The Art_KSI was successfully expressed in Escherichia coli and further purified by anion exchange column. The quality of the purified protein was confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis with homogeneity >95%
(data not shown). The precise molecular weight determined by LC/MS analysis showed the m/z= 13402 (M+H+), which is consistent with the molecular weight calculated from the amino acid composition of Art_KSI. Art_KSI possesses substantial activity with the value of kcat/Km = 1.12x107 (M-1s-1) when
△5-Androstene-3, 17-dione is used as substrate for assay. The dissociation constant (Kd) of steroid with Art_KSI was determined as described in literature63, 64 by the quenching yield of tyrosine fluorescence at λem= 307 nm (λex= 278 nm). The Kd value of 19-norandrostendinone was determined to be 10.9 μM. These results confirmed the protein folding of Art_KSI is maintained.
In order to convert the action of steroid binding into electric signal, a negatively charged ligand (the reporter) will need to be covalently labelled at the appropriate position of Art_KSI. The precursor of the reporter, named as mA51-mA51 (shown in Figure 3-3), was synthesized by coupling two molecules of 5-(2-aminoethylamino)-1-naphthalenesulfonic acid (1,5-EDANS) with one 4,4′-Dithiodibutyric acid. The Cys-86 residue, locating at the rim of the steroid-binding site, was designed to react with the mA51-mA51 through the thiol substitution to from a new disulfide bond between protein and the reporter (mA51).
The modified protein is designated as Art_KSI/mA51 and the success of chemical conjugation of Art_KSI was confirmed by LC/MS analysis as shown in Figure 3-3.
Figure 3-3. The mass spectroscopic analysis of Art_KSI conjugated with the reporter.
The measured molecular weight of Art_KSI/mA51 is 34770 ± 2 amu, which is consistent with the calculated value of 13768 amu (13402 amu for Art_KSI and 366 amu for mA51).
The Art_KSI/mA51 was further immobilized on the SiNW through Lys-125 residue (the C-terminal residue) or the amino group of N-terminus. Based on the inspection of protein structure, we predict that either way of immobilization will not cause steric hindrance for steroid binding. In principle, the reporter molecule can be expelled from the binding site and exposes to the surface of SiNW when steroid is present. To ensure the feasibility of this system, the binding affinity of the reporter should be taken into account. If the reporter strongly binds to protein, it will be hardly replaced by steroid. In contrast, a weak-binding reporter can not promise the application since the predominant portion of the reporter is outside the steroid-binding site. We chose 1,5-EDANS as the candidate of reporter for its specific, but moderate, binding affinity towards Art_KSI (Kd = 0.35 mM). The mA51 moiety on the Art_KSI/mA51 is hence presumably able to fit into the steroid-binding site. Indeed, the time-resolved fluorescence technique has been employed to study the anisotropy decay of Art_KSI/mA51. The anisotropy decay represents the degrees of freedom of the mA51 moiety in accordance to its location. The results revealed that that 75% of mA51 existed inside the steroid-binding cavity while 25% of it remained outside the cavity at equilibrium.
The employed process of protein immobilization was evaluated by examining the observed density of gold nanoparticles (AuNPs) on Si sample with 30 nm SiO2 thin film with different stages of surface modification, such as the treatment of APTES and further with BS3 and protein as shown in Figure 3-4.
Figure 3-4. The various stages of modification on SiO2 substrate and the corresponding SEM images after the treatment of AuNPs. The SEM images revealed the existence of AuNPs on the surface of substrate with the modification (a) by APTES, (b) by BS3, and (c) by KSI_126C.
Figure 3-4 also exhibited the SEM images of the AuNPs existed on different derivatized surfaces. Since AuNPs was synthesized by citrated reduction method,65 the negatively charged AuNPs can be expected to bind to the amine-derivatized surface effectively via the electrostatic interaction.29 After coupling with BS3, the amine-derivatized SiO2 surface was presumably converted into a sulfonated surface.
