Resonance light-scattering spectroscopic determination of protein with pyrocatechol violet by maimaiyeuem58595


									                                           Analytica Chimica Acta 444 (2001) 205–210

            Resonance light-scattering spectroscopic determination
                    of protein with pyrocatechol violet
                Xin Cong a , Zhong-Xian Guo b , Xiao-Xia Wang a , Han-Xi Shen a,∗
                                     a Department of Chemistry, Nankai University, Tianjin 300071, China
                       b   College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
                      Received 18 December 2000; received in revised form 11 April 2001; accepted 15 May 2001

   At pH 1.4–2.0 and in the presence of Triton X-100 the resonance light-scattering (RLS) spectrum of pyrocatechol violet
(PV) has a maximum peak at 341 nm with two wide bands centering on 399 and 468 nm, respectively. If micro protein
coexists in the system, the RLS intensities of PV at 399 and 468 nm are significantly increased by protein due to the binding
interaction between protein and PV. Based on this, a new assay for protein is proposed. The calibration graphs for bovine and
human serum albumin are linear up to 8.0 and 9.0 mg l−1 with 3σ detection limits of 5.2 and 6.9 × 10−5 mg l−1 , respectively.
The method is characterized by high sensitivity, short reaction period, simplicity and good stability of the chemical system.
Results for human serum, urine and saliva are in agreement with those obtained by the Bradford method with relative standard
deviations of 1.5–2.3% (n = 5). © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Protein assay; Resonance light-scattering spectroscopy; Pyrocatechol violet

1. Introduction                                                           is seriously interfered by coexisting substances while
                                                                          the bromocresol green method is liable to disturbance
   The quantitative determination of protein is consid-                   by turbidity and is not sensitive enough. Therefore, a
erably essential in biochemistry and clinical medicine.                   number of assays have been continuously reported in
Currently most of the widely used protein assays are                      recent years, such as those based on spectrophotome-
spectrophotometric, such as the Lowry [1], Bradford                       tric [9–12], fluorometric [13–17], chemiluminesence
[2,3], bromophenol blue [4,5] and bromocresol green                       [18] and electrochemical [19].
[6,7]. However, they all have some limitations in                            Since its first introduction to the quantitative de-
terms of sensitivity, selectivity, stability and simplic-                 termination of biomacromolecules in 1996 [20,21],
ity. For example, the Lowry procedure is complicated                      resonance light-scattering (RLS) technique has grad-
and subject to interference by coexisting substance                       ually gained regards from analytical chemists [22,23].
such as thiol reagents, some metal ions and amino                         It is characterized by high sensitivity, convenience
acids [8]. The bromophenol blue method has low                            in performance and simplicity in apparatus (usually
sensitivity with dynamic range of 10–80 mg l−1 and                        common spectrofluorometer). By using RLS spec-
                                                                          troscopy, organic dye probes used in spectrophotomet-
 ∗ Corresponding author. Tel.: +86-22-23502458;
                                                                          ric assays for protein and nucleic acids can become
fax: +86-22-23501705.                                                     much more sensitive. For instance, the dynamic
E-mail address: (H.-X. Shen).                        range for bovine serum albumin (BSA) of the RLS

0003-2670/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 0 0 3 - 2 6 7 0 ( 0 1 ) 0 1 1 2 6 - 6
206                            X. Cong et al. / Analytica Chimica Acta 444 (2001) 205–210

