Infra-red Spectroscopy and Crystal Structure. Five significant peaks of ferrocenium compound at 2050.63, 3099.75, 1413, 1007 and 852 cm-1 can be observed. The presence of a sharp stretching peak at frequency 2050.63 cm-1 is owing to the N=C=S. Meanwhile the peak at 3099.75 cm-1 is a characteristic for CH stretching of the cyclopentadienyl rings for ferrocenium group. The peak at 1413 cm-1 corresponds to the antisymmetrical C-C stretching while the peak at 852.17 cm-1 represents a CH out of plane mode for ferrocenium. The presence of all these significant peaks indicates that this ferrocenium compound has successfully formed17. The compound crystallized in trigonal crystal system with space group P-3, a = b = 18.1606(12) Å, c = 8.9808(6) Å, α = β = 90o, γ = 120o, Z = 1 and V = 2565.1(4). CCDC number for this compound is 1914296. From X-ray crystallography investigation, the titled compound consisted of six ferrocenium moiety, three hexa(isothiocyanato)iron(III) complexes and three hydronium group (Figure 2). The positive charge was stabilised by the presence of three hexa(isothiocyanato)iron(III) complexes and three hydronium group. All hexa(thiocyanato)iron(III) complexes in the compound adopt octahedral structure.
DNA binding of Hexaferrocenium Complex. The binding mode and the binding affinity between Hexaferrocenium complex and CT-DNA were performed using UV-Vis spectroscopic titrations. The UV-Vis absorption spectra of the title compound show two intense bands at 250 nm and 280 nm, respectively (Figure 3). The absorption band at 250 nm is assigned to the π-π* transition, which attributes to the localisation of molecular orbitals on the C=C group of hexaferrocenium cation. The absorption band at 280 nm is referring to the n-π* transition of C=N of hexa(isothiocyanato)iron(III) anion. Absorption measurements were carried out by using constant HexaFc complex concentration (3 × 10-5 M) while increasing the concentration of DNA until no changes could be seen on the UV-Vis spectrum. The spectrum of the HexaFc complex was recorded after each addition of the DNA.
Changes that could be observed in the spectrum were either hyperchromism (increased in absorption) or hypochromism (decreased in absorption). Hyperchromism occurs due to the secondary damage of the double helix structure of the DNA, causing the DNA to be single-stranded18,19. The occurrence of hypochromism is caused by the contraction of DNA in the helical axis. It is also affected by the transformation in DNA conformation20. Besides, the change in the wavelength of either the redshift (the absorption to the longer wavelengths) or the blue shift (the absorption change to the shorter wavelength) can also be observed.
The percentage of hypochromicity was calculated following Equation (1):
Based on the UV-Vis DNA titrations, this compound exhibited a change in hypochromism and redshift at 250 nm when DNA concentration increased. The hypochromicity shown was 40 %, and the redshift was 2.5 nm and in agreement with the established intercalators21,22. According to Wu et al.23 and Shahabadi et al.24, hypochromism event showed the binding strength of compounds towards DNA through intercalation mode. DNA binding through intercalation causes redshift. This involves strong overlapping interactions between the aromatic ligand chromophores of the metal complex with DNA bases25,26. Meanwhile, the effects of hyperchromism or hypochromism can be observed when the occurrence of electrostatic interaction or groove binding followed by a blue shift (hypsochrome effect) or a minor change in the wavelength absorption of the UV-Vis spectrum27. The intrinsic binding constant Kb of hexaferrocenium salt–DNA was determined according to Equation (2):
Where the apparent molar extinction coefficients, Δεap = |εA-εF|, εA = Aobserved/[Complex], Δε = |εB-εF|. εF and εB represent molar extinction coefficients for the free hexaferrocenium salt and the DNA bound hexaferrocenium complex, respectively. To interpret the binding affinity of the compound to DNA, the intrinsic binding constant Kb was discovered by recognising the changes of maximum absorption bands centred at around 250 nm region. From the plotted graph of [DNA]/ |εA-εF| versus [DNA], the y-intercept is equal to 1/ (|εB-εF| × Kb) whereas the slope is equal to 1/|εB-εF|. Kb values can be determined by dividing the slope value by the y-intercept. The binding constant (Kb) for HexaFc complex towards CT-DNA was 3.1 × 104 M-1. Also, this compound exhibited the same approximation value of binding constant with other reported ferrocene derivatives towards CT-DNA28,29.
