Graphene is a wonderful material applied in biosensor, due to its great conductivity. Thus, we oxidized the graphene of graphite by malic acid to decrease the layer density of graphite, and stratified the oxidized graphene from graphite by sonication. Then, we conjugated the antibody and aptamer on oxidized graphene to recognize the viral particles, as biosensor.
Graphene is an allotrope of carbon in the form of a two-dimensional carbon. Due to its sp^2 structure, the tremendous free electrons of graphene make the material have great conductivity. In this study, we develop an electrochemical sensor using carboxylated graphene modified Pt as an electrode. The carboxylated graphene modified Pt electrode has a great sensitivity to detect the change of electrical resistance when the electrode binds to target molecules.
Before using the carboxylated graphene modified Pt electrode for the detection of biomolecules, the tyrosine was conjugated on carboxylated graphene to test its characteristics. Due to the aromatic ring of tyrosine, the concentration of tyrosine is characterized by its absorbance at 274 nm, and the linking of tyrosine on the carboxylated graphene is determined by the amide bond formed with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysuccinimide(NHS). After we test the tyrosine modified carboxyl graphene, we use Western blotting to demonstrate the anti-VP1 binding to EV71. Then, the EV71 binds to the anti-VP1-modified carboxylated graphene tested by the electrochemistry.
Most commonly known influenza viruses are influenza virus A, influenza virus B, and influenza virus C. Among them, influenza virus A is the most virulent human pathogens that cause severe symptom. Influenza viruses are characterized by the different types of hemagglutinin (HA) and neuraminidase (NA) on the surface[4-6]. Hemagglutinin has 18 different antigens, and named H1 to H18 subtypes. Hemagglutinin is the primary protein binds to the receptor on the cell membrane, allowing the fusion of the viral envelope with the endosome membrane.
We then choose H1N1 influenza virus, one of the high epidemics, as our further detected pathogen. The aptamer is strand-specific for viruses. The influenza A/Puerto Rico/8/1934 H1N1 that we obtained could interact with the aptamer. HA proteins are exposed on the surface of the virus, and its molecular weight is estimated as 59.2 kDa. Overexpression of influenza A H1N1 (A/WSN/1933) HA protein was used for following aptamer interaction test as the negative control to demonstrate the specific binding pattern between aptamer and the targeted pathogen.
The aptamer was selected by HA protein from ssDNA (single strand DNA) library use SELEX process, the binding site of aptamers on the viruses is on HA protein. The interaction between this aptamer and virus, as well as HA protein, was determined by electrophoresis mobility shift assay (EMSA) in non-denaturing gel. In order to obtain enough quantity of ssDNA aptamers, multiple rounds of PCRs (polymerase chain reaction) were carried out. Hereafter, the binding of influenza A H1N1 (A/Puerto Rico/8/1934) and D26 aptamer was tested by electrochemistry.
1 g malic anhydride in 40 mL N-methyl-2-pyrrolidone (NMP) was added in three-necked flask and was heated to 90℃ for three hours. Meanwhile, 40 mg graphite was added in 40 mL NMP and sonicated for 20 minutes to suspend the graphite. Then, the graphite solution was added into the three-necked flask and reacted at 150℃ for 72 hours. 500 mL of 1 N HCl was added into the solution and stayed overnight. Next, the solution was centrifuged in 13,000 rpm for 5 minutes, removed the supernatant, and retrieved the precipitation. Finally, the sample was dialyzed to pH 7 and then, the solution containing of carboxylated graphene (GMA) nanoparticles was obtained.
The solid content of GMA solution (2.3 mg/mL) was obtained by evaporation. The quantification of carboxyl group was measured by titration and ionic conductivity. First, 0.1 N NaOH(aq) was prepared by 0.1 N potassium hydrogen phthalate (KHP). Then, 0.1 N NaOH(aq) was used to standardize HCl(aq) and the outcome was 0.79 N HCl(aq). Next, 2×10−3 N HCl titrated the GMA-NaOH(aq) solution, prepared by GMA solution added in 0.1 N NaOH(aq).
The structure of GMA was analyzed by Fourier transform infrared spectroscopy (FTIR), and the morphology of GMA nanoparticle was measured by scanning electron microscope (SEM) and transmission electron microscope (TEM).
