S 1 Supplementary Materials : Identification of Chinese Herbs Using a Sequencing-Free Nanostructured Electrochemical DNA Biosensor

Due to the nearly identical phenotypes and chemical constituents, it is often very challenging to accurately differentiate diverse species of a Chinese herbal genus. Although technologies including DNA barcoding have been introduced to help address this problem, they are generally time-consuming and require expensive sequencing. Herein, we present a simple sequencing-free electrochemical biosensor, which enables easy differentiation between two closely related Fritillaria species. To improve its differentiation capability using trace amounts of DNA sample available from herbal extracts, a stepwise electrochemical deposition of reduced graphene oxide (RGO) and gold nanoparticles (AuNPs) was adopted to engineer a synergistic nanostructured sensing interface. By using such a nanofeatured electrochemical DNA (E-DNA) biosensor, two Chinese herbal species of Fritillaria (F. thunbergii and F. cirrhosa) were successfully discriminated at the DNA level, because a fragment of 16-mer sequence at the spacer region of the 5S-rRNA only exists in F. thunbergii. This E-DNA sensor was capable of identifying the target sequence in the range from 100 fM to 10 nM, and a detection limit as low as 11.7 fM (S/N = 3) was obtained. Importantly, this sensor was applied to detect the unique fragment of the PCR products amplified from F. thunbergii and F. cirrhosa, respectively. We anticipate that such a direct, sequencing-free sensing mode will ultimately pave the way towards a new generation of herb-identification strategies.


e, Chinese he
bal medicine has been an important component in the holistic therapy application of modern clinical medicine.Its effectiveness and safety rely pretty much on the proper choice of herbs, which is severely affected by the availability of substandard or non-medicinal herbs in the market.For instance, medicinal plant bulbs of the genus Fritillaria are used for their lung-moistening, cough-relieving and phlegm-dissolving effects.However, the efficacy varies even within the ten species of the Fritillaria genus recorded as original medicinal herbs in the Chinese Pharmacopoeia [1].Unfortunately, they look identical or have highly similar chemical compositions, thus it is very hard to differentiate them.Although the emerging DNA barcode technology is capable of identifying the two Fritillaria species F. thunbergii and F. cirrhosa [2,3], it requires tedious sequencing of a large amount of samples, More importantly, its complicated data analysis needs a reference genome, which is not readily available in public databases.Therefore, the development of a simple, reliable and straightforward genotyping approach for identification of F. thunbergii and F. cirrhosa herbs is highly desirable to address these issues, if possible, at the specific DNA sequence level.

So far, sensors for DNA detection have been developed based on a variety of platforms, including field effect transistor [4,5], surface plasmon resonance spectroscopy [6], fluorescence [7,8], and electrochemical methods [9,10].For example, electrochemical DNA (E-DNA) sensors have been widely used to perform genetic diagnostics [11], DNA hybridization [12], and determination of electrochemical active molecules in Chinese herbal medicines [13,14].We note that electrochemical sensing technology has special advantages over other biosensors, such as low cost, simple operation, high sensitivity and specificity [15], which inspired us to construct a simple and sensitive E-DNA sensor to quickly probe the special DNA sequences for identification of F. thunbergii and F. cirrhosa.Generally, to improve sensitivity of an electrochemical sensor, graphene and gold nanoparticles (AuNPs) are often employed to construct E-DNA sensors owing primarily to their excellent conductivity, large surface area, remarkable electrocatalysis capacity and exceptional mechanical properties [16][17][18][19].They can work cooperatively to enlarge the current response, resulting in increased sensitivity [20,21].However, such a superior nanostructured signal amplification platform has not been applied in the identification of subtypes of Chinese herbs.

In this work, we constructed an E-DNA ensor with bottom-up electrodeposited nanocomposites, aiming to generate a rapid, accurate and quantitative readout of the target DNA in a sequencing-free fashion for direct discrimination between two species of the Fritillaria genus, F. thunbergii and F. cirrhosa.The nanoassembled sensing interface is readily fabricated via controllable electrochemical deposition of graphene and AuNPs, which clearly enhances the electron-transfer efficiency of the electrochemical redox molecule methylene blue (MB) [22,23].A layer of thiol-ended DNA probe is immobilized on the AuNPs surface via Au-S chemistry [24], which is perfectly complementary with the a 16-mer unique sequence in the spacer region of the 5S-rRNA of F. thunbergii.The hybridization event between the probe and the target sequences/single-stranded DNA amplified by asymmetric polymerase chain reaction (PCR) from real samples, allows direct readout of the two Chinese herbal Fritillaria species.Compared to the existing genosensors, the currently-developed E-DNA sensor has the following advantages: (1) the mature E-DNA sensing interface is readily fabricated via rapid, controllable and environmentally friendly electrochemical deposition rather than traditional toxic chemical reduction or time-consuming drop-casting method;

