- Email: [email protected]
Separation and determination of peptide hormones by capillary electrophoresis with laser-induced fluorescence coupled with transient pseudo-isotachophoresis preconcentration
Analytical Biochemistry 380 (2008) 297–302
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Separation and determination of peptide hormones by capillary electrophoresis with laser-induced fluorescence coupled with transient pseudo-isotachophoresis preconcentration Ying Chen a,b, Liangjun Xu b, Lan Zhang a, Guonan Chen a,* a b
Ministry of Education, Key Laboratory of Analysis and Detection Technology for Food Safety, and Department of Chemistry, Fuzhou University, Fuzhou, Fujian 350002, China Analytical and Testing Center, Fuzhou University, Fuzhou, Fujian 350002, China
a r t i c l e
i n f o
Article history: Received 19 April 2008 Available online 5 June 2008 Keywords: Capillary electrophoresis Laser-induced fluorescence Cholecystokinin tetrapeptide Neurotensin Neurotensin hexapeptide Neurokinin B Transient pseudo-isotachophoresis
a b s t r a c t A new method for the determination of the peptide hormones and their fragments by capillary electrophoresis (CE) with laser-induced fluorescence (LIF) detection and transient pseudo-isotachophoresis (pseudo-tITP) preconcentration was established in this study. The LIF detector used an argon ion laser with excitation wavelength at 488 nm and emission wavelength at 535 nm. Fluorescein isothiocyanate (FITC) was used as precolumn derivatization reagent to label cholecystokinin tetrapeptide (CCK-4), neurotensin (NT), neurotensin hexapeptide (NT8–13), and neurokinin B (NKB). Borate (10 mmol/L, pH 9.0) was selected as derivatization medium to get the high efficiency. When the addition of 70% (v/v) methanol and 1% (m/v) sodium chloride (NaCl) to the sample matrix, and with borate buffer (110 mM, pH 9.5) and 20% (v/v) methanol as running buffer, a preconcentration based on the pseudo-tITP afforded 100-fold improvement in peak heights compared with the traditional hydrodynamic injection (2.3% capillary volume). The detection limits (signal/noise = 3) based on peak height were found to be 0.04, 0.1, 0.2, and 0.08 nmol/L for NT8–13, NT, NKB, and CCK-4, respectively. The method was validated and applied to qualitative analysis of NT and NT8–13 in human cerebrospinal fluid sample. Ó 2008 Elsevier Inc. All rights reserved.
Neurotensin (NT)1 and its C-terminal hexapeptide fragment NT8– , neurokinin B (NKB), and cholecystokinin tetrapeptide (CCK-4) are 13 a group of peptide hormones that act as neurotransmitters and neuromodulators in the central nervous system (see Table 1). These peptide hormones are thought to be involved in various biological functions such as sensory information, antipsychotic, food intake control, and regulation of neuroendocrine processing [1–3]. These peptide hormones tend to exist at trace level in biological samples. Their abnormal secretion in the body may lead to severe pathological changes. Despite recent progress in peptide hormone research, many of the body mechanisms and the complete clinical and pharmacological aspects of these peptides remain complex. Therefore, trace amounts of these peptides in biological fluids must be determined
* Corresponding author. Fax: +86 591 83713866. E-mail address: [email protected] (G. Chen). 1 Abbreviations used: NT, neurotensin; NKB, neurokinin B; CCK-4, cholecystokinin tetrapeptide; RIA, radioimmunoassay; ELISA, enzyme-linked immunosorbent assay; HPLC, high-performance liquid chromatography; CE, capillary electrophoresis; LIF, laser-induced fluorescence; MS, mass spectrometry; ITP, isotachophoresis; CZE, capillary zone electrophoresis; pseudo-tITP, transient pseudo-isotachophoresis; CSF, cerebrospinal fluid; CCK-4HCl, cholecystokinin fragment 30–33 amide hydrochloride; FITC, fluorescein isothiocyanate; RSD, relative standard deviation. 0003-2697/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2008.06.002
