- Email: [email protected]
Determination of proteins with fullerol by a resonance light scattering technique
ANALYTICAL BIOCHEMISTRY Analytical Biochemistry 334 (2004) 297–302 www.elsevier.com/locate/yabio
Determination of proteins with fullerol by a resonance light scattering technique Guang-Chao Zhao*, Ping Zhang, Xian-Wen Wei, Zhou-Sheng Yang School of Chemistry and Materials Science, Anhui Normal University, Wuhu 241000, PeopleÕs Republic of China Received 10 June 2004 Available online 24 August 2004
Abstract Fullerol has been synthesized through the reaction of fullerene C60 with NaOH in aqueous solution by means of ultrasonic agitation and characterized by infrared and 1H-nuclear magnetic resonance spectroscopy. The fullerol obtained shows good solubility and excellent stability in water. A weak resonance light scattering (RLS) spectrum of fullerol was observed in aqueous solution. However, the intensity of the RLS signal could be enhanced in the presence of proteins, including bovine serum albumin (BSA), human serum albumin (HSA), pepsin (Pep), and lysozyme (Lys). Based on the enhancement of the RLS, a sensitive method for the determination of proteins has been established. The quantitative conditions were considered with regard to the effects of the pH, the ion strength, and the concentration of the fullerol. Under the optimum conditions, the intensity of the RLS was proportional to the concentration of proteins with the limits of detection of 9.7, 10.9, 57.4, and 8.5 ng mL1 for BSA, HSA, Pep, and Lys, respectively. Almost no interference can be observed from some amino acids, nucleic acids, and most of the metal ions. The model samples and human serum samples were determined satisfactorily with the proposed method. 2004 Elsevier Inc. All rights reserved. Keywords: Fullerol; Resonance light scattering; Proteins; Determination
Quantitative analysis of micro amounts of proteins is an importance task in many fields, such as molecular biology, biotechnology, and medical diagnostics. The common methods based on spectrophotometric and spectrofluorometric assays [1–8] have been widely used for the determination of proteins. However, it is still necessary to develop new methods to meet the requirements of fast and sensitive protein assay. Recently, a novel method based on the enhancement of the resonance light scattering (RLS)1 signal has been developed for the sensitive detection of biomolecules, including proteins and nucleic acids [9–14]. With either the common *
Corresponding author. Fax: +86 553 3869303. E-mail address: [email protected] (G.-C. Zhao). 1 Abbreviations used: RSL, resonance light scattering; BSA, bovine serum albumin; HSA, human serum albumin; Pep, pepsin; Lys, lysozyme; CTMAB, cetyltrimethylammonium bromide. 0003-2697/$ - see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2004.07.007
methods or the RLS technique, organic dyes with high molecular weights were often used as probes that can interact with biomolecules. As known, it is easy for most organic dyes to aggregate together in an aqueous solution, especially forming bipolymer or tripolymer, which limits their applications. Therefore, it is significant to find new stable materials as probes for the sensitive detection of proteins. Due to its remarkable photo- and electrochemical properties [15], Buckminsterfullerene (C60) has attracted much attention since its discovery in 1985. However, the insolubility of the fullerenes in water makes their applications in an aqueous solution difficult. To solve this problem, various water-soluble C60 derivatives have been synthesized [16–18] and have shown many potential applications in aqueous solution chemistry, electrochemistry, and biochemistry. Among these derivatives, fullerol, a polyhydroxylated fullerene derivative, shows
298
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
excellent solubility and stability in aqueous solution. In experiments we observed that fullerol dissolved in aqueous solution has a weak RLS spectrum. However, the RLS intensity could be enhanced strongly in the presence of proteins, such as BSA, HSA, Pep, and Lys. Based on the enhancements of the RLS signal, a sensitive assay of proteins was established with a nanogram limit of detection.
absorption at about 1600 cm1. In addition, a broad OH peak at d = 4.6 could be observed in the 1H-NMR spectrum. These results were the same as those described in the previous reports [17,18], which showed that the product is the fullerol. The solubility of the synthesized fullerol is more than 100 mg mL1. In the experiments, the working concentration of fullerol is 250 lg mL1, which was prepared by dissolving the fullerol into the doubly distilled water.
Experimental
General procedure
Apparatus
A mixture of 0.5 mL fullerol solution (250 lg mL1), 2.0 mL KH2PO4– H3PO4 buffer solution (0.1 mol L1), and an appropriate solution of protein was transferred into a 10-mL volumetric flask. Then the mixture was diluted to 10 mL with doubly distilled water, thoroughly mixed, and allowed to stand for 10 min. The mixture was then used for RLS measurements. The RLS spectra were recorded from 250.0 to 650.0 nm with the same excitation and emission wavelengths. The RLS intensities were measured at 360.0 nm with the same slit width of 10.0 nm for the excitation and emission.
The RLS spectra and the intensities were recorded and measured with an RF-540 spectrofluorometer (Shimadzu, Japan), while the absorption spectrum was obtained by using a Model UV-265 spectrophotometer (Shimadzu). A pHs-3C digital pH meter (Shanghai, China) was used to measure the pH values of aqueous solutions. Reagents and solutions Bovine serum albumin, human serum albumin, lysozyme, and pepsin were purchased from Sigma and used without further purification. Stock solutions of proteins were prepared by directly dissolving proteins in doubly distilled water. All working solutions were 50 lg mL1, prepared by diluting stock solutions. The C60 with purity of 99.9% was purchased from Tian-an (China). All other reagents were of analytical grade. The water used throughout was doubly distilled and all experiments were carried out under room temperature.
