Supporting Informations
fM to aM nucleic acid amplification for molecular diagnostics in a silicone-coated metal microfluidic bioreactor
Guoliang Huang 1,2, Qin Huang 2, Li Ma2, Xianbo Luo2, Biao Pang2, Zhixin Zhang2,Yuliang Wang1, Junqi Zhang1, Qi Li1, Rongxin Fu1, and Jiancheng Ye1
1Department of Biomedical Engineering, the School of Medicine, Tsinghua University, Beijing 100084, China.
2National Engineering Research Center for Beijing Biochip Technology, Beijing 102206, China.
Correspondence and requests for materials should be addressed to G.L.H. ()
Supporting Information
Results
Comparison of isothermal DNA amplification in uncoated and non-stick-coatedmetal micro-nanoliter fluidic chips. Using the same DNA template concentration of 1.3 fM (10-15 M) and 7-μL reaction mixtures in every bioreactor cell, isothermal DNA amplification in uncoated and non-stick coatingmetal micro-nanoliter fluidic chips was performed using our portable confocal detector. The contrast of DNA isothermal amplification curves is shown in Figure S1. Figure S1 (a) corresponds to isothermal DNA amplification in the uncoated metal micro-nanoliter fluidic chip, where the times at the second derivative inflexions of the exponential DNA amplification curves for the five bioreactor cells were 24.85, 27.29, 27.80, 28.18, and 29.53 min. The maximum of the time difference at these inflexions was ~4.68 min. At the top of the amplification curves, the fluorescence intensities of the five bioreactor cells ranged from 3600-6900; the percent deviation of the fluorescence intensity was ~91.7%. Figure S1 (b) is isothermal DNA amplification in the non-stick-coated metal micro-nanoliter fluidic chip, where the times at the second derivative inflexions of the exponential DNA amplification curves for the five bioreactor cells corresponded to 19.42, 19.45, 19.59, 19.60, and 19.92 min. The maximum of the time differences at these inflexions was 0.5 min. At the top of the amplification curves, the fluorescence intensities of the five bioreactor cells ranged from 4000-4500; the percent deviation of the fluorescence intensity was ~11%. Figure S1 (c) is the normalizing processing of the isothermal DNA amplification curves in Figure S1 (b), which is generally used in commercial RT-PCR setups.
Figure S1
Table S1
DNA template (fM)Time at the inflexion (min) / 1.3×102 / 1.3×101 / 1.3×100 / 1.3×10-1 / 1.3×10-2 / 1.3×10-3
In 7-μL non-stick-coated metal micro-nanoliter fluidic bioreactors measured using our Portable Confocal Detection System / 14.8 / 17.4 / 19.6 / 21.5 / 24.5 / 28.7
In25-μL Eppendorf tubes measured using an ABI 7700 Fast Real-Time PCR System / 16.3 / 18.4 / 20.8 / 23.7 / 28.9 / non amplification
Discussion
The fluorescent detection response of nucleic acid amplification is related to many factors, including the number of amplified double-stranded DNAs (dsDNAs), the number of denatured dsDNAs at the amplification temperature, the surface adsorption of the bioreactor, photo-bleaching, and background noise. The continuously increasing number of amplified dsDNAs produces a positive or upper dynamical amplified signal. However, the denatured dsDNA at 65°C, the surface adsorption of DNA by the bioreactor, and photo-bleaching from the excitation light lead to a negative or downward dynamical change of the amplified signal, which usually causes the detection response of the nucleic acid amplification to be unstable andlate1,2. Further, the background noise of the bioreactor material and surroundings produces a high original value for the dynamical amplified signal and low amplified efficiencyor low ratio of end signal to start signal, which usually decreases sensitivity. Therefore, general nucleic acid amplification reactions inEppendorf tubes and 96- or 384-well microplates is usually performed in ≥25-μL volumes to increase the signal of the amplified dsDNA far more than the lost signal from the negative factors, as well as to obtain a relative steady exponential mode for the amplified signal and good consistency for repeated detection. Indeed, in these conditions, the differences from the denatured dsDNA at the high amplification temperature, the surface adsorption of the bioreactor, photo-bleaching, and background noise can be ignored.
When the reaction volume is reduced to <10 μL, the above negative effects are significant. If the lost fluorescence signal from the denatured dsDNA is Eu, the increasing fluorescence signal from the amplified dsDNA is Ea, the decreased fluorescence signal from surface adsorption is Es, the reduced fluorescence signal from photo-bleaching is Ep, and the background signal from the bioreactor material and surroundings is Eb, then the dynamic measured fluorescence signal of nucleic acid amplification E can be calculated with the following formula (1):
E = Eo + Ea + Eb - Es - Eu - Ep (1)
where Eo is the original fluorescence signal of the mixtures, including the template DNA, sample, and reagents at 25°C. We report that the non-stick-coated metal micro-nanoliter fluidic chip and portable confocal detector decreasedthese possible negative effects, resulting in stablenucleic acid amplification, sensitive detection, and a rapid response in micro-nanoliter reaction volumes.