accurately so that their physiological effects and therapeutic measures can be better collected and understood. Traditionally, radioimmunoassay (RIA), enzyme-linked immunosorbent assay (ELISA), and immunohistochemical techniques have been the most widely used methods for the measurement of these peptides [4,5]. However, the most obvious disadvantage of RIA is the safety concern of using radioactive species. In addition, RIA involves labor-intensive procedures aside from sample preparation and lengthy reaction times. Although ELISA does not use radioactive reagents, some intrinsic problems involved in immunoassay, such as cross-reaction and nonspecific interference of coexistent compounds with the binding of the analytes, still need to be resolved. High-performance liquid chromatography (HPLC) with immunoassay or mass detection has been used to analyze these peptide hormones [6–8]. Nowadays, compared with HPLC, capillary electrophoresis (CE) has become a more attractive separation technique in peptide analysis due to its many advantages, including high efficiency, low waste production, and fast separation [9– 11]. Two basic approaches can be distinguished to improve detection sensitivity in CE either by using more sensitive detection methods, such as laser-induced fluorescence (LIF) and electrochemical mode, or by increasing analyte mass loading [12]. Recently, CE with different detection methods, such as mass
298
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
Table 1 Three-letter codes of analyzed peptide hormone of brain and intestine Peptides
Molecular mass
Three-letter codes of compound
NKB CCK-4 NT8–13 NT
1210.4 633.2 874.0 1672.0
Asp-Met-His-Asp-Phe-Phe-Val-Gly-Leu-Met-NH2 Trp-Met-Asp-Phe-NH2 Arg-Arg-Pro-Tyr-Ile-Leu pGlu-Leu-Tyr-Glu-Asn-Lys-Pro-Arg-Arg-Pro-Tyr-Ile-Leu
spectrometry (MS) and LIF, has been applied in peptide hormone analysis [13–20]. But to the best of our knowledge, CE–LIF has not been reported for simultaneous analysis of NT, NT8–13, CCK-4, and NKB. Sample stacking is an inherent and exclusive feature of CE that can be used directly in commercially available CE instruments without complicated steps or special equipment [21–24]. The coupling of isotachophoresis (ITP) to capillary zone electrophoresis (CZE) can be superior. ITP can be induced as a transient ITP during CZE by the composition of the sample itself or of the electrolyte system (i.e., the addition of leading/terminating ion). But ITP requires careful planning regarding the pH as well as leading and terminating ions and co-ions [25–31]. Shihabi developed a preconcentration method termed transient pseudo-isotachophoresis (pseudo-tITP) [32–34]. In pseudo-tITP, the inorganic ions in the sample act as leading ions, whereas organic solvents give the high field strength similar to the action of the terminating ions but without the rigid requirements of pH, concentration, or counter ions necessary in ITP. The aim of our work was to develop a sensitive method based on CE–LIF coupled with pseudo-tITP preconcentration for separation and determination of four peptide hormones. To optimize the CE method, the effects of experimental parameters, such as derivatization, methanol content in the buffer, and the amounts of NaCl and methanol in the sample matrix, were studied. Finally, this method was validated in terms of linearity, limit of detection, and repeatability and was applied to the analysis of human cerebrospinal fluid (CSF) samples. Materials and methods Chemicals NT, NT8–13, cholecystokinin fragment 30–33 amide hydrochloride (CCK-4HCl), NKB, and fluorescein isothiocyanate (FITC) were obtained from Sigma (St. Louis, MO, USA). Methanol (HPLC grade), sodium hydroxide, and sodium borate (analytical grade) were obtained from Shanghai Chemical Factory (Shanghai, China). Deionized water was obtained from a Milli-Q water purification system. Instrument A P/ACE MDQ CE instrument equipped with an LIF detector (Beckman, Fullerton, CA, USA) was used for the experiments. The