The GMA, suspended in the deionized water, was titrated to obtain the content of carboxyl group (8.05×10−6 mole/mL). One mL of GMA was centrifuged and re-suspended in 1 mL of activation buffer ( 0.1 M 2-(N-morpholino)ethanesulfonic acid (MES), 0.5 M NaCl ). 15 mg of EDC (7.8×10−5 mole) and 35 mg of sulfo-NHS (1.61×10−4 mole) were added and mixed at room temperature for 15 minutes.
The activated GMA, separated by centrifuged from the solution, then washed with 1 mL of MES buffer to remove sulfo-NHS, which had a strong absorption peak at 260−280nm.Re-suspended in 0.5 mL of MES buffer solution, the GMA was mixed with tyrosine, which is dissolved in 0.5 mL of PBS, at room temperature for two hours. The molar ratio of 1/1, 0.5/1 and 0.1/1 tyrosine to carboxyl group were prepared. The supernatant was separated by centrifuge for following quantification, further the tyrosine-modified GMA was examined by FTIR and electrochemical analysis.
100 μL of the supernatant of tyrosine-modified GMA was loaded into each well of 96-well glass plate, and scanned by enzyme-linked immuno-sorbent assay (ELISA) reader to measure the tyrosine contents.
Structures of tyrosine-modified GMA with different ratio tyrosine content were analyzed by FTIR.
Added with magnetic iron oxide (Fe3O4) nanoparticles, the tyrosine-modified GMA was coated on Pt electrode by an applied external magnetic field to form the magnetic tyrosine-modified GMA/Pt electrode. The response current of the magnetic tyrosine-modified-GMA/Pt electrode in 2 mM (K3[Fe(CN)6])/(K4[Fe(CN)6]) solution was used as the baseline. 2 μL magnetic iron nanoparticle was added to the tyrosine-modified-GMA to form magnetic tyrosine-modified-GMA by sonicated for 1 minute. Then, 35 μL magnetic tyrosine-modified-GMA was added into the cell and the solution was dried by vaporization. After dehydration, the magnetic tyrosine-modified-GMA was absorbed on the Pt electrode by an applied external magnetic field to form the magnetic tyrosine-modified GMA/Pt electrode. The electrochemical performance of the modified Pt electrode was measured by current-voltage (CV) scan for 50 runs. Then, the differential pulse voltammetry (dPV) and resistance (R) of the magnetic tyrosine-modified-GMA/Pt electrode with different tyrosine concentrations were measured.
To determine the VP1 of EV71, we employed the anti-VP1 antibody and anti-mouse as secondary antibody to check the VP1 of EV71. First, we prepared the human neuroblastoma SF268 cell infected EV71 for 18 and 24 hours. Then, Cell lysates were loaded into 1D SDS-PAGE and the band of protein was transferred on PVDF membrane.
Antibodies, anti-VP1 (1:5000), were linked on GMA for the antigen capturing. One mL of GMA was centrifuged and re-suspended in 1 mL of activation buffer (0.1 M MES, 0.5 M NaCl). 15 mg of EDC (7.8×10−5 mole) and 35 mg of sulfo-NHS (1.61×10−4 mole) were added as powder and mixed at room temperature for 15 minutes. Then, 1 mL anti-VP1 was added into the activated GMA for 2 hours. After 2 hours, the anti-VP1-modified GMA was centrifuged at 13,000 rpm for 5 minutes. After removing the supernatant, the anti-VP1-modified GMA was coated on Pt electrode. Then, the EV71 antigen was dropped on the anti-VP1-modified GMA/Pt electrode and its electrochemical performance was measured by CV scan.
Preparing aptamer-target interaction test, appropriate amount of aptamers were needed. Using primers (5’-GGGAGCTCAGAATAAACGCTCAAGGCACGGCATGTGTGGTATGTGGTGC-3’&5’-GATCCGGGCCTCATGTCGAACGAGTACAGGCACCACATACCACACATGC-3’, 10 μM ) to amplify the D26 aptamer (5’-CAAAAAGUUAGGCCAGCAAAUUGCGAGCUGAUCCGGUGACUGGCUACAGGAGGCCUUGUCCACGGCCGUAUU-3’), which was specific to influenza A H1N1 (A/California/07 /2009). Also, 25 μL dNTP (10 mM), 12.5 μL high-fidelity DNA polymerase, and 125 μL MgCl2 10X buffer (15 mM) were necessary. Each 50 μL of reaction mixture was amplified using the following conditions: 94℃ for 30 s, 55℃ for 30 s, and 72℃ for 30 s for 35 cycles to amplify the dsDNA product.