(2) the low-cost, sequencing-free E-DNA se sor is capable of achieving a simple and straightforward readout of inherent discrimination between Chinese herbal Fritillaria species (F.thunbergii and F. cirrhosa); (3) the developed E-DNA sensor shows a potential of multiplexed species-identification by using an integrated electrode array biosensor, leading to a promising direction in the authentication of confusable Chinese herb species in the future.


Materials and Methods


Apparatus and Reag

5 1 -ACGCTAGTATGGTCGTGATTCCTAGG-3 1 .


Fabrication of the Graphene and Gold Nanopar
icles Modified Glassy Carbon Electrode

Prior to fabrication, the bare GCE was polished sequentially with 1.0, 0.3 and 0.05 µm alumina slurries, then sonicated in ultrapure water and ethanol, respectively.Graphene oxide (GO) was synthesized via the classical Hummer's method [25,26].Next the obtained GO was used to synthesize reduced graphene oxide (RGO) on the surface of the bare GCE through one-step electrochemical reduction method by using cyclic voltammetry (CV) [27].The detailed procedures have been already reported in our group's previous work [10,28].The RGO/GCE was immersed in 3 mM HAuCl 4 and 0.1 M KNO 3 solution as a working electrode and chronoamperometry was employed to obtain AuNPs by electrochemical deposition for 50 s at ´0.2 V according to the previous protocol [11,16] with a few modifications.Afterwards, the AuNPs/RGO/GCE was dried with nitrogen and stor

at a 4 ˝C refrigerator until use.


Pr
be Immobilization and Hybridization

Before dropping, the capture probe was treated with TCEP for 1 h to avoid disulfide bond formation [29].Subsequently, 5 µL of 5 µM probe DNA was dropped on the surface of the cleaned AuNPs/RGO/GCE and incubated at 4 ˝C overnight, through which probe DNA was covalently immobilized on the AuNPs/RGO/GCE surface via Au-S binding.After that, the DNA-modified electrode was washed by 1 ˆPBS and pure water in turn to remove the unbound probe.Hence, the probe DNA-modified electrode was named as ssDNA/AuNPs/RGO/GCE.Then the DNA-modified electrode was immersed into 2 mM MCH for 1 h to fill the unoccupied region.After rinsed thoroughly with deionized water and dried with nitrogen, the hybridization process was performed by incubating various concentrations of complementary DNA on the probe modified electrode surface and incubated for 40 min at 37 ˝C [23].After hybridization, the modified electrode was denoted as dsDNA/AuNPs/RGO/GCE.Then the dsDNA-modified electrode was rinsed with ultrapure water and 1 ˆPBS containing 0.1% sodium dodecyl sulfate (SDS) twice to re

ve the unhybridized target DNA.


Accumulation of MB
and Electrochemical Measurements

After hybridization, the above modified electrode was immersed into the stirring 20 µM MB dissolved in 25 mM NaCl solution for 5 min [10] to allow MB to accumulate onto the surface-hybridized dsDNA.Subsequently, the electrode was cleaned carefully with 0.1% SDS solution and water to prevent the physisorption of MB.After dried with N 2 flow, the electrode was immersed into 20 mM Tris-HCl buffer (pH 7.0) as a working electrode for the electrochemical detection by differential pulse voltammetry (DPV).The reduction peak current caused by MB accumulation is closely correlated with the amount of target DNA hybridized with the probe.Therefore, the Sensors 2015, 15, 29882-29892 signal change (∆I = I dsDNA ´IssDNA ) before (I ssDNA ) and after (I dsDNA ) hybridization could be used for quantification [16,23].The DPV measurement parameters were as follows: potential range: 0.1~´0.6V; amplitude: 0.05 V; pulse width: 0.05 s; samp

width: 0.02 s; pulse period: 0.2 s and quiet time: 2 s.


