Q6

Q1

Q2

The Journal of Molecular Diagnostics, Vol. -, No. -, - 2013

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162

jmd.amjpathol.org

6364656667686970717273747576777879808182

TECHNICAL ADVANCECooperative Primers

2.5 MillioneFold Improvement in the Reduction ofNonspecific AmplificationBrent C. Satterfield

8384

From DNA Logix Inc, Bountiful, Utah

858687 Accepted for publication

C

a

P

h

888990919293

October 30, 2013.

Address correspondence toBrent C. Satterfield, DNALogix Inc, 585 W. 500 S., Ste.210, Bountiful, UT 84010.E-mail: info@dnalogix.net.

opyright ª 2013 American Society for Inve

nd the Association for Molecular Pathology.

ublished by Elsevier Inc. All rights reserved

ttp://dx.doi.org/10.1016/j.jmoldx.2013.10.004

9495969798

SC

The increasing need to multiplex nucleic acid reactions presses test designers to the limits of amplifi-cation specificity in PCR. Although more than a dozen hot starts have been developed for PCR to reduceprimer-dimer formation, none can stop the propagation of primer-dimers once formed. Even a smallnumber of primer-dimers can result in false-negatives and/or false-positives. Herein, we demonstrate anew class of primer technology that greatly reduces primer-dimer propagation, showing successfulamplification of 60 template copies with no signal dampening in a background of 150,000,000 primer-dimers. In contrast, normal primers, with or without a hot start, experienced signal dampening with asfew as 60 primer-dimers and false-negatives with only 600 primer-dimers. This represents more than a 2.5millionefold improvement in reduction of nonspecific amplification. We also show how a probe can beincorporated into the cooperative primer, with 2.5 times more signal than conventional fluorescentprobes. (J Mol Diagn 2013, -: 1e11; http://dx.doi.org/10.1016/j.jmoldx.2013.10.004)

99100

½F1�½F1�

Disclosure: DNA Logix Inc. is a commercial enterprise that owns therights to the cooperative primer technologies described herein. B.C.S. is theowner of DNA Logix Inc.

101102103104105106107108109110111112113114115116117118119120121122

In the short time since its inception, PCR has become analmost indispensable part of medical and diagnostic sci-ences.1,2 In PCR, a short DNA segment, called a primer,anneals to a target nucleic acid segment and is extended bythe polymerase, amplifying the DNA present. Althoughrules have been described for the selection of primers,3

primers often anneal to and amplify each other in an un-predictable manner, forming primer-dimers. This problem iscompounded in multiplexed tests, where the large numberof primers increases the probability of primer-dimer for-mation exponentially.3 Primer-dimers compete with thetarget for the primers and can result in signal dampening,false-negatives, or even false-positives if the nonspecificamplification product also interacts with a probe.

To try to prevent formation of primer-dimers, more than adozen hot start methods have been invented, each of whichlimits the activity of the polymerase before an initial denaturingstep.4e15 However, none of these can stop 100% of primer-dimer formation. Furthermore, the hot start protection is lostafter the initial denaturing step, allowingmoreprimer-dimers to

stigative Pathology

.

O 5.2.0 DTD � JMDI289_proof �

be formed and propagated during amplification. Because thenumber of possible primer-dimer products increases expo-nentially with the number of primers in the reaction accordingto the function

�n2 þ n=2

� ð1Þwhere n is the number of primers, highly multiplexed testscan be difficult to design. With the increasing demand forhighly multiplexed assays to diagnose cancers, drug resis-tance, and other disease, a solution must be found to theprimer-dimer problem.

Herein, we report on cooperative primers, the first tech-nology to prevent primer-dimer formation and propagationduring the actual amplification rounds (Figure 1). The primersconsist of short sequences that would ordinarily not amplifythe template. Because of the low primer melting temperature

123124

23 December 2013 � 5:30 pm � EO: JMD13_0073

Figure 1 Schematic showing cooperative primermechanisms. The short, low-Tm primer does notamplify unless the capture sequence binds first. Thepolyethylene glycol linker connecting the primer andthe capture sequence prevents the polymerase fromextending all the way through the primer to the cap-ture sequence, thus retaining the short primer speci-ficity in every round of amplification. Nonspecificamplification products that do not contain sequencescomplementary to the capture sequence, such asprimer-dimers, do not propagate. Because capturesequence hybridization is required during every roundof amplification for the primer to bind, it results in anexponential reduction of nonspecific amplicons incontrast to traditional hot starts that are effectiveonly before starting amplification. The polymeraseextension of the cooperative primer linked to the 50

end of the capture sequence (A) favors displacementof the capture sequence, whereas extension of thecooperative primer linked to the 30 end of the capturesequence (B) favors 50-30 exonuclease cleavage of thecapture sequence. Cleavage of the capture sequenceallows it to double as a sequence-specific probe,releasingafluorophoreonnuclease cleavage.Becausethe capture sequence must bind for the primer toextend, the efficiency of probe hybridization and theresultant cleavage are theoretically very high.

print&

web4C/FPO

Satterfield

125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186

187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235

(Tm), unless the capture sequence binds holding the primer inclose proximity to the template, the primers will not amplify.The polyethylene glycol linker connecting the primer and thecapture sequence prevents the polymerase from extendingthrough the capture sequence, retaining the primer specificityin each round of amplification. Nonspecific amplicons that donot have a complementary region to the capture sequence,such as primer-dimers, are not propagated. Because thisprocess is repeated in every round of amplification, it results inan exponential reduction of nonspecific amplification, incontrast to traditional hot starts, which are effective onlybefore starting the reaction. In addition, when the capturesequence is labeled, it can double as a sequence-specific probein reactions using a 50-30 exonuclease active polymerase.Because the capture sequence must bind for the primer toextend, theoretically 100% of primer extensions also result inprobe hybridization and cleavage, increasing the signal-to-noise ratio. We hypothesized that these cooperative con-structs would allow efficient amplification of the templatewhile simultaneously reducing the impact of primer-dimerson assay results.

½T1�½T1�

236237238239240241242243244245246247248

Materials and Methods

Oligonucleotide Design and Synthesis

Adaptingmath previously derived for cooperative probes16,17

and neglecting entropic/enthalpic penalties due to restrictingmobility from the linker, the effective equilibrium constantfor cooperative primers can be expressed as follows:

KeffZKprimer þKcap þPLKprimerKcapZCcap þCprimer þCboth

PTð2Þ

2SCO 5.2.0 DTD � JMDI289_proof �

where K is the equilibrium constant, with the added sub-scripts cap, primer, both, and eff referring to capturesequence, primer, both capture sequence and primer, andeffective, respectively.

