A DNA assay protocol in a lab-on-valve meso-fluidic system with detection by laser-induced fluorescence

Abstract

An automatic protocol for in-situ assay of dsDNA is presented by employing a micro-sequential injection lab-on-valve meso-fluidic system, which facilitates precise fluidic handling at the 0.1–10 µl level. Sub-nano-liter to a few micro-liters of DNA sample and ethidium bromide (EB) solutions were introduced into the meso-fluidic system, where EB binding onto DNA takes place and an intercalated DNA–EB adduct was formed, which was afterwards excited in the flow cell of the LOV by a 473 nm laser beam, and the emitted fluorescence was monitored in-situvia optical fibers. The experimental variables, i.e., pH of the buffer solution, the concentration and volume of EB solution, the reaction time and the fluid flow rates, were investigated. By loading 600 nl sample and 1.0 µl EB solution, a linear calibration graph was obtained within 0.03–3.0 µg ml⁻¹ (dsDNA), and a detection limit (3σ) of 0.009 µg ml⁻¹ was achieved, along with a sampling frequency of 60 h⁻¹ and a precision of 1.9% at the 1.0 µg ml⁻¹ level. The detection limit was further improved to 0.006 µg ml⁻¹ by increasing the sample volume to 2.0 µl. Plasmid DNA in E. Coli extraction and λ-DNA/Hind III in four synthetic samples were assayed by using this procedure. For the plasmid DNA, a good agreement with the documented UV method was obtained, while spiking recoveries for the synthetic samples were 95.6–103.4%.

---

Introduction

The quantification of small amounts of nucleic acid, in particular dsDNA (double stranded DNA), is a critical step in a wide variety of biological and diagnostic applications, such as the direct quantification of dsDNA concentration in biological materials for forensic analysis and the determination of the yields of purified DNA fragments for sub-cloning as well as the on-line monitoring of polymerase chain reaction processes.

Various methods for dsDNA quantification are available in the literature, including fluorimetry,^1,2 spectrophotometry,^3,4 chemiluminescence,^5,6 and electrochemical techniques.^7,8 Among those methodologies, laser induced fluorescence attracted extensive attention attributed to its high sensitivity and superior tolerance to interferences. The self fluorescence of dsDNA is seldom directly employed for this purpose because of its very low fluorescence quantum yield. Therefore, appropriate extrinsic fluorescence probes are demanded for improving the sensitivity of the DNA assay. At this juncture, a number of excellent fluorescent probes are commercially available so far, which encompass ethidium bromide.^9,10 Hoechst 33258,¹¹ YOYO-1,¹² M-hypocrellin A,¹³ PicoGreen¹⁴ and some trivalent lanthanide cations.^15,16 In batch mode assays, the most frequently encountered dilemma when using these probes is the high cost, as both the DNA sample and the reagent are rather expensive, and the consumption of sample and reagent volumes are usually large. Thus, downscaling the consumption of expensive/rare sample and/or reagent is obviously one of the most effective approaches in order to reduce the analytical expense. In addition, sample contamination is also regularly encountered which tends to cause further problems for trace level assays. Thus, on-line miniature systems with minimized sample and/or reagent consumption are highly demanded for DNA assay and/or bioassays.

The recently developed micro-sequential injection system with a lab-on-valve (LOV) unit,^17–21 which is also known as meso-fluidic system (where “meso” stands for the level of fluidic manipulation between “micro” and “macro”, and the system denotes a miniaturized system with a capacity of fluidic handling in the 0.1–10 µl scale),^22,23 is an excellent alternative for downscaling the level of sampling and/or sample pretreatment and facilitated novel applications.^24,25 The meso-fluidic system makes it readily feasible for precise manipulation of fluids at the 0.1–10 µl levels. This characteristic is especially beneficial for bioassays where the biochemical reagents/samples are rare or expensive.