Results and discussion RLS spectral characteristics Fig. 1 displays the RLS spectra of fullerol, BSA, and the mixture of fullerol with BSA. It can be seen that both fullerol and BSA have rather weak RLS
Preparation of fullerol According to the previous reports [16–18], the preparation of fullerol was improved by using a new phasetransfer catalyst, cetyltrimethyl ammonium bromide (CTMAB). A deep violet benzene solution of C60 (50 mg in 50 mL) was added to the aqueous NaOH (2 g in 2 mL water) containing 10 mg CTMAB and 5 drops of H2O2 (30%). After being ultrasonically agitated for 30 min at room temperature, the benzene solution turned colorless and the water phase turned reddish brown. The reddish brown aqueous layer was separated from the colorless organic layer and the product was precipitated through adding ethanol to the aqueous solution. This step was repeated a few times to ensure the complete removal of NaOH and CTMAB impurities. After being dried at 60 C, a dark brown solid product could be obtained. The IR spectrum of product showed a broad hydroxyl absorption at 3412 cm1, a CAO stretching absorption at 1070 cm1, and C‚C
Fig. 1. RLS spectra of the fullerol and BSA system in pH 3.5 phosphate buffer (1) 12.5 lg mL1 fullerol only; (2) 10.0 lg mL1 BSA only (dash); (3 to 5) 12.5 lg mL1 fullerol with different BSA; (3) 2.0 lg mL1 BSA; (4) 4.0 lg mL1 BSA; (5) 6.0 lg mL1 BSA.
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
signals in the scanning range from 250.0 to 650.0 nm when they exist separately in the buffer solution. However, dramatically enhanced RLS intensity can be observed when they are mixed. In addition, the RLS signal of the mixture increased with increasing BSA concentrations (from line 3 to line 5 of Fig. 1), while no obvious change could be observed with increasing fullerol concentration in the range of 1–40 lg mL1. BSA has a rather weak RLS signal even if its concentration is higher than 20 lg mL1, which is the same as reported in the literature [9]. Therefore, the enhancement of the RLS signal of the mixture of fullerol and BSA suggested that an interaction of fullerol with BSA had taken place. Compared with fullerol alone, both the RLS spectra of fullerol and the enhanced RLS spectra (from line 3 to 5) have similar features. All lines from 3 to 5 show three peaks at 360, 400, and 470 nm. Among these peaks, the intensity of the RLS signal located at 360.0 nm is the strongest. So, the wavelength at 360.0 nm was selected as the optimum for the determination of proteins with high sensitivity. On the other hand, other proteins including HSA, Lys, and Pep also show similar features when mixed with fullerol. Effects of pH and ionic strength In a general system, the enhancement of the RLS signal strongly depends on the pH and ionic strength of the buffer solution because they affect the construction and charge status of protein. The influence of pH on the RLS intensity of the fullerol–protein systems was investigated and the results are shown in Fig. 2.
299
As can be seen, the RLS intensity of the fullerol alone is independent of the pH value of the buffer solution and those of the fullerol–BSA and fullerol–HSA systems strongly depend on the pH of the solution. However, the intensities of the fullerol–Lys and fullerol–Pep systems seem to depend little on the pH. Repetitive experiments showed that the most suitable pH value for obtaining the maximum RLS intensity of the fullerol–BSA system was 3.5 while it was 4.5 for the fullerol–HSA system. This result is different from that of the dye–protein system [14], in which the same pH condition can be obtained for the dye–BSA and dye–HSA systems because they have almost the same isoelectric points (pI) (as shown in Table 1). The pH conditions for the other protein–fullerol systems were also tested and the results are shown in Table 1. The difference from the dye–protein system is that the most suitable pH condition of the fullerol–protein system seems to be independent of the isoelectric points of proteins. But the suitable pH should be less than the isoelectric points of proteins. Therefore, an appropriate pH value for determination of Pep in the pH range of 3.0–8.5 is hard to choose because its isoelectric point is 1.0. In our experiments, we selected pH 3.0 as for the determination of Pep. We also tested the effects of ionic strength on the RLS intensities of the fullerol–protein systems. Only a little decrease of RLS intensity was observed for 5 lg mL1 proteins when ionic strength increased from 0.01 to 0.15 mol L1. This result indicated that the ionic strength had almost no influence on the RLS when the ionic strength was less than 0.1 mol L1. It is also very different from the interaction of dye molecules with proteins, for which the RLS intensity of the system is strongly dependent on the ionic strength of the buffer solution. The effect of ionic strength on the RLS intensity is also considered strong evidence for the interaction of probes with proteins through electrostatic attraction. The effects of pH and ionic strength in this experiment suggested that the interaction mode of fullerol with proteins is different from that of dyes with protein. Effects of fullerol concentration and incubation time To obtain the higher sensitivity desired, the variations in incubation time and fullerol concentration were assessed. Fig. 3 shows the dependence of the intensity of the RLS signal on the fullerol concentration. Table 1 Optimum pH for protein assay
Fig. 2. Dependence of the RLS intensity on the pH of the system. Concentrations: fullerol, 12.5 lg mL1; proteins, 5 lg mL1. k = 360.0 nm. The blank is fullerol without proteins. The measurement was repeated in quintuplicate and values were averaged. The curves represent the average and standard deviation. Some error bars are obstructed by the symbols of the points.
Protein
pI
Optimum pH
BSA HSA Lysozyme Pepsin
4.8–4.9 4.7 11.0–11.2 1.0
3.5 4.5 5.0 3.0
300
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302 Table 2 Tolerance of nonprotein substances on the determination of BSA Substance +
Fig. 3. Effect of fullerol concentration on the RLS intensity of the system. Conditions: 5 lg mL1 BSA in 20 mmol1, pH 3.5 phosphate buffer solution. The measurement was repeated in triplicate and values were averaged. The curves represent the average and standard deviation. Some error bars are obstructed by the symbols of the points.
