Eco RI(Gln111), suggesting that amplicon contamination was extremely low for this sample.

After the initial recording of FP values Eco RI(Gln111) was added to each sample and FP values were again recorded at 37oC. Eco RI(Gln111) significantly increased the FP values for SDA samples containing target M. tuberculosis DNA, expanding the difference between them and the no M.tuberculosis DNA and control samples. However, the SDA samples containing M.tuberculosis DNA did not achieve the same level of polarization enhancement observed in the simpler hybridization experiment (Fig. 3 ), even for the SDA samples reaching probe saturation (10 3 -10 5 genomes). In Figure 3 we observed an ~180 mP increase upon addition of Eco RI(Gln111) to the double-stranded oligodeoxynucleotide, whereas the largest polarization increases were only 35-51 mP for the SDA experiment in Figure 4 . We suspect this is due to Eco RI(Gln111) binding to the non-target DNA in the SDA samples, which is a combination of the added human DNA (100 ng) and background products generated during SDA. Protein partitioning between the detector probe and other DNA is probably also responsible for the extended time necessary to achieve binding equilibrium for these SDA samples. Unlike the simpler experiment in Figure 3 , where polarization readings stabilized within a few minutes after addition of Eco RI(Gln111), the SDA samples required 3.5 h to reach binding equilibrium. Also in contrast to the simpler experiment using the detector probes and their oligodeoxynucleotide complements (Fig. 3 ), Eco RI did not increase FP values for SDA reactions performed with detector probe D 2 (data not shown), again presumably due to its lower binding affinity ( 10 ) and partitioning between the detector probe and other DNA.

This background DNA issue may also be responsible for the larger error in polarization values for the samples containing Eco RI(Gln111) in Figure 4 . For example, we expected the 10 3 -10 5 genome samples to exhibit the same polarization value in the presence of Eco RI(Gln111) as observed in its absence. Comparable values were also expected for the no M.tuberculosis DNA sample and the `no SDA' sample. Perhaps varying amounts of non-target DNA in these samples is responsible for the higher variability. We did not attempt to increase association kinetics and achieve binding saturation by adding more protein, because higher protein levels approach a practical limitation. This variability issue will have to be resolved before widespread application of the technique can be undertaken. We expect a new form of SDA ( 13 ), which operates at higher temperatures and produces less background amplification, will go a long way in this regard.

DISCUSSION

We have combined SDA with FP detection using a fluorescein-labelled detector probe. Samples containing as few as 10 M.tuberculosis genomes were detectable using a very simple protocol. Probe hybridization to the amplified target occurs simultaneously with SDA, as indicated by subsequent FP measurement. We were able to enhance the sensitivity of the system by including Eco RI(Gln111), which binds specifically to the double-stranded detector probe.

We are currently working toward simultaneous SDA and FP detection in a homogeneous closed-tube format. Such a system offers many advantages over current post-amplification methods. A homogeneous, simultaneous method is not only faster and simpler because it involves fewer sample manipulations, but it also guards against the false positive problem associated with accidental contamination with previous amplification products (amplicons), because there is no need to open samples after amplification. Another advantage of a simultaneous format is that it provides a baseline measurement at the start of SDA (time zero), thereby obviating extensive control samples to account for background fluorescence. Since the only critical variable changing during simultaneous SDA and detection is the concentration of the amplified product, it is the change in polarization that is important, not the initial or final values.

Previously we discovered that the Eco RI recognition site must be very close to the fluorescein label at the 5'-end of the oligodeoxynucleotide for the protein to increase polarization of the double-stranded form ( 2 ). The detector probe in the current study contains a recognition site just 1 nt from the 5'-fluorescein. Positioning the recognition site 5 nt from the 5'-fluorescein abolishes the effect of Eco RI binding ( 2 ). This suggests that polarization enhancement by the protein may not in fact derive from an increase in exclusion volume, but rather from direct contact between protein and dye. Fluorescein binding to the protein may restrict its motion. Alternatively, direct fluorescein-protein contact may not be responsible, but close proximity between dye and protein may be necessary due to helical flexibility. In the case of the oligodeoxynucleotide with the recognition sequence 5 nt from the 5'-fluorescein ( 2 ) restricted motion at the recognition site may not translate to the 5'-fluorescein due to flexibility in the intervening helix. Regardless of the explanation, the observations urge caution when it comes to monitoring protein binding to oligonucleotides using FP.

We suspect that the greater ability of Eco RI(Gln111) (versus Eco RI) to increase polarization is related to its higher binding affinity ( 10 ). Unlike the catalytic situation of DNA cleavage in the presence of Mg 2+ , the current FP system requires stoichiometric protein binding. The ability of a restriction enzyme to bind specifically to its double-stranded recognition site is related to its absolute and relative affinities for the recognition site and random sequence DNA. In comparison with Eco RI, the respective Eco RI(Gln111) binding affinities are ~1000- and 100-fold greater for the recognition site and random sequence DNA, so Eco RI(Gln111) generally binds tighter and is ~10-fold more specific for the recognition sequence ( 10 ). Eco RI(Gln111) is probably more effective at enhancing polarization because of its higher specificity and absolute affinity for the detector probe recognition site, which is located just 1 nt from the end of the helix, an unfavourable location considering that the footprint of Eco RI extends beyond the recognition site ( 11 ) and the fact that Eco RI uses flanking sequences to scan for the site in a one-dimensional manner ( 12 ). [We assume Eco RI(Gln111) has similar traits.] In comparison with samples containing just oligodeoxynucleotides (Fig. 3 ), slower protein binding was observed under SDA conditions, which includes 100 ng human DNA. This probably reflects partitioning between the oligodeoxynucleotide probe and random sequence DNA.