In the present paper, the meso-fluidic system incorporating a LOV unit was demonstrated to be very effective for the dsDNA assay with minimized sample consumption, i.e., a few hundred nano-liters. The procedure was developed based on the substantial fluorescence enhancement of ethidium bromide when combining with nucleic acids at a low concentration level. The employment of laser induced fluorescence significantly improved the detection sensitivity. The fluorescence enhancement is proportional to the concentration of dsDNA in a certain range, i.e., 0.03–3.0 µg ml⁻¹ (dsDNA), and thus provided the basis for quantification. The LOV meso-fluidic system not only minimized sample/reagent processing and consumption, but also facilitated in-situ real time monitoring of the reaction process with satisfactory sensitivity.

Experimental

Instrumentation

Experiments were performed by employing a FIAlab 3000 sequential injection system (FIAlab Instruments, Inc., Bellevueo, WA, USA) equipped with a 250 µl syringe pump and a LOV unit, the configuration of which was detailed elsewhere.²⁶ The flow manifold set-up of the LOV-meso-fluidic system is depicted in Fig. 1. All the external channels, including the holding coil, were narrowbore PTFE tubing (id = 0.5 mm) connected to the LOV unit with PEEK nuts/ferrules (Upchurch Scientific, Oak Harbor, WA, USA). The flow cell in the LOV unit was furnished with optical fibers of 200 µm diameter (OceanOptics, Inc., Dunedin, FL, USA) to monitor the laser induced fluorescence right at the inlet of the flow cell. A 10 mW MBL-II laser diode (473 nm, New Industries Optoelectronics Tech. Co., Changchun, China) laser beam was employed as excitation light source via the optical fibers.

Fig. 1 Flow manifold of the micro-sequential injection LOV meso-fluidic system for dsDNA quantification with laser-induced fluorescence.

A PMT-FL fluorometer (OceanOptics) was used to monitor the emission light intensity at 610 nm. The flow cell of the PMT-FL was demounted in order to incorporate the optical fiber from the flow cell in the LOV unit. The entire system, including the meso-fluidic system as well as the PMT-FL fluorometer, was controlled with a single computer by running the FIAlab software for Windows, version 5.9.163.

Reagents

All the reagents used were at least of analytical reagent grade, and 18 MΩ ion-free water was used throughout.

A 1.0 mg ml⁻¹ ethidium bromide (EB) (Life technologies, Gaithersburg, MD, USA) stock solution was prepared, working solutions of different concentrations were obtained by step-wise dilution of the stock solution.

A TE buffer (10 mM Tris, 1 mM EDTA, pH = 8.5) was prepared by mixing 20 ml 0.05 mol l⁻¹ Tris-HCl (pH = 8.5) with 10 ml 0.01 mol l⁻¹ EDTA solution (Tris: Merck, Darmstadt; HCl and EDTA: Shenyang Chemicals Co., China), the mixture was diluted to 100 ml with deionized water.

λ-DNA/Hind III (Huamei Biological Engineering Co., Luoyang, China) was used as purchased, and working solutions of different concentrations were further diluted with the TE buffer (10 mM Tris, 1 mM EDTA, pH = 8.5).

Operating procedure

200 µl of deionized water was first aspirated into the syringe pump, which was afterwards used as carrier for delivering sample and reagent zones into the flow cell for measuring fluorescence or subsequently directed to waste. Consequently, appropriate amounts of EB solution and dsDNA sample solution were sequentially aspirated into the LOV system at a flow rate of 1.0 µl s⁻¹. Thereafter, the syringe pump was immediately set to propel the stacked EB/sample zones forward into the flow cell at a flow rate of 0.5 µl s⁻¹. The intrinsic fluorescence intensity of EB (I₀) as well as that of EB intercalated with dsDNA (I) was monitored by the PMT-FL fluorometer (Em = 610 nm) under identical conditions. A calibration graph was thus derived based on the relationship between the enhancement of fluorescence intensity ΔI(I − I₀), and the concentration of dsDNA.