As can be seen, the RLS intensity of fullerol alone is independent of its concentration in the range of 1– 40 lg mL1. However, under a constant BSA concentration, the RLS intensity of the mixture of fullerol–BSA is dependent on the fullerol concentration in the same concentration range. When fullerol concentration is 12.5 lg mL1, the RLS intensity of the system is the strongest. Other proteins showed a similar relationship. So, 12.5 lg mL1 of fullerol solution is selected as the optimum concentration for the determination of proteins. The experiment also shows that the RLS intensities of the fullerol–proteins reach their maximum value after 5 min at room temperature (25 C) and only a little decrease of the RLS intensity can be observed after 10 h, which suggests that the stability of the fullerol–protein system is satisfactory for determination of protein. In addition, the RLS data are independent of the addition sequence of reagents in this experiment.
Concentration
3
1.0 · 10 M 1.0 · 103 M 1.0 · 104 M 1.0 · 104 M 1.0 · 105 M 1.0 · 105 M 1.0 · 105 M 1.0 · 105 M 1.0 · 105 M 1.0 · 106 M 1.0 · 107 M 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1
Na , Cl NHþ 4 ; Cl Ba2+, Cl Mg2+, Cl Cu2þ ; SO2 4 Mn2þ ; SO2 4 Cd2+, Cl Zn2þ ; SO2 4 Cr3þ ; SO2 4 2+ Co , Cl Hg2+, Cl L -Lysine L -Cystine DL -Tyrosine L -Leucine DL -Cysteine L -Valine Fish sperm DNA Calf thymum DNA
Change of Ia (%) 0.8 0.9 1.0 2.2 18.3 2.2 4.5 1.4 2.3 1.5 1.2 10.6 1.4 3.0 15.4 2.3 1.1 12.0 3.6
Concentrations: fullerol, 12.5 lg mL1; BSA, 5 lg mL1, pH 3.5. Average value is from three measurements. Ia is the light scattering intensity.
the RLS determination of proteins. The standard solution containing 12.5 lg mL1 fullerol and 5.0 lg mL1 BSA was premixed with the foreign substances, and then the RLS intensity of the system was recorded according to the general procedure and compared with that of the standard solution itself. The results are summarized in Table 2. As can be seen from Table 2, the common metal ions such as Ba2+, Mg2+, and Na+ can be allowed with relatively higher concentration (even to 1.0 · 104 mol L1) under the tolerance level of 3%, whereas Co2+ and Hg2+ can be allowed only at very low concentration levels (lower than 1.0 · 105 mol L1). In addition, some amino acids and nucleic acids such as L -cystine, DL -tyrosine, and calf thymus DNA can be allowed at relatively high concentrations. The result indicates that fullerol has a promising application in practical testing of proteins.
Interference of nonprotein substance
Calibration curves and analytical application
We also investigated the effects of foreign substances including metal ions, amino acids, and nucleic acids on
Under the above optimum conditions, a good linear relationship between the RLS intensity and the
Table 3 Analytical parameters for the determination of different proteins Protein
Linear range (lg mL1)
Linear regression equation (C; lg mL1)
Determination limit (3r; ng mL1)
Correlation coefficient (r)
BSA HSA Pep Lys
0.1–10.0 0.1–14.0 0.5–14.0 0.5–18.0
DI = 5.15 + 7.75c DI = 0.98 + 6.86c DI = 1.49 + 1.31c DI = 2.21 + 8.83c
9.7 10.9 57.4 8.5
0.9951 0.9979 0.9939 0.9997
Concentrations: fullerol, 12.5 lg mL1; k = 360.0 nm.
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
concentration of proteins was obtained over a wide range. All analytical parameters are summarized in Table 3. As can be seen from Table 3, different proteins showed different response characteristics. Among the proteins investigated, Lys, BSA, and HSA had stronger responses but Pep had a rather weak one. With the calibration curves, four model samples including metal ions, amino acids, and nucleic acids were determined. The results are shown in Table 4.
Table 4 Results of determinations in model samples Sample (lg mL1)
Main additivesa
Found (lg mL1)
Recovery (% n = 5)
RSD (% n = 5)
BSA 5.0
Zn2+, Ba2+, Co2+, Na+ L -Lys, DL -Cys, L -Val Fish sperm DNA, Mg2+, Hg2+ L -Cys, DL -Tyr, Mn2+, Cr3+
4.80
96.7–104.5
1.8
5.91
98.1–102.1
1.3
6.07
97.9–101.8
2.1
4.92
97.2–103.4
1.5
HSA 6.0 Lys 6.0 Pep 5.0 a
The concentration of additives is as follows: L -Lys, DL -Cys, L -Val, fish sperm DNA, L -Cys, DL -Tyr, 20 lg mL1; Ba2+, Na+, Mg2+, 1.0 · 104 M; Zn2+, Cr3+, Mn2+, 1.0 · 105 M; Co2+, 1.0 · 106 M; Hg2+, 1.0 · 107 M.
301
From Table 4, it can be seen that the proteins in these model samples can be detected with satisfactory results. The present method was also applied to quantify the total protein in human serum samples. First, the standard human serum, which was used to construct the calibration curve, was obtained by mixing 30 normal serum samples, followed by the determination of the total content of protein using the Bradford method [28]. The construction of the calibration curve and the analysis of the serum samples were then performed according to the procedures described for the determination of model samples. Human serum samples with unknown concentration, which were provided by the Second Renmin Hospital of Wuhu, were totally diluted 10,000-fold with doubly distilled water. The determination results are shown in Table 5. It is clear that the results of the determination are in good agreement with the clinical data from the hospital. The results indicate that the proposed method has the potential for sensitive and reliable determination of proteins in human serum samples. Compared with the sensitivity of other RLS methods for determination of protein, the specific advantage of this method is shown in Table 6. It can be seen that this method using fullerol as RLS sensor has high sensitivity. In addition, this method has some other advantages, such as rapid reaction, simple operation, and high stability. Therefore this method will be a valuable tool for studying the biological properties of proteins
Table 5 Determination of proteins in human serum samples This method
Protein founda (mg mL1)
Recovery (%; n = 5)
RSD (%; n = 5)
Clinical datab
Sample 1 Sample 2 Sample 3
74.1 73.2 76.2
97.5–101.9 94.3–102.6 98.7–104.1
2.2 4.7 3.8
73.5 73.7 75.3
Fullerol concentration: 12.5 lg mL1. a Average value is from five measurements. b Clinical data provided by the hospital.