Characterisation of DNA Biosensor. We observed the characteristic modification when SiNSs immobilised porcine DNA probe and interacted with complementary DNA molecules. The XRD patterns of layer-by-layer DNA biosensor are shown in figure 4. The diffractogram of SPE exhibited a band at 25.6°. The intensity band and the percentage of crystallinity decreased after AuNPs and SiNSs were deposited on SPE. But the intensity band and the percentage of crystallinity were slightly increased after treatment with glutaraldehyde (GA)30. The intensity band and the percentage of crystallinity continued to decrease with the addition of the porcine DNA probe and the complementary DNA. After the addition of HexaFc complex, the intensity and the percentage of crystallinity began to increase due to the presence of iron metal from the complex.
FESEM examined the morphology of the DNA biosensor. FESEM can provide a clear view of the structure of the DNA biosensor. Mapping with EDX made it possible to see the specifics of the DNA biosensor behaviour. Particularly after immobilisation of the aminated probes and hybridisation with complementary DNA. Figure 5 shows the morphology of the stepwise fabrication of the DNA biosensor. From the FESEM micrograph (Figure 5a), SiNSs were spherical with diameter between 20 to 200 nm while AuNPs with diameter less than 100 nm. Both SiNSs and AuNPs were dispersed with high homogeneity. More DNA probes could be immobilised on silica nanospheres than flat surface silica, hence an increase the biosensor sensitivity31. The DNA hybridisation in the presence of complementary DNA involving the intercalation with hexaferrocenium complex can be confirmed from FESEM studies (Figures 5). Thus, a noticeable change on the biosensor surface before and after interaction with a complementary porcine DNA was observed.
The dispersive energy X-ray (EDX) elementary mapping analysis was employed to detect the distribution of the different elements present in the biosensor32. This biosensor consists of carbon, oxygen, silicon and aurum elements. The elemental mappings of Au (yellow), C (red), Si (blue) and O (green) were observed and have shown to be well distributed in the biosensor (Figure 5a and 5b). As can be seen, nitrogen element was appeared after DNA immobilisation (Figure 5c) and it is derived from the aminated DNA probe (N-H). The presence of the nitrogen atom on the biosensor caused a higher attenuation of X-ray and gave a better contrast on the image. The N element increases in figure 5d due to the formation of double-stranded DNA33. The process of DNA hybridisation is successful and was proved by the addition of a nitrogen element in figure 5d. Meanwhile, Figure 5e shows the presence of Fe element after HexaFc complex intercalation is performed. The combination of FESEM micrograph and EDX mapping supports the observation of stepwise fabrication of DNA biosensor, including DNA immobilisation and DNA hybridisation process.
Electrochemical Studies. The electrochemical study of HexaFc complex solution was inspected by cyclic voltammetry (CV) to determine their electron transfer properties. Figure 6 shows every layer of modified SPE that were tested by CV with HexaFc complex solution as a redox-active test probe. The sigmoid curve from the biosensor response proposed that the majority of radial diffusion occur on the electrode surface34. Table 1 shows the peak separation between anodic and cathodic peak potential (ΔEP) and anodic and cathodic peak current ratio (IPA/IPC). The separation between anodic and cathodic peak relates to ion resistance involved in the redox reaction35.