PZPo �Ccap �Cprimer �Cboth; TZTo �Ccap �Cprimer �Cboth

ð3Þwhere Po and To are initial primer and target concentrations,respectively, PL refers to the local primer concentration[PL Z 1 molecule/(volume swept out by linker length �Avogadro’s number)], and C is the concentration of thesequence indicated by the subscript that is hybridized to thetemplate. From this equation, the effective primer bindingefficiency (Eff ) in the initial rounds of amplification (eg,Po >> To) can be calculated as follows:

EffZCprimer þCboth

T0Z

�Kprimer þPLKprimerKcap

�P0

1þKeff P0ð4Þ

Exploration of predicted enthalpy and entropy values toevaluate the equilibrium constants in equation (2) revealsthat primers with predicted Tms 7�C to 12�C below thereaction temperature can still amplify with up to 99% effi-ciency when coupled with capture sequences with an equalor greater Tm and when separated by three or fewer hexa-ethylene glycol sequences (HEGs). When six HEG linkersseparate the primer and the capture sequence, 99% effi-ciencies can still be obtained with predicted primer Tms 4�Cto 7�C below the reaction temperature.For proof of concept, cooperative primers were designed

to the human beta-actin gene, the mitochondrial sequence ofPlasmodium spp., and the rpoB D516V mutation in Myco-bacterium tuberculosis (Table 1). The primer sequence was

jmd.amjpathol.org - The Journal of Molecular Diagnostics23 December 2013 � 5:30 pm � EO: JMD13_0073

Cooperative Primers

249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310

311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371

designed with a Tm 3�C to 9�C below the reaction tem-perature. The capture sequence was designed with a Tm 3�Cto 11�C below the reaction temperature or 5�C and 7�Cabove the reaction temperature. The capture sequence wasalso labeled with a FAM/Dabcyl FRET pair to determine anoptimal method of incorporating a probe into the coopera-tive primer. One of the primers, PfcF inv 62HP, had hairpinsecondary structure intentionally designed into the capturesequence to improve quenching of the fluorophore. Suffi-cient HEGs were used as linkers for the primer and thecapture sequence to bind in a rigid double helix. Three or sixHEGs (for cooperative primers with the linker attached tothe 50 or 30 end of the capture sequence, respectively) wereused to link the capture sequence to the primer.

For the cooperative primers with the linker attached to the50 end of the capture sequence, a blocking carbon chain wasplaced at the 30 end of the capture sequence to prevent itfrom being used as a primer. To attach the linker to the 50

end of the capture sequence (eg, with the 50 end of theprimer linked to the 50 end of the capture sequence), theprimer had to be synthesized with an inverted linkage(Table 1). Reverse amidites were used to switch the strandpolarity at this linkage using otherwise standard solid-phasesynthesis for oligonucleotide manufacture. A reverse ami-dite comprises a dimethoxytrityl or similar protecting groupat the 30-hydroxyl of a deoxynuceloside and a phosphategroup at the 50-hydroxyl of a deoxynucleoside such that the50 region of the reverse amidite is amenable for linkage to a50-OH or 30-OH of another nucleoside. Sequences madefrom inverted bases function like normal primers. Theorientation affects only the direction of binding (Figure 1A).

Control primers for determining normal primer amplifi-cation efficiency and susceptibility to primer-dimers weredesigned with a Tm 5�C to 7�C above the reaction tem-perature. This gave the primers just enough affinity tomaximize binding to the template for good amplificationefficiency without unduly increasing the likelihood ofspurious product formation. An additional set of controlprimers was identical to the short primer sequence in thecooperative primer and was designed to verify that coop-erative primers cannot amplify properly without the capturesequence. Primer-dimers, by definition, were designed asthe perfect complement of each primer set, with 30 endstouching each other. Synthesis of cooperative primers wasperformed by Biosearch Technologies (Petaluma, CA). Dualhigh-performance liquid chromatography purification wasused to purify the cooperative primers. Template was syn-thesized by either Biosearch Technologies or IntegratedDNA Technologies (Coralville, IA) with no purification.

Cooperative Primer PCR Efficiency

Human beta-actin real-time PCR was run by making amaster mix with a 250-nmol/L final concentration of eachprimer/probe (b-act P, b-act cF, b-act cR), a 5-mmol/L finalconcentration of MgCl2, and an additional 0.275 U per

The Journal of Molecular Diagnostics - jmd.amjpathol.orgSCO 5.2.0 DTD � JMDI289_proof �

reaction of GoTaq polymerase in GoTaq colorless mastermix (Promega Corp., Madison, WI). Dilutions of templatewere made using 600,000, 6000, 60, and 0 copies. The re-action was run on the StepOne system (Applied Biosystems,Foster City, CA) and included a 20-second denaturing stepat 95�C followed by 45 cycles at 95�C for 1 second and55�C for 20 seconds. Reactions were run in duplicate. A logplot of the concentration versus the CT was used to find theslope from which to calculate PCR efficiency.

Impact of Primer-Dimers

Plasmodium falciparum PCR was run by making a mastermix with a 250-nmol/L final concentration of each primerfor cooperative primers (PfcF inv62, PfcR inv) or normalprimers (PfnF, PfnR, PfP), a 5-mmol/L final concentrationof MgCl2, and an additional 0.275 U per reaction of GoTaqpolymerase in GoTaq colorless master mix (PromegaCorp.). Sixty copies of template were added to each reactionand 0, 60, 600, 6000, 600,000, 6,000,000, 60,000,000,150,000,000, or 150,000,000,000 primer-dimers wereplaced in each. Each reaction was run in duplicate. Thereaction with normal primers was repeated using GoTaq hotstart and GoTaq hot start colorless master mix. The re-actions were run on the StepOne system (Applied Bio-systems) and included a 20-second denaturing step at 95�Cfollowed by 50 cycles at 95�C for 1 second and 55�C for 20seconds. After amplification, the PCR products were run ona 2.2% FlashGel system (Lonza Inc., Walkersville, MD) andwere imaged.

As a follow-up to verify that the results from the coop-erative primers were due to cooperativity, the cooperativeprimers were made without a capture sequence (Pf Low TmF and Pf Low Tm R). Combined with the probe (PfP) andusing the same master mix and thermocycling conditions asnoted previously herein, 0, 60, 6000, 600,000, or 6,000,000copies of template were added to the reaction. Reactionswere run in duplicate.