LIF detector used an argon ion laser for excitation at 488 nm and emission at 535 nm. Data collection, processing, and analysis were performed with system 32 Karat software (Beckman) and recorded on an IBM-compatible personal computer. Electrophoresis was performed in bare fused-silica capillary (60 cm [effective length 50 cm] 75 lm i.d., Rui-feng, Yongnian, Hebei, China). C18–ODS cartridges (500 mg/6 ml) were obtained from Agilent Technologies (Palo Alto, CA, USA). Borate buffer (10 mmol/L, pH 9.0) was obtained by titrating 10 mmol/L borate with 10 mmol/L boric acid to pH 9.0. For FITC derivatization reaction, 5 ll standard peptides mixed solution, 5 ll FITC (in acetone), and 90 ll borate buffer (10 mmol/L, pH 9.0) were blended together. The derivatized solution was kept in the dark for 16 h. The obtained solution was diluted prior to injection. Borate buffer (110 mmol/L, pH 9.5) was obtained by titrating 110 mmol/L borate with 1 mol/L sodium hydroxide to pH 9.5. The buffer solution was filtered through a 0.45-lm pore-size membrane filter and degassed before use. Conditioning of new capillary was performed by first flushing with Milli-Q water for 10 min, followed by 20 min with 1.0 M sodium hydroxide and finally with Milli-Q water for 10 min at high pressure (138 kPa, 20 psi). Between runs,the capillary was rinsed with 0.1 mol/L sodium hydroxide for 2 min, with Milli-Q water for 2 min, followed by the running buffer for 2 min at 138 kPa. Preparation of CSF sample CSF samples were collected from a patient with craniocerebral trauma and subarachnoid hemorrhage (provided by the Department of Neurology, Fuzhou General Hospital, Fuzhou, China). The procedures used for obtaining peptide from CSF sample include deproteination and solid-phase extraction using C18–ODS cartridges. After 2 ml of CSF was collected in a 10-ml polyethylene centrifuge tube, 2 ml of acetonitrile was added to the sample. The sample mixture was centrifuged at 5000g for 5 min. Supernatant was passed through the C18–ODS cartridge. The C18–ODS cartridge was washed with 3 ml of water, and the tested peptides were eluted using 3 ml of 50% (v/v) methanol/1% (v/v) aqueous acetic acid. The eluate of CSF sample was dried by N2 and reconstituted in 20 ll of derivatization buffer and then taken for derivatization.
Results and discussion Optimization of fluorescence derivatization conditions Because the maximum excitation and emission wavelengths of derivants formed by FITC and analytes matched those of the used Ar+ laser (with 488 nm as the excitation wavelength and 535 nm as the emission wavelength), FITC was selected as the precolumn label reagent. Fig. 1 shows the scheme of the derivatization reaction between FITC and amino acid. Several effect
Fig. 1. Scheme of the derivatization reaction between FITC and amino acid.
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
factors on the labeling, such as buffer pH, concentration of buffer, reaction time, and concentration of FITC, were investigated in detail to get the maximum labeling efficiency. The effect of pH on the labeling efficiency was investigated in the range of 7.0 to 10.0, and the results show that the peak heights are increased with the increasing pH up to 9.0; after that, the peaks are decreased. Therefore, sodium borate buffer (pH 9.0) was selected as the derivatization buffer. The influence of the concentrations of sodium borate buffer (5– 50 mmol/L, pH 9.0) on fluorescent intensity were studied, and the results show that the maximum reaction yield was achieved at a borate concentration of 10 mmol/L. Thus, 10 mmol/L borate solution was used for subsequent experiments. By comparing the peak heights of derivatives at different concentrations of FITC, the result show that a high concentration and excessive amount of FITC was required to obtain the high efficiency for labeling, with 2.5 10 4 mol/L FITC being found as optimal. The reaction time for derivatization was investigated by comparing peak heights of derivatives at various time intervals. The results show that the derivatization reaction could be completed to get a maximum peak height in 16 h at room temperature and could remain at this level for approximately 40 h. In subsequent experiments, 24 h was used as the analyte derivatization time. Fig. 2 shows the electropherogram of four peptide hormones under the optimal conditions. Enhancing sensitivity by pseudo-tITP When the volume of sample injected was increased over 3% of the capillary volume, the separation deteriorated rapidly. To allow trace amount determination of these analytes in biological sample, we investigated a stacking mode based on a pseudo-tITP