Figure S1 shows an obvious change in the nucleic acid amplification from the surface adsorption in uncoated and non-stick-coated metal micro-nanoliter fluidic bioreactors. In Figure S1 (a), the uncoated metal micro-nanoliter fluidic chip displays a larger difference both in the times at the second derivative inflexion of the exponential DNA amplification curves (~4.7 min) and the relative fluorescence intensity (~91.7%). In Figure S1 (b), when the metal micro-nanoliter fluidic bioreactor was coated with the silicone non-stick material, the surface of the bioreactor became smooth and inert to DNA molecules, effectively eliminating surface adsorption (Es), which makes the nucleic acid amplification stable with little difference in the times at the second derivative inflexion of the exponential DNA amplification curves (~0.5 min) and the relative fluorescence intensity (~11%).
Figure S2 displays our portable confocal detector, whichcancollect fluorescence near the optical diffraction limit and effectively limit the background noise (Fig. S5). The fluorescence from an object on the focal plane can be focused into a small spot and transmitted with >92% intensity to the detector behind a pinhole filter (PH) at the focal plane of the imaging lens set (L2). The fluorescence from other off-focus objects creates a large dispersive spot with a very low transmitted efficiency behind the PH (e.g., the transmitted efficiency at the off-focus position of 18 μm is only ~1.13%). The detection plane for the fluorescence of the bioreactor material and surrounding is farther than the off-focus position of 50 μm, and thus, the background noise (Eb) from the bioreactor material and surroundings is approximately 0 and can be ignored.
The photo-bleaching from the excitation lightis directly proportional to 1-e-t/T, where T is half of the attenuant period and t is the irradiation time with the excitation light3. In our portable confocal detector, the micro-nanoliter fluidic bioreactor was moved by a rotary scanning stage (RS), and every bioreactor cell in one measuring period of 30 s was irradiated for <10 ms. Therefore, the photo-bleaching from the excitation light (Ep) was approximately 0. High amplification efficiency and a negligible effect from denatured dsDNA can be obtained by optimizing the amplification conditions, including the temperature (65°C) and the addition of 0.5 mg/ml BSA and 6 mM MgS04, which renders the lost fluorescence signal from denatured dsDNA (Eu) stable as a constant.
After all of the effects from the surface adsorption of the bioreactor, the denatured dsDNA at the amplification temperature, the photo-bleaching from the excitation light, and the background from the bioreactor material and surroundings are limited, and equation (1) can is simplified as:
E = Ea + C (2)
where C isa constant,C = Eo - Eu. Compared to equation (1), equation (2) indicates that the dynamic fluorescence signal E of the nucleic acid amplification is directly proportional to the increased fluorescence signal Ea of amplified dsDNAs and can immediatelyrespond to amplified dsDNAs in real time. Thus, the sensitivity can be improved when the negative factors are eliminated.
Figure S2
Methods
Design of primers. Oligonucleotide primers were designed for the isothermal amplified assay according to the sequence of the Mycoplasma pneumoniae P1 gene from GenBank (Accession No. M18639). Six primers, including two loop primers (LF and LB), two outer primers (F3 and B3), and two inner primers (FIP and BIP), were designed to recognize eight distinct regions on the target sequence (Fig. S3). BIP consists of the complementary sequence of B1 and the sense sequence of B2, FIP is comprised of the complementary sequence of F1 and the sense sequence of F2. The primer sequences are listed in Figure S4.
Figure S3
Figure S4
Portable real-time fluorescent confocal detector. To detect the fluorescent signal of DNA amplification in real time with high sensitivity and low background, a new portable confocal detector was developed as shown in Figure S5. In Figure S5, an objective set L1 with a focal length of 13 mm and numerical aperture of NA = 0.72 was designed, which consists of seven lenses, including two doublets, and uses only three glass materials, ZK7, ZF2, and ZK11. The excited light from a white LED was first collimated by an aspherical lens L1 and filtered by the band pass filter F1 with a central wavelength of 480 nm and bandwidth of 30 nm. Then, the excited light was focused by the objective set L2 to illuminate the bioreactor cells of the micro-nanoliter fluidic chip MC on the rotary scanning stage RS. The intensity of the excited light was adjusted to adapt to the different applications of the chip by the attenuator A1. A dichroic mirror D1 was used to reflect the fluorescence from the bioreactor cell and to penetrate the excited light from the LED. The fluorescence from Sybase Green inserting into the dsDNA in the bioreactor cell was excited by the LED and initially collected by the objective set L2, reflected by D1, and focused onto the detector PMT (HAMAMATSU, Japan) by the imaging lens set L3 with a focal length of 22 mm. A filter F2 with a 520-nm central wavelength and 40-nm bandwidth was used to penetrate the fluorescence and limit the excited light. The pinhole PH was used to filter the farraginous light from the environment and the off-focused fluorescence from the material of the micro-nanoliter fluidic chip. Finally, the fluorescence signal was transferred to a computer by an A/D processor. The temperature controller TCD was used to administer the heater HF with an accuracy of 0.1°C and speed of raising the temperature of 1°C/s by the temperature feedback of sensor S in the process of DNA isothermal amplification. A multiple-axis moving driver MAMD was used to supervise the rotary scanning stage RS with angle accuracy of 0.01°.