Effect of Complex Samples

To analyze performance in complex samples, 5000 copiesof P. falciparum DNA were spiked into 0, 0.6, 1.2, or 3 mgof human gDNA (BioChain Institue Inc., Newark, CA). Acooperative primer with a long capture sequence (PfcFinv62) and a cooperative primer with a short capturesequence (PfcF) were compared with normal primers (PfnFand PfnR). Each master mix had a 500-nmol/L final con-centration of primers, a 5-mmol/L final concentration ofMgCl2, and an additional 0.275 U per reaction of GoTaqpolymerase in GoTaq colorless master mix (PromegaCorp.). Each reaction was run in duplicate. The reactionswere run using the ABI 7500 system (Applied Biosystems)and included a 20-second denaturing step at 95�C followedby 50 cycles at 95�C for 3 seconds and 55�C for 32 seconds.

3

372

23 December 2013 � 5:30 pm � EO: JMD13_0073

Table 1 Oligonucleotide Sequences Q5

Name Sequences*Tmprimer

Tmcapture

Beta-actin (amplification efficiency)Normal primers/probesb-act P 50-[FAM]TGTGGCCGAGGACTTTGACGGC[BHQ1]-30

Cooperative primersb-act cF 30-AGTGGCAAGGTC-50[Sp18][Sp18][Sp18]50-GGTGACAGCAGTC

[Sp3]-3052.3 50.8

b-act cR 30-TAGGATTTTCGGTG-50[Sp18][Sp18][Sp18]50-GCAAGGGACTTCC[Sp3]-30

48.1 51.4

TemplatesBeta actin 50-AGGATTTAAAAACTGGAACGGTGAAGGTGACAGCAGTCGGTTGGAGCG

AGCATCCCCCAAAGTTCACAATGTGGCCGAGGACTTTGATTGCACATTGTTGTTTTTTTAATAGTCATTCCAAATATGAGATGCGTTGTTACAGGAAGTCCCTTGCCATCCTAAAAGCCACCCCA-30

P. falciparum (impact of primer-dimers and probe selection)

Normal primers/probesPfnF 50-CGCATCGCTTCTAACGGTGA-30

PfnR 50-GAAGCAAACACTAGCGGTGGAA-30

PfP 50-[FAM]ACTCTCATTCCAATGGAACCTTGTTCAAGTTCAAACCATTGGAA[DABC]-30

Cooperative primers/probesPfcF inv 30-TCGCTACGCA-50[Sp18][Sp18][Sp18]50-[FAM]ACGGTGAACTC

TCA[DABC]-3046.6 52.8

PfcF inv62 30-TCGCTACGCA-50 [Sp18][Sp18][Sp18] 50-[T(FAM)]ACGGTGAACTCTCATTCCA[DABC]-30

46.6 62.0

PfcF inv62HP 30-TCGCTACGCA-50[Sp18][Sp18][Sp18]50-[T(FAM)]ACGGTGAACTCTCATTCCACCG[DABC]-30

46.6 62.0

PfcF 50-[FAM]ACGGTGAACTCTCA[DABC][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]ACGCATCGCT-30

46.6 52.8

PfcR inv 30-ATTGACATACCTGC-50[Sp18][Sp18][Sp18]50-AGCAAGTGGAATGTT [Phos]-30

47.3 53.2

Low-Tm primers minus capturesequencePf Low Tm F 50-ACGCATCGCT-30 46.6Pf Low Tm R 50-CGTCCATACAGTTA-30 47.3

TemplatesNormal primer-dimer 50-GAAGCAAACACTAGCGGTGGAATCACCGTTAGAAGCGATGCG-30

Cooperative primer-dimer 50-CGTCCATACAGTTAAGCGATGCGT-30

P. falciparum 50-CCAGCTCACGCATCGCTTCTAACGGTGAACTCTCATTCCAATGGAACCTTGTTCAAGTTCAAATAGATTGGTAAGGTATAGTGTTTACTATCAAATGAAACAATGTGTTCCACCGCTAGTGTTTGCTCTAACATTCCACTTGCTTATAACTGTATGGACG-30

M. tuberculosis (SNP differentiation)Normal primers/probesMTb P 50-[FAM]CGCCGCGATCAAGGAGTTCGCG[BHQ1]-30

Cooperative primers/probesMTb cF 30-ACACTAGCGGAG-50[Sp18][Sp18][Sp18]50-CGCAGACGTTGAT [Phos]-30 46.5 52.5MTb cR1 50-[CF 560]TGGACCATGAATTGGCT[BHQ1][Sp18][Sp18][Sp18]

[Sp18][Sp18][Sp18]CAGCGGGTTGTT-3050.9 59.5

MTb cR2 50-[CF 560]TGGaCCATGAATTGG[BHQ1][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]CAGCGGGTTGTT-30

50.9 52.5

MTb cR3 50-[CF 560]CATGAATTGGCTCAGCTG[BHQ1][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]GGGTTGTTCTGGA-30

48.7 59.3

MTb cR4 50-[CF 560]CATGAATTGGCTCAGCTG[BHQ1][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]CGGGTTGTTCTAGA-30

44.8 59.3

(table continues)

Satterfield

4 jmd.amjpathol.org - The Journal of Molecular Diagnostics

373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434

435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496

SCO 5.2.0 DTD � JMDI289_proof � 23 December 2013 � 5:30 pm � EO: JMD13_0073

½F2�½F2�

Table 1 (continued )

Name Sequences*Tmprimer

Tmcapture

MTb cR5 50-[CF 560]TGGACCATGAATTGG [BHQ1][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]AGCGGGTTGTT-30

48.4 52.5

MTb cR6 50-[CF 560]TGGACCATGAATTG[BHQ1] [Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]CAGCGGGTTGTT-30

50.9 47.9

MTb cR7 50-[CF 560]TGGaCCATGAATT[BHQ1][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]CAGCGGGTTGTT-30

50.9 43.3

MTb cR8 50-[CF 560]CATGAATTGGCTCAGCTG[BHQ1][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]GGGTTGTTCTcGa-30

34.8 59.3

MTb cR9 50-[CF 560]CATGAATTGGCTCAGCTG[BHQ1][Sp18][Sp18][Sp18][Sp18][Sp18][Sp18]GGGTTcTTCTGGA-30

29.5 59.3

TemplatesMTb WT 50-CGTGGAGGCGATCACACCGCAGACGTTGATCAACATCCGGCCGGTGGTC

GCCGCGATCAAGGAGTTCTTCGGCACCAGCCAGCTGAGCCAATTCATGGACCAGAACAACCCGCTGTCGGGGTTGACCCACAAGCGCCGACTGTCGGCGCTGGGGCCCGGCGGTCTGTCACGTGAGCGTGCCGGGCTGGAGGTCCGCGA-30