method with bromophenol blue was reported to be                prepared by mixing 50 ml of 1 mol l−1 hydrochloric
0.34–18.7 mg l−1 while the spectrophotometric pro-             acid and 53.5 ml of 1.0 mol l−1 sodium acetate, dilut-
cedure can be used just for the protein concentrations         ing the solution to 250 ml with water, and adjusting to
greater than 10 mg l−1 [21]. It is also true for ethyl         pH 1.8. A 2.0% (v/v) Triton X-100 solution was ob-
violet in the determination of DNA [24]. Up-to-date            tained by dissolving 2.0 ml of Triton X-100 (Sigma)
several RLS procedures have been prescribed for                and diluting to 100 ml with water. The working solu-
protein assays using organic dye [21,25–28] or metal           tion of Coomassie brilliant blue G-250 (CBB G-250,
complex [29].                                                  Fluka) was prepared by the recommended procedure
   Pyrocatechol violet PV, a very usual dye, was used          [3]. Unless otherwise mentioned all chemicals were
in its complex form with molybdenum [30] or tin                of analytical grade or the best grade commercially
[31] for spectrophotometric determination of protein.          available. Deionized water was used throughout.
However, PV itself has little practical usefulness in
protein spectrophotometry due to poor sensitivity.             2.3. Procedure
Here, PV is found a highly sensitive RLS probe for
protein and a simple and practical method for pro-                Into a 10 ml colorimetric tube, 1.0 ml of 2.0% Tri-
tein quantification is described on the basis of the            ton X-100, 1.0 ml of Walpole buffer (pH 1.8), 0.30 ml
enhancement effect on the RLS intensity of PV in the           of 1.0 × 10−3 mol l−1 PV solution and an appropriate
presence of Triton X-100.                                      volume of sample or protein working solution were
                                                               placed in turn. The mixture was diluted to 10 ml with
                                                               water and mixed thoroughly. The RLS spectrum was
2. Experimental                                                obtained by synchronously scanning the excitation
                                                               and emission monochromators ( λ = 0.0 nm) of
2.1. Apparatus                                                 RF-540 spectrofluorometer in the wavelength region
                                                               from 300 to 500 nm. The excitation and emission slits
  RLS intensities were measured with 10 mm quartz              were set to 2.0 nm. The RLS intensity was measured
cells on a Shimadzu RF-510 spectrofluorometer                   at the maximum wavelength (399 nm) with RF-510
(Kyoto, Japan). RLS spectra were obtained with a               spectrofluorometer. The enhancement of the RLS in-
Shimadzu RF-540 spectrofluorometer. pH values were              tensity ( I) is the difference in intensities between
measured with a pH S-2 acidimeter (Shanghai Second             sample (I) and blank (I0 ).
Analytical Instrument Plant, Shanghai, China).

2.2. Reagents                                                  3. Results and discussion

   Appropriate amounts of human serum albumin                  3.1. Absorption and resonance
(HSA) and BSA (Sino-American Biotechnology,                    light-scattering spectra
Tianjin, China) and cytochrome c from yeast (Cyt c,
Shanghai Lizhu Dongfeng Biotechnology, Shanghai,                  Based on the reaction of protein, PV and metal
China) were directly dissolved in water to prepare             ions including molybdenium and tin in the presence
stock solutions and stored at 0–5◦ C. The working so-          of surfactant, Fujita et al. [30,31] described two spec-
lutions were obtained by diluting the stock solutions          trophotometric procedures for the protein determina-
with water just prior to use. A 386.4 mg mass of PV            tion. According to their discussion, protein reacts with
purchased from Tianjin First Chemicals Plant, Tianjin,         metal–PV complex to develop a ternary complex.
China, was dissolved and diluted to 100 ml with wa-            Although the interaction of PV and protein occurs at
ter to prepare a 1.0 × 10−2 mol l−1 solution. When             acidic media due to the electrostatic binding, it has
stored in a brown flask or in the dark the solution is          not been proposed for protein spectrophotometric de-
stable for several months. A PV working solution of            termination yet. The main reason can be seen from
1.0 × 10−3 mol l−1 was prepared by diluting the stock          the absorption spectra of PV in the absence and pres-
solution with water. A pH 1.8 Walpole buffer was               ence of protein as shown in Fig. 1. The absorption
                                       X. Cong et al. / Analytica Chimica Acta 444 (2001) 205–210                                      207

Fig. 1. Absorption spectra against water of (a) 1.0 × 10−4 mol l−1
PV; (b) 8.0 mg l−1 BSA and (c) 1.0 × 10−4 mol l−1 PV-8.0 mg l−1
BSA at pH 1.8 and in the presence of 0.2% Triton X-100.