We have developed a FP detection system that should be generally applicable to a range of nucleic acid probe assays that include other amplification strategies, such as PCR. In addition, a number of restriction enzymes are probably applicable. We have tested a few other restriction enzymes and found that although they enhance polarization of a double-stranded oligodeoxynucleotide in a clean system as does Eco RI, they too are ineffective in the presence of background DNA (data not shown). We have not tried rare cutting restriction enzymes (e.g. Not I) in the hope that they would be more specific, as suggested by one of the manuscript's reviewers. As previously mentioned, we expect that a new form of SDA ( 13 ), which operates at higher temperatures and produces less background amplification, will improve the specificity of protein binding to the hybridized detector probe. For all restriction enzymes tested to date we placed the recognition site at the 5'-end of the oligodeoxynucleotide, very near the fluorophore, so we do not know if this is a necessary conditions for all restriction enzymes, as is the case for Eco RI ( 2 ).

ACKNOWLEDGEMENTS

We thank Paul Modrich and David Wright for supplying Eco RI(Gln111).

REFERENCES

1 Murakami,A., Nakaura,M., Nakatsuji,S., Tran-Cong,Q. and Makino,K. (1991) Nucleic Acids Res., 19, 4097-4102. MEDLINE Abstract

2 Kumke,M.U., Li,G., McGown,L.B., Walker,G.T. and Linn,C.P. (1995) Anal. Chem., 67, 3945-3951.

3 Cook,J., Holtom,G. and Lu, P. (1990) Anal. Biochem., 190, 331-339. MEDLINE Abstract

4 LeTilly,V. and Royer,C.A. (1993) Biochemistry, 32, 7753-7758. MEDLINE Abstract

5 Walker,G.T., Fraiser,M.S., Schram,J.L., Little,M.C., Nadeau,J.G. and Malinowski,D.P. (1992) Nucleic Acids Res., 20, 1691-1696. MEDLINE Abstract

6 Walker,G.T. (1993) PCR Methods Appl., 3, 1-6. MEDLINE Abstract

7 Spargo,C.A., Haaland,P.D., Jurgensen,S.R., Shank,D.D. and Walker,G.T. (1993) Mol. Cell. Probes, 7, 395-404. MEDLINE Abstract

8 Walker,G.T., Nadeau,J.G., Spears,P.A., Schram,J.L., Nycz,C.M. and Shank,D.D. (1994) Nucleic Acids Res., 22, 2670-2677. MEDLINE Abstract

9 Thierry,D., Cave,M.D., Eisenach,K.D., Crawford,J.T., Bates,J.H., Gicquel,B. and Guesdon,J.L. (1990) Nucleic Acids Res., 18, 188. MEDLINE Abstract

10 Wright,D.J., King,K. and Modrich,P. (1989) J. Biol. Chem., 264, 11816-11821. MEDLINE Abstract

11 Lu,A.L., Jack,W.E. and Modrich,P. (1981) J. Biol. Chem., 256, 13200-13206. MEDLINE Abstract

12 Terry,B.J., Jack,W.E. and Modrich,P. (1985) J. Biol. Chem., 260, 13130-13137. MEDLINE Abstract

13 Spargo,C.A., Fraiser,M.S., Van Cleve,M., Wright,D.J., Nycz,C.M., Spears,P.A. and Walker,G.T. (1996) Mol. Cell. Probes, in press.

Return

* To whom correspondence should be addressed
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
H. P. Morgan, P. Estibeiro, M. A. Wear, K. E.A. Max, U. Heinemann, L. Cubeddu, M. P. Gallagher, P. J. Sadler, and M. D. Walkinshaw
Sequence specificity of single-stranded DNA-binding proteins: a novel DNA microarray approach
Nucleic Acids Res., May 11, 2007; 35(10): e75 - e75.
[Abstract] [Full Text] [PDF]

Home page
 Genome ResHome page
X. Chen, L. Levine, and P.-Y. Kwok
Fluorescence Polarization in Homogeneous Nucleic Acid Analysis
Genome Res., May 1, 1999; 9(5): 492 - 498.
[Abstract] [Full Text]

Home page
 Clin. Chem.Home page
L. J. Kricka
Nucleic Acid Detection Technologies — Labels, Strategies, and Formats
Clin. Chem., April 1, 1999; 45(4): 453 - 458.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (96K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (23)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Walker, G.
Right arrow Articles by Nadeau, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Walker, G.
Right arrow Articles by Nadeau, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?