- Email: [email protected]
Determination of nucleic acids using phosphin 3R as a fluorescence probe
Analytica Chimica Acta 394 (1999) 177±184
Determination of nucleic acids using phosphin 3R as a ¯uorescence probe Qing-Zhi Zhua,*, Huang-Hao Yanga, Dong-Hui Lib, Jin-Gou Xua a
The Key Laboratory of Analytical Sciences of MOE, Department of Chemistry, Xiamen University, Xiamen 361005, People's Republic of China b Cancer Research Center, Xiamen University, Xiamen 361005, China Received 13 November 1998; received in revised form 24 March 1999; accepted 27 March 1999
Abstract A novel ¯uorimetric method has been developed for rapid determination of DNA and RNA with phosphin 3R (PR) as a ¯uorescence probe, based on the ¯uorescence quenching of PR in the presence of DNA or RNA. Maximum ¯uorescence quenching is observed in the pH range 7.0±8.4, with maximum excitation and emission wavelength at 468 and 505 nm, respectively. Under optimal conditions, the calibration graphs are linear up to 2.0 mg/ml for both calf thymus DNA (CT DNA) and salmon DNA (SM DNA), and up to 1.6 mg/ml for yeast RNA, respectively. The corresponding detection limits are 5.0 ng/ ml for CT DNA, 6.0 ng/ml for SM DNA and 13.0 ng/ml for yeast RNA. CT DNA could be determined in the presence of 20% (w/w) yeast RNA, and the relative standard deviation of six replicate measurements is 1.00% for a solution containing 400 ng/ ml of CT DNA. Three real samples were determined with satisfactory results. The interaction mechanism for the binding of PR to DNA is also studied; the results of absorption spectra and thermal denaturation experiments suggested the interaction between PR and DNA to be intercalative in nature. # 1999 Elsevier Science B.V. All rights reserved. Keywords: Fluorescence quenching; Phosphin 3R; DNA; RNA
1. Introduction The quantitative analysis of nucleic acids is very important since it is often used as a reference for measurements of other components in biological samples. Generally, direct use of the natural ¯uorescence emission properties of nucleic acids for their structure and dynamic studies, and ¯uorimetric determinations has been limited [1,2] due to the low ¯uorescence quantum yield of native DNA. Therefore, some extrin*Corresponding author. Tel.: +86-592-2182442; fax: +86-5922188054; e-mail: [email protected]
sic ¯uorescence probes have been employed to study DNA. A number of ¯uorimetric methods for the determination of nucleic acids have been established [3±8] based on the reaction in which the double helix strands of DNA molecules are intercalated by or interact with probes such as ethidium bromide [3], mithramycin [4], bisimidazole (Hoechst 33258) [5], and 40 ,6-diamidino-2-phenylindole [6]. Furthermore, the trivalent lanthanide cations such as Tb(III) [9±11], Eu(III) [12], La(III) [13] and some dimeric asymmetric cyanine dyes [14,15] have been developed as ¯uorescence probes of the structure and function of nucleic acids in recent years. However, the main
0003-2670/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 3 - 2 6 7 0 ( 9 9 ) 0 0 3 0 9 - 8
178
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
trum was performed on a Shimadzu UV-240 ultraviolet±visible spectrophotometer. All the pH measurements were made with a Model PHS-301 pH meter 631 (Xiamen, China). 2.2. Reagents
Fig. 1. The structure of phosphin 3R.
limitation of these methods is that the lanthanide metals (such as Tb3 and Eu3) and cyanine dyes (such as TOTO, YOYO etc.) are very expensive. More recently, we have studied the interaction of some inexpensive photoactive ¯uorescence probes with nucleic acids and developed the sensitive ¯uorimetric methods for determination of nucleic acids based on the in-situ photochemical ¯uorescence reaction [16,17]. However, the photo-induced ¯uorescence reaction needed a long time to reach equilibrium. Therefore, it is considered important to develop another sensitive and rapid method for determination of DNA using an inexpensive probe. Phosphin 3R (PR) is a very inexpensive dye (Fig. 1), which displays strong ¯uorescence emission at 505 nm with maximum excitation at 468 nm. In neutral medium, its ¯uorescence is signi®cantly quenched in the presence of nucleic acids. We have therefore employed PR as a ¯uorescence probe and developed a sensitive ¯uorimetric method for the determination of nucleic acids. The use of PR as a ¯uorescence probe leads to a particularly inexpensive, simple and sensitive system, permitting a limit of detection of 5.0 ng/ml for CT DNA, 6.0 ng/ml for SM DNA and 13.0 ng/ml for yeast RNA, respectively. The mechanism for the binding of PR to DNA is also studied. 2. Experimental 2.1. Apparatus A Hitachi 650-10S spectro¯uorimeter equipped with a plotter unit and a 1 cm quartz cell was used for ¯uorescence measurements. The absorption spec-
All chemicals were of analytical reagent grade or higher purity available. All aqueous solutions were made up in distilled, deionized water. Commercially prepared calf thymus DNA, salmon DNA and yeast RNA, obtained from Sino-American Biotechnology (Shanghai, China), were directly dissolved in water at a ®nal concentration of 200 mg/ml and stored at 48C. These solutions were diluted to 1.0 mg/ml with water to be used as working solutions. A PR stock solution (1.010ÿ3 mol/l) was prepared by dissolving the appropriate weight of PR (Beijing Chemical Reagent, Beijing, China) into 100 ml of water and stored in the dark. This solution was diluted to 1.010ÿ4 mol/l with water as working solution. A pH 8.0 Tris±HCl buffer solution was prepared by mixing 50 ml of 0.1 mol/l Tris and 29.2 ml of 0.1 mol/l HCl, then diluted to 100 ml with water. 2.3. Procedure Transfer 2.0 ml of buffer solution (pH 8.0) and 0.5 ml of PR solution (1.010ÿ4 mol/l) to a 10 ml standard ¯ask. Add a known volume of DNA (or RNA) standard solution. Dilute to the mark with water and mix. Measure the relative ¯uorescence intensities of the reagent blank (F0) and the mixed solution (F) at 505 nm with excitation at 468 nm. Plot a calibration curve of the ¯uorescence quenching (F0ÿF) vs. the concentration of nucleic acids. 2.4. Treatment of real samples The treatment of real samples was similar to the method reported previously [18]. In order to obtain DNA and RNA from the samples completely, a modi®ed procedure was used. 0.1 g of dry pollen (or 0.3 g of honey) was mixed with 1 ml of 5% perchloric acid solution in a mortar box. The mixture was ground to a homogeneous mass at 08C, and centrifuged at 08C for 10 min (3000 r.p.m.). The precipitate was then collected and washed with 1 ml of 5% perchloric acid at