MTb D516V 50-CGTGGAGGCGATCACACCGCAGACGTTGATCAACATCCGGCCGGTGGTCGCCGCGATCAAGGAGTTCTTCGGCACCAGCCAGCTGAGCCAATTCATGGTCCAGAACAACCCGCTGTCGGGGTTGACCCACAAGCGCCGACTGTCGGCGCTGGGGCCCGGCGGTCTGTCACGTGAGCGTGCCGGGCTGGAGGTCCGCGA-30

The boldfaced bases in the probes are the bases added to the 30 end to form the stem. [FAM] or [CF 560] is the fluorophore FAM or Cal Fluor 560 at the 50 endof the probes. [Sp18] is an 18-atom polyethylene glycol linker (eg, a HEG linker approximately six nucleotides in length). [Sp3] and [Phos] block extension atthe 30 end of some of the capture sequences and are a three-carbon linker and a phosphate molecule, respectively. [DABC] is the dabcyl quencher, and [BHQ1]is the Black Hole Quencher 1.

*Some of the cooperative primers have inverted bases; those that do indicate it with the 50e30 orientation.

Cooperative Primers

497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558

559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619

After amplification, the PCR products were run on a 2.2%FlashGel system (Lonza Inc.) and were imaged.

Incorporating a Probe into Cooperative Primers

P. falciparum real-time PCR was run by making a mastermix with a 250-nmol/L final concentration of each primer(PfcF inv, PfcF inv62, PfcF inv62HP, or PfcF with PfcRinv), a 5-mmol/L final concentration of MgCl2, and anadditional 0.275 U per reaction of GoTaq polymerase inGoTaq colorless master mix (Promega Corp.). Five million,600,000, 50,000, 500, or 0 copies of template were added toeach reaction. The reaction was run on the StepOne system(Applied Biosystems) and included a 20-second denaturingstep at 95�C followed by 45 cycles at 95�C for 1 second and55�C for 20 seconds. Each reaction was run in duplicate.Because of higher fluorescent signals, the capture sequencewith the 30 end attached to the linker was selected as themethod for making the probe in the remaining experiments.

The cooperative primers PfcF and PfcR inv, selectedpreviously herein, were then used with the same cyclingconditions to analyze 5 mL of 1:1000 dilutions (approxi-mately 50 pg of purified genetic material, frozen for 4 years)from Plasmodium samples obtained from BEI ResourcesRepository (Manassas, VA), including P. falciparum St.Lucia, Plasmodium vivax Panama in Aotus, Plasmodiumbrasilianum Peruvian III in Saimiri, Plasmodium cynomolgibastianellii in Rhesus, P. cynomolgi Smithsonian in Rhesus,Plasmodium fragileetype strain in Rhesus, Plasmodiumsimium Howler in Saimiri, Plasmodium knowlesi H strain,

The Journal of Molecular Diagnostics - jmd.amjpathol.orgSCO 5.2.0 DTD � JMDI289_proof �

and P. falciparum FCR-3/Gambia clone D-4, knobless.Negative controls were included, and all the tests were run induplicate.

Single Nucleotide Polymorphism Differentiation

M. tuberculosis real-time PCR for the D516Vmutation in therpoB gene conferring rifampicin resistance was run bymaking a master mix with a 250-nmol/L final concentrationof each primer/probe (MTb cF, MTb P, and one of MTb cR1,MTb cR2, MTb cR3, MTb cR4, MTb cR5, MTb cR6, MTbcR7,MTb cR8, or MTb cR9), a 5-mmol/L final concentrationof MgCl2, and an additional 0.275 U per reaction of GoTaqpolymerase in GoTaq colorless master mix (Promega Corp.).Fifty thousand copies of template (MTbWT or MTb D516V)were added to each reaction. Each reaction was run induplicate. The reaction was run using the ABI 7500 system(Applied Biosystems) and included a 20-second denaturingstep at 95�C followed by 45 cycles at 95�C for 3 seconds and55�C for 32 seconds. The CTs were automatically determinedby the machine with a threshold of 10,000, and the DRn wastaken from cycle 45 of the exported data.

Results

Cooperative Primer PCR Efficiency

To evaluate the concept for cooperative primers, the firststep was to see whether they would even amplify and, if so,with what efficiency (Figure 2). The beta-actin forward

5

620

23 December 2013 � 5:30 pm � EO: JMD13_0073

½F3�½F3�

½F4�½F4�

½F5�½F5�

Figure 2 Cooperative primer amplification efficiency. Cooperativeprimer amplification of 6e6, 6e4, and 6e2 copies of beta-actin DNA run induplicate and the standard curve with data points and standard deviations.Calculated amplification efficiency was 110%.

Figure 4 Effect of hot start on primer-dimer (P-D) propagation in re-actions with normal primers. A hot start does not stop propagation of low-copy-number P-Ds spiked into the reaction with normal primers. Sixty P-Dscompeted with template amplification, and 600 P-Ds eclipsed templateamplification, resulting in a false-negative. Each concentration was run induplicate.

Satterfield

621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682

683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728

cooperative primer (b-act cF) had a Tm that was 2.7�Cbelow the reaction temperature, and the reverse cooperativeprimer (b-act cR) had a Tm 6.9�C below the reactiontemperature. Even so, the primer set amplified with nearperfect efficiency, with a calculated amplification efficiencyof 110%.

Figure 3 Cooperative primer versus normal primer resistance to primer-dimers (P-Ds). Gels of PCR products amplifying 60 copies of P. falciparumDNA (the causative agent of malaria) with and without spiked-in P-Ds fornormal primers and cooperative primers. Each was run in duplicate. Theaddition of just 600 P-Ds in the reaction with normal primers eclipsed theamplification of P. falciparum. This was in contrast to cooperative primers,where even 600,000 P-Ds did not decrease template amplification.