Therefore, the deposition of AuNPs became rare due to the repulsion force, as shown in Figure 3-4. KSI_126C, a mutant with an extra cysteine residue added at the C-terminus, was further immobilized on this substrate through the reaction of lysine
residue with the BS3-activated surface. The resulting substrate was treated with AuNPs. If the immobilization of KSI_126C is effective, the deposition of AuNPs will be largely increased presumably owing to the formation of Au-S bond and/or the electrostatic interaction between the two substances. The high density of AuNPs layout (Figure 3-4C) clearly demonstrated the efficiency of KSI immobilization under the condition employed in the case of SiNW.
On the other hand, the channel surfaces were also modified by using an enzyme which was through a self-assembly process to recognize biomolecule as presented in Figure 3-5.
Figure 3-5. The multiple surface modified processes for enzyme immobilization on the p-type UTB-FETs surface.
The hydrophilic surfaces of channel were initially terminated with hydroxyl groups by UV-ozone plasma treatment, and the excess organic reagents were removed by UV-ozone plasma cleaning. The organic monolayer with amino groups
(AEAPTMS) was covalently linked to the surface via Si-O-Si bonds. Upon electrostatic interaction, 15 nm gold nanoparticles (AuNPs) anchored easily to the silane-coupler treated substrate with amino groups (AEAPTMS) via the carboxyl groups on the AuNPs surface. The carboxyl groups surrounded AuNPs were utilized to conjugated with the cross-linker dicyclohexyl carbodiimide (DCC) and then to produce the key intermediate: the O-acylisourea, which can be viewed as a carboxylic ester with an activated leaving group. After DCC activation, Art_KSI was covalently linked to the surface of AuNPs through the formation of peptide bonds by N-terminal amino acid of protein sequences or Lysine. Conductance measurements were performed on 14 UTB-FETs devices and shown reproducible. Figure 3-6A plots I–V curves following various chemical modification at a fixed bottom gate bias of Vg = -10 V. The inset displays the optical image of the device structure of a UTB-FETs.
The channel length, channel width and channel thickness of UTB-FETs are 10 µm, 5 µm and 10~20 nm respectively.
-1.0 -0.5 0.0 0.5 1.0
Figure 3-6A Ids vs. Vds measurements on the same p-type UTB-FETs after deferent chemical modifications; insert: the optical image of p-type UTB-FETs devices.
Dimensions, width= 5 m , length= 10 m, thickness= 10 nm.
The conductance of UTB-FETs devices responded to the induction of charge on the channel surface. AEAPTMS modified the silicon dioxide layer by silanol groups to generate an amino-derivatized surface which changed the surface potential of the UTB-FETs from negative to positive and affected the conductance of UTB-FETs device from 3.09 μS to 1.74 μS, as plotted in Figure 3-6A. (circles).
Negatively charged AuNPs and protonated amino-group of AEAPTMS cross-linked to each other on the channel surface by electrostatic interaction. The binding of negatively charged AuNPs onto the amino-derivatized surface acts as a negative gate voltage, which increases the conductance by the accumulation of holes in the channel. Most importantly, these results showed that the attachment of about
10μm
6.5*104 AuNPs substantially enhances the conductance of a UTB-FETs (the increase is revealed by the difference between the circular and trianglular data points in Figure 3-6A. The increases in the conductance of the UTB-FETs upon AuNPs attachment is simply an electrostatic gating effect of AuNPs. After AuNPs bound to the amino-terminal surface, the conductance increased from 1.74 μS to 2.72 μS.
Figure 3-6A plots the electronic characteristic of the DCC modified device, DCC was attached to the UTB-FETs surface via covalent bonds with the carboxyl groups upon the AuNPs surface, depleting the electronic holes in the UTB-FETs channel.
Since the carboxyl groups were substituted by DCC conjugation, the conductance of UTB-FETs decreased from 2.72 μS to 1.90 μS. The chemical modification of the channel surface dramatically affected the conductance of UTB-FETs devices. To characterize the reproducibility of these observations, we have made similar measurements on over 14 devices. The histogram of conductance measurements was shown in Figure 3-6B. All devices behaved with same tendency in the changes of conductance after modifications.
