Gold nanoparticle-assisted polymerase chain reaction: effects of surface ligands, nanoparticle shape and material

Institute of Biochemistry and Physiology Academy of Sciences, 13 Prospekt Entuzi khlebtsov@ibppm.ru Saratov Scientic and Research Veterinary I Sciences, 6 Ulitsa 53 Strelkovoj Divizii, Sara Saratov National Research State Univers 410012, Russia † Electronic supplementary information as-prepared gold and silica nanoparticl as-prepared and functionalized NPs be chemical structures and IUPAC name nano-PCR specicity and efficiency; the e 710 bp target from A. brasilense and a 1 effect of AuNPs–CTAB, AuNPs–PDDA and of AuNPs on two-round PCR model 1; the of a long 1156 bp target from PCR model Cite this: RSC Adv., 2016, 6, 110146


Introduction
The polymerase chain reaction (PCR) has become a standard biomolecular technique, which is commonly used for sensitive and rapid DNA detection.2][3] However, the PCR can be error-prone and the specicity and efficiency of PCR can be unsatisfactory without proper optimization. 4As a result of the poor purication of template DNA, primer-template mismatches, spontaneous formation of primer dimers, and non-optimized PCRmixture ratio and thermocycling conditions, the formation of the target amplicon may be accompanied by non-specic side products called PCR artifacts.To x this problem, several enhancers have been introduced: (1) various additives in PCR mixture, such as betaine, 5 dithiothreitol, 6 dimethyl sulfoxide, 7 single-stranded DNA-binding protein (SSB); 8 (2) instrumental design, including development of thermocyclers with precise heating/cooling rates; 9,10 (3) optimization of PCR system through optimal Mg 2+ concentration, cycle numbers and proper primer design; [11][12][13] (4) enzyme modication; 14,15 and (5) new touchdown 16 and nested 17 PCR strategies.Although these tools improve the PCR outcome, yet they are not all-purpose and the optimization protocol can be case dependent.
During the past decade, various nanomaterials have been used for PCR enhancement, including gold nanoparticles (AuNPs), 18 silver nanoparticles and graphene oxide, 19 reduced graphene oxide, 19 titanium dioxide and quantum dots, 20 upconversion nanoparticles, 21 fullerenes (C 60 ), 22 carbon nanoparticles and nanotubes, 23 carbon nanopowder, 24 magnetic nanoparticles, 25 semiconductor nanomaterials, 26,27 dendrimers, 28 etc.Nevertheless, AuNPs are one of the promising candidates for use as PCR enhancers owing to their controlled geometrical parameters, reproducible synthesis and functionalization protocols.Recently, several hypothetic mechanisms of AuNP-assisted PCR were proposed to account for: (1) the SSBlike mechanism 29 and electrostatic interactions between AuNPs and PCR components; 30 (2) the high surface-to-volume ratio and non-specic adsorption of PCR components; [28][29][30][31] and (3) the thermal properties of AuNPs. 32However, none of the suggested mechanisms can be considered convincing.
Here, we elucidated the role of the particle functionalization with neutral (PVP and mPEG-SH) and cationic (PDDA and CTAB) polymer ligands and role of the nanoparticle material by performing PCR with gold and silica particles of comparable size and charge, other PCR conditions being unchanged.In addition, the nanoparticle shape effect was examined with CTAB stabilized nanorods.All nanoparticles were tested with two PCR diagnostic models: (1) the nitrogen xation (NifD) gene from Azospirillum brasilense Sp7 bacteria; and (2) the polymorphic ompA gene encoding major outer membrane protein of Chlamydia trachomatis.At an optimal concentration of citratestabilized 16 nm negatively charged spherical AuNPs (0.4 nM), the PCR specicity and efficiency was greatly enhanced, whereas the negatively charged PVP-and PEG-stabilized AuNPs did not reveal any enhancing properties.Also, no specic PCR enhancement was observed with positively charged AuNPs and gold nanorods.By contrast to 16 nm negatively charged citratestabilized AuNPs, silica nanoparticles with the same charge and comparable 20 nm size have no impact on PCR efficiency.These observations show an important role of the nanoparticle material and surface modication in PCR enhancement, and could be helpful for further mechanistic studies.