6

729730

SCO 5.2.0 DTD � JMDI289_proof �

Impact of Primer-Dimers

Once the amplification efficiency was shown to be sufficient,we tested the central hypothesis behind cooperative primers:the ability to eliminate propagation of primer-dimers. Wespiked up to 600,000 primer-dimers (100 fmol/L final con-centration) into a malaria reaction with only 60 copies ofP. falciparum template (Figure 3). There was no visibleamplification of spiked-in primer-dimers or dampening of theamplification product when cooperative primers were used.In contrast, the addition of only 600 primer-dimers to a re-action with normal primers resulted in false-negatives. Next,we determined whether a traditional antibody-mediated hotstart could stop primer-dimer propagation with regularprimers (Figure 4). However, the results were similar with orwithout hot start. Primer-dimers were already competingwith the template with only 60 primer-dimers and completelyeclipsed the template amplification with 600 primer-dimers.Hot starts are capable only of preventing primer-dimer for-mation during reaction setup. Once primer-dimers form, theycannot stop primer-dimer propagation.In a subsequent experiment, we attempted to find the limit

of specificity for cooperative primers (Figure 5). Only oncethe primer-dimer concentration was within an order ofmagnitude of the primer concentration with 150,000,000,000primer-dimers spiked into the 10-mL reaction were primer-dimers finally formed and the product eclipsed. In contrast,it took only 600 primer-dimers to eclipse the P. falciparumamplification product using normal primers with or without ahot start. This is consistent with our experience that a 10-foldexcess of a competitive reaction product is sufficient to resultin false-negatives when using normal primers. Cooperativeprimers were spiked with up to 150,000,000 primer-dimers,

Figure 5 Gel showing maximum resistance of cooperative primers toprimer-dimers (P-Ds). Sixty copies of P. falciparum DNA visibly amplify withno trace of P-D amplification in a background of 150,000,000 P-Ds. However,when P-D concentrations were spiked in at concentrations within an order ofmagnitude of the primer concentration (25 nmol/L or 150,000,000,000 P-Dsin a 10-mL reaction), amplification of the product is finally eclipsed, and P-Dsare formed. Each concentration was run in duplicate.

jmd.amjpathol.org - The Journal of Molecular Diagnostics

731732733734735736737738739740741742743744

23 December 2013 � 5:30 pm � EO: JMD13_0073

½T2�½T2�

½F6�½F6�

½F7�½F7�

Table 2 Summary of Amplification Results in gDNA

gDNA (mg) 52.8�C capture 62�C capture Normal primers

0 Amplified Amplified Amplified0.6 Amplified Amplified Amplified1.2 Amplified Failed Amplified3.0 Failed Failed Failed

Cooperative primers and normal primers amplify the target in a back-ground of 0.6 mg of gDNA but not in a background of 3.0 mg of gDNA. Thecooperative primer with a longer capture sequence (Tm of 62�C) was not aseffective as the cooperative primer with a shorter capture sequence (Tm of52.8�C) at amplifying the target in a background of gDNA.

print&web4C=FPO

Figure 6 Cooperative primers with Integrated DNA Technologiesprobes. Labeled cooperative primers (blue) or normal hybridization probes(red) were used for real-time detection of 5,000,000, 50,000, 500, or0 copies of P. falciparum template. Each was run in duplicate. Labeledcapture sequences in cooperative primers had a fluorescent signal 2.5 timeshigher than that of normal hybridization probes, although the capturesequence had a Tm below the reaction temperature.

Cooperative Primers

745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806

807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867

2.5 million times more primer-dimers than template, with nosigns of dampening of the amplification product.

To verify that the result was truly due to cooperativity, thecooperative primers were created without a capture sequence.These isolated, low-Tm primers did not amplify primer-dimers but were also incapable of detectably amplifying evenup to 600,000 copies of P. falciparumDNA (data not shown).In contrast, cooperative primers exhibited efficient amplifi-cation and detection of low copy numbers, demonstrating theimportance of the cooperative link to the capture sequence.

Effect of Complex Samples

To evaluate cooperative primers in complex samples, twodifferent cooperative primers, one with a long capturesequence (PfcF inv62) and the other with a short capturesequence (PfcF), were compared with normal primers PfnFand PfnR (Table 2). Five thousand copies of PlasmodiumDNA were spiked into 0.6, 1.2, or 3 mg of human gDNA. Allthe primers were able to successfully amplify the Plasmo-dium sequence in a background of 0.6 mg of gDNA, but thecooperative primer with the 62�C-Tm capture sequence didnot amplify template in a background of 1.2 mg of gDNA.The short capture sequence cooperative primer and thenormal primers did not amplify the template in a backgroundof 3 mg of gDNA.

Incorporating a Probe into Cooperative Primers

Because probes incorporated into primers have shown rela-tively high signal to noise in the past,18 we attempted toincorporate a probe into the cooperative primer. This wasdone by labeling the capture sequence. First, inverted primerswere attached to the 50 end of capture sequences. To char-acterize the effectiveness of labeling the capture sequence,three different capture sequences were used. The first had aTm below the reaction temperature (PfcF inv), the second hada Tm above the reaction temperature (PfcF inv62), and thethird formed a hairpin to encourage greater quenching (PfcFinv 62HP). However, very little signal was observed fromany of these primers (data not shown), and electrophoreticgels showed that very few of the primers were cleaving thecapture sequence [Figure 3 (the barely visible bands belowthe amplicon of the cooperative primers)].

The Journal of Molecular Diagnostics - jmd.amjpathol.orgSCO 5.2.0 DTD � JMDI289_proof �

We hypothesized that conformational strain from thelinker was lifting the 50 end of the capture sequence andcausing the polymerase to displace the sequence rather thancleave it (Figure 1). Consequently, if we moved the strainfrom the 50 end to the 30 end (eg, by changing where thelinker was attached), the polymerase might cleave the cap-ture sequence with greater efficiency. On testing this hy-pothesis (with PfcF), the fluorescent signal rose dramatically(Figure 6). Although the labeled capture sequence had a Tmbelow the reaction temperature, the signal was still 2.5 timeshigher than the signal from normal hybridization probes.

The cooperative primers with an incorporated probe werethen run on a series of Plasmodium samples obtained fromBEI Resources Repository, including P. falciparum St.Lucia, P. vivax Panama in Aotus, P. brasilianum PeruvianIII in Saimiri, P. cynomolgi bastianellii in Rhesus, P. cyn-omolgi Smithsonian in Rhesus, P. fragileetype strain inRhesus, P. simium Howler in Saimiri, P. knowlesi H strain,and P. falciparum FCR-3/Gambia clone D-4, knobless(Figure 7). The pan-plasmodium primers were able to suc-cessfully detect DNA from each species.