Table 6 Comparison of the sensitivity for the determination of protein of the present assay with that of the reported RLS method Probe
Protein
Limits of detection (ng mL1)
Reference
Fullerol ACAP CPA-mA T(5-ST)P PSbMo Amaranth DBMCAPA Quercetin Arsenazo-DBN SLS
BSA/HSA BSA BSA BSA BSA BSA/HSA BSA BSA/HSA BSA/HSA BSA/HSA
9.7/10.9 68.0 52.7 26.4 21.0 22/24 30 59.8/44.4 67.4/44.8 12.8/21.6
This method [19] [20] [21] [22] [23] [24] [25] [26] [27]
ACAP, 4-azochromotoropic acid phenylfluorome; CPA-mA, m-acetylchlorophosphomazo; T(5-ST)P, a,b,c,d-tetrakis(5-sulfothienyl)porphine; PSbMo, PsbMo heteropoly blue; DBMCAPA, dibromonethylchlorophosphonazo; SLS, sodium lauryl sulfate.
Conclusions This investigation shows that fullerol dissolved in aqueous solution shows excellent resonance light scattering characteristics for the determination of proteins. Due to good stability, fullerol can be expected to be a new probe used in bioanalysis. Furthermore, this investigation also shows the application potentials of fullerene derivatives in biology. Acknowledgments We thank the Anhui Provincial Natural Science Foundation (No. 01044702) and Excellent Young Scholar Foundation (No. 04046065), the Education Department (No. 2004kj174) of Anhui Province, and the National Natural Science Foundation (No. 20271002, key Project No. 20490210) of P.R. China for financial support. References [1] D-A. Zhang, Experimental Handbook for Biomacromolecules, Jilin University Press, Changchun, China, 1991 p. 323. [2] K. Berggren, T.H. Steinberg, W.M. Lauber, J.A. Carroll, Anal. Biochem. 276 (1999) 129–143.
302
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
[3] C. Kemper, K. Berggren, Z.J. Diwu, W.F. Patton, Electrophoresis 22 (2001) 881–889. [4] A. Manfred, Kessler, A. Meinitzer, W. Petek, Clin. Chem. 43 (1997) 996–1002. [5] M.A. Kessler, M.R. Hubmann, B.A. Dremel, O.S. Wolfbeis, Clin. Chem. 38 (1992) 2089–2093. [6] Y-J. Wei, K-A. Li, S-Y. Tong, Talanta 44 (1997) 923–930. [7] J-X. Zhang, P.J. Halling, Anal. Biochem. 188 (1990) 9–13. [8] Y-J. Wei, K-A. Li, S-Y. Tong, Talanta 43 (1996) 1–10. [9] Y-F. Li, C-Z. Huang, X-H. Huang, M. Li, Anal. Sci. 16 (2000) 1249–1254. [10] R.F. Pasternack, C. Bustamante, P.J. Collings, P.J. Collings, J. Am. Chem. Soc. 115 (1993) 5393–5397. [11] C-Z. Huang, K-A. Li, S-Y. Tong, Anal. Chem. 69 (1997) 514– 518. [12] Y-F. Li, K-A. Li, S-Y. Tong, Talanta 55 (2001) 669–675. [13] C-X. Yang, Y-F. Li, C-Z. Huang, Anal. Sci. 19 (2003) 211– 215. [14] Y-F. Li, C-Z. Huang, M. Li, Anal. Sci. 18 (2002) 177–181. [15] U.S. Schubert, C.H. Weidl, P. Rapta, E. Harth, Chem. Lett. 28 (1999) 949–950.
[16] L.Y. Chiang, R.B. Upasani, J.W. Swirezewski, J. Am. Chem. Soc. 114 (1992) 10154–10156. [17] J. Li, A. Takeuchi, M. Ozawa, X.H. Li, J. Chem. Soc. Chem. Commun. (1993) 1784–1786. [18] T-B. Li, K-X. Huang, X-H. Li, H-Y. Jiang, J. Li, Chem. J. Chin. Univ. 19 (1998) 858–860. [19] Z.X. Guo, H.X. Shen, Anal. Chim. Acta 408 (2000) 177–182. [20] Q.F. Li, S.H. Liu, H.Y. Zhang, X.G. Chen, Z.D. Hu, Anal. Lett. 34 (2001) 1133–1138. [21] C.Z. Huang, Y.F. Li, J.G. Mao, D.G. Tan, Analyst 123 (1998) 1401. [22] Z.X. Guo, H.X. Shen, Spectrochim. Acta A 55 (1999) 2919–2925. [23] X.X. Wang, H.X. Shen, Y.M. Hao, Fenxi Huaxue 28 (2000) 1388. [24] Q.F. Li, H.Y. Zhang, C.X. Xue, X.G. Chen, Z.D. Hu, Spectrochim. Acta 56A (2000) 2465–2470. [25] P. Feng, X.L. Hu, C.Z. Huang, Anal. Lett. 32 (1999) 1323. [26] Q.F. Li, X.G. Chen, H.Y. Zhang, C.X. Xue, Z.D. Hu, Analyst 125 (2000) 1483. [27] R.T. Liu, J.H. Yang, C.X. Sun, X. Wu, L. Li, Z.M. Li, Anal. Bioanal. Chem. 377 (2003) 375–379. [28] Y.J. Wei, K.A. Li, Talanta 43 (1993) 1–5.