Characterization of nanoparticles
The following as-prepared and surface-functionalized AuNPs were used: negatively charged citrate-stabilized spherical gold nanoparticles (AuNP); nanoparticles capped with neutral polymers PVP and mPEG-SH (AuNP-PVP and AuNP-PEG-SH, respectively) and cationic polymers CTAB and PDDA (AuNP-CTAB and AuNP-PDDA, respectively); gold nanorods (AuNR-662, 662 stands for the longitudinal plasmon resonance maximum wavelength) with aspect ratio of about 2.3; and spherical silica nanoparticles (SiO 2 NP).The acronym denitions and chemical structures of polymeric ligands are given in ESI le (Section S1).† Fig. 1 shows illustrative TEM images of citrate-stabilized AuNPs, gold nanorods, and silica nanospheres.Extinction spectra of nanoparticles are shown in Fig. S1 (ESI).† Table 1 summarizes the values of the zetapotentials 2 (mV), absorption maximum wavelengths of asprepared NPs in aqueous solution (for nanorods, the longitudinal resonance is indicated); the average TEM and DLS sizes; the nanoparticle molar concentrations normalized to the extinction A max ¼ 1; and the optimal concentrations and maximal specicity and efficiency of different PCR additives used in our PCR systems.
It follows from Table 1 that the average TEM and DLS diameters of citrate NPs (16 AE 0.8 nm and 17 AE 1.2 nm, respectively) have no signicant difference within the standard deviation (SD) of about 1 nm.By contrast, aer functionalization with PVP, PEG-SH, CTAB, and PDDA polymeric ligands, the hydrodynamic diameters increase signicantly by 6.8, 10.4, 3.6, and 7.4 nm.We consider these data as strong evidence for successful attachment of ligands.The second evidence comes from drastic change in colloidal stability of NPs before and aer functionalization.Fig. S2 (ESI †) shows extinction spectra and photos of citrate-stabilized and ligand-stabilized AuNPs before and aer addition of 0.1 M NaCl salt.As expected, the addition of salt to the citrate-stabilized AuNPs results in immediate aggregation which is evident from changes in suspension color and extinction spectrum.By contrast, the addition of the same amount of salt to the ligand-stabilized AuNPs does not induce any aggregation phenomena accompanied by color and spectral changes.What is more, the ligand-stabilized NPs do not aggregate even in salt environment under strong temperature variations during PCR cycles (see below, Fig. 4).