Single Nucleotide Polymorphism Differentiation

Finally, we analyzed the ability of these efficient, primer-dimer resistant, cooperative primers to differentiate singlenucleotide polymorphisms (SNPs). Cooperative primers

7

868

23 December 2013 � 5:30 pm � EO: JMD13_0073

½F8�½F8�½T3�½T3�

Figure 7 Cooperative primers with multiple Plasmodium species. Pan-plasmodium cooperative primers were able to successfully detect multiplespecies of Plasmodium parasites, including P. falciparum St. Lucia, P. vivaxPanama in Aotus, P. brasilianum Peruvian III in Saimiri, P. cynomolgibastianellii in Rhesus, P. cynomolgi Smithsonian in Rhesus, P. fragileetypestrain in Rhesus, P. simium Howler in Saimiri, P. knowlesi H strain, andP. falciparum FCR-3/Gambia clone D-4, knobless. Each was run in duplicate,and only the negative control was negative.

print&

web4C=FPO

Satterfield

869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930

931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975

were designed to the rpoB gene D516V mutation, which ispresent in up to 7.4% of rifampicin-resistant M. tuberculosisisolates in India.19 Two different strategies were used: theamplification-refractory mutation system (ARMS) methodand labeled capture sequence differentiation. Both methodsresulted in the ability to differentiate SNPs similar to stan-dard primers and probes (Figure 8 and data summary inTable 3).

For the probe-based method, cooperative primer MTbcR6 gave the best ratio of fluorescent signals between themutant and wild-type strains. For the ARMS-based method,MTb cR8 gave the best difference in CT values. Both areshown in Figure 8.

Figure 8 SNP differentiation with cooperative primers. Cooperative primers dirifampicin resistance (blue) and without the SNP (red) using probe-based differentbased method with the SNP under the 30 end of the primer (right panel).

8SCO 5.2.0 DTD � JMDI289_proof �

Discussion

Despite the development of more than a dozen hot starts, nocurrently existing method has the ability to stop the propa-gation of primer-dimers once formed (an example ofantibody-mediated hot start is shown in Figure 4). We havecreated low-Tm primers that ordinarily would not amplifythe target or primer-dimers. However, on linking the primersto a capture sequence, they are capable of amplifying thetarget with up to 100% efficiency (Figure 2). Spuriousamplification products, such as primer-dimers, which do nothave a complementary sequence to the capture sequence, arenot propagated (Figures 3 and 5).The principle of cooperativity is what enables the coop-

erative primers to function. The capture sequence binds tothe target, holding the primer in close proximity to thetemplate. This increases the effective concentration byseveral orders of magnitude, which increases the number ofcollisions between the primer and the target DNA, shiftingits effective Tm above that of the reaction temperature andallowing extension to occur. In this manner, the target DNAis amplified, but primer-dimers and other artifacts not pos-sessing a region complementary to the capture sequence arenot propagated (Figure 3).Cooperative primers are not the first real-time PCR tech-

nology to make use of cooperativity. We previously reportedon Tentacle Probes (Arcxis Biotechnologies, Pleasanton,CA), which were built based on cooperativity and led to a100- to 200-fold improvement in kinetics and up to a 10,000-fold improvement in concentration-independent speci-ficity.16,17 However, Tentacle Probes are nonextendable anddo not solve the problem of primer-dimers. Note that hereinwe used cooperativity to make another multiple order ofmagnitude improvement in PCR, only this time with respectto resistance to amplification of primer-dimers. In fact,cooperative primers reduced propagation of primer-dimersand corresponding false-negatives by up to 250 millionefold compared with normal primers, with or without a hotstart. This is such a strong reduction in sensitivity to primer-dimers that it would take nearly every primer in the reactionforming a primer-dimer during reaction setup to cause

fferentiate between tuberculosis complex with the rpoB D516V SNP causingiation with the SNP under the capture sequence (left panel) and the ARMS-

jmd.amjpathol.org - The Journal of Molecular Diagnostics

976977978979980981982983984985986987988989990991992

23 December 2013 � 5:30 pm � EO: JMD13_0073

3

Table 3 Summary of SNP Differentiation Methods

D516V primername Method

PrimerDTm (�)

ProbeDTm (�) DCT

DRnVar/DRnWT

MTb cR1 Probe (4.1) 4.5 1.68 3.01MTb cR2 Probe (4.1) (2.5) 4.79 3.35MTb cR5 Probe (6.6) (2.5) 4.83 1.89MTb cR6 Probe (4.1) (7.1) 3.83 3.67MTb cR7 Probe (4.1) (11.7) 5.30 3.62MTb cR3 ARMS (6.3) 4.3 4.43 NAMTb cR4 ARMS (10.2) 4.3 5.99 NAMTb cR8 ARMS (20.2) 4.3 7.57 NAMTb cR9 ARMS (25.5) 4.3 7.13 NA

Each primer is listed together with whether it uses ARMS- or probe(labeled capture sequence)-based differentiation, the number of degreesthe predicted Tm for the primer or probe is above or below the reactiontemperature (values below the reaction temperature are shown in paren-thesis), the difference between mutant and wild-type CT values, and theratio of the mutant and wild-type fluorescence.NA, not available.

Cooperative Primers

9939949959969979989991000100110021003100410051006100710081009101010111012101310141015101610171018101910201021102210231024102510261027102810291030103110321033103410351036103710381039104010411042104310441045104610471048104910501051105210531054

1055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094109510961097109810991100110111021103110411051106110711081109111011111112111311141115

primer-dimererelated signal dampening or false-negatives.This type of catastrophic failure is extremely unlikely. Toforce amplification failure in Figure 5, the primer-dimersspiked into the reaction had 100% overlap with the fulllength of the forward and reverse cooperative primers. Evenso, it took a nearly equivalent concentration of primer-dimersto primers manually spiked into the reaction to force ampli-fication failure. In other words, properly designed coopera-tive primers should be all but impervious to the effects ofprimer-dimers.

Although the cooperative primers exhibit incrediblespecificity in reduction of primer-dimers, care must still betaken in the design of primers that may be used with sam-ples containing large amounts of gDNA. The short primingregion by itself is not sufficient to guarantee specificity inthe presence of complex sequences. Capture sequences witha Tm above the reaction temperature, such as the 62�Ccapture sequence in Table 2, have the potential to increasemispriming. This is because a large capture sequence ismore likely to nonspecifically bind to the genomic DNA,providing sufficient stability for the short priming region toamplify the nontarget gDNA sequence after the initialmispriming event.

Even when the capture sequence is designed optimally,such as the 52.8�C capture sequence in Table 2, cooperativeprimers still do not solve all the problems of normal primers.Despite reducing primer-dimer propagation up to 250millionefold more effectively than normal primers, thecooperative primers were still no more effective than normalprimers in reducing mispriming of gDNA.

In the present study, we used a single annealing tem-perature of 55�C with just one type of master mix. We findthis temperature convenient because it is compatible withsingle-tube reactions for RT-PCR, where both the reversetranscriptase step and the annealing step during polymeri-zation are performed at the same temperature. The principal

The Journal of Molecular Diagnostics - jmd.amjpathol.orgSCO 5.2.0 DTD � JMDI289_proof �

behind cooperative primers suggests that they should becompatible with any reaction temperature and in a variety ofdifferent salt or primer concentrations. However, slighttweaking of the general design parameters (eg, the ideal Tmfor the primer and the capture sequence) will likely benecessary. Further work will be needed to fully explore therange of conditions possible and the optimum design pa-rameters for each in real-time PCRs.