Determination of proteins with fullerol by a resonance light scattering technique Guang-Chao Zhao*, Ping Zhang, Xian-Wen Wei, Zhou-Sheng Yang School of Chemistry and Materials Science, Anhui Normal University, Wuhu 241000, PeopleÕs Republic of China Received 10 June 2004 Available online 24 August 2004
Abstract Fullerol has been synthesized through the reaction of fullerene C60 with NaOH in aqueous solution by means of ultrasonic agitation and characterized by infrared and 1H-nuclear magnetic resonance spectroscopy. The fullerol obtained shows good solubility and excellent stability in water. A weak resonance light scattering (RLS) spectrum of fullerol was observed in aqueous solution. However, the intensity of the RLS signal could be enhanced in the presence of proteins, including bovine serum albumin (BSA), human serum albumin (HSA), pepsin (Pep), and lysozyme (Lys). Based on the enhancement of the RLS, a sensitive method for the determination of proteins has been established. The quantitative conditions were considered with regard to the effects of the pH, the ion strength, and the concentration of the fullerol. Under the optimum conditions, the intensity of the RLS was proportional to the concentration of proteins with the limits of detection of 9.7, 10.9, 57.4, and 8.5 ng mL1 for BSA, HSA, Pep, and Lys, respectively. Almost no interference can be observed from some amino acids, nucleic acids, and most of the metal ions. The model samples and human serum samples were determined satisfactorily with the proposed method. 2004 Elsevier Inc. All rights reserved. Keywords: Fullerol; Resonance light scattering; Proteins; Determination
Quantitative analysis of micro amounts of proteins is an importance task in many fields, such as molecular biology, biotechnology, and medical diagnostics. The common methods based on spectrophotometric and spectrofluorometric assays [1–8] have been widely used for the determination of proteins. However, it is still necessary to develop new methods to meet the requirements of fast and sensitive protein assay. Recently, a novel method based on the enhancement of the resonance light scattering (RLS)1 signal has been developed for the sensitive detection of biomolecules, including proteins and nucleic acids [9–14]. With either the common *
Corresponding author. Fax: +86 553 3869303. E-mail address: [email protected] (G.-C. Zhao). 1 Abbreviations used: RSL, resonance light scattering; BSA, bovine serum albumin; HSA, human serum albumin; Pep, pepsin; Lys, lysozyme; CTMAB, cetyltrimethylammonium bromide. 0003-2697/$ - see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2004.07.007
methods or the RLS technique, organic dyes with high molecular weights were often used as probes that can interact with biomolecules. As known, it is easy for most organic dyes to aggregate together in an aqueous solution, especially forming bipolymer or tripolymer, which limits their applications. Therefore, it is significant to find new stable materials as probes for the sensitive detection of proteins. Due to its remarkable photo- and electrochemical properties [15], Buckminsterfullerene (C60) has attracted much attention since its discovery in 1985. However, the insolubility of the fullerenes in water makes their applications in an aqueous solution difficult. To solve this problem, various water-soluble C60 derivatives have been synthesized [16–18] and have shown many potential applications in aqueous solution chemistry, electrochemistry, and biochemistry. Among these derivatives, fullerol, a polyhydroxylated fullerene derivative, shows
298
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
excellent solubility and stability in aqueous solution. In experiments we observed that fullerol dissolved in aqueous solution has a weak RLS spectrum. However, the RLS intensity could be enhanced strongly in the presence of proteins, such as BSA, HSA, Pep, and Lys. Based on the enhancements of the RLS signal, a sensitive assay of proteins was established with a nanogram limit of detection.
absorption at about 1600 cm1. In addition, a broad OH peak at d = 4.6 could be observed in the 1H-NMR spectrum. These results were the same as those described in the previous reports [17,18], which showed that the product is the fullerol. The solubility of the synthesized fullerol is more than 100 mg mL1. In the experiments, the working concentration of fullerol is 250 lg mL1, which was prepared by dissolving the fullerol into the doubly distilled water.
Experimental
General procedure
Apparatus
A mixture of 0.5 mL fullerol solution (250 lg mL1), 2.0 mL KH2PO4– H3PO4 buffer solution (0.1 mol L1), and an appropriate solution of protein was transferred into a 10-mL volumetric flask. Then the mixture was diluted to 10 mL with doubly distilled water, thoroughly mixed, and allowed to stand for 10 min. The mixture was then used for RLS measurements. The RLS spectra were recorded from 250.0 to 650.0 nm with the same excitation and emission wavelengths. The RLS intensities were measured at 360.0 nm with the same slit width of 10.0 nm for the excitation and emission.
The RLS spectra and the intensities were recorded and measured with an RF-540 spectrofluorometer (Shimadzu, Japan), while the absorption spectrum was obtained by using a Model UV-265 spectrophotometer (Shimadzu). A pHs-3C digital pH meter (Shanghai, China) was used to measure the pH values of aqueous solutions. Reagents and solutions Bovine serum albumin, human serum albumin, lysozyme, and pepsin were purchased from Sigma and used without further purification. Stock solutions of proteins were prepared by directly dissolving proteins in doubly distilled water. All working solutions were 50 lg mL1, prepared by diluting stock solutions. The C60 with purity of 99.9% was purchased from Tian-an (China). All other reagents were of analytical grade. The water used throughout was doubly distilled and all experiments were carried out under room temperature.