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
08C for 10 min, followed by centrifugation. This procedure was repeated three times in order to remove acid-soluble proteins. 5 ml of ethyl alcohol±ethyl ether±chloroform mixing solvent (2:2:1, v/v) was added to the precipitate, the mixture was allowed to stand for 15 min at room temperature to extract phosphatide, then centrifuged (3000 r.p.m.) to remove the extraction solution. The procedure was repeated four times, and the ®nal precipitate of nucleic acids was dried in vacuum drier. 3.0 ml of 1.0 mol/l NaOH solution was added to the precipitate to hydrolyze nucleic acids at 258C for 8 h. The mixture was centrifuged and the centrifugate was collected, then the relict was washed with 1.0 ml of 1.0 mol/l NaOH twice, followed by centrifugation to remove the residue. All the centrifugates were mixed and acidi®ed by adding 1.0 ml of 6.0 mol/l HCl solution, and then left at 48C overnight to precipitate DNA from the solution. After DNA was separated by centrifugation, the centrifugate, RNA moiety, was quantitatively transferred into a 25 ml standard ¯ask and diluted to mark with water. Then, 5.0 ml of 1.0 mol/l perchloric acid solution was added to the DNA precipitate to hydrolyze DNA at 758C for 20 min. The hydrolysate was centrifuged and the centrifugate (DNA moiety) was transferred into a 25 ml standard ¯ask and diluted to mark with water. The DNA and RNA levels in the real samples were determined according to the procedure described in Section 2.3. 3. Results and discussion 3.1. Spectral characteristics of fluorescence The uncorrected excitation and emission spectra of PR are shown in Fig. 2. In neutral medium, PR has an emission band located at 505 nm with an excitation peak at 468 nm. When DNA (or RNA) is added, the excitation and emission maxima of the PR±DNA system are similar to that of free PR, but the ¯uorescence intensity is signi®cantly quenched. On the other hand, RNA can similarly quench the ¯uorescence of PR; however, its quenching ability is lower than that of DNA. These results indicated that PR can be used as a new ¯uorescence probe for sensitive determination of DNA. In this paper, the maximum excitation peak at
179
Fig. 2. Excitation (a) and emission (b) spectra of free PR (solid curves) and in the presence of yeast RNA (dot-dashed curves) and CT DNA (dashed curves). PR: 5.010ÿ6 mol/l; CT DNA: 400 ng/ ml; RNA: 400 ng/ml.
468 nm and the emission peak at 505 nm were used for ¯uorescence intensity measurements. 3.2. Optimization of the general procedure The experimental results indicated that maximum and constant ¯uorescence quenching was produced when the PR concentration was in the range of 4.010ÿ6±6.010ÿ6 mol/l. In this work, a PR concentration of 5.010ÿ6 mol/l was recommended. The effect of pH on the ¯uorescence quenching of the system was studied. The ¯uorescence quenching reached a maximum in the pH range of 7.0±8.4. Therefore, a pH of 8.0 was recommended. This value was obtained by addition of 2.0 ml buffer solution per 10 ml of the ®nal solution. The in¯uence of incubation time on ¯uorescence quenching was also investigated. The results showed that the maximal ¯uorescence quenching was immediately reached when the solutions were mixed and remained constant for at least 4 h. In this work, the ¯uorescence intensity was directly measured after the solutions were mixed,
180
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
and an additional incubation time was not needed. For RNA, the optimum conditions were similar to that of the DNA system. 3.3. Interference of foreign substances The effects of various foreign substances on the determination of 400 ng/ml CT DNA by the described procedure were studied and compared with another method, which uses 8-hydroxyquinoline/lanthanum complex as a probe [13]. The results are shown in Table 1. It can be seen that the coexisting purine or pyrimidine bases, even when present in large excess, do not interfere with the determination of DNA in the proposed method. Furthermore, compared with 8hydroxyquinoline/lanthanum probe, proteins only give less interference in our method. However, some foreign ions, such as Pb2, Co2, Ni2 seem to show low tolerance levels. 3.4. Calibration graphs The calibration graphs for the determination of DNA (or RNA) were constructed under the optimal
conditions. The results, shown in Fig. 3, exhibit good linear relationships between the extent of ¯uorescence quenching and the concentrations of DNA (or RNA). All the analytical parameters are presented in Table 2. From Table 2 it can be seen that the sensitivity for the determination of CT DNA is higher than that for SM DNA and yeast RNA with the sequence of CT DNA > SM DNA > yeast RNA. An interesting phenomenon is that there are two linear ranges in the calibration curves for both DNA and RNA; one linear range is between 0 and 100 ng/ml DNA (or RNA), the other is over 100±2000 ng/ml DNA (100± 1600 ng/ml RNA). Furthermore, there are good linear relationships between the ¯uorescence quenching and the concentrations of DNA (or RNA) in both ranges, and the sensitivities at low concentration are higher than those at high concentration. This phenomenon was also observed for other methods [12,19]. The limit of detection (LOD) was given by the equation, LODKs0/S, where K is a numerical factor chosen according to the con®dence level desired, s0 is the standard deviation of the blank measurements (n9) and S is the sensitivity of the calibration graph. Here a value of 3 for K was used.