Results and discussion

The choice of buffer solution and the sampling sequence

TE (10 mM Tris, 1 mM EDTA) buffer was selected as it is one of the most suitable for diluting dsDNA sample solution.² It was demonstrated previously that EB binding with DNA occurs only within a pH range where the structure of the dsDNA is not disrupted,^9,10 the effect of pH on the EB–DNA intercalation in the meso-fluidic system was thus explored. Experimental results indicated that the intrinsic fluorescence intensity of EB as well as that intercalated with dsDNA remained virtually unchanged when the pH of buffer varied within the range of 7.1–9.0, this observation was consistent with that reported in the literature.⁹ A pH of 8.5 was therefore selected for further experiments.

Experiments showed that the enhancement of fluorescence intensity of the EB-DNA intercalated adduct depended strongly on the sampling sequence of EB and DNA solutions in the present system. It was illustrated that under identical experimental conditions, the sensitivity was significantly improved by aspirating EB followed by DNA solution into the meso-fluidic system. The two zones during their transport to the measuring cell undergo dispersion and thereby overlap each other promoting the reaction. This sampling sequence was thus adopted for the ensuing investigations.

The concentration and volume of EB solution

The concentration of the intercalating reagent has a significant effect on the fluorescence intensity of the EB–DNA intercalated adduct, as intrinsically EB emits fluorescence at an identical wavelength. That is to say, the fluorescence arising form the formation of the intercalated species is superimposed on that emitted by EB molecules. The concentration and volume of EB solution should thus be carefully optimized. Fig. 2 illustrates the effect of EB concentration on the net fluorescence intensity from the EB–DNA adduct when using 1.0 µl EB solution of 0.5–3.0 µg ml⁻¹, while the concentration/volume of the DNA sample was fixed as 1.0 µg ml⁻¹/2.0 µl. As a result, a significant enhancement of the fluorescence intensity was observed with the increase of EB concentration up to 2.0 µg ml⁻¹, while afterwards a decline was observed. This might be ascribed to the fact that with an affixed amount of DNA, the fluorescence intensity after intercalation is a measurement of the DNA concentration only when all the DNA binding sites are occupied by EB. The number of EB molecule with a concentration of 0.5 µg ml⁻¹ was insufficient for binding all the DNA fragments. At this stage, an increase of EB concentration gave rise to more intercalated species, and thus an increment of the fluorescence intensity. A concentration of 2.0 µg ml⁻¹ seems to provide sufficient EB molecules to occupy the binding sites, and a maximum fluorescence from the intercalated DNA was achieved. A further increase of EB concentration resulted only in a raise of its intrinsic fluorescence, which submerged the emission from the intercalated species, this together with concentration quenching caused a decline in the recorded signal. An EB concentration of 2.0 µg ml⁻¹ was thus adopted for the ensuing experiments.

Fig. 2 The effect of EB concentration. Volume of EB solution: 1.0 µl; concentration/volume of λ-DNA/Hind III: 1.0 µg.ml⁻¹/2.0 µl; pH = 8.5; loading flow rate: 0.5 µl s⁻¹.

In the present case, the degree of the intercalation reaction depends strongly on the stacked EB and DNA containing sample zones to disperse into each other. The effect of the volume of EB solution was thus further investigated by fixing its concentration at 2.0 µg ml⁻¹, along with a sample zone volume of 2.0 µl. Fig. 3 shows the relationship between the recorded fluorescence intensity derived from the intercalated species and the volume of EB solution within 0.6–2.5 µl. The results illustrate that a maximum was attained with 1.0 µl of EB solution, while further increasing the EB volume did not contribute to the intercalating reaction and the fluorescence intensity curve was leveled off. This observation illustrates that for the EB-λ-DNA/Hind III intercalation system, the binding sites were fully occupied with a stoichiometric concentration of EB and DNA.