Mechanisms of AuNP enhancing effect
Effect of nanoparticles on non-optimized PCR system was investigated by performing PCR with gold and silica particles of comparable size and charge, without changing the typical PCR conditions such as reagent concentrations and amplication prole.
To evaluate the overall ability of AuNPs to enhance PCR in terms of specicity and yield (efficiency), two PCR models were examined: (1) nitrogen xation (NifD) gene from A. brasilense Sp7 bacterial culture (A. brasilense DNA detection, 710 bp amplicon), and (2) chromosomal ompA gene of C. trachomatis (157 to 160 bp target sequences for the variable domain 2 of MOMP).For brevity, we will refer to PCR models 1 and 2, respectively.
First, an optimal concentration of 16 nm citrate-stabilized AuNPs was found to be 0.4 nM by adding AuNPs in various concentrations to both PCR systems (Fig. S3, ESI †).This result is in good agreement with previous reports for quite different PCR models. 29,32Fig. 2a illustrates the specicity of AuNP-assisted PCR performed with model 1.Numerous nonspecic bands (smears) are observed for duplicate lanes 1 and 2 without nanoparticles, performed by using DNA template mixed with heterogeneous DNA fragments from similar microorganisms, whereas the addition of citrate-stabilized AuNPs at an optimal 0.4 nM concentration completely inhibits all non-target bands.Fig. 2b exemplies the efficiency of AuNP-assisted PCR, when no signicant bands can be observed without nanoparticles (duplicate lanes 1 and 2) in the case of a low DNA template concentration, performed with ten-fold diluted initial template DNA.It should be emphasized, that even for this unoptimized system, the addition of AuNPs at optimal 0.4 nM concentration results in a great PCR yield enhancement.To the best of our knowledge, this is the rst demonstration of the yield-enhanced AuNP-PCR for A. brasilense Sp7 (NifD) gene.Additionally, we have observed the enhancing effect of AuNPs on two-round error-prone PCR system by taking the PCR product as a template for the second round of PCR amplication (Fig. S5, ESI †).Moreover, we have shown that AuNPs could be used to improve PCR specicity while amplifying the long PCR product (Fig. S6, ESI †) which is in good agreement with a recent study by Zhou et al. 33 Finally, in the third test with the model 1 (Fig. 2c), the AuNPs were added aer amplication step to verify whether the AuNPs affect the amplication step themselves (as in Fig. 2a  and b) or the post-amplication PCR outcome, just before gel electrophoresis.As shown on Fig. 2c, there was no effect of AuNPs added aer amplication step without nanoparticles and the smeared nonspecic bands look similarly for lanes 1 and 2. This gives strong evidence for the rst assumptioncitrate-stabilized AuNPs enhance the PCR outcome just during the amplication step.
One of hypothetic mechanisms of nano-PCR 29 is that AuNPs exhibit SSB-like behavior related to greater affinity of singlestranded DNA-binding protein for ssDNA compared to dsDNA.However, this explanation remains unanswered the following question: how many AuNPs per one PCR tube are needed to perform efficient PCR amplication caused by an SSB-like moiety alone?Below we provide simple estimations based on our experimental conditions.A typical PCR mixture, used in our experiments with optimal concentration 0.4 nM of AuNPs, 29 contained 2.5 U of Taq DNA-polymerase, equal to 10 ng or 3.5 nM of enzyme (the activity of 250 000 U Taq-DNA polymerase is equal to 4 ng or 43 fmol of 94 kDa M w protein).A rough estimation shows that the molar ratio between Taq DNApolymerase and AuNPs is about 1 : 10.The number of DNA templates, which exponentially grows in each of 35 PCR cycles,  is at least one order greater than the number of DNApolymerase molecules per one PCR tube.This means that the molar ratio DNA : AuNPs is about 1 : 1, which is quite insufficient to ensure AuNP action as an SSB-like substances. 34onsequently SSB-like behavior of gold nanoparticles seems to be not relevant to AuNP-assisted PCR enhancement.
To investigate a possible role of surface functionalization, we used AuNPs stabilized with two neutral polymer ligands such as PVP (AuNPs-PVP) and mPEG-SH (AuNPs-PEG-SH).According to Table 1 data, the both polymers do not change the sign and, in some extent, the value of nanoparticle zeta-potential.Surprisingly enough, the enhancing effect of AuNPs-PVP and AuNPs-PEG-SH on PCR was negligible (Fig. 3).
In particular, non-specic bands were present even at high concentrations of nanoparticles, and the PCR tubes remained pink colored aer amplication (Fig. 4).In some cases, the addition of a high nanoparticle amount resulted in complete inhibition of PCR (see, e.g., lanes 11 and 7 in Fig. 3a and b, respectively).We note that all these experiments included reagent controls with free polymers added to PCR tubes to verify the polymer effect alone (see lanes SA in Fig. 3).Thus, the functionalization of AuNP surface with neutral polymers, which does not change the initial negative charge of AuNPs, results in complete suppression of the PCR enhancement.This observation can be attributed to the steric hindrance for biomolecular PCR components by protective PVP or mPEG-SH layers.In particular, these stabilizers prevent the nanoparticle aggregation aer PCR amplication (Fig. 4b).By contrast, the stabilization of AuNPs with small citrate ions results in great enhancing PCR outcome (Fig. 2a and b), but does not prevent nanoparticle aggregation at the end of PCR (Fig. 4a).
The next set of experiments was aimed at understanding the role of AuNP functionalization with positively charged ligands.These experiments were motivated by a recent study, 30 revealing a crucial role of the particle charge in PCR enhancing.Specically, the optimal enhancing concentration of positivelycharged PDDA-capped AuNPs was shown 30 to be 1.54 pM, which is 260 times lower than the optimal 400 pM concentration of negatively charged citrate-stabilized AuNPs. 29The goal of our experiment was twofold: ( 1 ) to verify the enhancing role of PDDA ligand for PCR models that differs from those investigated in ref. 30; (2) to examine the enhancing role of positive nanoparticle charge with CTAB ligand, which differs from PDDA.Clearly, the affirmative results for both points would expand the reported observations 30 and conrm a universal role of the positive AuNP charge in PCR enhancing.However, our experiments with models 1 and 2 gave negative answers for both questions.Fig. 5a demonstrates the presence of non-target bands for very small (0.001, 0.01 pM) and high (10, 100 pM) concentrations of positively charged CTAB-stabilized AuNPs.Furthermore, for the same experimental model, no signicant enhancement of PCR outcome was observed with PDDA-capped AuNPs (Fig. 5b) at variation of nanoparticle concentration from 0.001 pM to 5 pM.Finally, when we used the positively charged CTAB-stabilized gold nanorods, the results were the same: no PCR enhancing or even complete inhibition.This illustrates insignicant role of the nanoparticle shape, at least for the PCR models examined here.Similar negative results were obtained for second PCR model, i.e. with nitrogen xation (NifD) gene from A. brasilense Sp7 (Fig. S4, ESI †).Thus, our experiments with two PCR models and with two cationic polymers did not  conrm any efficient PCR enhancing reported previously 30 for a different PCR model and positively charged AuNPs with PDDA ligand.
The precise reasons for observed discrepancy between our results and previously reported ones are unclear at present.Furthermore, the observed effects at trace 0.001 AuNP concentrations cannot be explained in terms of existing mechanistic models.In any case, we can conclude that the enhancing role of positively charged AuNPs is not ubiquitous.Instead, it can be ligand-dependent and PCR model-dependent.
Another proposed mechanism of AuNP-assisted PCR is related to high thermal conductivity of AuNPs as compared to water. 32This can change the amplication conditions in local environment around gold nanoparticles, thus changing the local conditions in "nanoreactor".On the other hand, it has been supposed 30 that Au nanoparticles in PCR mixture act as small nanoreactors, in which the local concentration of primers and templates is increased owing to electrostatic interactions between AuNPs and PCR components.In this model, the particle charge is important, irrespective of the particle material.To verify these models, we fabricated silica nanospheres with the 20 nm size and À31.2 mV zeta-potential close to those for citrate-stabilized AuNPs (16 nm and À29.9 mV, respectively).If the nanoparticle charge plays a crucial role, one would expect a similar PCR enhancement for both particle types.If the material heat properties are important, one would expect quite different PCR outcomes.Fig. 6 shows that the second alternative agrees with experimental data.Indeed, we observed dramatic difference in PCR results for gold and silica nanoparticles with close parameters except for their heat properties.Of course, the chemical structure and physical properties of the nanoparticle surface should also be considered for ultimate conclusions.