Table 1 Tolerance of foreign substances and its comparison with another method Substance
BSA HSA IgG Thymine Adenine Cytosine Guanine EDTA SDS Al3, chloride Pb2, nitrate Ba2, chloride Zn2, chloride Mg2, sulphate Co2, chloride Ni2, nitrate Cu2, chloride Mn2, nitrate a b
Coexisting concentration (10ÿ7 M)
Relative error (%)
This methoda
This methoda
Tong's methodb
ÿ1.9 ÿ3.3 ÿ4.1 ÿ6.2 ÿ3.9 ÿ4.0 ÿ6.1 ÿ4.0 ÿ4.3 ÿ8.1 ÿ7.7 ÿ2.4 ÿ6.0 ÿ1.7 ÿ4.6 ÿ6.0 ÿ5.9 ÿ1.1
8.2 5.6 7.3 ÿ8.6 ÿ6.7 ÿ4.1 ÿ2.2 ± ± ± ÿ5.9 ± ÿ5.2 ÿ5.8 ÿ9.6 ÿ9.7 ÿ7.0 ÿ7.7
0.29 0.29 60.0 63.6 73.5 77.5 53.0 200.0 5.0 1.0 0.5 1.0 2.0 5.0 0.2 0.5 20.0 40.0
Concentration of CT DNA is 0.4 mg/ml. Ref. [13], concentration of CT DNA is 2.0 mg/ml.
Tong's methodb 0.04 0.04 0.04 50.0 50.0 500.0 5.0 ± ± ± 5.0 ± 10.0 100.0 8.0 3.0 10.0 25.0
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
181
Fig. 3. Calibration curves for CT DNA (&), SM DNA (*) and yeast RNA (~). The conditions are described in Section 2.3.
3.5. Comparison of the methods
3.6. Interaction of PR with DNA
Some characteristics of the proposed method and other methods for nucleic acid determinations are summarized in Table 3. Compared to classical dyes like ethidium and Hoechst 33258, the proposed method has higher sensitivity. In addition, it is more sensitive and rapid than the method that uses lanthanide cations as ¯uorescence probes. On the other hand, the cyanine dyes, such as TOTO, YOYO, etc., are more sensitive compared to PR, but they show narrow linear ranges and longer incubation times; in addition, these dyes are very expensive.
3.6.1. Absorption spectral studies In Fig. 4 the absorption spectra of PR in the presence and absence of CT DNA are shown. The absorption spectra of PR with and without CT DNA clearly show that the absorbance of PR decreased in the presence of CT DNA. The hypochromism was suggested to be due to a strong interaction between the intercalating chromophore and DNA bases [20±22]. Since the strength of this electronic interaction is expected to decrease as the third power of the distance of separation between the chromophore and the DNA bases [22], the observed hypochromism suggested a
Table 2 Analytical parameters for the determination of nucleic acids Nucleic acid
Linear range (ng/ml)
Linear regression equation (C: ng/ml)
CT DNA
0±100 100±2000 0±100 100±2000 0±100 100±1600
F0.47C0.5 F0.08C45.2 F0.37C0.8 F0.07C34.0 F0.17C0.2 F0.03C19.1
SM DNA Yeast RNA
LOD (ng/ml) 5.0 6.0 13.0
r 0.9993 0.9980 0.9990 0.9953 0.9994 0.9998
182
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
Table 3 Comparison of methods for determination of DNA Methods 3
Ethidium bromide Hoechst 332585 Tb3-phenanthroline10 Eu3-tetracycline12 TOTO15 YOYO15 This method
Incubation time (min)
LOD (ng/ml)
Linear range (mg/ml)
RSD (%)
± ± 20±30 ± 20±30 20±30 Noa
10 10 100 10 0.5 0.5 5
± 0±15 0.4±15 0.02±1.0 0.0005±0.1 0.0005±0.1 0±2.0
± ± 3.0 3.0 ± ± 1.0
a The maximal fluorescence quenching was immediately reached when the solutions were mixed; the additional incubation time was not needed.
close proximity of the chromophore to the DNA bases. In addition to the decrease in the absorbance of PR, a small red shift was also observed in the spectra. These spectral changes are consistent with the intercalation of the dye into the DNA base stack [20,21]. 3.6.2. The effects of dsDNA and ssDNA on the quenching of PR fluorescence Double strand DNA (dsDNA) was converted into single strand DNA (ssDNA) with the opening of its double helix by incubation at 1008C for 10 min and immediate cooling in ice-water. The effects of dsDNA and ssDNA on the ¯uorescence quenching of PR were
performed according to the procedure. The results are summarized in Table 4. If the dye is intercalated into the helix stack of dsDNA, the quenching of the ¯uorescence of the dye by ssDNA is smaller than that by dsDNA [19], because the helix stacks of dsDNA were destroyed along with the denaturation. However, if the dye binds to DNA in a non-intercalative mode (such as groove binding), the denaturation of DNA would have little effect on the ¯uorescence quenching of the dye. From Table 4, it can be seen that the quenching of the ¯uorescence of PR by dsDNA is much higher than that by ssDNA. This phenomenon further con®rmed that PR is intercalated into the helix stack of dsDNA. 3.7. Determination of nucleic acids in synthetic samples As described above, the sensitivity of the determination of CT DNA is higher than that of RNA, namely, CT DNA has a much greater ability to quench the ¯uorescence of PR. So, it is expected that CT DNA Table 4 The effects of dsDNA and ssDNA on the quenching of PR fluorescence Concentration of DNA (mg/ml)
Types of DNA
F0/F
0.1
dsDNA ssDNA dsDNA ssDNA dsDNA ssDNA dsDNA ssDNA
1.13 1.02 1.17 1.05 1.30 1.13 1.42 1.21
0.2 0.4 Fig. 4. Absorption spectra of free PR (solid curve) and in the presence of CT DNA (dashed curve). PR: 1.010ÿ4 mol/l; CT DNA: 30 mg/ml.