Fig. 3 The effect of EB volume. Concentration of EB solution: 2.0 µg ml⁻¹; concentration/volume of λ-DNA/Hind III: 1.0 µg ml⁻¹/2.0 µl; pH = 8.5; loading flow rate: 0.5 µl s⁻¹.

An EB concentration of 2.0 µg ml⁻¹ along with a volume of 1.0 µl was therefore employed throughout the present study even when the sample volume was further downscaled to 0.6 µl, considering that too small an EB volume tended to cause a significant dispersion into the adjacent carrier stream.

The reaction time and loading flow rate

Different sample and reagent loading flow rates correspond to various DNA–EB intercalation times. In the present case, appropriate sample and reagent zones were sequentially aspirated into the LOV unit, which were afterwards propelled immediately into the flow cell for fluorescence measurement. Experiments indicated that the DNA–EB intercalating reaction is not a very fast one, it takes a few seconds to reach a maximum readout. Further investigations illustrated that a slower flow was preferential for improving the fluorescence intensity. When a flow rate of 0.5 µl s⁻¹ was employed, it took ca. 4 s to deliver the sample/reagent zones into the measuring point. Further increasing the loading flow rate resulted in a decline of the fluorescence intensity attributed perhaps to an insufficient reaction time. On the other hand, the gain by employing an even lower flow rate, i.e., 0.25 µl s⁻¹, was virtually negligible. This might be assigned to the fact that the profit by increasing the reaction time was diluted by the dispersion of EB and sample zones into the adjacent carrier stream.

The effect of reaction time was further exploited by using stopped flow with a fixed flow rate at 0.5 µl s⁻¹. It can be seen from Fig. 4 that a slight increase of the fluorescence intensity was obtained with a stopped flow time of 1–2 s, while afterwards the gain was offset by the increase of the dispersion of sample/reagent zones into the adjacent carrier stream. Furthermore, for a longer reaction time, the fluorescence intensity was even dominated by the dispersion effect owing to the very small sample/reagent zones and thus a significant decrease was recorded.

Fig. 4 The effect of stopped flow time. Concentration/volume of EB solution: 2.0 µg ml⁻¹/1.0 µl; concentration/volume of λ-DNA/Hind III: 1.0 µg ml⁻¹/2.0 µl; pH = 8.5; loading flow rate: 0.5 µl s⁻¹.

As a compromise, a flow rate of 0.5 µl s⁻¹ was adopted for the ensuing experiments, while stopped flow was not employed. A sampling frequency of 60 h⁻¹ was achieved at these conditions.

The effect of foreign species

For real world biological samples, the matrix compositions are usually quite complex, which might cause interferences for the assay of trace level DNAs, attributed to both their disturbance on the DNA–EB intercalation reaction as well as the quenching effect. The potential interfering effects of some of those species frequently encountered in biological samples were investigated in the present system. For the assay of 1.0 µg ml⁻¹ λ-DNA/Hind III, no interfering effects were observed arising from alkali, alkaline earth and heavy metal species at the concentration levels found in body fluids. It is also worth mentioning that EB binding onto the DNA fragment occurs within a wide range of NaCl concentration without changing the number of binding sites.⁹ It is thus not necessary to incorporate a sample pretreatment procedure in order to eliminate the salt effect as encountered in a previously established spectrophotometric procedure.²⁶ This feature is highly beneficial for biological samples, as whenever a pretreatment step is included, although appropriate dilution could attenuate the matrix effects, it cannot be used excessively because DNA contents in body fluids were mostly quite low.

The effects of some other important species were also investigated, and the results are summarized in Table 1.