1.0x10-6 2.0x10-6 3.0x10-6 4.0x10-6
Figure 3-6B. Histogram of conductance of the multiple surface modified p-type UTB-FETs.
The exposure of an Art_KSI_mA51 modified UTB-FETs device to a 19-norandrostendione solution (10 μM) increased the conductance by 265 nS (~8 %), as plotted in Figure 3-7B. The increase in the exposure of mA51 in buffer produced an effective negative gate on p-type UTB-FETs due to competitive inhibition between mA51 and 19-norandrostendione. This special sensing model can be adopted to analyze the non-charged labeled analyte with steroids in our UTB-FETs devices. On the other hand, control experiments don’t show any considerable change (or difference) in conductance with or without 19-NA, which shows in Figure 3-7A. The increasing conductance confirmed that Art_KSI protein was bound strongly via the formation of peptide bonds on the silicon surface. However, the conductance of Art_KSI-modified UTB-FETs cannot be affected by 19-NA involving in this control experiment. All measurements of conductance were performed on 14 UTB-FETs devices. Figure 3-7C shows changes of conductance (1-G/G ) in the p-type
UTB-FETs using 10 μM target analytes, where G0 is the conductance of UTB-FETs before supply of target analytes and G is the conductance of UTB-FETs after supply of target analytes. When target analytes were introduced on the channel surface of UTB-FETs, the UTB-FETs modified by Art_KSI_mA51 shows the obviously change in conductance than that of Art_KSI-modified UTB-FET. Table I presents conductance changes after the chemical modifications and biological detection.
Figure 3-7A. Ids-Vds curve of the p-type UTB-FETs. The electronic response of Art_KSI modified on p-type UTB-FETs.
0.0 0.2 0.4 0.6 0.8 1.0
0.0 1.0x10-6 2.0x10-6 3.0x10-6 4.0x10-6
I ds (A)
Vds (V) DCC
Art_KSI 19-NA
Figure 3-7B. Ids-Vds curve of the p-type UTB-FETs. The electronic response of Art_KSI_mA51 modified on p-type UTB-FETs.
Figure 3-7C. Ids-Vds curve of the p-type UTB-FET. (1-G/G0) electronic response in the p-type UTB-FET in detection of 10 μM target analytes; G is the conductance of modified Art_KSI_mA51 or Art_KSI, G0 is conductance when target analytes are introduced.
-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15
Conductance Change (1-G/G 0) Art_KSI
Art_KSI_mA51
Table 3-1. Comparison of changes in conductance due to surface modifications of the p-type UTB-FETs.
GAEAPTMS-GSiO2 GAuNPS-GAEAPTMS GDCC-GAuNPS G Art_KSI_mA51-GDCC G19-NT –GArt_KSI_mA51
ΔG -1100 nS 827 nS -337 nS 767 nS 265 nS
In case, the conductance of Art_KSI/mA51-labeled SiNW-FETs will be modulated if the charge-state of the surface is changed. Since no existing reference is available to justify the influence of steroid in the present system, we ensured the observed signals were derived from the binding of steroid in Art_KSI/mA51 by comparing the responses of SiNW-FETs modified by BS3 and further by Art_KSI.
The effects of 19-NA on those devices were shown in Figure 3-8. The electrical response of SiNW-FETs was measured in 0.1mM Tris buffer.
Figure 3-8. The response of conductance of various SiNW-FETs with the presence of 19-NA. The arrows indicated the time point of the addition of 19-NA to SiNW-FETs
labeled (a) by Art_KSI and (b) by Art_KSI/mA51 with the concentration: (1) 0.0013 fM, (2) 0.013 fM, (3) 0.13 fM, (4) 1.3 fM, (5) 13 fM, (6) 130 fM, and (7) 1300 fM.