                                                                       Fig. 2. Resonance light-scattering spectra of (a) PV; (b) 2.0 mg l−1
                                                                       BSA, (c) 4.0 mg l−1 BSA; (d) PV-2.0 mg l−1 BSA and (e) PV-
difference between such two situations is too small to                 BSA in the presence of 0.2% Triton X-100 and at pH 1.8. All the
explore a relatively sensitive procedure. However, we                  concentration of PV is 3.0 × 10−5 mol −1 .
can find that the case is different for RLS spectrum
and intensity.
   Fig. 2 shows that the RLS spectrum of PV has a                      isoelectric points of BSA, HSA and cytochrome c are
sharp and maximum peak at 341 nm and two wide                          4.6–4.7, 4.8 and >10, respectively [33]. The dissoci-
bands centering on 399 and 468 nm, respectively, with                  ated sulfonic group in PV can interact with positively
relatively small intensity. When protein coexists in                   charged groups in proteins, such as protonated amino
the system, the RLS spectrum of PV has a significant                    group. This makes PV and protein bind through elec-
change. The peak at 341 nm disappears, and that at                     trostatic forces. Meanwhile, the aromatic rings of PV
468 nm becomes sharp. Such a phenomenon of this                        and protein have interaction due to non-electrostatic
system is different from those in reported systems. The                effects, such as hydrophobic and Van der Waals forces.
RLS intensity of PV is greatly enhanced by the pres-                   Therefore, many PV dye molecules can position at
ence of BSA in the examined range 300–500 nm, and                      the non-specific binding sites of a protein molecule,
maximum enhancement is present around 399 nm. On                       which leads to the aggregation of PV molecules on
the other hand, when PV is absent the difference of                    the surface of protein molecule. The aggregation
RLS intensity between various BSA concentrations is                    will produce relatively large particles, and result in
very small. Therefore, the enhancement effect of pro-                  the enhancement effect of protein on RLS intensity
tein can serve as a basis for protein assay.                           of PV.
   Such an enhancement effect of protein on RLS in-
tensity of PV can be understood in view of molecular                   3.2. Optimization of reaction conditions
structure. PV is an acid dye with triphenylmethane
structure. A PV molecule has hydrophilic groups,                          In weakly acidic medium PV, a negatively charged
i.e. a sulfonic and 3-phenol hydroxyl groups, and a                    dye binds with positively charged protein, and the
planar and hydrophobic triphenylmethane ring. At                       reaction is obviously affected by acidity. As Fig. 3
pH 1.8, PV and protein are negatively and positively                   shows, the enhancement of RLS intensity of PV by
charged, respectively, since pKal of PV is <1 [32] and                 BSA reaches a maximum and remains constant in
208                                  X. Cong et al. / Analytica Chimica Acta 444 (2001) 205–210

Fig. 3. Impact of pH on enhancement on RLS intensity of              Fig. 4. Effect of PV concentration on RLS intensity enhancement
3.0 × 10−5 mol l−1 PV by 4.0 mg l−1 BSA in the presence of 0.2%      by 4.0 mg l−1 BSA in the presence of 0.2% Triton X-100.
Triton X-100.