Results and discussion RLS spectral characteristics Fig. 1 displays the RLS spectra of fullerol, BSA, and the mixture of fullerol with BSA. It can be seen that both fullerol and BSA have rather weak RLS
Preparation of fullerol According to the previous reports [16–18], the preparation of fullerol was improved by using a new phasetransfer catalyst, cetyltrimethyl ammonium bromide (CTMAB). A deep violet benzene solution of C60 (50 mg in 50 mL) was added to the aqueous NaOH (2 g in 2 mL water) containing 10 mg CTMAB and 5 drops of H2O2 (30%). After being ultrasonically agitated for 30 min at room temperature, the benzene solution turned colorless and the water phase turned reddish brown. The reddish brown aqueous layer was separated from the colorless organic layer and the product was precipitated through adding ethanol to the aqueous solution. This step was repeated a few times to ensure the complete removal of NaOH and CTMAB impurities. After being dried at 60 C, a dark brown solid product could be obtained. The IR spectrum of product showed a broad hydroxyl absorption at 3412 cm1, a CAO stretching absorption at 1070 cm1, and C‚C
Fig. 1. RLS spectra of the fullerol and BSA system in pH 3.5 phosphate buffer (1) 12.5 lg mL1 fullerol only; (2) 10.0 lg mL1 BSA only (dash); (3 to 5) 12.5 lg mL1 fullerol with different BSA; (3) 2.0 lg mL1 BSA; (4) 4.0 lg mL1 BSA; (5) 6.0 lg mL1 BSA.
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
signals in the scanning range from 250.0 to 650.0 nm when they exist separately in the buffer solution. However, dramatically enhanced RLS intensity can be observed when they are mixed. In addition, the RLS signal of the mixture increased with increasing BSA concentrations (from line 3 to line 5 of Fig. 1), while no obvious change could be observed with increasing fullerol concentration in the range of 1–40 lg mL1. BSA has a rather weak RLS signal even if its concentration is higher than 20 lg mL1, which is the same as reported in the literature [9]. Therefore, the enhancement of the RLS signal of the mixture of fullerol and BSA suggested that an interaction of fullerol with BSA had taken place. Compared with fullerol alone, both the RLS spectra of fullerol and the enhanced RLS spectra (from line 3 to 5) have similar features. All lines from 3 to 5 show three peaks at 360, 400, and 470 nm. Among these peaks, the intensity of the RLS signal located at 360.0 nm is the strongest. So, the wavelength at 360.0 nm was selected as the optimum for the determination of proteins with high sensitivity. On the other hand, other proteins including HSA, Lys, and Pep also show similar features when mixed with fullerol. Effects of pH and ionic strength In a general system, the enhancement of the RLS signal strongly depends on the pH and ionic strength of the buffer solution because they affect the construction and charge status of protein. The influence of pH on the RLS intensity of the fullerol–protein systems was investigated and the results are shown in Fig. 2.
299
As can be seen, the RLS intensity of the fullerol alone is independent of the pH value of the buffer solution and those of the fullerol–BSA and fullerol–HSA systems strongly depend on the pH of the solution. However, the intensities of the fullerol–Lys and fullerol–Pep systems seem to depend little on the pH. Repetitive experiments showed that the most suitable pH value for obtaining the maximum RLS intensity of the fullerol–BSA system was 3.5 while it was 4.5 for the fullerol–HSA system. This result is different from that of the dye–protein system [14], in which the same pH condition can be obtained for the dye–BSA and dye–HSA systems because they have almost the same isoelectric points (pI) (as shown in Table 1). The pH conditions for the other protein–fullerol systems were also tested and the results are shown in Table 1. The difference from the dye–protein system is that the most suitable pH condition of the fullerol–protein system seems to be independent of the isoelectric points of proteins. But the suitable pH should be less than the isoelectric points of proteins. Therefore, an appropriate pH value for determination of Pep in the pH range of 3.0–8.5 is hard to choose because its isoelectric point is 1.0. In our experiments, we selected pH 3.0 as for the determination of Pep. We also tested the effects of ionic strength on the RLS intensities of the fullerol–protein systems. Only a little decrease of RLS intensity was observed for 5 lg mL1 proteins when ionic strength increased from 0.01 to 0.15 mol L1. This result indicated that the ionic strength had almost no influence on the RLS when the ionic strength was less than 0.1 mol L1. It is also very different from the interaction of dye molecules with proteins, for which the RLS intensity of the system is strongly dependent on the ionic strength of the buffer solution. The effect of ionic strength on the RLS intensity is also considered strong evidence for the interaction of probes with proteins through electrostatic attraction. The effects of pH and ionic strength in this experiment suggested that the interaction mode of fullerol with proteins is different from that of dyes with protein. Effects of fullerol concentration and incubation time To obtain the higher sensitivity desired, the variations in incubation time and fullerol concentration were assessed. Fig. 3 shows the dependence of the intensity of the RLS signal on the fullerol concentration. Table 1 Optimum pH for protein assay
Fig. 2. Dependence of the RLS intensity on the pH of the system. Concentrations: fullerol, 12.5 lg mL1; proteins, 5 lg mL1. k = 360.0 nm. The blank is fullerol without proteins. The measurement was repeated in quintuplicate and values were averaged. The curves represent the average and standard deviation. Some error bars are obstructed by the symbols of the points.
Protein
pI
Optimum pH
BSA HSA Lysozyme Pepsin
4.8–4.9 4.7 11.0–11.2 1.0
3.5 4.5 5.0 3.0
300
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302 Table 2 Tolerance of nonprotein substances on the determination of BSA Substance +
Fig. 3. Effect of fullerol concentration on the RLS intensity of the system. Conditions: 5 lg mL1 BSA in 20 mmol1, pH 3.5 phosphate buffer solution. The measurement was repeated in triplicate and values were averaged. The curves represent the average and standard deviation. Some error bars are obstructed by the symbols of the points.