When the micro-nanoliter fluidic chip was used, the inlet and outlet holes were first opened using an Eppendorf tip with light pressure. Then, the mixtures of the circular probes, reagents, and DNA samples were injected into the bioreactor cells of the micro-nanoliter fluidic chip from the inlet hole using the tip. Third, the inlet hole and the outlet hole were sealed with a thin PC film from ABI Corporation. Fourth, the thin PC film, exceeding the edge of the chip or covering the center fixed hole of the chip, was trimmed for detection. Fifth, the channel between any two adjacent bioreactor cells was obstructed by heating compression to keep all bioreactor cells independent of each other during the process of isothermal amplification. Finally, the micro-nanoliter fluidic chip was placed into the portable confocal detector (Fig. S5) for isothermal DNA amplification and gene-specific identification.
Figure S5
References
- Wang, M. et al. Accelerated Photobleaching of a Cyanine Dye in the Presence of a Ternary Target DNA, PNA Probe, Dye Catalytic Complex: A Molecular Diagnostic. Anal. Chem.81, 2043–2052 (2009).
- Koek, M.M. et al. Metabolic Profiling of Ultrasmall Sample Volumes with GC/MS: From Microliter to Nanoliter Samples. Anal. Chem.82, 156–162 (2010).
- Huang, G.L. et al. Fluorescence photobleaching properties for microarray chips. Chinese Journal of Luminescence. Chinese Journal of Luminescence27, 259-264 (2006). (in Chinese)
Figure caption
Figure S1Comparison of DNA amplification in uncoated and non-stick-coated metal micro-nanoliter fluidic chips. (a) Isothermal DNA amplification in the uncoated metal micro-nanoliter fluidic chip, which displays obvious differences among the five parallel bioreactors. (b) Isothermal DNA amplification in the non-stick-coated metal micro-nanoliter fluidic chip, where the amplification in the five parallel bioreactors displays good consistency and the time difference at the second derivative inflexions of the exponential DNA amplification curves for the five bioreactors are within 0.5 min. (c)Normalizing processing of the isothermal DNA amplification curves in (b).
Table S1 Comparison of the times at the inflexions of the second derivative of the exponential DNA amplification curves between our advanced metal micro-nanoliter fluidic bioreactors and in the general 25-μL Eppendorftubes.
Figure S2 Comparison of fluorescence collection on the focal plane to the off-focus position forthe developed portable confocal detector. The horizontal axis corresponds to the radius of the optical spot from the center of the optical axis, and the vertical axis is the relative transmitted intensity. The fluorescent intensity on the focal plane can be collected with a high efficiency (92%) near the optical diffractive limit.There was a low transmitted efficiency of 3.46% at the off-focus position of 8 μm and lower transmitted efficiency of 1.13% at the off focus position of 18 μm.
Figure S3 Diagram of the primers used. Constructions of the inner primers BIP and FIP are displayed. F1c and B1c are complementary sequences to F1 and B1, respectively.
Figure S4 Partial nucleotide sequences of the P1 gene of M. pneumoniae and the primers used for amplification. Arrows indicate positions of specific primers. The left arrow indicates that a complementary sequence is used for the primer. The right arrow indicates that a sense sequence is used for the primer.
Figure S5 Portable confocal detector. The white light LED had a power of 120 mW, aspherical lens L1 had a focal length of 20 mm, A1 is an attenuator, F1 and F2 are two bandpass filters, D1 is a dichroic mirror, and L2 is an objective. MC is the microfluidic chip, S is the temperature sensor, RS is the rotary scanning stage, HF is the heater, L3 is an imaging lens set, PH is the pinhole, and PMT is a photo-electronic detector. A/D is a 16-bit analog-to-digital processor, TCD is the temperature controller, MAMD is a multiple-axis moving driver, and a computer was used to manage the process of isothermal DNA amplification and display the amplification curves in real time.
1