Table 1 Tolerant levels for foreign species^a

Substance	Tolerance level	Substance	Tolerance level
a Concentration of λ-DNA/Hind III is 1.0 µg ml^−1.b v/v.
BSA	0.2%	GuHCl	0.5%
SDS	0.01%	Urea	1.0 mol l^−1
CTAB	1.0%	Isopropanol	1.0%^b
Triton X-100	0.1%^b	CHCl3	1.0%^b
BSA	0.1%	Phenol	0.1%^b
EDTA	20 mmol l^−1	H2PO4^−	50 mmol l^−1

---

The performance of the proposed procedure

Fig. 5 illustrates the fluorescent spectra of EB along with those intercalated with various concentrations of λ-DNA/Hind III, recorded under the optimal experimental conditions. Table 2 summarizes the performance data of the present approach for dsDNA assay by employing the meso-fluidic system, as well as a documenting batch procedure based on EB–DNA intercalation with detection by fluorescence.¹⁰

Fig. 5 The fluorescent spectra recorded under different DNA concentrations. Concentration/volume of EB solution: 2.0 µg ml⁻¹/1.0 µl; concentration/volume of λ-DNA/Hind III: 1.0 µg ml⁻¹/2.0 µl; pH = 8.5; loading flow rate: 0.5 µl s⁻¹.

Table 2 Characteristic performance data of the meso-fluidic system for dsDNA assay

This procedure:
Linear calibration range	0.03–3.0 µg ml^−1
Sampling frequency	60 h^−1
DNA sample consumption	0.6 µl (1.0 µg ml^−1)
EB solution consumption	1.0 µl (2.0 µg ml^−1)
RSD (1.0 µg ml^−1)	1.9%
Detection limit (3σ): 0.6 µl sample	0.009 µg ml^−1
2.0 µl sample	0.006 µg ml^−1
The documented UV procedure (ref. 10):
DNA sample consumption	500 µl (5.0 µg ml^−1)
EB solution consumption	500 µl (25 µg ml^−1)
Detection limit	0.02 µg ml^−1

---

It is obvious that the present approach not only significantly downscaled the DNA sample consumption to 600 nl with respect to 500 µl in the documented procedure, but also provided a superior detection limit of 0.009 µg ml⁻¹ as compared to 0.02 µg ml⁻¹ for the batch mode procedure. The detection limit was further improved to 0.006 µg ml⁻¹ by employing a sample volume of 2.0 µl.

The practical applicability of the proposed procedure was demonstrated and validated by analyzing plasmid DNA in E. Coli extraction and λ-DNA/Hind III in four synthetic samples. The plasmid DNA pGEM-T Easy extracted from Escherichia coli was directly determined by the proposed procedure. An agreement between the obtained content and that determined by a documented UV methodology²⁷ was achieved. Synthetic samples were prepared by spiking appropriate amounts of foreign species in a solution of 1.0 µg ml⁻¹ λ-DNA/Hind III. The obtained results summarized in Table 3 showed satisfactory spiking recoveries.

Table 3 Determination of plasmid DNA in E.Coli extraction and λ-DNA/Hind III in synthetic samples

Samples	Spiked foreign species^a	DNA content/µg ml^−1		RSD^b (%)
		Known	Found
a BSA: 10 µg ml^−1; Ca^2+: 10 µmol l^−1; Mg^2+: 10 µmol l^−1; urea: 10 µmol l^−1; GuHCl: 100 µg ml^−1; EDTA: 10 mmol l^−1; H2PO4^−: 10 µmol l^−1; CTAB: 10 µg ml^−1; SDS: 10 µg ml^−1; Triton X-100: 0.01% (v/v).b Average of 5 determinations.c µg µl^−1, determined by UV spectrophotometry.
Plasmid DNA extractant		0.40^c	0.38
Synthetic sample 1	BSA, Ca^2+, Mg^2+	1.0	0.96	2.4
Synthetic sample 2	BSA, Urea, GuHCl	1.0	0.97	1.9
Synthetic sample 3	BSA, EDTA, H2PO4^−	1.0	1.00	2.4
Synthetic sample 4	SDS, CTAB, Triton X-100	1.0	1.03	1.4

---

Conclusions

The described protocol for dsDNA assay by employing a meso-fluidic system is characterized by minimized fluid consumption, i.e., only 600 nl of sample volume is required. This allows considerable reduction in sample consumption in comparison with conventional flow analysis procedure, not to say the batch mode method based on the same EB–DNA intercalation reaction, which consumes 800 times more DNA sample. This provides a promising avenue for bioassays, especially for routine analysis with very limited amount of sample volume. In the mean time, the employment of a LOV unit also facilitated in-situ monitoring of the reaction process. Thus, it might open a new path to carry out separation and purification of nucleic acids and some other species of biological interest by using the meso-fluidic system with in-situ fluorescence detection.