As can be seen, the RLS intensity of fullerol alone is independent of its concentration in the range of 1– 40 lg mL1. However, under a constant BSA concentration, the RLS intensity of the mixture of fullerol–BSA is dependent on the fullerol concentration in the same concentration range. When fullerol concentration is 12.5 lg mL1, the RLS intensity of the system is the strongest. Other proteins showed a similar relationship. So, 12.5 lg mL1 of fullerol solution is selected as the optimum concentration for the determination of proteins. The experiment also shows that the RLS intensities of the fullerol–proteins reach their maximum value after 5 min at room temperature (25 C) and only a little decrease of the RLS intensity can be observed after 10 h, which suggests that the stability of the fullerol–protein system is satisfactory for determination of protein. In addition, the RLS data are independent of the addition sequence of reagents in this experiment.
Concentration
3
1.0 · 10 M 1.0 · 103 M 1.0 · 104 M 1.0 · 104 M 1.0 · 105 M 1.0 · 105 M 1.0 · 105 M 1.0 · 105 M 1.0 · 105 M 1.0 · 106 M 1.0 · 107 M 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1 20 lg mL1
Na , Cl NHþ 4 ; Cl Ba2+, Cl Mg2+, Cl Cu2þ ; SO2 4 Mn2þ ; SO2 4 Cd2+, Cl Zn2þ ; SO2 4 Cr3þ ; SO2 4 2+ Co , Cl Hg2+, Cl L -Lysine L -Cystine DL -Tyrosine L -Leucine DL -Cysteine L -Valine Fish sperm DNA Calf thymum DNA
Change of Ia (%) 0.8 0.9 1.0 2.2 18.3 2.2 4.5 1.4 2.3 1.5 1.2 10.6 1.4 3.0 15.4 2.3 1.1 12.0 3.6
Concentrations: fullerol, 12.5 lg mL1; BSA, 5 lg mL1, pH 3.5. Average value is from three measurements. Ia is the light scattering intensity.
the RLS determination of proteins. The standard solution containing 12.5 lg mL1 fullerol and 5.0 lg mL1 BSA was premixed with the foreign substances, and then the RLS intensity of the system was recorded according to the general procedure and compared with that of the standard solution itself. The results are summarized in Table 2. As can be seen from Table 2, the common metal ions such as Ba2+, Mg2+, and Na+ can be allowed with relatively higher concentration (even to 1.0 · 104 mol L1) under the tolerance level of 3%, whereas Co2+ and Hg2+ can be allowed only at very low concentration levels (lower than 1.0 · 105 mol L1). In addition, some amino acids and nucleic acids such as L -cystine, DL -tyrosine, and calf thymus DNA can be allowed at relatively high concentrations. The result indicates that fullerol has a promising application in practical testing of proteins.
Interference of nonprotein substance
Calibration curves and analytical application
We also investigated the effects of foreign substances including metal ions, amino acids, and nucleic acids on
Under the above optimum conditions, a good linear relationship between the RLS intensity and the
Table 3 Analytical parameters for the determination of different proteins Protein
Linear range (lg mL1)
Linear regression equation (C; lg mL1)
Determination limit (3r; ng mL1)
Correlation coefficient (r)
BSA HSA Pep Lys
0.1–10.0 0.1–14.0 0.5–14.0 0.5–18.0
DI = 5.15 + 7.75c DI = 0.98 + 6.86c DI = 1.49 + 1.31c DI = 2.21 + 8.83c
9.7 10.9 57.4 8.5
0.9951 0.9979 0.9939 0.9997
Concentrations: fullerol, 12.5 lg mL1; k = 360.0 nm.
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
concentration of proteins was obtained over a wide range. All analytical parameters are summarized in Table 3. As can be seen from Table 3, different proteins showed different response characteristics. Among the proteins investigated, Lys, BSA, and HSA had stronger responses but Pep had a rather weak one. With the calibration curves, four model samples including metal ions, amino acids, and nucleic acids were determined. The results are shown in Table 4.
Table 4 Results of determinations in model samples Sample (lg mL1)
Main additivesa
Found (lg mL1)
Recovery (% n = 5)
RSD (% n = 5)
BSA 5.0
Zn2+, Ba2+, Co2+, Na+ L -Lys, DL -Cys, L -Val Fish sperm DNA, Mg2+, Hg2+ L -Cys, DL -Tyr, Mn2+, Cr3+
4.80
96.7–104.5
1.8
5.91
98.1–102.1
1.3
6.07
97.9–101.8
2.1
4.92
97.2–103.4
1.5
HSA 6.0 Lys 6.0 Pep 5.0 a
The concentration of additives is as follows: L -Lys, DL -Cys, L -Val, fish sperm DNA, L -Cys, DL -Tyr, 20 lg mL1; Ba2+, Na+, Mg2+, 1.0 · 104 M; Zn2+, Cr3+, Mn2+, 1.0 · 105 M; Co2+, 1.0 · 106 M; Hg2+, 1.0 · 107 M.
301
From Table 4, it can be seen that the proteins in these model samples can be detected with satisfactory results. The present method was also applied to quantify the total protein in human serum samples. First, the standard human serum, which was used to construct the calibration curve, was obtained by mixing 30 normal serum samples, followed by the determination of the total content of protein using the Bradford method [28]. The construction of the calibration curve and the analysis of the serum samples were then performed according to the procedures described for the determination of model samples. Human serum samples with unknown concentration, which were provided by the Second Renmin Hospital of Wuhu, were totally diluted 10,000-fold with doubly distilled water. The determination results are shown in Table 5. It is clear that the results of the determination are in good agreement with the clinical data from the hospital. The results indicate that the proposed method has the potential for sensitive and reliable determination of proteins in human serum samples. Compared with the sensitivity of other RLS methods for determination of protein, the specific advantage of this method is shown in Table 6. It can be seen that this method using fullerol as RLS sensor has high sensitivity. In addition, this method has some other advantages, such as rapid reaction, simple operation, and high stability. Therefore this method will be a valuable tool for studying the biological properties of proteins
Table 5 Determination of proteins in human serum samples This method
Protein founda (mg mL1)
Recovery (%; n = 5)
RSD (%; n = 5)
Clinical datab
Sample 1 Sample 2 Sample 3
74.1 73.2 76.2
97.5–101.9 94.3–102.6 98.7–104.1
2.2 4.7 3.8
73.5 73.7 75.3
Fullerol concentration: 12.5 lg mL1. a Average value is from five measurements. b Clinical data provided by the hospital.