1.0
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184 Table 5 Determination of CT DNA in synthetic samples
Table 7 Determination of DNA and RNA in real samples
Sample No.
Composition of samples (ng/ml)
Recovery of DNAa (%)
RSD (%)
1 2 3 4 5 6
DNA (400) DNA (800) DNA(800)RNA(40) DNA(800)RNA(80) DNA(800)RNA(160) DNA(800)RNA(400)
102 101 103 104 105 127
1.00 1.23 1.75 2.45 2.39 1.95
a
could be measured in the presence of RNA. The determinations of CT DNA in synthetic samples, which contained various concentrations of RNA, were carried out according to the experimental procedure. The results are shown in Table 5. It can be observed that CT DNA could be determined in the presence of up to 20% RNA (w/w) with satisfactory results. These results show that the proposed method has a certain selectivity of DNA over RNA, and this method may be applied to determine DNA in normal samples which contain low RNA concentration, i.e. RNA=DNA < 20% (w/w). However, if RNA is present in high levels in the samples, it would interfere with the determination of DNA seriously. The above procedure was also applied to the determination of RNA in four synthetic samples (Table 6). Table 6 shows that the presence of CT DNA seriously interferes with the determination of RNA even when the concentration ratio of CT DNA to RNA in the samples is 5%. 3.8. Application to DNA and RNA determination in real samples The DNA and RNA levels in pollen and honey samples were determined by the proposed method. Table 6 Determination of yeast RNA in synthetic samples Composition of samples (ng/ml)
1 2 3 4
RNA (400) RNA (800) RNA(800)DNA(40) RNA(800)DNA(80)
a
Average of six determinations.
Sample No.
Pollen 1 Pollen 2 Honey
DNA levels (mg/g)a
RNA levels (mg/g)a
This Spectrophotomethod metryb
This method
Spectrophotometryc
0.40 0.47 0.07
8.18 11.41 1.97
7.80 11.80 2.04
0.37 0.48 0.07
a
Average of six determinations. Ref. [23], diphenylamine used as a chromogenic reagent, DNA was determined at 595 nm. c Ref. [18], RNA was directly determined at 260 nm. b
Average of six determinations.
Sample No.
183
Recovery of RNAa (%) 96 102.5 129.2 168.8
Since the large quantities of coexisting substances in real samples, such as proteins, ions, etc., may interfere with determination of nucleic acids, in addition to which DNA interferes with the determination of RNA too, it is necessary to separate DNA and RNA from the interfering substances and prepare DNA and RNA extraction solutions, respectively. The analytical results obtained by the proposed method are summarized in Table 7. In order to prove the possibility of using this method for analysis of real samples, these real samples were also analyzed by two conventional spectrophotometric methods [18,23] (Table 7). It can be seen that the values obtained by the different methods are in good agreement. Acknowledgements The ®nancial support of the National Natural Science Foundation of China (no. 29775021), the Natural Science Foundation of Fujian Province (no. B9810004) and the foundation of The Key Laboratory of Analytical Sciences of MOE are gratefully acknowledged.
References RSD (%) 1.5 2.5 2.6 2.7
[1] [2] [3] [4] [5] [6] [7]
S. Udenfriend, P. Zaltzman, Anal. Biochem. 3 (1962) 49. H.C. Borresen, Acta Chem. Scand. 17 (1963) 921. J.-B. Le Pecq, C. Paoletti, Anal. Biochem. 17 (1966) 100. B.T. Hill, Anal. Biochem. 70 (1976) 635. M.V. Lankeker, L.C. Gheyssens, Anal. Lett. 19 (1986) 615. J. Kapuscinski, B. Skoczylas, Anal. Biochem. 83 (1977) 252. Q.Z. Zhu, F. Li, X.Q. Guo, J.G. Xu, W.Y. Li, Y.B. Zhao, Analyst 122 (1997) 937.
184
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
[8] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Anal. Lett. 30 (1997) 527. [9] M.D. Topal, J.R. Fresco, Biochem. 19 (1980) 5531. [10] Y.X. Ci, Y.Z. Li, W.B. Chang, Anal. Chim. Acta 248 (1991) 589. [11] Y.B. Zhao, D.Y. Wang, X.Q. Guo, J.G. Xu, Anal. Chim. Acta 353 (1997) 329. [12] Y.X. Ci, Y.Z. Li, X.J. Liu, Anal. Chem. 67 (1995) 1785. [13] C.Z. Huang, K.A. Li, S.Y. Tong, Anal. Lett. 29 (1996) 1705. [14] H.S. Rye, S. Yue, D.E. Wemmer, M.A. Quesada, R.P. Haugland, R.A. Mathies, A.N. Glazer, Nucleic Acids Res. 20 (1992) 2803. [15] H.S. Rye, J.M. Dabora, M.A. Quesada, R.A. Mathies, A.N. Glazer, Anal. Biochem. 208 (1993) 144. [16] W.Y. Li, X.Q. Guo, J.G. Xu, Q.Z. Zhu, Y.B. Zhao, Anal. Chim. Acta 340 (1997) 291.
[17] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Anal. Lett. 30 (1997) 245. [18] Z.L. Zhang, Zhiwu Shenglixue Shiyan Zhidao (Chinese), Gaodengjiaoyu Chubanshe, 1990, pp. 186±189. [19] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Spectrochim. Acta 53 (1997) 781. [20] E.C. Long, J.K. Barton, Acc. Chem. Res. 23 (1990) 271. [21] H.M. Berman, P.R. Young, Annu. Rev. Biophys. Bioeng. 10 (1981) 87. [22] C. Cantor, P.R. Schimmel, Biophysical Chemistry, vol. 2, Freeman, San Francisco, 1980, p. 398. [23] C.Q. Wang, Y.L. Li, D.C. Li, J.Q. Chen, X.W. Zhou, F.Y. Hao, Z.R. Liao, H.S. Yuan, Gaoji Shengwu Huaxue Shiyan Jiaocheng (Chinese), Beijing Daxue Chubanshe, 1994, pp. 69±71.