Mixed phenol chloroform (25:24) 1:1 in volume to the PCR product. Centrifuged the mixture with 13,000 rpm for 5 mins, and transfered the supernatant to new Eppendorf tubes. Repeated the steps for the supernatant. 3 M NaOAc (10% volume of supernatant) and 100% EtOH (2.5 times volume of supernatant) were added to the supernatant from the second time centrifugation, and mixed them well. Then, Put the mixture in the −80℃ refrigerator overnight. Centrifuged the mixture with 13,000 rpm for 5 minutes, then discarded the supernatant. 1ml 70 % EtOH was added to the pellet, then centrifuged the mixture with 13,000 rpm for 5 minutes, discarded the supernatant. Evaporated the solution until powder form of DNA acquired, re-dissolved the PCR product with ddH2O, with the least volume needed.
Suspended in ddH2O, GMA using the titration method, the carboxyl content of GMA was 8.05×10−6 mole/mL. One mL of GMA was added in three-necked flask and evaporated at 90℃. Then, we added 1 mL of thionyl chloride (98% SOCl2) into the flask and reacted with GMA refluxing at 70℃ for 12 hours, after the excess of SOCl2 was removed by drying. Next, the D26 aptamer solution, which was made by PCR, was removed the water by evaporation, and re-suspended in distilled pyridine. Then, the D26 aptamer (py) was added into the flask to react with acyl-chloride GMA for 6 hours at room temperature. After the reaction was complete, the D26 aptamer-modified GMA was transferred into eppendorf tube and successively washed by 1 mL of acetone and 1 mL of ddH2O for 3 times. Then, the D26 aptamer-modified GMA was heated at 95℃ for 5 minutes to denature the aptamer and its complementary strand of D26 aptamer-modified GMA into ssDNA. We carefully maintained the temperature of solution at 95℃. Thus, the solution was spun down by rotor. The supernatant was repeatedly discarded by 200 mL and heated at 95℃ for 5 minutes, which we were in order to insure the DNA of D26 aptamer-modified GMA is ssDNA on GMA. Then, we had the confidence to name the reactant, D26 aptamer-modified GMA, though there were maybe still half of complementary strand of D26 aptamer on GMA.
We designed primers (5’-GCGCGGATCCATGAAGGCTTTTGTACTAG-3’ and 5’-GCGCAAGCTTGATGCATATTCTGCACTGC-3’) with BamHI and HindIII restriction enzyme digesting sites to amplify H1N1 A/WSN/33(H1N1) hemagglutinin (HA protein) sequence from pPOLI-Rib-WSN-HA plasmids (SSR lab, CGU, TW) to get our insert sequence, followed by cloning into the pMAL-p5X vector. The HA PCR product and destination vector, pMAL-p5x, were then digested with BamHI and HindIII, followed by ligation to obtain the Maltose Binding Protein (MBP)-HA protein. Maltose binding protein was used as a tag that we ccould further purify this MBP-HA protein by one-step affinity purification.
The pMAL-p5X-HA plasmid was then transformed into DH5α competent cells for further confirmation. The prepared plasmids were digested with BamHI and HindIII to confirm the inserted HA sequences.
Next, we retransformed this plasmid into BL21 competent cells to produce MBP-HA protein. After incubation the cells at 37℃ overnight, the single colony cells was picked for further incubation with IPTG at 37℃ for 3 hours.
The transformed cells were harvested and lysed to obtain the proteins, and further analyzed by SDS-PAGE. As shown in Figure 13a, there was no obvious band with the molecular size at around 100 kDa, instead, many visible bands were observed at the bottom of the gel. Since the SDS-PAGE result could only separate the proteins by their molecular weights, we need to perform another experiment in order to confirm the expressed protein more specific.
To determine the expressed HA protein in the E. Coli expression system, we employed the anti-MBP specific antibody and anti-mouse as secondary antibody to check whether the MBP-HA protein was expressed. With lots of bands on the membrane, there ought to be only one band corresponding to the size of MBP-HA fusion protein (Figure 13b). To solve this, we used lower temperature to conduct induction during mega preparation to reduce the protease activity.
We again cultivated large amount of bacteria followed by lysis the bacteria by using sonication and then centrifuged to recover the protein extracts. The MBP-HA fusion protein was then purified on the column (Column PD-10, Empty, GE Healthcare) with anti-MBP resin. The measurements of eluted fractions indicated that one of six fractions contained high amount of protein (Table 2).