Preparation of DNA Samples an
Asymmetric Polymerase Chain Reaction (PCR) Amplification

Fresh leaves of F. thunbergii (which were respectively obtained from Pan'an Country (c1), Zhejiang Province; Linzhi Country (c2), Xizang Autonomous Region; Anguo Chinese herbal medicine market (c3), Hebei Province) and F. cirrhosa (Hehuachi Chinese herbal medicine market (b), Chengdu.Sichuan Province) were ground into powder in liquid nitrogen.DNA extraction was conducted using a plant genomic DNA kit following the manufacturer's instructions.The purity of extracted genomic DNA was evaluated by measuring OD 260 /OD 280 value using a spectrometer.Then the DNA s ock solution was kept frozen before next PCR process.

The asymmetric PCR was performed as described before [30] with a minor change.The reaction system was in a final volume of 50 µL mixture, which contained 1 µL of formamide, 5 µL of 10 ˆPCR buffer (200 mM Tris-HCl, pH 8.4, 500 mM KCl), 4 µL of 25 mM MgCl 2 , 2 µL of DNA extracted as template, 1 µL of 10 mM dNTPs, 2.5 µL of 0.1 µM forward primer, 2.5 µL of 10 µM reverse primer, 1.5 µL of 1 units/µL Taq DNA polymerase (Shanghai Sangon Biotech Co. Ltd., Shanghai, China), and 30.5 µL sterilized double-deionized water.Finally, the mixed liquids were blocked with 25 µL paraffin oil.PCR was performed on a thermal cycler (Bio-RAD T100 TM , Bio-Legend BioTech Co. Ltd., Shanghai, China) with the following settings: 4 min at 95 ˝C for initial denaturation, followed by 40 cycles of 95 ˝C for 30 s (denaturation), 60 ˝C for 30 s (annealing), 72 ˝C for 45 s (extension), Final extension was performed at 72 ˝C for 10 min, then the reaction was hold at 4 ˝C.To analyze, 7 µL of PCR products and 3 µL of SYBR Green I (Shanghai GENEray Biotech Co. Ltd., Shanghai, China) containing autoclaved water and 5 ˆloading buffer (1:99:150) were loaded in 1.5% agarose gel prepared with 1 ˆTAE bu

bserved under ultraviolet light.


Results and Discussion


Tailored Electrochemica
Sensor for Sequence-Specific Detection of Fritillaria DNA

A nanostructured E-DNA biosensor based on graphene-AuNPs functionalization for the sequence-specific detection of DNA sequences in herbal plant (Fritillaria species) is illustrated in Scheme 1. RGO is first assembled on the surface of GCE by one-step electrochemical reduction method to enhance the electrochemical response.Then AuNPs are electrodeposited on the RGO layer for further signal amplification [31,32].Prior to detection, capture probe DNA is tethered onto the AuNPs surface via a strong Au-S bond, followed by MCH blocking to fill the unoccupied region, which renders the probe a favorable orientation.To facilitate the next rapid and accurate sequence-specific readout of target DNA, and even PCR amplicons in complex matrices, several issues need to be addressed.Firstly, to reduce the assay cost and simplify the procedure, the redox MB is employed as a straightforward, label-free and cost-effective electrochemical indicator to generate the readable DPV signals, showing remarkable current changes before and after DNA hybridization.Secondly, the core parameters of this E-DNA sensor, including sensitivity, selectivity, reproducibility and stability, require an optimization by using the designed standard target DNA.Thirdly, to distinguish the F. thunbergii and F. cirrhosa, asymmetric PCR based on different amounts of double-primer is used to yield single-stranded DNA (ssDNA) as targets.The detailed procedure involves the following steps: (i) the total DNA of Fritillaria species plants is separately extracted for subsequent denaturation to single strands; (ii) after annealing, different amounts of double-primer are incorporated to produce double-stranded DNA (dsDNA); (iii) as the forward primer is exhausted, one strand of the amplified dsDNA can act as a template to amplify the desired target DNA with the assistance of reverse primer and manifold cycles.The yielded ssDNA target sequence could