Typical time–dependent data were obtained from the output of Art_KSI/mA51-labeled SiNW-FETs after introducing various concentration of 19-NA. The conductance of BS3- and Art_KSI-labeled SiNW-FETs remained unchanged with the addition of 19-NA up to 1300 fM, indicating the background perturbation of 19-NA is insignificant (Figure 3-8A, the data of BS3-modified devices not shown). Upon addition of different concentration of 19-NA, the conductance of Art_KSI/mA51-labeled SiNW-FETs rapidly increased to a constant value (Figure 3-8B). A higher concentration of 19-NA resulted in a stronger conductance, suggesting that 19-NA competed with mA51 for the steroid-binding site in Art_KSI/mA51. The negatively charged mA51, in which its charge may be compensated by protein when it is bound, is expelled to expose in the solution near SiNW. The correlation between the increased conductance and the applied 19-NA was shown in Figure 3-9. A good linear correlation can be found when the concentration of 19-NA is greater than 0.013 fM. In summary, the conductance changes observed from Art_KSI/mA51-labeled SiNW-FETs correspond to the specific binding of 19-NA to Art_KSI/mA51. The sensitivity of 19-NA detection can reach the level of sub-femto molar. This is the first successful demonstration that a SiNW-FETs can be used for sensing a non-charged analyte by integrating the technique of protein engineering.
Figure 3-9. The linear correlation of the conductance change in Art_KSI/mA51-labeled SiNW-FETs with respect to the applied concentration of 19-NA. Note that the abscissa is a logarithmic value of 19-NA concentration in femto molar. Each data point was the average of 130 time of measurements. G0 = condunctance at 0.0013 fM 19-NA, G = conductance at different concentration 19-NA, G△ = G-G0.
The above-mentioned study of non-charged steroid detection on SiNW-FETs, has been carried out by complex bio-and chem-design. To improve the complex designed for non-charged analytes detection on SiNW-FETs, a novel immunobioassay system has been utilized to our biosensing platform.
As a powerful but inexpensive alternative to conventional approaches, noncompetitive immunoassays, such as sandwich ELISA, have recently been adopted
extensively to determine the concentration of antigen. Such as open–sandwich enzyme-linked immunoassay (OS-ELISA)67-74 which is a novel immunoassay approach that employs the antigen dependency of the interaction between the separated heavy chain (VH) and light chain (VL) of an antibody variable region.
Without antigen, the two fragments are prone to dissociate; while in the presence of antigen, they associate owing to increased interaction by the bridging antigen. In a number of OS-ELISA system, the fragments of display antibody have been developed on phage surface, utilized to select high affinity binders from the libraries of various sources. OS-ELISA has been successfully demonstrated by the study of small molecule and peptide detection.
In this study, non-charged Bisphenol A has been used to demonstrate the possibility of the application-based OS-ELISA on SiNW-FETs. Silicon oxide surface has been modified by the straightforward manner in experimental section. VL firstly conjugated with BS3-activated silicon oxide surface and then introduced the analytes (0.334 mg/mL BPA mixed with 1.126 μg/mL VH) into our system. Figure 3-10A shown that the obvious conductance change (△condutance = 1.5 nS) which was caused by bridging antigen. However, without antigen (BPA), VL difficultly associated with VH to perturb the charged density on the surface of silicon nanowire. The result was shown in Figure 3-10B. On the other hand, as lock of antibody (VH), the intrinsic limitation of FET restricts noncharged BPA detection on SiNW-FETs, which was shown in Figure 3-10C.
(A)
(B)
(C)
Figure 3-10. The application of open – sandwich bioassay for noncharged analyte BPA detection by using SiNW-FETs platform. (A) The response of conductance of various SiNW-FETs with the presence of BPA and VH. (B) The response of conductance of various SiNW-FETs with the presence of VH.only. (C) The response of conductance of various SiNW-FETs with the presence of BPA only.
3-4. Conclusion
By integration of protein engineering and nanotechnology, the novel uncharged-steroid detection has been successfully applied on Field Effect Transistor.
The detection limitation is approach to sub-femto molar. On the hand, the open sandwich immunoassay system has firstly demonstrated on SiNW-FETs by using of bisphenol A deteciotn..