Determination of nucleic acids using phosphin 3R as a ¯uorescence probe Qing-Zhi Zhua,*, Huang-Hao Yanga, Dong-Hui Lib, Jin-Gou Xua a
The Key Laboratory of Analytical Sciences of MOE, Department of Chemistry, Xiamen University, Xiamen 361005, People's Republic of China b Cancer Research Center, Xiamen University, Xiamen 361005, China Received 13 November 1998; received in revised form 24 March 1999; accepted 27 March 1999
Abstract A novel ¯uorimetric method has been developed for rapid determination of DNA and RNA with phosphin 3R (PR) as a ¯uorescence probe, based on the ¯uorescence quenching of PR in the presence of DNA or RNA. Maximum ¯uorescence quenching is observed in the pH range 7.0±8.4, with maximum excitation and emission wavelength at 468 and 505 nm, respectively. Under optimal conditions, the calibration graphs are linear up to 2.0 mg/ml for both calf thymus DNA (CT DNA) and salmon DNA (SM DNA), and up to 1.6 mg/ml for yeast RNA, respectively. The corresponding detection limits are 5.0 ng/ ml for CT DNA, 6.0 ng/ml for SM DNA and 13.0 ng/ml for yeast RNA. CT DNA could be determined in the presence of 20% (w/w) yeast RNA, and the relative standard deviation of six replicate measurements is 1.00% for a solution containing 400 ng/ ml of CT DNA. Three real samples were determined with satisfactory results. The interaction mechanism for the binding of PR to DNA is also studied; the results of absorption spectra and thermal denaturation experiments suggested the interaction between PR and DNA to be intercalative in nature. # 1999 Elsevier Science B.V. All rights reserved. Keywords: Fluorescence quenching; Phosphin 3R; DNA; RNA
1. Introduction The quantitative analysis of nucleic acids is very important since it is often used as a reference for measurements of other components in biological samples. Generally, direct use of the natural ¯uorescence emission properties of nucleic acids for their structure and dynamic studies, and ¯uorimetric determinations has been limited [1,2] due to the low ¯uorescence quantum yield of native DNA. Therefore, some extrin*Corresponding author. Tel.: +86-592-2182442; fax: +86-5922188054; e-mail: [email protected]
sic ¯uorescence probes have been employed to study DNA. A number of ¯uorimetric methods for the determination of nucleic acids have been established [3±8] based on the reaction in which the double helix strands of DNA molecules are intercalated by or interact with probes such as ethidium bromide [3], mithramycin [4], bisimidazole (Hoechst 33258) [5], and 40 ,6-diamidino-2-phenylindole [6]. Furthermore, the trivalent lanthanide cations such as Tb(III) [9±11], Eu(III) [12], La(III) [13] and some dimeric asymmetric cyanine dyes [14,15] have been developed as ¯uorescence probes of the structure and function of nucleic acids in recent years. However, the main
0003-2670/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 3 - 2 6 7 0 ( 9 9 ) 0 0 3 0 9 - 8
178
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
trum was performed on a Shimadzu UV-240 ultraviolet±visible spectrophotometer. All the pH measurements were made with a Model PHS-301 pH meter 631 (Xiamen, China). 2.2. Reagents
Fig. 1. The structure of phosphin 3R.
limitation of these methods is that the lanthanide metals (such as Tb3 and Eu3) and cyanine dyes (such as TOTO, YOYO etc.) are very expensive. More recently, we have studied the interaction of some inexpensive photoactive ¯uorescence probes with nucleic acids and developed the sensitive ¯uorimetric methods for determination of nucleic acids based on the in-situ photochemical ¯uorescence reaction [16,17]. However, the photo-induced ¯uorescence reaction needed a long time to reach equilibrium. Therefore, it is considered important to develop another sensitive and rapid method for determination of DNA using an inexpensive probe. Phosphin 3R (PR) is a very inexpensive dye (Fig. 1), which displays strong ¯uorescence emission at 505 nm with maximum excitation at 468 nm. In neutral medium, its ¯uorescence is signi®cantly quenched in the presence of nucleic acids. We have therefore employed PR as a ¯uorescence probe and developed a sensitive ¯uorimetric method for the determination of nucleic acids. The use of PR as a ¯uorescence probe leads to a particularly inexpensive, simple and sensitive system, permitting a limit of detection of 5.0 ng/ml for CT DNA, 6.0 ng/ml for SM DNA and 13.0 ng/ml for yeast RNA, respectively. The mechanism for the binding of PR to DNA is also studied. 2. Experimental 2.1. Apparatus A Hitachi 650-10S spectro¯uorimeter equipped with a plotter unit and a 1 cm quartz cell was used for ¯uorescence measurements. The absorption spec-
All chemicals were of analytical reagent grade or higher purity available. All aqueous solutions were made up in distilled, deionized water. Commercially prepared calf thymus DNA, salmon DNA and yeast RNA, obtained from Sino-American Biotechnology (Shanghai, China), were directly dissolved in water at a ®nal concentration of 200 mg/ml and stored at 48C. These solutions were diluted to 1.0 mg/ml with water to be used as working solutions. A PR stock solution (1.010ÿ3 mol/l) was prepared by dissolving the appropriate weight of PR (Beijing Chemical Reagent, Beijing, China) into 100 ml of water and stored in the dark. This solution was diluted to 1.010ÿ4 mol/l with water as working solution. A pH 8.0 Tris±HCl buffer solution was prepared by mixing 50 ml of 0.1 mol/l Tris and 29.2 ml of 0.1 mol/l HCl, then diluted to 100 ml with water. 2.3. Procedure Transfer 2.0 ml of buffer solution (pH 8.0) and 0.5 ml of PR solution (1.010ÿ4 mol/l) to a 10 ml standard ¯ask. Add a known volume of DNA (or RNA) standard solution. Dilute to the mark with water and mix. Measure the relative ¯uorescence intensities of the reagent blank (F0) and the mixed solution (F) at 505 nm with excitation at 468 nm. Plot a calibration curve of the ¯uorescence quenching (F0ÿF) vs. the concentration of nucleic acids. 2.4. Treatment of real samples The treatment of real samples was similar to the method reported previously [18]. In order to obtain DNA and RNA from the samples completely, a modi®ed procedure was used. 0.1 g of dry pollen (or 0.3 g of honey) was mixed with 1 ml of 5% perchloric acid solution in a mortar box. The mixture was ground to a homogeneous mass at 08C, and centrifuged at 08C for 10 min (3000 r.p.m.). The precipitate was then collected and washed with 1 ml of 5% perchloric acid at