Scorpion Qprimers may be somewhat similar in appearanceto cooperative primers, but they are not cooperative.18

Scorpion primers bind to themselves after the primer isextended rather than cooperatively binding to the template.Because there is no cooperative capture sequence, a low-Tmprimer would not work in a Scorpion primer.

Dual-priming oligonucleotides are the only example of aprimer that could be considered cooperative in function.20 Ashort primer of 6 to 12 bases is linked to a high-Tm capturesequence connected via a series of deoxyinosines. The dif-ference between the function of the dual-priming oligonu-cleotide and the present cooperative primer is that the capturesequence in the former binds upstream from the primer ratherthan downstream. Because the linker is composed of deox-yinosines, the polymerase extends through the entire primer,including the capture sequence, enabling high-efficiencyPCR. However, this same principle prevents it from beingable to stop propagation of primer-dimers once formed. If anonextendable linker were to be used in the dual-primingoligonucleotide, then the first-round amplification productwould not include a complementary region to the capturesequence, preventing further amplification. In contrast, thecooperative primers in the present work maintain affinity andspecificity during every round of amplification owing to theunique mechanism of the capture sequence binding down-stream from the primer.

Given that the capture sequence in the cooperative primermust bind for the primer to extend, it seemed convenient tolabel the capture sequence and use it as a probe. This couldeliminate the need to identify a third conserved region in agiven target or the need to add a separate probe to the re-action mix. Furthermore, the potential to design the capturesequence with a Tm below the reaction temperature couldmake it impervious to false-positives from spurious ampli-fication products, seeing as the capture sequence cannotlight up unless the primer binds to a region of the targetimmediately upstream from it. All these ideas lent them-selves toward incorporating a probe directly into the coop-erative primer.

However, when we labeled the capture sequence attachedto the inverted primer, there was very little fluorescence.This did not change even for high-Tm capture sequences orcapture sequences with hairpin structures. We noticed thatthe electrophoretic gels, like the one shown in Figure 3, hada faint band underneath the amplicon for cooperativeprimers. This band was consistently the length of theamplicon minus the capture sequence(s). Thus, it seemedthat something about the inverted cooperative primer caused

9

1116

23 December 2013 � 5:30 pm � EO: JMD13_0073

Satterfield

11171118111911201121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178

1179118011811182118311841185118611871188118911901191119211931194119511961197119811991200120112021203120412051206120712081209121012111212121312141215121612171218121912201221122212231224122512261227122812291230123112321233123412351236123712381239

the polymerase to favor displacing the capture sequencerather than cleaving it. This, in turn, would allow amplifi-cation but would prevent generation of a signal.

We hypothesized that the strain of the linker on the 50 endof the capture sequence was what caused the polymerase tofavor displacement. Consequently, a cooperative primer thatwas attached to the capture sequence at the 30 end mightallow for efficient 50-30 nuclease activity by removing thestrain from the 50 end. As it turned out, multiple primerswith the linker attached to the 30 end of the capture sequencegenerated large fluorescent signal, even when the capturesequence was as much as 11.7�C below the reaction tem-perature (Figures 6 and 8). In fact, the fluorescent signal wasso high that it was more than 2.5 times the signal of normalhybridization probes. It could be thought that the signal washigher purely because of better quenching efficiency on theshort capture sequence versus the relatively long hybridi-zation probe. However, the hybridization probe in theseexperiments used secondary structure, placing the quencherwithin seven bases of the fluorophore, theoretically giving itmore efficient quenching than the short capture sequence inthe cooperative primer. Although the short distance betweenthe quencher and the fluorophore on the capture sequencecertainly contributes to the signal, it does not seem to be theonly factor. It is possible that some of the increased fluo-rescence is because the primer does not have to competewith the synthesized template for hybridization to the targetsequence the way that a hybridization probe does.

Differentiating SNPs using ARMS- and probe-based dif-ferentiation seemed to yield similar results for cooperativeprimers as for normal primers and probes. The 30 mismatchon the primer delayed the CT by approximately four cycles.An intentional mismatch added to the third base from the 30

end in addition to the SNP under the 30 end delayed the CT byapproximately six to seven cycles (Table 3). These are typicalresults for ARMS-based methods using normal primers.21e23

Probe-based discrimination showed strong signal for an exactmatch, with a weak signal for a mismatch (Figure 5). This istypical of results for dual-labeled hybridization probes.24e26

Although we were able to obtain signal and amplificationfrom all the designs that were tried, some seemed to be moreefficient than others. This has led to a few general obser-vations for reasonably effective design for reaction tem-peratures at 55�C. For best results with cooperative primerscontaining three HEG linkers (eg, inverted primers attachedto the 50 end of the capture sequence), a Tm 5�C to 7�Cbelow the reaction temperature should be used. For bestresults with cooperative primers containing six HEG linkers(eg, primers attached to the 30 end of the capture sequence),a Tm for both oligonucleotides 4�C to 6�C below the re-action temperature is desirable. Although results can beobtained with lower Tms, the risk of losing either signal oramplification efficiency grows dramatically. Higher Tmdesigns can also be used, but, in theory, the lower the Tm ofthe primer, the more resistant it will be to amplification ofspurious products. Also, for SNP differentiation, the lower

10SCO 5.2.0 DTD � JMDI289_proof �

the Tm in the target range, the better. High-Tm capturesequences could potentially be used to provide additionalstability for highly polymorphic targets, but additional workneeds to be done to explore this application.

Conclusion

More than a dozen hot starts have been developed to reduceprimer-dimer formation during PCR setup; however, noneof these can prevent primer-dimer formation after the initialhot start or the propagation of primer-dimers once formed.With increasing multiplexing demands, the need for a realsolution to primer-dimers is only increasing. Cooperativeprimers are the first instance of a technology that preventsthe propagation of primer-dimers. The implications to thefield of life sciences, which relies on the use of PCR, aretremendous. We believe that this technology will provideresearchers with the ability to rapidly and easily createmultiplexed tests almost without hindrance from primer-dimers.