Engineering Synergisitc Biosensing Interface

The morphologies of the RGO and the AuNPs/RGO nanocomposite were observed y SEM.As shown in Figure 1A, a typically wrinkled and crumpled sheet structure was obtained for the RGO, which is in accordance with the

eviously described report [33].After electrod
position, AuNPs were densely dispersed on the graphene surface with a diameter of approximately 30 nm (Figure 1B), confirming that AuNPs have been successfully fabricated.Furthermore, CV was employed to characterize the stepwise fabrication process of the electrochemical DNA biosensor, in which Fe(CN)6 3−/4− was chosen as a redox probe.As shown in Figure 1C, the peak current of bare GCE (curve a) was obviously smaller than that of RGO/GCE (curve b).After AuNPs were assembled on the surface of RGO/GCE, the peak current distinctly increased (curve c).All these enhancement consequences came from the excellent conductivity of graphene and AuNPs as reported [34,35].When the probe DNA was modified on AuNPs/RGO/GCE via Au-S bonding, the peak current sharply decreased (curve d), and further declined after blocked by MCH (curve e).When the probe DNA was hybridized with target DNA, the conductivity of the electrode was clearly reduced (curve f), indicating that negatively charged phosphate backbone of DNA has limited the electron transfer of Fe(CN)6 3−/4− because of electrostatic repulsion interaction as dsDNA is more negatively charged than ssDNA [36].


Optimization of Experimental Conditions

It is reported that experimental parameters such as probe concentration and hybridization time have great influence on the accuracy and sensitivity of an electrochemical sensor [37,38].Hence, the influence of these two factors

as investigated in this work.Figure 2A s
ows that the DPV peak current of MB increased with the probe concentrations ranging from 1 μM to 5 μΜ, then sharply decreased and tended to be saturated at 7 μΜ and 10 μΜ. Figure 2B shows the effect of hybridization time on the DPV response.With the increasing hybridization time from 10 min to 60 min, the reduction peak current Scheme 1. Schematic illustration of the fabrication process for the E-DNA biosensor and the asymmetric PCR procedure.


Engineering Synergisitc Biosensing Interface

The morphologies of the RGO and the AuNPs/RGO nanocomposite were observed by SEM.As shown in Figure 1A, a typically wrinkled and crumpled sheet structure was obtained for the RGO, which is in accordance with the previously d

cribed report [33].After electrodeposition, A
NPs were densely dispersed on the graphene surface with a diameter of approximately 30 nm (Figure 1B), confirming that AuNPs have been successfully fabricated.Furthermore, CV was employed to characterize the stepwise fabrication process of the electrochemical DNA biosensor, in which Fe(CN) 6 3´{4´w as chosen as a redox probe.As shown in Figure 1C, the peak current of bare GCE (curve a) was obviously smaller than that of RGO/GCE (curve b).After AuNPs were assembled on the surface of RGO/GCE, the peak current distinctly increased (curve c).All these enhancement consequences came from the excellent conductivity of graphene and AuNPs as reported [34,35].When the probe DNA was modified on AuNPs/RGO/GCE via Au-S bonding, the peak current sharply decreased (curve d), and further declined after blocked by MCH (curve e).When the probe DNA was hybridized with target DNA, the conductivity of the electrode was clearly reduced (curve f), indicating that negatively charged phosphate backbone of DNA has limited the electron transfer of Fe(CN) 6 3´{4´b ecause of electrostatic repulsion interaction as dsDNA is more negatively charged than ssDNA [36].


Optimization of Experimental Conditions

It is reported that experimental parameters such as probe concentration and hybridization time have great influence on the accuracy and sensitivity of an electrochemical sensor [37,38].Hence, the influence of these two factors was investi

ted in this work.Figure 2A shows that th
DPV peak current of MB increased with the probe concentrations ranging from 1 µM to 5 µM, then sharply decreased and tended to be saturated at 7 µM and 10 µM. Figure 2B shows the effect of hybridization time on the DPV response.With the increasing hybridization time from 10 min to 60 min, the reduction peak current suddenly enlarged and reached to a maximum value at 40 min, and then Sensors 2015, 15, 29882-29892 slightly decreased.As a result, 5 µM probe concentration and 40 min hybridization time were selected as the optimal conditions throughout the total experiment.

Sensors 2015, 15 7 suddenly enlarged and reached to a maximum value at 40 min, and then slightly decreased.As a result, 5 μΜ probe concentration and 40 min hybridization time were selected as the optimal conditions throughout the total experiment.Sensors 2015, 15 7 suddenly enlarged and reached to a maximum value at 40 min, and then slightly decreased.As a result, 5 μΜ probe concentration and 40 min hybridization time were selected as the optimal conditions throughout the total experiment.