Conclusion
In conclusion, we have studied the effects of AuNP concentration, charge, shape and material on the PCR efficiency and specicity.For two new PCR systems that were not studied previously, we have obtained high specic and efficient enhancement of target amplicons at optimal citrate-stabilized AuNP concentration about 0.4 nM.3][34][35][36] Functionalization of citrate-stabilized AuNPs with neutral PVP and PEG-SH polymers results in elimination of enhancing properties, although the particle zeta-potentials remain the original sign and values.This phenomenon can be related to steric hindrance effects that prevent a proper interaction of PCR molecular components with nanoparticles.Functionalization of AuNPs with cationic polymers PDDA and CTAB leads to expected reverse of the particle charge sign.However, in contrast to previous report on positive  For all nanoparticle concentrations from 0.4 to 20 nM (lanes 2-6), the PCR outcomes are similar to those obtained without nanoparticles (lane 1).Thus, for both PCR models, the effect of negatively charged silica nanoparticles is not significant as compared with that for citratestabilized AuNPs.The normalized quantities of PCR specificity (S L ) and efficiency (E L ) for each lane are indicated on the bottom panels.
enhancing properties of PDDA capped particles, we have obtained non-distinguishable enhancing effect.The particle shape plays negligible role as demonstrated by comparison of PCR outcomes with CTAB-coated gold nanospheres and nanorods.We have performed also PCR amplications with two gold and silica particles of similar size and charge to show that the gold particles greatly enhance PCR results whereas silica counterparts do not.This implies a possible positive role of metallic heat properties in local environment of PCR nanoreactor.To summarize, further studies are needed to understand the underlying physicochemical mechanisms of nanoparticleenhanced PCR, including the role of nanoparticle size, charge, surface functionalization, and the dynamic ratio between nanoparticle concentration and concentration of PCR system components.