299
method first described by Shihabi [31,32]. Miscible organic solvents, such as methanol, ethanol, isopropanol, and acetonitrile, have low electric conductance. When these organic solvents are added to the sample matrix to provide the high electric field strength necessary for sample zone sharpening, especially in the presence of salts, a stacking similar to transient ITP is expected. In this method, a 110-mM borate solution at pH 9.5 was selected as running buffer; it was known that at the high pH derivatized peptides would migrate as anions. The chloride anion that had high mobility could act as leading ion. Borate ion might act as terminating ion. The leading ion rapidly migrates ahead of the analytes, leaving behind an area of higher field strength. This difference in the field strength accelerates the stacking process before the analytes enter the separation buffer. Our studies show that when the sample was dissolved in a mixture of 70% (v/v) methanol and 1% (m/v) NaCl, a sample volume of 50% of the capillary can be injected without band broadening. As shown in Fig. 3, we obtained an increase of peak height by approximately 100-fold with regard to the traditional injection procedure (Fig. 2A). Optimal conditions, such as the effect of methanol and NaCl concentrations in the sample matrix, must prevail for stacking. Effect of the percentage of methanol in the sample matrix We studied the effect of the percentage of methanol in the sample matrix in the presence of 1% (m/v) NaCl on the peptide separation and on the sensitivity. As demonstrated in Fig. 3, the sensitivity is clearly improved with increasing percentages of methanol. This is because increasing the amount of methanol in the sample matrix may increase the field strength in the sample zone, which in turn will accelerate the peptide stacking and
Fig. 2. Electropherograms of four peptide hormones derivatized with FITC using LIF detection. (A) Four peptides. (B) FITC blank. running buffer: borate buffer (pH 9.5, 110 mM) and 20% (v/v) methanol; separation voltage: 25 kV; injection: 3.45 kPa for 10 s. 1, NT8–13 (5.0 10 9 mol/L); 2, NT (1.0 10 8 mol/L); 3, NKB (5.0 10 9 mol/L); 4, CCK-4 (5.0 10 9 mol/L).
300
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
Fig. 3. Effect of percentage of methanol in the sample matrix. The concentration of NaCl in the sample matrix was held constant at 1% (m/v). Injection: 34.5 kPa for 25 s; separation voltage: 20 kV; running buffer: borate buffer (pH 9.5, 110 mM) and 20% (v/v) methanol. Other conditions are the same as in Fig. 2. RFU, relative fluorescence units.
improve the sensitivity. Based on sensitivity and separation, 70% (v/v) methanol was chosen in the sample matrix. Effect of the percentage of NaCl in the sample matrix From a practical point of view, this type of stacking is dependent on the concentration of both the salt and the buffer ionic strength. The influence of the concentration of NaCl was then studied, and the results show that change in the concentration of NaCl in sample matrix should affect the analyte ion velocities and, consequently, the stacking. Increasing the NaCl concentration in the sample to 0.5% (m/v) would improve the stacking. The stacking was maintained by increasing the NaCl concentration to 1% (m/ v), whereas increasing of NaCl concentration to 3% (m/v) would decrease the stacking. Effect of sample volume Sample solutions containing four peptides (NT8–13 [5.0 10– 9 M], NT [1.0 10 8 M], NKB [5.0 10 9 M], and CCK-4 [5.0 10 9 M]) were injected into the capillary at different time intervals under the injection pressure of 34.5 kPa. The peak heights were proportional to the injection time between 5 and 25 s. The longer injection time meant that it took a longer time to stack the samples. In addition, peak heights would decrease when the injection time was longer than 25 s. Thus, an injection time of 25 s at 34.5 kPa was chosen.