Sensitivity

Under optimal experiment conditions, the sensitivity of the DNA sensor was explored by DPV.As seen in Figure 3A, the peak current of MB increased gradually with the concentrations of complementary target DNA concentrations rangi

from 100 fM
to 10 nM, and the linear correlation relationship between ∆I (I dsDNA ´IssDNA ) and the logarithm of target DNA concentration were displayed in Figure 3B.The linear regression equation was ∆I (µA) = 3.042 logC (M) + 42.90 (R = 0.9921).The detection limit was estimated to be 11.7 fM (S/N = 3).The comparison of this Sensors 2015, 15,[29882][29883][29884][29885][29886][29887][29888][29889][29890][29891][29892] proposed method with other works is illustrated in Table S1.These results state that the designed electrochemical DNA sensor is more sensitive than most other sensors.This is probably because of the synergetic signal amplification caused by modification of AuNPs and RGO composite on the electrode.Du et al. fabricated an electrode with individual RGO material for DNA detection based on peptide nucleic acid-DNA hybridization, and achieved a lower sensitivity than that of this work.Subsequently, Zhang and Jiang developed the other DNA biosensor based on AuNPs/RGO.The sensitivity was improved by about one order of magnitude, but was still lower than that of this work.This may be ascribed to the short electro-deposition time (20 s) of AuNPs, leading to a low density of AuNPs for the probe immobilization.From the abovediscussed results, this nanostructured E-DNA biosensor has shown good sensitivity that could recognize low concentrations of target DNA easily.

As seen in Figure 3A, the peak current of MB increased gradually with the concentrations of complementary target DNA concentrations ranging from 100 fM to 10 nM, and the linear correlation relationship between ΔI (IdsDNA − IssDNA) and the logarithm of target DNA concentration were displayed in Figure 3B.The linear regression equation was ΔI (μA) = 3.042 logC (M) + 42.90 (R = 0.9921).The detection limit was estimated to be 11.7 fM (S/N = 3).The comparison of this proposed method with other works is illustrated in Table S1.These results state that the designed electrochemical DNA sensor is more sensitive than most other sensors.This is probably because of the synergetic signal amplification caused by modification of AuNPs and RGO composite on the electrode.Du et al. fabricated an electrode with individual RGO material for DNA detection based on peptide nucleic acid-DNA hybridization, and achieved a lower sensitivity than that of this work.Subsequently, Zhang and Jiang developed the other DNA biosensor based on AuNPs/RGO.The sensitivity was improved by about one order of magnitude, but was still lower than that of this work.This may be ascribed to the short electro-deposition time (20 s) of AuNPs, leading to a low density of AuNPs for the probe immobilization.From the abovediscussed results, this nanostructured E-DNA biosensor has shown good sensitivity that could recognize low concentrations of target DNA easily.


Selectivity

As specificity is a very important property of a DNA biosensor, three different sequences including complementary DNA, single-base mismatch DNA, and non-c

plementary D
A, respectively, were employed to test this sensor under the same conditions.As shown in Figure 4, the ΔI (IdsDNA − IssDNA) value of complementary DNA was apparently higher than that of single-base mismatch DNA.When hybridized with non-complementary DNA, the response was found to be negligible.These results declare that the proposed electrochemical biosensor could distinguish single nucleotide from complementary target with excellent selectivity.


Selectivity

As specificity is a very important property of a DNA biosensor, three different sequences including complementary DNA, single-base mismatch DNA, and non-compl

entary DNA,
espectively, were employed to test this sensor under the same conditions.As shown in Figure 4, the ∆I (I dsDNA ´IssDNA ) value of complementary DNA was apparently higher than that of single-base mismatch DNA.When hybridized with non-complementary DNA, the response was found to be negligible.These results declare that the proposed electrochemical biosensor could distinguish single nucleotide from complementary target with excellent selectivity.


Reproducibility and Stability

In practical applications the reproducibility of a biosensor is an extremely vital standard to judge whether a biosensor is reliable or not.Her

n, five parallel AuNPs/RGO com
osite modified electrodes were modified with probe DNA and applied to detect 10 pM target DNA, respectively, and the relative


Reproducibility and Stability

In practical applications the reproducibility of a biosensor is an extremely vital standard to judge whether a biosensor is reliable or not.Herei

five parallel AuNPs/RGO compo
ite modified electrodes were modified with probe DNA and applied to detect 10 pM target DNA, respectively, and the relative standard deviation (RSD) of the peak current was found to be 5.85%, suggesting that the biosensor has excellent reproducibility.The stability of the fabricated electrodes was also investigated by applying the DNA-modified electrodes with 10 pM complementary sequence and measuring the response of the sensor every day in 6 days.As shown in Figure S1, the DPV response changed negligibly within 6 days, denoting an excellent stability of this sensor.