Synthesis of nanoparticles
Spherical AuNPs with 16 nm average diameter were fabricated by reduction of HAuCl 4 with sodium citrate as described by Grabar et al. 37 and were used as a template for subsequent capping with various surface ligands.Briey, 25 mL of 38.8 mM sodium citrate was added quickly to boiled 250 mL of 1 mM water solution HAuCl 4 , which resulted in a change in solution color from pale yellow to deep red.
To obtain polymer-capped AuNPs, 38 the as-prepared AuNPs were mixed with appropriate amount of each surface agent such as mPEG-SH, PVP or PDDA respectively, incubated for some time with subsequent centrifugation and resuspension in MQ water.A detail description of ligand capping procedures are given in ESI le.† Gold nanorods (AuNRs) with length and diameter about 70 nm and 25 nm, respectively, were fabricated according to Ratto et al. 39 with minor modications. 40First, gold seed particles were prepared by adding aqueous ice-cold sodium borohydride (10 mM, 0.1 mL) to a mixed aqueous solution of CTAB (0.1 M, 1 mL) and HAuCl 4 (10 mM, 0.025 mL) and were vigorously stirred for about 2 minutes.Aer sequential addition of 28 mL of 100 mM AA and 12 mL of two-hours-aged gold seeds to a growth solution (5 mL of 0.1 M CTAB, 250 mL of 0.01 M HAuCl 4 , and 100 mL of 4 mM AgNO 3 ), the mixture was incubated for 24 h at 25 C. Finally, 16 mL of 10 mM ascorbic acid was added in three portions every 24 h.AuNRs were allowed to grow overnight without stirring at 30 C.Then, as-prepared AuNRs were repeatedly centrifuged and redispersed overnight in 1 mM CTAB.
Highly monodisperse SiO 2 NPs with 20 nm TEM average diameter were synthesized following the multistep seedmediated growth technique in an aqueous solution of L-argi- nine. 41Briey, 9.1 mg of L-arginine was added to 6.9 mL of water in a standard 20 mL scintillation vial under magnetic stirring.Then, 0.45 mL of cyclohexane was accurately added to the top of the solution and the mixture was heated to 60 C. Further, 0.55 mL of TEOS was added to the top layer of cyclohexane and the mixture was allowed to react for 20 h.A key tip for synthesizing high-quality monodisperse SiO 2 NPs is to keep the cyclohexane and water parts unmixed, ensuring very slow addition of TEOS to the reaction mixture.

Characterization
Extinction spectra of as-prepared and ligand-capped nanoparticles solutions were measured with a Specord BS-250 and Specord S-300 spectrophotometers (Analytik, Jena, Germany).Transmission electron microscopy (TEM) images were obtained using Libra-120 transmission electron microscope (Carl Zeiss, Jena, Germany) at the Simbioz Center for the Collective Use of Research Equipment in the Field of Physical-Chemical Biology and Nanobiotechnology at the IBPPM RAS.The average diameters of as-prepared and polymer-capped AuNPs and silica nanoparticles and their zeta-potentials were measured with a Zetasizer Nano ZS device (Malvern, UK).  2. Bacteria were cultured at 37 C in a liquid malate-salt medium (MSM) as described previously. 44The starting cell density of the 18 h cultures for DNA extraction was estimated by absorbance measurements on a Specord S-300 spectrophotometer at 660 nm. 1 mL of the bacterial suspension (2 Â 10 7 cells per mL) aer centrifugation (Minispin, Eppendorf) at 5000g for 10 min was suspended in TE buffer (Tris HCl, 10 mM, EDTA, 1 mM).A genomic DNA puri-cation kit (Thermo Scientic, Lithuania) was used for DNA extraction following the manufacturer's instruction.The kit is based on selective detergent-mediated DNA precipitation of crude lysate from different sample sources, including bacterial cells.The extracted DNA samples were further resuspended into 100 mL of MQ water.The extracted DNA concentration and purity were estimated by UV-spectrophotometry on a Specord BS-250 spectrophotometer by A 260 /A 280 and A 260 /A 230 ratio, respectively, as indicative of nucleic acids purity (A 260 /A 280 $ 1.8 and A 260 /A 230 ¼ 2). 4 DNA quality was examined in 1% agarose gel electrophoresis.The DNA samples were stored at À20 C until use in PCR.
Model 2 and model 3. C. trachomatis genomic DNA.The autoclaved C. trachomatis positive DNA from clinical samples (C.trachomatis DNA) were received from FSBSI Saratov SRVI, the Department Zoo-and Zoo-anthroponotic diseases and were used as DNA templates without further purication.The target was chromosomal ompA gene of C. trachomatis genovars (Gen-Bank accession numbers: M58938, DQ064281, AF352789, AF063196, AY535104 (ref.45 and 46)).The experimental synthetic primers (Syntol, Russia) are listed in Table 2.The DNA samples were stored at À20 C until use in PCR.