Optimization of separation conditions Sodium borate solution (110 mM, pH 9.5) was selected as running buffer and was found to be a good compromise among relatively low joule heat, short analysis time, and high selectivity. When pseudo-tITP was used, the effect of the content of methanol in the running buffer on the separation was studied. Without methanol, CCK-4 and NKB cannot be separated from the big background peaks of FITC. As a result, only NT and NT8–13 can be detected without methanol in the running buffer. Baseline separation of the four peptides was obtained with 20% (v/v) methanol in the running buffer. Further increases in the content of methanol in the running buffer will lead to increased analysis time and decreased sensitivity. The optimal conditions for separation and preconcentration of four analytes were a running buffer consisting of 110 mM sodium
borate (pH 9.5)–methanol (80:20, v/v), sample matrix methanol (70%, v/v), and NaCl (1%, m/v). A hydrodynamic injection of 25 s and 34.5 kPa was selected, corresponding to an injection volume of 50% the total capillary. Due to the large injection volume and the mechanism of pseudo-tITP, the four peptides were baseline separated and the peak heights were enhanced approximately 100-fold. Method validation The linearity of the method was assessed by preparing standards of different concentrations. Each standard was analyzed in triplicate under the experimental conditions mentioned above. The linear response range and the detection limits for NT8–13, NT, NKB, and CCK-4 under the optimal conditions are listed in Table 2. The detection limit was defined as the minimum analyte concentration yielding a signal/noise ratio equal to 3. The precision of the method was determined by measuring the repeatability of migration times and peak heights for each analyte. The precisions for the peak heights and migration times of NT8–13, NT, NKB, and CCK-4 were evaluated by the relative standard deviation (RSD) of replicate experiments (n = 6) with the concentrations of 5.0
Table 2 Regression equations, linearities, and detection limits Peptide
Regression equationa
R2
Linear range (nmol/L)
Detection limit (nmol/L)
NT8–13 NT NKB CCK-4
y = 238855x 48646 y = 67587x 13525 y = 63808x 18370 y = 85934x 31931
0.9945 0.9972 0.9998 0.9974
0.25–10 0.25–10 0.30–10 0.40–10
0.04 0.10 0.20 0.08
a
The variable y is peak height and x is concentration (in nmol/L).
Table 3 Determination of NT and NT8–13 and recovery of four peptides in human CSF sample Specie
Peptide
Concentration (nmol/L ± SD) (n = 3)
Added (nmol/L)
Found (nmol/L)
Recovery (%)
RSD (%) (n = 3)
CSF
NT NT8–13 NKB CCK
2.48 ± 0.03 0.57 ± 0.04
6.00 6.80 1.00 1.00
5.84 6.08 0.88 0.84
97.4 89.2 87.9 84.4
3.3 4.5 3.2 3.0
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
301
Fig. 4. Electropherogram of the human CSF sample. Sample matrix and CE conditions are the same as in Fig. 3.
10 9 mol/L for NT8–13, 1.0 10 8 mol/L for NT, 5.0 10 9 mol/L for NKB, and 5.0 10 9 mol/L for CCK-4. The results show that the RSD of migration times was less than 2%. The precisions for the peak heights were in the range of 3 to 6%. The results indicate that the proposed method has excellent precision. Application to human CSF sample The human CSF sample was collected from the patient with craniocerebral trauma subarachnoid hemorrhage. The human CSF sample was pretreated as described in Materials and Methods. Because the concentrations of NKB and CCK-4 in CSF are beyond their detection limits in our method, they cannot be detected. Table 3 shows the determination of NT and NT8–13 and the recovery of four peptides in the human CSF sample. A typical electropherogram of NT and NT8–13 in the CSF sample is shown in Fig. 4. The identity of the peaks was confirmed by adding the standard derivatized peptides to the sample. The multiple peaks near NT may be the metabolites of NT or the impurities in the CSF sample. We also experimented with the recoveries of the four peptides in the CSF sample spiked with different concentrations of NT, NT8–13, NKB, and CCK-4, and the results show that the recoveries of the four peptides ranged from 84.4 to 97.4%. Conclusions The CE–LIF combined with pseudo-tITP preconcentration allows the four peptide hormones (NT, NT8–13, NKB, and CCK) to be baseline separated and an improvement of peak heights of 100-fold with regard to the traditional hydrodynamic injection (2.3% capillary volume). We used this pseudo-tITP with LIF detection for the determination of NT and NT8–13 in a human CSF sample with satisfactory results. Acknowledgments This project was financially supported by the National Nature Sciences Foundation of China (20735002, 20575011) and the Key Natural Sciences Foundation of Fujian Province, China (D0520001). References [1] M.D. Davis, C.B. Nemeroff, A.K. Sen, T. Lee (Eds.), Neurotensin, dopamine, and schizophrenia, Receptors and Ligands in Psychiatry, Cambridge University Press, Cambridge, UK, 1988, pp. 167–186. [2] C. Hultsch, B. Pawelke, R. Bergmann, F. Wuest, Synthesis and evaluation of novel multimeric neurotensin8-13 analogues, Bioorg. Med. Chem. 14 (2006) 5913–5920.