All of the purified fractions were loaded onto the SDS-PAGE to separate the proteins. As shown in Figure 13c, the second fraction (Figure 13c, lane 2) had band match the size of the MBP-HA fusion protein, while the other fractions (Figure 13c, lanes 1, 3-5) won’t exhibit any clear band, and that content of pellet (Figure 13c, lane 6) has various sizes of bands.
The carboxylation of graphite causes the delamination of graphite layers. Thus, by sonication, the GMA is stratified from graphite as identified by SEM and TEM (Figure 1).Figure 1 shows that graphite (Figure 1a) was delaminated into the high surface area of the GMA sheet (Figure 1b). Closer examination of magnetic GMA (MGMA) revealed Fe3O4 attached to GMA sheets with sphere-like morphology and a size of 6.9 nm (Figure 1c). Water dispersion of graphite, GMA, and magnetic GMA were shown in (Figure 2). It indicates that GMA and magnetic GMA have very good dispersion properties. The magnetic GMA can be attracted by an applied magnetic field as shown in (Figure 3), indicating that Fe3O4 nanoparticles were attached to GMA sheets and consisting with the result of (Figure 1c) .
The mole of carboxyl group of GMA is measured by titration as shown in (Figure 4). The ion-mobilities of [H+] and [OH−] are larger than other ions. Therefore, the ionic conductivity drops to the lowest point when the solution is titrated to equivalence point. As a result, we can use the ionic conductivity to check the equivalence point as shown in (Figure 4).
The GMA is added into the excessive base (0.1 N NaOH; 3 mL) to neutralize the carboxyl group of GMA. The solution is then titrated by 2×10−3 N HCl(aq). The first equivalence point is the neutralization of excessive OH- in GMA solution, and the second equivalence point is the neutralization of the carboxyl group of GMA. Thus, the amount of carboxyl group of GMA can be calculated based on the equivalence points (Figure 4a). The two equivalence points are at 130 mL and 190 mL, respectively, which is consistent with the inflection points of ionic conductivity (Figure 4b). The obtained carboxyl group per mg of GMA is 8.31×10−6 mole.
Based on the structure of GMA, the double bond absorption peak of FTIR spectrum is located at 1719 cm−1 (Figure 5). Peaks of 1869 and 1777 cm−1 are the symmetry and asymmetry of carbonyl group, respectively, and the peak at 1418 cm−1 is the scissor vibration of CH2 bond. The FTIR result confirms that malic acid is bonded with the graphene to form GMA.
Table 1 shows the mole ratio of tyrosine to carboxyl group of tyrosine-modified GMA. The concentration ratios of 0.1/1 and 0.5/1 samples were too low to be distinguished from the control group by this detection method. However, the absorbance of sample with concentration ratio of 1/1 was detectable and its precise concentration was calculated by the standard calibration curve (Figure 6). The bonding efficiency of tyrosine to carboxyl group was obtained by calculating the tyrosine quantity difference in moles before and after the reaction, and made it divided by carboxyl group. So, a 35 mole% bonding efficiency means 35 mole% of the carboxyl group of GMA is conjugated with tyrosine.
The functional groups of tyrosine-modified GMA samples with different concentrations of tyrosine were measured by FTIR (Figure 5). The sample with tyrosine/GMA of 1/1 has the largest absorption peak at 3467 cm−1 than the other two samples (0.1/1 and 0.5/1). Thus, it indicates there is more tyrosine bonded with GMA.
After 50 runs of CV scans, the detected current of the sample with tyrosine/GMA of 1/1 becomes smaller (Figure 7). Meanwhile, the CV peaks of differential pulse voltammetries (dPVs) decrease with increasing tyrosine contents (Figure 8). This result indicates the bonding of tyrosine increases the electrical resistance of GMA.
The human neuroblastoma SF268 cells were infected by EV71 for 18 and 24 hours. Hsp90β is the internal control. According to the results (Figure 9), lane 1 has presented that VP1 has been bound by anti-VP1 and VP3 has been shown as well. Lane 2 has the same consequence as lane 1. Thus, we confirm VP1 is bound by anti-VP1.
The GMA modified Pt electrode linked with antibody, VP-1, is used to detect the antigen- EV71. The CV and dPV scans of GMA modified Pt electrode with and without capturing EV71 antigens is shown in (Figures 10 and 11), respectively. When the EV71 binds to the anti-VP1, both of current and voltage get smaller. Thus, the EV71 binds to the anti-VP1-modified GMA/Pt electrode leading to the increase of electrical resistance. The electrical resistance of GMA was increases when the biomolecular are bonded to it.