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
08C for 10 min, followed by centrifugation. This procedure was repeated three times in order to remove acid-soluble proteins. 5 ml of ethyl alcohol±ethyl ether±chloroform mixing solvent (2:2:1, v/v) was added to the precipitate, the mixture was allowed to stand for 15 min at room temperature to extract phosphatide, then centrifuged (3000 r.p.m.) to remove the extraction solution. The procedure was repeated four times, and the ®nal precipitate of nucleic acids was dried in vacuum drier. 3.0 ml of 1.0 mol/l NaOH solution was added to the precipitate to hydrolyze nucleic acids at 258C for 8 h. The mixture was centrifuged and the centrifugate was collected, then the relict was washed with 1.0 ml of 1.0 mol/l NaOH twice, followed by centrifugation to remove the residue. All the centrifugates were mixed and acidi®ed by adding 1.0 ml of 6.0 mol/l HCl solution, and then left at 48C overnight to precipitate DNA from the solution. After DNA was separated by centrifugation, the centrifugate, RNA moiety, was quantitatively transferred into a 25 ml standard ¯ask and diluted to mark with water. Then, 5.0 ml of 1.0 mol/l perchloric acid solution was added to the DNA precipitate to hydrolyze DNA at 758C for 20 min. The hydrolysate was centrifuged and the centrifugate (DNA moiety) was transferred into a 25 ml standard ¯ask and diluted to mark with water. The DNA and RNA levels in the real samples were determined according to the procedure described in Section 2.3. 3. Results and discussion 3.1. Spectral characteristics of fluorescence The uncorrected excitation and emission spectra of PR are shown in Fig. 2. In neutral medium, PR has an emission band located at 505 nm with an excitation peak at 468 nm. When DNA (or RNA) is added, the excitation and emission maxima of the PR±DNA system are similar to that of free PR, but the ¯uorescence intensity is signi®cantly quenched. On the other hand, RNA can similarly quench the ¯uorescence of PR; however, its quenching ability is lower than that of DNA. These results indicated that PR can be used as a new ¯uorescence probe for sensitive determination of DNA. In this paper, the maximum excitation peak at
179
Fig. 2. Excitation (a) and emission (b) spectra of free PR (solid curves) and in the presence of yeast RNA (dot-dashed curves) and CT DNA (dashed curves). PR: 5.010ÿ6 mol/l; CT DNA: 400 ng/ ml; RNA: 400 ng/ml.
468 nm and the emission peak at 505 nm were used for ¯uorescence intensity measurements. 3.2. Optimization of the general procedure The experimental results indicated that maximum and constant ¯uorescence quenching was produced when the PR concentration was in the range of 4.010ÿ6±6.010ÿ6 mol/l. In this work, a PR concentration of 5.010ÿ6 mol/l was recommended. The effect of pH on the ¯uorescence quenching of the system was studied. The ¯uorescence quenching reached a maximum in the pH range of 7.0±8.4. Therefore, a pH of 8.0 was recommended. This value was obtained by addition of 2.0 ml buffer solution per 10 ml of the ®nal solution. The in¯uence of incubation time on ¯uorescence quenching was also investigated. The results showed that the maximal ¯uorescence quenching was immediately reached when the solutions were mixed and remained constant for at least 4 h. In this work, the ¯uorescence intensity was directly measured after the solutions were mixed,
180
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
and an additional incubation time was not needed. For RNA, the optimum conditions were similar to that of the DNA system. 3.3. Interference of foreign substances The effects of various foreign substances on the determination of 400 ng/ml CT DNA by the described procedure were studied and compared with another method, which uses 8-hydroxyquinoline/lanthanum complex as a probe [13]. The results are shown in Table 1. It can be seen that the coexisting purine or pyrimidine bases, even when present in large excess, do not interfere with the determination of DNA in the proposed method. Furthermore, compared with 8hydroxyquinoline/lanthanum probe, proteins only give less interference in our method. However, some foreign ions, such as Pb2, Co2, Ni2 seem to show low tolerance levels. 3.4. Calibration graphs The calibration graphs for the determination of DNA (or RNA) were constructed under the optimal
conditions. The results, shown in Fig. 3, exhibit good linear relationships between the extent of ¯uorescence quenching and the concentrations of DNA (or RNA). All the analytical parameters are presented in Table 2. From Table 2 it can be seen that the sensitivity for the determination of CT DNA is higher than that for SM DNA and yeast RNA with the sequence of CT DNA > SM DNA > yeast RNA. An interesting phenomenon is that there are two linear ranges in the calibration curves for both DNA and RNA; one linear range is between 0 and 100 ng/ml DNA (or RNA), the other is over 100±2000 ng/ml DNA (100± 1600 ng/ml RNA). Furthermore, there are good linear relationships between the ¯uorescence quenching and the concentrations of DNA (or RNA) in both ranges, and the sensitivities at low concentration are higher than those at high concentration. This phenomenon was also observed for other methods [12,19]. The limit of detection (LOD) was given by the equation, LODKs0/S, where K is a numerical factor chosen according to the con®dence level desired, s0 is the standard deviation of the blank measurements (n9) and S is the sensitivity of the calibration graph. Here a value of 3 for K was used.