Acknowledgements

The authors are indebted to financial support from the National Natural Science Foundation of China (NSFC, 20375007), the China Postdoctoral Science Foundation, and the Natural Science Foundation of Liaoning Province (20042011).

References

1. E. S. Morozkin, P. P. Laktionov, E. Y. Rykova and V. V. Vlassov, Anal. Biochem., 2003, 322, 48.
2. F. Vitzthum, G. Geiger, H. Bisswanger, H. Brunner and J. Bernhagen, Anal. Biochem., 1999, 276, 59.
3. A. A. Killeen, Microchem. J., 1995, 52, 333.
4. M. Samuel, M. Q. Lu, C. J. Pachuk and C. Satishchandran, Anal. Biochem., 2003, 313, 301.
5. Y. J. Ma, M. Zhou, X. Y. Jin, Z. Y. Zhang, X. L. Teng and H. Chen, Anal. Chim. Acta, 2004, 501, 25.
6. H. Chen, M. Zhou, X. Y. Jin, Y. M. Song, Z. Y. Zhang and Y. J. Ma, Anal. Chim. Acta, 2003, 478, 31.
7. N. A. El-Maali and J. Wang, Sens. Actuators B, 2001, 76, 211.
8. R. H. Yang, K. M. Wang, D. Xiao, K. Luo and X. H. Yang, Anal. Chim. Acta, 2001, 432, 135.
9. J. B. Le Pecq and C. Paoletti, Anal. Biochem., 1966, 17, 100.
10. U. Karsten and A. Wollenberger, Anal. Biochem., 1972, 46, 135.
11. R. Rago, J. Mitchen and G. Wilding, Anal. Biochem., 1990, 191, 31.
12. J. M. Butler, B. R. McCord, J. M. Jung, M. R. Wilson, B. Budowle and R. O. Allen, J. Chromatogr. B, 1994, 658, 271.
13. Q. Z. Zhu, F. Li, X. Q. Guo, J. G. Xu and W. Y. Li, Analyst, 1997, 122, 937.
14. P. Murakami and M. T. McCamanl, Anal. Biochem., 1999, 274, 283.
15. Y. X. Ci, X. L. Hu, Y. Z. Li and W. B. Chang, Chin. J. Anal. Chem., 1992, 20, 1083.
16. Y. X. Ci, Y. Z. Li and X. J. Liu, Anal. Chem., 1995, 67, 1785.
17. J. Ruzicka, Analyst, 2000, 125, 1053.
18. A. D. Carroll, L. Scampavia and J. Ruzicka, Analyst, 2002, 127, 228.
19. C. M. Schulz and J. Ruzicka, Analyst, 2002, 127, 1293.
20. J. H. Wang and E. H. Hansen, Trends Anal. Chem., 2003, 22, 225.
21. J. H. Wang and E. H. Hansen, J. Anal. At. Spectrom., 2001, 16, 1349.
22. J. H. Wang, Anal. Bioanal. Chem., 2005, 381, 809.
23. J. H. Wang and Z. L. Fang, Chin. J. Anal. Chem., 2004, 32, 1401.
24. Y. Chen and J. Ruzicka, Analyst, 2004, 129, 597.
25. H. Erxleben and J. Ruzicka, Analyst, 2005, 130, 469.
26. X. W. Chen, J. H. Wang and Z. L. Fang, Talanta, 2005, 67, 227.
27. J. Sambrook, E. F. Fritsch and T. Maniatis, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor, New York, 1989.

---