In this part, to demonstrate the SBLIA and its efficacy on AuNP colorimetry, our system is applied to short strand DNA detection with a target DNA concentration of 0,
Table 1. Comparison of the proposed SBLIA method and the SCB method
Technology System Cost Resolution
doi:10.6342/NTU201700804
32
20, 40, 80 and 120 nM. The target DNA is 15-mer in length and is complementary to a 15-mer probe strand.
The probe ssDNA is always 200 nM for all samples regardless of the target complementary DNA concentration In this experiment, the SBLIA works under the optimized parameter, i.e. f=2.5 kHz, τ = 60 ms and Vaudio=40 mV. Finally, the SBLIA-AuNP colorimetry results are compared with a commercial UV-Visible spectrometer.
UV-Visible spectrometer being generally considered as gold standard tool for colorimetry detection. The spectrometer was set to have an averaging time around 60 ms in order to compare both system on a fair basis. A typical UV-Visible spectrum for AuNP samples with different concentrations of target DNA is shown in Fig. 15 (a). As revealed by the spectrum, the absorbance between 550 nm to 700 nm increases as target DNA concentration increases. This corresponds to a AuNP sample color change as shown in Fig. 15 (b) from red to dark purple, when target DNA concentration increases. As explained earlier, this is due to the fact that AuNP tends to aggregate in salt environment when dsDNA presents. This aggregation leads to change in the AuNP sizes which alters the localized surface plasmon resonance spectrum.
Fig. 15 (c) demonstrates a typical DNA SBLIA-AuNP colorimetry sensogram. Judging on the sensogram, SBLIA system shows the similar monotonic change as in UV-Visible spectrometer, the modulation amplitude of the beam drop by 33% from 0 nM to 120 nM
doi:10.6342/NTU201700804
33
of target DNA. This is again due to the increasing absorbance of the AuNP sample along with increasing dsDNA concentration. In order to further compare the results from SBLIA-AuNP sensogram and UV-Visible spectrometer, the above mentioned detection is repeated four times for both systems and calibration curves are then established. The data from our SBLIA system is converted into the absorption unit (AU) for constructing the calibration curve. The noise is measured in each set of test and presented as averaged results. The calibration curve can be found in Fig. 15 (d). For both systems, the absorbance at 650 nm (A650) has a very similar linear dependence to the concentration (C,
in unit of nano-molar) of target DNA. As indicated from the curve, the calibration curve is A650=1.44x10-3C for the SBLIA and A650=1.1x10-3C for the UV-Visible system. Since both systems use a cuvette with the same optical path length, this result is reasonable judging from the Beer-Lambert’s law. Further investigation of the systematic noise indicates that the noise form the SBLIA system is 4.3 times smaller than the UV-Visible spectrometer, i.e. 3.7 × 10−4 AU compared to 1.6 × 10−3 AU.
Finally, based on the obtained noise and the calibration curves, the LOD is then calculated to be 0.77 nM for our system which is 5.7 times lower than 4.36 nM achieved by the UV-Visible system.
The importance of the obtained results, beyond the figure of improvement, lies in the insights that they provide into the context and future development of smartphone based
doi:10.6342/NTU201700804
34
diagnostic. Firstly and most importantly, the SBLIA proposed herein has high potential for development of further optical based sensor technologies. Although not explicitly used herein, the phase detection noise of our SBLIA system is down to 0.2 degree which could lead to high quality smartphone based phase sensitive detection (Fig. 16). For
Fig. 15 SBLIA-AuNP colorimetry DNA sensing
(a) UV-Visible spectrum of the AuNP sample solution. The sample solution contains 200 nM of ssDNA with different concentration of target DNA. (b) An image of the sample solutions and a scheme of the AuNP DNA sensing mechanism. (c) SBLIA-AuNP colorimetry DNA sensing amplitude sensogram. The concentration of target DNA is marked and the blue spike in the figure indicate the spike due to change of sample. (d) Calibration curve for the SBLIA-AuNP colorimetry system (blue trace) and UV-Visible spectrometer (black trace).
doi:10.6342/NTU201700804
35
example, through proper optical design, SBLIA could be applied to build smartphone based LIA enhanced fluorescence tag bio-sensor23, phase sensitive detection of plasmonic sensor 24 and even holographic measurement (multi-channel phase detection). Hence,
while smart AuNP colorimetry is undoubtedly an interesting approach to pursue, we believe that the SBLIA itself should spur a wide range of distinct applications in sensing.
Secondly, we note that the ratiometric measurement using a 650 nm/520 nm absorbance ratio is not used here as in our previous report since audio channel provide only one input25. As is well known, using 520 nm absorbance to normalized 650 nm absorbance may further enhance the batch-to-batch reproducibility of AuNP colorimetry.
In the future, this may be addressed by incorporating new electronic design6 or by extending SBLIA onto an external embedded system where multiple analog receivers are available. The latter design could add flexibility to SBLIA data resolution, by adding a
Fig. 16 phase signal extracted with SBLIA.
doi:10.6342/NTU201700804
36
number of detection channels. Such approach would also be well adapted to smartphone models where audio jack is absent.
Finally, the ultimate performances of smartphone based diagnostic is still an on-going debate8 and important design factors need to be considered for applying SBLIA-AuNP system in on-site disease screening. These factors include complex background of physiological sample26 and the consequent pre-processing issue27. Still, some research results report smart diagnostic devices have already reached world health organization (WHO) gold standards3 or other gold standards9, and have been applied to on-site application or disease screening4. As exemplified by this work, we are expecting more and more smartphone based diagnostic tools with equal or even better performances as compared to some of their laboratory relatives.