Table 1 Tolerance of foreign substances and its comparison with another method Substance
BSA HSA IgG Thymine Adenine Cytosine Guanine EDTA SDS Al3, chloride Pb2, nitrate Ba2, chloride Zn2, chloride Mg2, sulphate Co2, chloride Ni2, nitrate Cu2, chloride Mn2, nitrate a b
Coexisting concentration (10ÿ7 M)
Relative error (%)
This methoda
This methoda
Tong's methodb
ÿ1.9 ÿ3.3 ÿ4.1 ÿ6.2 ÿ3.9 ÿ4.0 ÿ6.1 ÿ4.0 ÿ4.3 ÿ8.1 ÿ7.7 ÿ2.4 ÿ6.0 ÿ1.7 ÿ4.6 ÿ6.0 ÿ5.9 ÿ1.1
8.2 5.6 7.3 ÿ8.6 ÿ6.7 ÿ4.1 ÿ2.2 ± ± ± ÿ5.9 ± ÿ5.2 ÿ5.8 ÿ9.6 ÿ9.7 ÿ7.0 ÿ7.7
0.29 0.29 60.0 63.6 73.5 77.5 53.0 200.0 5.0 1.0 0.5 1.0 2.0 5.0 0.2 0.5 20.0 40.0
Concentration of CT DNA is 0.4 mg/ml. Ref. [13], concentration of CT DNA is 2.0 mg/ml.
Tong's methodb 0.04 0.04 0.04 50.0 50.0 500.0 5.0 ± ± ± 5.0 ± 10.0 100.0 8.0 3.0 10.0 25.0
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
181
Fig. 3. Calibration curves for CT DNA (&), SM DNA (*) and yeast RNA (~). The conditions are described in Section 2.3.
3.5. Comparison of the methods
3.6. Interaction of PR with DNA
Some characteristics of the proposed method and other methods for nucleic acid determinations are summarized in Table 3. Compared to classical dyes like ethidium and Hoechst 33258, the proposed method has higher sensitivity. In addition, it is more sensitive and rapid than the method that uses lanthanide cations as ¯uorescence probes. On the other hand, the cyanine dyes, such as TOTO, YOYO, etc., are more sensitive compared to PR, but they show narrow linear ranges and longer incubation times; in addition, these dyes are very expensive.
3.6.1. Absorption spectral studies In Fig. 4 the absorption spectra of PR in the presence and absence of CT DNA are shown. The absorption spectra of PR with and without CT DNA clearly show that the absorbance of PR decreased in the presence of CT DNA. The hypochromism was suggested to be due to a strong interaction between the intercalating chromophore and DNA bases [20±22]. Since the strength of this electronic interaction is expected to decrease as the third power of the distance of separation between the chromophore and the DNA bases [22], the observed hypochromism suggested a
Table 2 Analytical parameters for the determination of nucleic acids Nucleic acid
Linear range (ng/ml)
Linear regression equation (C: ng/ml)
CT DNA
0±100 100±2000 0±100 100±2000 0±100 100±1600
F0.47C0.5 F0.08C45.2 F0.37C0.8 F0.07C34.0 F0.17C0.2 F0.03C19.1
SM DNA Yeast RNA
LOD (ng/ml) 5.0 6.0 13.0
r 0.9993 0.9980 0.9990 0.9953 0.9994 0.9998
182
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
Table 3 Comparison of methods for determination of DNA Methods 3
Ethidium bromide Hoechst 332585 Tb3-phenanthroline10 Eu3-tetracycline12 TOTO15 YOYO15 This method
Incubation time (min)
LOD (ng/ml)
Linear range (mg/ml)
RSD (%)
± ± 20±30 ± 20±30 20±30 Noa
10 10 100 10 0.5 0.5 5
± 0±15 0.4±15 0.02±1.0 0.0005±0.1 0.0005±0.1 0±2.0
± ± 3.0 3.0 ± ± 1.0
a The maximal fluorescence quenching was immediately reached when the solutions were mixed; the additional incubation time was not needed.
close proximity of the chromophore to the DNA bases. In addition to the decrease in the absorbance of PR, a small red shift was also observed in the spectra. These spectral changes are consistent with the intercalation of the dye into the DNA base stack [20,21]. 3.6.2. The effects of dsDNA and ssDNA on the quenching of PR fluorescence Double strand DNA (dsDNA) was converted into single strand DNA (ssDNA) with the opening of its double helix by incubation at 1008C for 10 min and immediate cooling in ice-water. The effects of dsDNA and ssDNA on the ¯uorescence quenching of PR were
performed according to the procedure. The results are summarized in Table 4. If the dye is intercalated into the helix stack of dsDNA, the quenching of the ¯uorescence of the dye by ssDNA is smaller than that by dsDNA [19], because the helix stacks of dsDNA were destroyed along with the denaturation. However, if the dye binds to DNA in a non-intercalative mode (such as groove binding), the denaturation of DNA would have little effect on the ¯uorescence quenching of the dye. From Table 4, it can be seen that the quenching of the ¯uorescence of PR by dsDNA is much higher than that by ssDNA. This phenomenon further con®rmed that PR is intercalated into the helix stack of dsDNA. 3.7. Determination of nucleic acids in synthetic samples As described above, the sensitivity of the determination of CT DNA is higher than that of RNA, namely, CT DNA has a much greater ability to quench the ¯uorescence of PR. So, it is expected that CT DNA Table 4 The effects of dsDNA and ssDNA on the quenching of PR fluorescence Concentration of DNA (mg/ml)
Types of DNA
F0/F
0.1
dsDNA ssDNA dsDNA ssDNA dsDNA ssDNA dsDNA ssDNA
1.13 1.02 1.17 1.05 1.30 1.13 1.42 1.21
0.2 0.4 Fig. 4. Absorption spectra of free PR (solid curve) and in the presence of CT DNA (dashed curve). PR: 1.010ÿ4 mol/l; CT DNA: 30 mg/ml.
1.0
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184 Table 5 Determination of CT DNA in synthetic samples
Table 7 Determination of DNA and RNA in real samples
Sample No.