Direct Identification of F. thunbergii and F. cirrhosa Using E-DNA Sensor

To demonstrate whether the proposed electrochemical DNA sensor could be applied to differentiate real tr

itional Chinese medicinal plant samples selectively and accurately, asymme
ric PCR products amplified from F. thunbergii and F. cirrhosa were used as detection targets.The agarose gel electrophoresis analysis diagram of the obtained PCR products was shown in Figure 5A.The PCR amplification products from F. thunbergii (band c1, c2 and c3) showed light bands between 500-600 bp, which correlates with the result (564 bp) reported in the literature [3].However, the PCR products of F. cirrhosa (band b) appeared a band apparently at 500 bp, which is obviously different from that of F. thunbergii.This phenomenon might result from the deficiency of specific target DNA in F. cirrhosa genes to the immobilized probe.All these results are in good agreement with the previous report [30].

Before electrochemical detection, PCR products were heated in hot water (95 ˝C) for 5 min, and immediately chilled in ice for another 5 min to obtain denatured ssDNA.After that, the den tured PCR products were applied to the probe-modified electrodes for the electrochemical detection.Figure 5B reveals the DPV responses of the DNA biosensor to the different PCR amplified products.Curve a (I = 13.25 µA) was the blank control, in which the probe-immobilized electrochemical biosensor was incubated with 20 mM Tris-HCl buffer (pH 7.0) without any hybridization.After hybridization with the different PCR products and incubated in MB, the peak currents of the PCR products obtained from F. thunbergii (curve c1, c2 and c3) and F. cirrhosa (curve b) were 22.42 µA, 21.65 µA, 22.03 µA and 14.30 µA, respectively.The DPV responses of 3 kinds of F. thunbergii obtained from different locations were comparably similar, while the DPV responses of F. thunbergii were distinctly higher than that of F. cirrhosa.As the probe DNA assembled on the proposed sensor is specific to F. thunbergii, only the PCR products from this species could hybridize with the probe and cause remarkable signals, while the sequences of the PCR products from F. cirrhosa are not complementary to the immobilized probe, leading to smaller signals.These results indicate that the electrochemical DNA sensor could specifically distinguish two close species of Fritillaria by detecting the PCR products amplified from F. thunbergii and F. cirrhosa, which is in good agreement with the results obtained from the gel electrophoresis.Therefore, this kind of electrochemical DNA sensor has shown a great potential as the analysis technique for plant identification.


Conclusions

A nanosensing interface-based E-DNA biosensor has been developed for identification of Fritillaria species at DNA level in a sequencing-free manner.Both graphene and AuNPs were a sembled on a GCE electrode by an electrochemical reduction method, showing outstanding amplification properties.MB was utilized as a label-free indicator for the hybridization events.The proposed biosensor showed superior sensitivity (a detection limit of 11.7 fM), selectivity, reproducibility and stability, and was successfully applied to detect the asymmetric PCR products amplified from real F. thunbergii and F. cirrhosa samples, respectively.The results confirm that this sensor could distinguish these two close Fritillaria species accurately and reliably.Further, the obtained favorable linear range and low detection limit of target sequence can provide quantitative information, showing a great potential to carry out a dir

t, PCR-free
dentification of closely resembled subspecies in untreated herbal extracts.It is anticipated that


Conclusions

A nanosensing interface-based E-DNA biosensor has been developed for identification of Fritillaria species at DNA level in a sequencing-free manner.Both graphene and AuNPs were assembled on a GCE electrode by an electrochemical reduction method, showing outstanding amplification properties.MB was utilized as a label-free indicator for the hybridization events.The proposed biosensor showed superior sensitivity (a detection limit of 11.7 fM), selectivity, reproducibility and stability, and was successfully applied to detect the asymmetric PCR products amplified from real F. thunbergii and F. cirrhosa samples, respectively.The results confirm that this sensor could distinguish these two close Fritillaria species accurately and reliably.Further, the obtained favorable linear range and low detection limit of target sequence can provide quantitative information, showing a great potential to carry out a direct, PCR-free identification of closely resembled subspecies in untreated herbal extracts.It is anticipated that such a differentiating strategy would be extended to achieve multiplexed Chinese herbal species-identification simultaneously by using a nanostructured electrode array biosensor as a complementation.

6 Scheme 1 .
61
Scheme 1. Schematic illustration of the fabrication process for the E-DNA biosensor and the asymmetric PCR procedure.


Figure 1 .
1
Figure 1.SE