Acknowledgments

We thank Dr. Michael R. Caplan (Arizona State University,Tempe, AZ) for critical review of the manuscript.The following reagents were obtained through the MR4 as

part of the BEI Resources Repository, National Institute ofAllergy and Infectious Diseases, NIH: Genomic DNA-P.falciparum St. Lucia, MRA-331G; P. vivax Panama in Aotus,MRA-343G; P. brasilianum Peruvian III in Saimiri, MRA-349G; P. cynomolgi bastianellii in Rhesus, MRA-350G; P.cynomolgi Smithsonian in Rhesus, MRA-351G; P. fragile-type strain in Rhesus, MRA-352G; P. simium Howler in Sai-miri, MRA-353G; P. knowlesi H strain, MRA-456 G; and P.falciparum FCR-3/Gambia clone D-4, knobless, MRA-739G.

References

1. Mullis K, Faloona F, Scharf S, Saiki R, Horn G, Erlich H: Specificenzymatic amplification of DNA in vitro: the polymerase chain reac-tion. Cold Spring Harb Symp Quant Biol 1986, 51(Pt 1):263e273

2. Mullis KB: The unusual origin of the polymerase chain reaction. SciAm 1990, 262:56e61, 64e65

3. Vallone PM, Butler JM: AutoDimer: a screening tool for primer-dimerand hairpin structures. Biotechniques 2004, 37:226e231

4. Faloona F, Weiss S, Ferre F, Mullis K: Direct detection of HIV se-quences in blood: high-gain polymerase chain reaction. Abstract 1019.Presented at the Sixth International Conference on AIDS, 1990 June20-24. San Francisco, CA

5. Chou Q, Russell M, Birch DE, Raymond J, Bloch W: Prevention ofpre-PCR mis-priming and primer dimerization improves low-copy-number amplifications. Nucleic Acids Res 1992, 20:1717e1723

6. Kaijalainen S, Karhunen PJ, Lalu K, Lindstrom K: An alternative hotstart technique for PCR in small volumes using beads of wax-embedded reaction components dried in trehalose. Nucleic Acids Res1993, 21:2959e2960

7. Rapley R: Enhancing PCR amplification and sequencing using DNA-binding proteins. Mol Biotechnol 1994, 2:295e298

jmd.amjpathol.org - The Journal of Molecular Diagnostics

1240

23 December 2013 � 5:30 pm � EO: JMD13_0073

4

Cooperative Primers

124112421243124412451246124712481249125012511252125312541255125612571258125912601261126212631264126512661267126812691270

127112721273127412751276127712781279128012811282128312841285128612871288128912901291129212931294129512961297129812991300

8. Kaboev OK, Luchkina LA, Tret’iakov AN, Bahrmand AR: PCR hotstart using primers with the structure of molecular beacons (hairpin-like structure). Nucleic Acids Res 2000, 28:E94

9. Kainz P, Schmiedlechner A, Strack HB: Specificity-enhanced hot-startPCR: addition of double-stranded DNA fragments adapted to theannealing temperature. Biotechniques 2000, 28:278e282

10. Kermekchiev MB, Tzekov A, Barnes WM: Cold-sensitive mutants ofTaq DNA polymerase provide a hot start for PCR. Nucleic Acids Res2003, 31:6139e6147

11. Barnes WM, Rowlyk KR: Magnesium precipitate hot start method forPCR. Mol Cell Probes 2002, 16:167e171

12. Koukhareva I, Haoqiang H, Yee J, Shum J, Paul N, Hogrefe RI,Lebedev AV: Heat activatable 3’-modified dNTPs: synthesis andapplication for hot start PCR. Nucleic Acids Symp Ser (Oxf) 2008:259e260

13. Lebedev AV, Paul N, Yee J, Timoshchuk VA, Shum J, Miyagi K,Kellum J, Hogrefe RI, Zon G: Hot start PCR with heat-activatableprimers: a novel approach for improved PCR performance. NucleicAcids Res 2008, 36:e131

14. Kellogg DE, Rybalkin I, Chen S, Mukhamedova N, Vlasik T,Siebert PD, Chenchik A: TaqStart antibody: “hot start” PCR facilitatedby a neutralizing monoclonal antibody directed against Taq DNApolymerase. Biotechniques 1994, 16:1134e1137

15. Birch DE: Simplified hot start PCR. Nature 1996, 381:445e44616. Satterfield BC, Bartosiewicz M, West JA, Caplan MR: Surpassing

specificity limits of nucleic acid probes via cooperativity. J Mol Diagn2010, 12:359e367

17. Satterfield BC, West JA, Caplan MR: Tentacle probes: eliminatingfalse positives without sacrificing sensitivity. Nucleic Acids Res 2007,35:e76

The Journal of Molecular Diagnostics - jmd.amjpathol.orgSCO 5.2.0 DTD � JMDI289_proof �

18. Whitcombe D, Theaker J, Guy SP, Brown T, Little S: Detection ofPCR products using self-probing amplicons and fluorescence. NatBiotechnol 1999, 17:804e807

19. Suresh N, Singh UB, Arora J, Pant H, Seth P, Sola C, Rastogi N,Samantaray JC, Pande JN: rpoB gene sequencing and spoligotyping ofmultidrug-resistant Mycobacterium tuberculosis isolates from India.Infect Genet Evol 2006, 6:474e483

20. Chun JY, Kim KJ, Hwang IT, Kim YJ, Lee DH, Lee IK, Kim JK: Dualpriming oligonucleotide system for the multiplex detection of respi-ratory viruses and SNP genotyping of CYP2C19 gene. Nucleic AcidsRes 2007, 35:e40

21. Newton CR, Graham A, Heptinstall LE, Powell SJ, Summers C,Kalsheker N, Smith JC, Markham AF: Analysis of any point mutationin DNA: the amplification refractory mutation system (ARMS).Nucleic Acids Res 1989, 17:2503e2516

22. Germer S, Holland MJ, Higuchi R: High-throughput SNP allele-frequency determination in pooled DNA samples by kinetic PCR.Genome Res 2000, 10:258e266

23. Ye S, Dhillon S, Ke X, Collins AR, Day IN: An efficient procedure forgenotyping single nucleotide polymorphisms. Nucleic Acids Res 2001,29:E88e8 Q

24. Yao Y, Nellaker C, Karlsson H: Evaluation of minor groove bindingprobe and Taqman probe PCR assays: influence of mismatches andtemplate complexity on quantification. Mol Cell Probes 2006, 20:311e316

25. Tyagi S, Bratu DP, Kramer FR: Multicolor molecular beacons forallele discrimination. Nat Biotechnol 1998, 16:49e53

26. Marras SA, Kramer FR, Tyagi S: Multiplex detection of single-nucleotide variations using molecular beacons. Genet Anal 1999, 14:151e156

1123 December 2013 � 5:30 pm � EO: JMD13_0073