the pH range 1.4–2.0. So pH 1.8 is recommended.                      chain groups in molecules. Among examined nonionic
Furthermore, the effect of buffer nature on I was ex-                and polymeric surfactants, Triton X-100 has the best
amined, and several kinds of buffers, such as Walpole,               enhancing effect, as shown in Table 1. Further in-
potassium chloride–HCl and glycine–HCl, gave the                     vestigation showed that a suitable volume of 2.0% is
almost same result. Walpole buffer of pH 1.8 was                     0.6–2.0 ml; 1.0 ml was used subsequently. A reason-
chosen, and in a final 10 ml solution, 1.0 ml of this                 able explanation of these phenomena is assumed to be
buffer was suitable.                                                 the binding process between protein and Triton X-100
   PV concentration affects the RLS intensity en-                    by Van der Waals forces and hydrophobic interactions.
hancement of the system. When PV concentration is                       As shown in Table 2, the addition order of reagents
too low, a specified amount of protein cannot interact                affects the enhancement of protein on the RLS inten-
fully with PV, in other word, limited PV molecules                   sity. The best order is final addition of protein (or PV)
cannot occupy all non-specific binding sites of protein               after the addition and mixing of Triton X-100, buffer
coexisting in the system. On the other hand, the rel-                and PV (or protein). The reason for this phenomenon
atively strong absorption of highly concentrated PV                  is not clear yet and worthy of further study.
weakens the RLS intensity of the system. As shown
in Fig. 4, the enhancement of protein on RLS inten-
sity of PV is small when PV concentration is very                    Table 1
low. It increases with the increasing of PV concentra-               Influence of surfactant on enhancement effect of 4.0 mg l−1 BSA
                                                                     on the RLS intensity of 3.0 × 10−5 mol l−1 PV
tion, and reaches a maximum and remains constant
when PV concentration is in the range 2.8 × 10−5                     Surfactant (concentration)        λmax (nm)        I
to 3.5 × 10−5 mol l−1 . Thus, a PV concentration of                  Polyvinyl alcohol (0.2% m/v)      387            31.6
3.0 × 10−5 mol l−1 was chosen.                                       Tween 80 (0.2% v/v)               399            19.4
   The experiments showed that surfactant has an ef-                 Tween 20 (0.2% v/v)               399            17.9
                                                                     Emulsifier OP (0.2% v/v)           399            23.7
fect on the system. Cationic and anionic surfactants
                                                                     Triton X-100 (0.2% v/v)           399            44.1
interfere with the determination due to charged long
                                          X. Cong et al. / Analytica Chimica Acta 444 (2001) 205–210                                    209

Table 2                                                                   3σ limits of detection for BSA and HSA are 5.2 and
Influence of the addition order of reagents on the RLS enhancement         6.9 × 10−5 mg l−5 , respectively (here σ represents
by BSA
                                                                          the standard deviation of 11 blank measurements).
Addition order of reagents                         I                      This procedure can be used for quantification of cy-
Triton X-100, buffer, PV, BSA                    44.7                     tochrome c with a linear range up to 9.0 mg l−1 and
Triton X-100, buffer, BSA, PV                    44.3                     detection limit of 5.8 × 10−5 mg l−1 . The linear re-
PV, buffer, Triton X-100, BSA                    44.1                     gression equation for cytochrome c is as follows:
PV, BSA, Triton X-100, buffer                    12.8                       I = 0.825 + 11.9 C (mg l−1 ) with a correlation
PV, Triton X-100, BSA, buffer                    34.6
Triton X-100, PV; BSA, buffer                    35.8
                                                                          coefficient of 0.9994 (n = 10). From the dynamic
buffer, PV, BSA, Triton X-100                    11.5                     ranges and detection limits for serum albumins, it is
                                                                          very clear that this method is much more sensitive
                                                                          than most of currently used and recently prescribed
                                                                          spectrophotometric [1–12] and fluorometric [16] pro-
   After the dilution of the mixture of protein and                       cedures, and at least comparably sensitive with most
reagents, the RLS intensity of the system reaches a                       of RLS methods [21,25–28].
maximum within 5 min and remains constant for at
least 3.0 h. An incubation time of 10 min was used in                     3.4. Effect of coexisting substances
the study. Therefore, the system allows ample time to
measure the RLS intensity enhancements of a great                           Under the optimum conditions various non-protein
number of samples.                                                        coexisting substances were examined for interfer-
                                                                          ence. For 4.0 mg l−1 BSA, the following metal ions
3.3. Calibration curves and sensitivity                                   and amino acids at given concentrations in mol l−1
                                                                          and mg l−1 , respectively, caused a recovery within
   The calibration graphs for BSA, HSA and cy-                            94.0–105.6%: Ca2+ , 5.0 × 10−5 ; Cd2+ , 3.0 × 10−5
tochrome c were constructed under the optimized                           Co2+ , 3.0 × 10−5 Cu2+ , 5.0 × 10−5 Fe3+ , 5.0 × 10−5
conditions. The linear range for BSA is 0–8.0 mg l−1                      Hg2+ , 1.0×10−5 Mg2+ , 5.0×10−5 Mn2+ , 5.0×10−5 ;
and the linear regression equation is as follows:                         Ni2+ , 5.0×10−5 ; Pb2+ , 1.0×10−5 ; Zn2+ , 1.0×10−5 ;
  I = −2.24 + 11.7 C (mg l−1 ), with a correlation                        l-Val, 20; l-His, 40; l-Cys, 40; l-Lys, 20; l-Arg,
coefficient of 0.9994 (n = 9). The dynamic range                           40; l-Try, 20; l-Pro, 50; l-Gln, 50; l-Ser, 40; l-Leu,
for HSA is 0–9.0 mg l−1 and the linear regression                         40; l-Asp, 40; l-Tyr, 20; Gly, 50. The presence of
equation is as follows: I = 3.72 + 10.9 C (mg l−1 )                       glucose, sodium dodecylsulfate and cetyltrimethy-
with a correlation coefficient of 0.9997 (n = 10). The                     lammonium bromide at 20, 18 and 18 mg l−1 resulted