Table 6 Comparison of the sensitivity for the determination of protein of the present assay with that of the reported RLS method Probe
Protein
Limits of detection (ng mL1)
Reference
Fullerol ACAP CPA-mA T(5-ST)P PSbMo Amaranth DBMCAPA Quercetin Arsenazo-DBN SLS
BSA/HSA BSA BSA BSA BSA BSA/HSA BSA BSA/HSA BSA/HSA BSA/HSA
9.7/10.9 68.0 52.7 26.4 21.0 22/24 30 59.8/44.4 67.4/44.8 12.8/21.6
This method [19] [20] [21] [22] [23] [24] [25] [26] [27]
ACAP, 4-azochromotoropic acid phenylfluorome; CPA-mA, m-acetylchlorophosphomazo; T(5-ST)P, a,b,c,d-tetrakis(5-sulfothienyl)porphine; PSbMo, PsbMo heteropoly blue; DBMCAPA, dibromonethylchlorophosphonazo; SLS, sodium lauryl sulfate.
Conclusions This investigation shows that fullerol dissolved in aqueous solution shows excellent resonance light scattering characteristics for the determination of proteins. Due to good stability, fullerol can be expected to be a new probe used in bioanalysis. Furthermore, this investigation also shows the application potentials of fullerene derivatives in biology. Acknowledgments We thank the Anhui Provincial Natural Science Foundation (No. 01044702) and Excellent Young Scholar Foundation (No. 04046065), the Education Department (No. 2004kj174) of Anhui Province, and the National Natural Science Foundation (No. 20271002, key Project No. 20490210) of P.R. China for financial support. References [1] D-A. Zhang, Experimental Handbook for Biomacromolecules, Jilin University Press, Changchun, China, 1991 p. 323. [2] K. Berggren, T.H. Steinberg, W.M. Lauber, J.A. Carroll, Anal. Biochem. 276 (1999) 129–143.
302
G.-C. Zhao et al. / Analytical Biochemistry 334 (2004) 297–302
[3] C. Kemper, K. Berggren, Z.J. Diwu, W.F. Patton, Electrophoresis 22 (2001) 881–889. [4] A. Manfred, Kessler, A. Meinitzer, W. Petek, Clin. Chem. 43 (1997) 996–1002. [5] M.A. Kessler, M.R. Hubmann, B.A. Dremel, O.S. Wolfbeis, Clin. Chem. 38 (1992) 2089–2093. [6] Y-J. Wei, K-A. Li, S-Y. Tong, Talanta 44 (1997) 923–930. [7] J-X. Zhang, P.J. Halling, Anal. Biochem. 188 (1990) 9–13. [8] Y-J. Wei, K-A. Li, S-Y. Tong, Talanta 43 (1996) 1–10. [9] Y-F. Li, C-Z. Huang, X-H. Huang, M. Li, Anal. Sci. 16 (2000) 1249–1254. [10] R.F. Pasternack, C. Bustamante, P.J. Collings, P.J. Collings, J. Am. Chem. Soc. 115 (1993) 5393–5397. [11] C-Z. Huang, K-A. Li, S-Y. Tong, Anal. Chem. 69 (1997) 514– 518. [12] Y-F. Li, K-A. Li, S-Y. Tong, Talanta 55 (2001) 669–675. [13] C-X. Yang, Y-F. Li, C-Z. Huang, Anal. Sci. 19 (2003) 211– 215. [14] Y-F. Li, C-Z. Huang, M. Li, Anal. Sci. 18 (2002) 177–181. [15] U.S. Schubert, C.H. Weidl, P. Rapta, E. Harth, Chem. Lett. 28 (1999) 949–950.
[16] L.Y. Chiang, R.B. Upasani, J.W. Swirezewski, J. Am. Chem. Soc. 114 (1992) 10154–10156. [17] J. Li, A. Takeuchi, M. Ozawa, X.H. Li, J. Chem. Soc. Chem. Commun. (1993) 1784–1786. [18] T-B. Li, K-X. Huang, X-H. Li, H-Y. Jiang, J. Li, Chem. J. Chin. Univ. 19 (1998) 858–860. [19] Z.X. Guo, H.X. Shen, Anal. Chim. Acta 408 (2000) 177–182. [20] Q.F. Li, S.H. Liu, H.Y. Zhang, X.G. Chen, Z.D. Hu, Anal. Lett. 34 (2001) 1133–1138. [21] C.Z. Huang, Y.F. Li, J.G. Mao, D.G. Tan, Analyst 123 (1998) 1401. [22] Z.X. Guo, H.X. Shen, Spectrochim. Acta A 55 (1999) 2919–2925. [23] X.X. Wang, H.X. Shen, Y.M. Hao, Fenxi Huaxue 28 (2000) 1388. [24] Q.F. Li, H.Y. Zhang, C.X. Xue, X.G. Chen, Z.D. Hu, Spectrochim. Acta 56A (2000) 2465–2470. [25] P. Feng, X.L. Hu, C.Z. Huang, Anal. Lett. 32 (1999) 1323. [26] Q.F. Li, X.G. Chen, H.Y. Zhang, C.X. Xue, Z.D. Hu, Analyst 125 (2000) 1483. [27] R.T. Liu, J.H. Yang, C.X. Sun, X. Wu, L. Li, Z.M. Li, Anal. Bioanal. Chem. 377 (2003) 375–379. [28] Y.J. Wei, K.A. Li, Talanta 43 (1993) 1–5.