PCR and gel electrophoresis
All PCR reactions were conducted in triplicate with a T-100 Thermal Cycler (Bio-Rad, USA).PCR reagents were mixed to a nal 30 mL reaction volume in 200 mL thin-walled tubes (Scientic Specialties, Inc., USA).Full PCR mixture in absence of nanoparticles was used as blank in all experiments.Pure double-distilled water (PCR-grade) as template was used for negative control in all experiments.
Each PCR tube with sample 1 DNA (A.brasilense Sp7) as a template contained: 3 mL 10Â PCR buffer, 0.6 mL 10 mM dNTPs, 3 mL 25 mM MgCl 2 , 0.5 mL 5 U mL À1 Taq DNA polymerase, 2 mL template containing 120-200 ng DNA, 50 pmol of each primer (0.3 mL nifD-up and 0.3 mL nifD-do primer), and 15 mL of appropriate NPs (varying in concentrations and particle type) and lled up to the 30 mL nal volume with MQ water.The PCR procedure was as follows: 5 min at 94 C for predenaturation, followed by 35 cycles of 30 s at 94 C, 30 s at 52 C, and 30 s at 72 C. Then the cycling was terminated aer 10 min nal elongation step at 72 C.
Each PCR tube with sample 2 DNA (C.trachomatis) as a template contained: 3 mL 10Â PCR buffer, 0.6 mL 10 mM dNTPs, 3 mL 25 mM MgCl 2 , 0.5 mL 5 U mL À1 Taq DNA polymerase, 2 mL template containing 120-200 ng DNA, a multiplexbroad-spectrum PCR primer mix containing 1 mL multiple forward primer (equimolar mix of 10 pmol momp-fw1, momp-fw2, momp-fw3) and 1.2 mL multiple reverse primer (equimolar mix of 10 pmol momp-rv1, momp-rv2, momp-rv3, momp-rv4), and 15 mL of appropriate NPs (varying in concentrations and particle type) and lled up to the 30 mL nal volume with MQ water.The PCR procedure was as follows: 2 min preheating step at 94 C followed by 38 cycles of amplication (30 s at 94 C, 30 s at 55 C, and 30 s at 72 C) and a nal 10 min elongation step at 72 C. Puried DNA of genovar E as a template was used as positive control in all experiments.
Each PCR tube with sample 3 DNA (C.trachomatis) as a template contained: 3 mL 10Â PCR buffer, 0.6 mL 10 mM dNTPs, 1.8 mL 25 mM MgCl 2 , 0.15 mL 5 U mL À1 Taq DNA polymerase, 1 mL template containing 120-200 ng DNA, 50 pmol of each primer (1 mL F1 and 1 mL B11), and 15 mL of AuNPs in serial dilutions and lled up to the 30 mL nal volume with MQ water.The PCR procedure was as follows: 2 min preheating step at 92 C followed by 35 cycles of amplication (45 s at 94 C, 45 s at 55 C, and 60 s at 72 C) and a nal 10 min elongation step at 72 C. Puried DNA of genovar E as a template was used as positive control in all experiments.
Aer the PCR amplication, all the samples were stored at 4 C before gel-electrophoresis.The PCR products were analyzed by horizontal electrophoresis system (SE 1, Helicon) with a voltage set at 125 V for 65 min (Elf-4 power supplier, DNAtechnology) with EtBr staining.Briey, 3.5 mL of each PCR product was mixed with 0.5 mL 6Â loading dye prior to loading in the wells of an 1.5% agarose gel with EtBr containing in 1Â TBE buffer (10 mM Tris, 10 mM boric acid, 1 mM EDTA, pH 8.0).The uorescence of the bands was visualized by UV transillumination (312 nm, Vilber Lourmart), the gel electrophoregrams were obtained with a Canon 350D digital camera equipped with an orange lter to minimize the UV-lamp background lighting.