Table 3
Determination results of human fluids samples (n = 5)
Sample                  This method                                          CBB assay                                           t-testa (t)

                        Foundb                          RSD(%)c              Foundb                     RSD(%)c
Serum 1                  73.2   ±   2.1                 2.3                    74.5   ±   1.9           2.1                      1.27
Serum 2                  68.8   ±   1.5                 1.8                    69.1   ±   1.2           1.4                      0.47
Urine 1                  48.9   ±   1.3                 2.1                    50.0   ±   1.2           2.0                      1.72
Urine 2                  53.3   ±   1.5                 2.3                    53.9   ±   1.4           2.1                      0.80
Saliva 1                 684    ±   13                  1.5                    689    ±   2.1           2.5                      0.56
Saliva 2                 779    ±   18                  1.9                    790    ±   24            2.4                      1.02
   a  t0.95,8 = 2.31, t0.90,8 = 1.86.
   b                                   ¯                                                       ¯
      The found result is expressed as x ± (tP ,n−1 S/ n), where tP ,n−1 = t0.95,4 = 2.78, and x and s are the mean and standard deviation,
respectively. The protein concentration in serum is expressed in mg ml−1 , and that in other samples is in mg l−1 .
    c Relative standard deviation.
210                             X. Cong et al. / Analytica Chimica Acta 444 (2001) 205–210