Table 1 display the potential difference (∆EP) increase in the electrode order of AuNPs/SiNSs-SPE<AuNPs-SPE<bare SPE<SiNSs-SPE as the electron transfer rate decreased at the electrode surface in the electrode order of AuNPs/SiNSs-SPE>AuNPs-SPE>bare SPE>SiNSs-SPE. As the electron transfer rate decreased, CV became more expansive, and ΔEp value increased. According to Monk36, the reversible redox system can be determined by peak potential differences and IPA/IPC values ~1. The bare SPE and the SiNSs-SPE depict greater peak separations (Figure 6). This may be attributed to the sluggish electron conductivity of the bare electrode, and the non-conductive properties of the SiNSs, which resulted in low electron transfer on the electrode surface. Therefore, the difference between anodic and cathodic peaks can be an indication of the resistance of electron transfer of the electrode35. The value of ΔEP generally decreases due to the presence of AuNPs of good electrical properties because AuNPs enhanced the electron transfer rate36. The current oxidation-reduction peak (IPA/IPC) ratio of AuNP / SiNSs-SPE was close to 1. It shows that this system is still reversible even though the electrode has been modified.
Table 1. Electrodynamics data about anodic peak potential (EPA), cathodic peak potential (EPC), potential difference (∆Ep), anodic peak current (IPA), cathodic peak current (IPC) and anodic to cathodic peak current ratio (IPA/IPC) of HexaFc with different surface-modified working electrodes.
|
EPA (V)
|
EPC (V)
|
∆EP (V)
|
IPA (A)
|
IPC (A)
|
|IPA/IPC|
|
Bare SPE
|
0.26566
|
0.18524
|
0.08042
|
4.5673 × 10-6
|
-3.0541 × 10-6
|
1.4955
|
AuNPs- SPE
|
0.26566
|
0.18997
|
0.07569
|
6.4508 × 10-6
|
-7.0527 × 10-6
|
0.9147
|
SiNSs-SPE
|
0.26093
|
0.17578
|
0.08515
|
5.9079 × 10-6
|
-3.0966 × 10-6
|
1.9079
|
AuNPs/SiNSs-SPE
|
0.27039
|
0.19943
|
0.07096
|
6.191 × 10-6
|
-6.1335 × 10-6
|
1.0094
|
The scan rate study was conducted from 0.008 until 0.1 Vs-1 in the K3[Fe(CN)6] redox indicator (Figure 7a). The K3[Fe(CN)6] system is an appropriate and valued tool for monitoring the characteristics of the modified electrode37. According to figure 7a, the peak current for oxidation and reduction increases proportionally with the scan rate from 0.008 until 0.1V/s. Figure 7b demonstrates the cyclic voltammetry plot at different scan rates. Polarisation increase with the increasing of scan rate, and it develops wide and distorted oxidation and reduction peak. Other than that, the increasing scan rate will lead to the increasing of voltage change among anodic and cathodic peaks38.
Scan rate study was also performed in HexaFc complex from 0.008 until 0.08 Vs-1 to investigate the electrochemical process of the complex. Figure 7c displays that the redox peak current increases with the scan rate from 0.008 until 0.08 Vs-1. However, there is no much difference in the redox peak current for the scan rates within 0.09 and 0.1 Vs-1. This result suggests that HexaFc complex is irreversible at the higher scan rate. The scan rates from 0.008 until 0.08 Vs-1 were more reversible, proof by the linear increase in the current redox (Figure 7d). According to the larger peak area of the CV curves, more material was set based on the Faraday Law39.
Analytical Performance. SiNSs is a non-conductive material and has been used as DNA probe immobilisation sites. The low conductivity of SiNSs reduces biosensor performance. AuNPs have therefore been used to overcome this problem. AuNPs can enhance the potential of electron transfer from the redox indicator and improve the biosensor sensitivity37. Glutaraldehyde has been utilised as a link between porcine DNA probe and to the amine groups of SiNSs. The pH buffer also plays an essential role in producing DNA biosensor because it provides a suitable environment for DNA hybridisation process31,40. Generally, an acidic or basic environment can cause DNA damage31. All the parameters in the optimised condition were used to fabricate porcine DNA biosensor and shown in Table 2.