PCR product quantication
The performance of PCR additives was quantied according to recent report 30 through calculations of two densitometric quantities, which are termed specicity and yield efficiency.The gel electrophoregram images were analyzed with an open ImageJ soware and the output data were represented as the means and standard deviations (n ¼ 3).A detail description of calculations is given in ESI le.† In brief, the ratio of the densitometric value of the specic target band to that of all bands amplied by PCR was dened as the specicity of amplication.By this denition, if there are no smears in PCR products, the maximal value of specicity equals 1.The ratio of the densitometric intensity of a specic target band to that for 500 bp of DNA marker was used as a measure of the yield efficiency.For brevity, we use the term "efficiency".If the obtained efficiency is $1, the PCR additive is considered as high efficient for PCR optimization.In all gures, molecular weight markers (M) represented the same SM0383 (ThermoFisher Scientic).
The optimal concentration of PCR additives was dened as a concentration that maximizes the brightest of specic target band.

Fig. 2
Fig. 2 Effect of citrate-stabilized AuNPs on PCR amplification of 710 bp region of NifD gene from A. brasilense Sp7 (model 1).Lanes M stand for DNA markers.(a) Illustration of the AuNP-assisted PCR specificity.The duplicate smeared lanes 1 and 2 were obtained without AuNPs, lane 3 corresponds to addition of 0.4 nM AuNPs.(b) Illustration of the AuNP-assisted PCR high-yield efficiency.No significant bands were observed without AuNPs (duplicate lanes 1 and 2), whereas the addition of 0.4 nM AuNPs results in clear detection of the target 710 bp band.(c) Lanes 1 and 2 correspond to PCR without AuNPs and with AuNPs added after amplification step.The normalized quantities of PCR specificity (S L ) and efficiency (E L ) for each lane are indicated on the bottom panels.

Fig. 3
Fig. 3 (a).Effect of AuNPs capped with neutral polymers mPEG-SH and PVP on amplification of a 160 bp target from C. trachomatis.Symbols SA stand for addition of the capping polymers alone at a concentration corresponding to their concentration in 0.4 nM AuNP solution, symbol M designates DNA markers.No specific enhancement was obtained for both polymers.(b) Effect of AuNP-PEG-SH particles on amplification of a 710 bp target from A. brasilense Sp7 (model 1).With an increase in nanoparticle concentration, a gradual inhibition of all bands is observed, up to visual disappearance of the target band at 10 nM concentration.The normalized quantities of PCR specificity (S L ) and efficiency (E L ) for each lane are indicated on the bottom panels.

Fig. 4
Fig. 4 Photos of test-tubes with citrate-(a) and PVP-stabilized (b) AuNPs before and after PCR amplification.

Fig. 5
Fig. 5 Effect of AuNPs-CTAB (a), AuNPs-PDDA (b) and AuNRs-662 (c) on PCR amplification of a 160 bp target from model 2.The symbol M stands for DNA markers.Thus, the effect of ligand-capped AuNPs is not significant as compared with that for citrate-stabilized AuNPs.The normalized quantities of PCR specificity (S L ) and efficiency (E L ) for each lane are indicated on the bottom panels.

Fig. 6
Fig. 6 (a) Effect of silica nanoparticles on amplification of a 160 bp target from PCR model 2. The symbol M stands for DNA markers, the symbol P designates a positive control with a purified DNA of genovar E as a template.No single distinct bands were observed at all nanoparticle concentrations from 0.4 to 200 nM (lanes 3-7).(b) Effect of silica nanoparticles on amplification of a 710 bp target from PCR model 1.For all nanoparticle concentrations from 0.4 to 20 nM (lanes 2-6), the PCR outcomes are similar to those obtained without nanoparticles (lane 1).Thus, for both PCR models, the effect of negatively charged silica nanoparticles is not significant as compared with that for citratestabilized AuNPs.The normalized quantities of PCR specificity (S L ) and efficiency (E L ) for each lane are indicated on the bottom panels.

Table 1
Parameters of nanoparticles used as additives in PCR systems

Table 2
Oligonucleotide primers used in PCR experiments