in a recovery of 97.4, 94.7 and 90.4%, respectively.            References
Glucose, amino acids and metal ions, able to complex
with PV, have relatively high tolerable limits. There-           [1] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, J.
fore, compared with common dye-binding methods,                      Biol. Chem. 193 (1951) 265.
                                                                 [2] M.M. Bradford, Anal. Biochem. 72 (1976) 248.
the present procedure is selective.
                                                                 [3] T. Zor, Z. Selinger, Anal. Biochem. 236 (1996) 302.
                                                                 [4] R. Flores, Anal. Biochem. 88 (1978) 605.
3.5. Sample determination                                        [5] K. Jung, E. Nickel, M. Pergande, Clin. Chim. Acta 187 (1990)
                                                                 [6] F.L. Rodkey, Clin. Chem. 11 (1965) 478.
   The present method was applied to quantify total
                                                                 [7] B.T. Doumas, W.A. Watson, H.G. Biggs, Clin. Chim. Acta
protein in human body fluids including serum, urine                   31 (1971) 87.
and saliva. Human serum samples, obtained from the               [8] G.L. Peterson, Anal. Biochem. 100 (1979) 201.
Nankai University Hospital, were stored at 0–5◦ C                [9] H.S. Soedjak, Anal. Biochem. 220 (1994) 142.
and diluted 1000-fold with deionized water just be-             [10] A.A. Waheed, P.D. Gupta, Anal. Biochem. 233 (1996) 249.
                                                                [11] K. Zhu, K.A. Li, S.Y. Tong, Anal. Lett. 29 (1996) 575.
fore the determination. Human saliva samples were               [12] Z.X. Guo, Y.M. Hao, X. Cong, H.X. Shen, Anal. Chim. Acta
diluted 10-fold to become sample’s working solu-                     403 (2000) 225.
tions. Fresh human urine samples were readily used              [13] J. Nakamura, S. Igarashi, Anal. Lett. 29 (1996) 981.
as working solutions for protein assay. Table 3 sum-            [14] N. Li, K.A. Li, S.Y. Tong, Anal. Biochem. 233 (1996) 151.
marizes the results, which are very close to those, ob-         [15] J. Yuan, K. Matsumoto, J. Biomed. Anal. 15 (1997) 1397.
                                                                [16] M.A. Kessler, A. Meinitzer, O.S. Wolfbeis, Anal. Biochem.
tained by the Bradford method. The t-test estimation                 248 (1997) 180.
showed there are no significant differences between              [17] D.H. Li, H.H. Yang, H. Zhen, Y. Fang, Q.Z. Zhu, J.G. Xu,
the results obtained by the proposed method and                      Anal. Chim. Acta 401 (1999) 185.
Bradford assay at the confidence probability of 0.95             [18] K. Tsukagoshi, Y. Okumura, H. Akasaka, R. Nakajima, T.
or 0.90.                                                             Hara, Anal. Sci. 12 (1996) 525.
                                                                [19] H.M. Zhang, Z.W. Zhu, N.Q. Li, Fresenius’ J. Anal. Chem.
                                                                     363 (1999) 408.
                                                                [20] C.Z. Huang, K.A. Li, S.Y. Tong, Anal. Chem. 68 (1996) 2259.
4. Conclusions                                                  [21] C.Q. Ma, K.A. Li, S.Y. Tong, Anal. Biochem. 239 (1996) 86.
                                                                [22] Y.F. Li, C.Z. Huang, X.L. Hu, Chinese J. Anal. Chem. 26
                                                                     (1998) 1508.
   The new protein assay described here is based on
                                                                [23] Y.T. Wang, K.A. Li, S.Y. Tong, Chem. J. Chinese Univ. 21
the interaction of protein and PV, a very cheap and                  (2000) 1491.
commercially available dye, and utilizes the reso-              [24] T.J. Li, H.X. Shen, Chem. J. Chinese Univ. 19 (1998) 1570.
nance light-scattering technique. The enhancement               [25] C.Z. Huang, K.A. Li, S.Y. Tong, Anal. Sci. 13 (1997) 263.
effect of Triton X-100 is also involved. The method             [26] C.Z. Huang, Y.F. Li, J.G. Mao, D.G. Tan, Analyst 123 (1998)
is highly sensitive, reproducible, relatively free from
                                                                [27] Q.F. Li, X.G. Chen, H.Y. Zhang, C.X. Xue, S.H. Liu, Z.D.
interference, with reaction rapidity and good sta-                   Hu, Analyst 125 (2000) 1483.
bility. Therefore, it is advantageous over most of              [28] Z.X. Guo, H.X. Shen, Anal. Chim. Acta 408 (2000) 177.
existing protein assays. It has been used in the detec-         [29] Z.X. Guo, H.X. Shen, Spectrochim. Acta A 55 (1999)
tion of protein in human fluids yielding satisfactory                 2919.
                                                                [30] Y. Fujita, I. Mori, S. Kitano, Chem. Pharm. Bull. 32 (1984)
                                                                [31] Y. Fujita, I. Mon, T. Matsuo, Bunseki Kagaku 44 (1995) 733.
                                                                [32] Y.E. Zhang, H.S. Zhang, Z.H. Chen, Handbook of Modern
Acknowledgements                                                     Chemical Reagents, Part IV (Chromogenic Reagents for
                                                                     Inorganic Ions), Chemical Industry Press, Beijing, PR China,
                                                                     1989, p. 340.
   The authors thank gratefully the National Natural            [33] P.X. Tang, Z.J. Tao, O.R. Qi, J.J. Xu, Handbook of Modern
Science Foundation of China for its financial aid for                 Chemical Reagents, Part 1 (Biochemical Reagents), Chemical
the project (Grant No. 29875011).                                    Industry Press, Beijing, PR Chian, 1990, pp. 262–422.

To top