Composition of samples (ng/ml)
Recovery of DNAa (%)
RSD (%)
1 2 3 4 5 6
DNA (400) DNA (800) DNA(800)RNA(40) DNA(800)RNA(80) DNA(800)RNA(160) DNA(800)RNA(400)
102 101 103 104 105 127
1.00 1.23 1.75 2.45 2.39 1.95
a
could be measured in the presence of RNA. The determinations of CT DNA in synthetic samples, which contained various concentrations of RNA, were carried out according to the experimental procedure. The results are shown in Table 5. It can be observed that CT DNA could be determined in the presence of up to 20% RNA (w/w) with satisfactory results. These results show that the proposed method has a certain selectivity of DNA over RNA, and this method may be applied to determine DNA in normal samples which contain low RNA concentration, i.e. RNA=DNA < 20% (w/w). However, if RNA is present in high levels in the samples, it would interfere with the determination of DNA seriously. The above procedure was also applied to the determination of RNA in four synthetic samples (Table 6). Table 6 shows that the presence of CT DNA seriously interferes with the determination of RNA even when the concentration ratio of CT DNA to RNA in the samples is 5%. 3.8. Application to DNA and RNA determination in real samples The DNA and RNA levels in pollen and honey samples were determined by the proposed method. Table 6 Determination of yeast RNA in synthetic samples Composition of samples (ng/ml)
1 2 3 4
RNA (400) RNA (800) RNA(800)DNA(40) RNA(800)DNA(80)
a
Average of six determinations.
Sample No.
Pollen 1 Pollen 2 Honey
DNA levels (mg/g)a
RNA levels (mg/g)a
This Spectrophotomethod metryb
This method
Spectrophotometryc
0.40 0.47 0.07
8.18 11.41 1.97
7.80 11.80 2.04
0.37 0.48 0.07
a
Average of six determinations. Ref. [23], diphenylamine used as a chromogenic reagent, DNA was determined at 595 nm. c Ref. [18], RNA was directly determined at 260 nm. b
Average of six determinations.
Sample No.
183
Recovery of RNAa (%) 96 102.5 129.2 168.8
Since the large quantities of coexisting substances in real samples, such as proteins, ions, etc., may interfere with determination of nucleic acids, in addition to which DNA interferes with the determination of RNA too, it is necessary to separate DNA and RNA from the interfering substances and prepare DNA and RNA extraction solutions, respectively. The analytical results obtained by the proposed method are summarized in Table 7. In order to prove the possibility of using this method for analysis of real samples, these real samples were also analyzed by two conventional spectrophotometric methods [18,23] (Table 7). It can be seen that the values obtained by the different methods are in good agreement. Acknowledgements The ®nancial support of the National Natural Science Foundation of China (no. 29775021), the Natural Science Foundation of Fujian Province (no. B9810004) and the foundation of The Key Laboratory of Analytical Sciences of MOE are gratefully acknowledged.
References RSD (%) 1.5 2.5 2.6 2.7
[1] [2] [3] [4] [5] [6] [7]
S. Udenfriend, P. Zaltzman, Anal. Biochem. 3 (1962) 49. H.C. Borresen, Acta Chem. Scand. 17 (1963) 921. J.-B. Le Pecq, C. Paoletti, Anal. Biochem. 17 (1966) 100. B.T. Hill, Anal. Biochem. 70 (1976) 635. M.V. Lankeker, L.C. Gheyssens, Anal. Lett. 19 (1986) 615. J. Kapuscinski, B. Skoczylas, Anal. Biochem. 83 (1977) 252. Q.Z. Zhu, F. Li, X.Q. Guo, J.G. Xu, W.Y. Li, Y.B. Zhao, Analyst 122 (1997) 937.
184
Q.-Z. Zhu et al. / Analytica Chimica Acta 394 (1999) 177±184
[8] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Anal. Lett. 30 (1997) 527. [9] M.D. Topal, J.R. Fresco, Biochem. 19 (1980) 5531. [10] Y.X. Ci, Y.Z. Li, W.B. Chang, Anal. Chim. Acta 248 (1991) 589. [11] Y.B. Zhao, D.Y. Wang, X.Q. Guo, J.G. Xu, Anal. Chim. Acta 353 (1997) 329. [12] Y.X. Ci, Y.Z. Li, X.J. Liu, Anal. Chem. 67 (1995) 1785. [13] C.Z. Huang, K.A. Li, S.Y. Tong, Anal. Lett. 29 (1996) 1705. [14] H.S. Rye, S. Yue, D.E. Wemmer, M.A. Quesada, R.P. Haugland, R.A. Mathies, A.N. Glazer, Nucleic Acids Res. 20 (1992) 2803. [15] H.S. Rye, J.M. Dabora, M.A. Quesada, R.A. Mathies, A.N. Glazer, Anal. Biochem. 208 (1993) 144. [16] W.Y. Li, X.Q. Guo, J.G. Xu, Q.Z. Zhu, Y.B. Zhao, Anal. Chim. Acta 340 (1997) 291.
[17] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Anal. Lett. 30 (1997) 245. [18] Z.L. Zhang, Zhiwu Shenglixue Shiyan Zhidao (Chinese), Gaodengjiaoyu Chubanshe, 1990, pp. 186±189. [19] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Spectrochim. Acta 53 (1997) 781. [20] E.C. Long, J.K. Barton, Acc. Chem. Res. 23 (1990) 271. [21] H.M. Berman, P.R. Young, Annu. Rev. Biophys. Bioeng. 10 (1981) 87. [22] C. Cantor, P.R. Schimmel, Biophysical Chemistry, vol. 2, Freeman, San Francisco, 1980, p. 398. [23] C.Q. Wang, Y.L. Li, D.C. Li, J.Q. Chen, X.W. Zhou, F.Y. Hao, Z.R. Liao, H.S. Yuan, Gaoji Shengwu Huaxue Shiyan Jiaocheng (Chinese), Beijing Daxue Chubanshe, 1994, pp. 69±71.