Table 2. Optimised parameter of porcine DNA biosensor
Parameters
|
Optimum amount
|
Amount of gold nanoparticles
|
0.05 mg
|
Amount of silica nanospheres
|
0.04 mg
|
% of glutaraldehyde
|
10
|
Potassium phosphate buffer concentration
|
0.05 M
|
Potassium phosphate buffer pH
|
pH 7
|
Ionic strength
|
1 M
|
Concentration of DNA probe
|
1 µM
|
Immobilization time
|
24 hours
|
Concentration of HexaFc complex
|
2 × 10-5 M
|
The performance of the biosensor was tested in a sodium phosphate buffer solution, 0.05 M containing complementary porcine DNA with different concentration and 1 M Na+ ion at pH 7. Figure 8 shows a good linearity response of complementary porcine DNA from 1 × 10-6 µM until 1 × 10-3 µM with the correlation coefficient, R2=0.9642. Limit of detection (LOD) for complementary porcine DNA was determined at 4.83 × 10-8 µM. The LOD of the porcine DNA biosensor was calculated following the three times of the standard deviation of the biosensor response in the linear range divided by the direct calibration code41. The average reproducibility relative standard deviation for each calibration point of this biosensor is excellent with most values below 4.5 % (n=3).
Figure 9a exhibits the DPV response before and after the probe-complementary DNA hybridisation process. The DNA electrode based on AuNPs/SiNSs-SPE showed a 100 % matching of the complementary-porcine DNA sequences. The strong and sharp of the DPV peak at a potential ~0.22 V revealed that the HexaFc complex had been successfully intercalated into the porcine DNA double-stranded and it suggested that DNA probe is immobilised and hybridised with complementary-DNA. DNA probe covalently immobilised with SiNSs. No DPV peak was found at potential ~0.22 V for single-stranded porcine DNA probe, and only a small peak was obtained after this biosensor exposed to beef and chicken DNA. This new research will be beneficial for determining porcine DNA in food products by applying new HexaFc complex as an electrochemical porcine DNA indicator.
This DNA biosensor was tested with raw pork meat from several markets and supermarkets around Bangi, Malaysia. It was discovered that the signal from raw pork was found to have a high current value compared to a DNA probe (Figure 9b).
Performance Comparison with Other Reported Porcine DNA Biosensors
Table 3 displays a performance comparison between the constructed porcine DNA biosensor and the other electrochemical porcine DNA biosensors. This comparison is needed for the validation of the constructed biosensor. The constructed porcine DNA biosensor showed substantial improvements in high linear range response and low detection limit compared to other DNA immobilisation matrixes, such as disposable electrochemical printed chips42, gold nanoparticles43, AuNP/NBA-NAS44 and graphene biochips45. This outcome is attributed to the large surface area of silica nanospheres (SiNSs) for DNA immobilisation sites. When compared with DNA hybridisation indicator that employed ferrocenium compounds9,10, the biosensor developed here has comparable LOD, but it did not require any procedure for ferrocene labelling, i.e. chemical attachment of ferrocene compounds to the DNA. This advantage is because the ferrocenium complex used in this work detects DNA hybridisation by intercalation.
Table 3. Comparison of performance between the DNA biosensor reported here with other biosensors using different ferrocene and non-ferrocene based compounds as an indication of hybridisation
DNA electrode materials
|
DNA label
|
Linear range
(µM)
|
LOD
(µM)
|
Reference
|
SPE/AuNPs/SiNSs
|
HexaFc complex
|
1 × 10-6-1 × 10-3
|
4.83 × 10-8
|
This work
|
Fc-acid-OMPA
|
Ferrocene oligomer
|
1.0 × 10-9-1.0 × 10-4
|
2.0 × 10-9
|
[9]
|
Graphene
|
Tetraferrocene
|
2.0 × 10-8-2.0 × 10-3
|
8.2 × 10-9
|
[10]
|
Gold nanoparticles-DNA bioconjugate
|
Methylene blue
|
7.35 × 10-3-3.68 × 10-1
|
4.26 × 10-2
|
[43]
|
SPE/AuNP/NBA-NAS
|
Ruthenium(II) complex
|
1.0 × 10-7-1.0 × 10-2
|
-
|
[44]
|
Graphene biochips
|
Ruthenium hexamine
|
-
|
6.89
|
[45]
|
Disposable electrochemical printed chips
|
Hoechst 33258
|
-
|
1.34
|
[42]
|