The same position of single band in gel electrophoresis (Figure 12) was confirmed that the obtained PCR product is ds-D26 aptamer, and it’s suitable for further conjugation on modified graphene.
To provide a strong evidence that the aptamer part of the aptamer-GMA complex accounts for the specificity of this rapid test platform, we examine whether the capability of aptamer to capture the target virus through its binding site on HA protein with non-denaturing gel electrophoresis (data not shown). There is no retarded band on any lane shown, which differs from our expectation. For D26 aptamer (which sieves from ssDNA library with strand influenza A/California/07/2009), the results match the expectation that it has no interaction with strand of A/Puerto Rico/8/1934. For Aptamer 2 which sieves from ssDNA library with strand A/California/07/2009’s HA protein, the EMSA fails to show the interaction between the aptamer and virus, as well as with HA protein, under several different binding conditions. The reannealing of dsDNA after denatured in 94℃ may occur rapidly before interaction with the added virus or HA protein, thus shows no interaction with them.
We in vitro constructed H1N1 HA protein in the pMAL-p5X vector, followed by expression this vector in the E. Coli by IPTG induction to produce HA recombinant protein. In the upper panel of (Figure 13a), there is no obvious band located in around 100 kDa while there are some shown at the bottom. Thus, we speculate that these bands are various degradation products derived from the intact HA protein.
Using anti-MBP as primary antibody and anti-mouse as secondary antibody to carry out western blot analysis of recombinant HA protein, the lane with IPTG induction has a strong specific band between 95 to 130 kDa (Figure 13b), corresponding to the expected size 104 kDa, revealing that MBP-HA fusion protein was truly expressed. Combined the results of SDS-PAGE and western-blot of the recombinant HA protein, we predicted that the MBP is a stable protein while HA is a less stable one, which is easily been degraded by cellular proteases.
Completion of protein purification and obtained the eluted fractions, not only testing the protein concentration, but also proving the protein size in the fractions is important. SO we further conduct the SDS-PAGE analysis of purified recombinant HA protein fractions (Figure 13c).
Modification of tyrosine on GMA has been proved feasibly by FTIR, SEM and TEM, which is further confirmed by electrochemical analysis. The GMA modified Pt electrode can be used for the detection of biomolecules by the electrochemical sensing. We have shown that the electrical resistance of anti-VP1 modified GMA/Pt electrode increases with the capturing of EV71 antigen.
The electrical resistance of aptamer is much higher than tyrosine due to molecular size, we can infer that GMA/Pt electrode linked with aptamer should show a distinct increasing on electrical resistance. We conjugated the aptamer to GMA, and tested the electrical resistance on virus-aptamer-GMA complex. Since the aptamers were amplified using PCR, the interaction of aptamers and viruses may interfere by the presence of complementary strand of aptamer, which result in undetectable interaction between aptamer-GMA complex and virus. Efforts will be in development of this platform model. Besides, We are going to use HA protein we inducted to compare the outcome of virus.
 P. Held. (2003). "Quantitation of Peptides and Amino Acids with a Synergy™HT using UV Fluorescence. " Biotek application note.
Fischer, M. J. (2010). "Amine coupling through EDC/NHS: a practical approach." Methods Mol Biol 627: 55-73.
Yang, H. W., et al. (2014). "Combined detection of cancer cells and a tumor biomarker using an immunomagnetic sensor for the improvement of prostate-cancer diagnosis." Adv Mater 26(22): 3662-3666.
 Enami, M. and K. Enami (1996). "Influenza virus hemagglutinin and neuraminidase glycoproteins stimulate the membrane association of the matrix protein." J Virol 70(10): 6653-6657.
Gamblin, S. J. and J. J. Skehel (2010). "Influenza hemagglutinin and neuraminidase membrane glycoproteins." J Biol Chem 285(37): 28403-28409.
Gottschalk, A. (1957). "Neuraminidase: the specific enzyme of influenza virus and Vibrio cholerae." Biochim Biophys Acta 23(3): 645-646.
 Keyi Liu, X. F. (2015). "A Label-free Aptasensor for Rapid Detection of H1N1 Virus based on Graphene Oxide and Polymerase-aided Signal Amplification." Journal of Nanomedicine & Nanotechnology 06(03).