[3] S. Khan, I. Liberzon, J.L. Abelson, Effect of repeat exposure on neuroendocrine and symptom responses to pentagastrin, Psychiatry Res. 126 (2004) 189–195. [4] J.E. Rivier, H.L. Lawrence, M.H. Perrin, M.R. Brown, Neurotensin analogues: Structure–activity relationships, J. Med. Chem. 20 (1977) 1409–1412. [5] J.F. Rehfeld, Accurate measurement of cholecystokinin in plasma, Clin. Chem. 44 (1998) 991–1001. [6] J. Gobom, K.O. Kraeuter, R. Persson, H. Steen, P. Roepstorff, R. Ekman, Detection and quantification of neurotensin in human brain tissue by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, Anal. Chem. 72 (2000) 3320–3326. [7] S. Merani, R.M. Palmour, J. Bradwejn, I. Berezowska, F.J. Vaccarino, J. Gutkowska, Development of a sensitive and specific assay system for cholecystokinin tetrapeptide, Peptides 18 (1997) 869–875. [8] G.A. Qureshi, I. Bednar, Q. Min, P. Södersten, J. Silberring, F. Nyberg, M. Thörnwall, Quantitation and identification of two cholecystokinin peptides, CCK-4 and CCK-8, in rat brain by HPLC and fast atom bombardment mass spectrometry, Biomed. Chromatogr. 7 (1993) 251–255. [9] P.G. Righetti, Capillary electrophoretic analysis of proteins and peptides of biomedical and pharmacological interest, Biopharm. Drug Dispos. 22 (2001) 337–351. [10] V. Kasicka, Recent advances in capillary electrophoresis and capillary electrochromatography of peptides, Electrophoresis 24 (2003) 4013–4046. [11] V. Kasicka, Recent developments in capillary electrophoresis and capillary electrochromatography of peptides, Electrophoresis 27 (2006) 142–175. [12] M. Urbanek, L. Krivankova, P. Bocek, Stacking phenomena in electromigration: From basic principles to practical procedures, Electrophoresis 24 (2003) 466–485. [13] E.M.J. Hoyes, U. Bondesson, D. Westerlund, P.E. Andren, Simultaneous analysis of endogenous neurotransmitters and neuropeptides in brain tissue using capillary electrophoresis–microelectrospray–tandem mass spectrometry, Electrophoresis 20 (1999) 1527–1532. [14] S. Xia, L. Zhang, P. Tong, M. Lu, W. Liu, G. Chen, Determination of peptide hormones of brain and intestine by CE with ESI–MS detection, Electrophoresis 28 (2007) 3268–3276. [15] P.E. Andren, R.M. Caprioli, Determination of extracellular release of neurotensin in discrete rat brain regions, Brain Res. 845 (1999) 123–129. [16] F. Benavente, E. Balaguer, J. Barbosa, V. Sanz-Nebot, Modelling migration behavior of peptide hormones in capillary electrophoresis–electrospray mass spectrometry, J. Chromatogr. A 1117 (2006) 94–102. [17] V. Kasicka, Recent developments in CE and CEC of peptides, Electrophoresis 29 (2008) 179–206. [18] V. Solinova, V. Kasicka, D. Koval, T. Barth, A. Ciencialova, L. Zakova, Analysis of synthetic derivatives of peptide hormones by capillary zone electrophoresis and micellar electrokinetic chromatography with ultraviolet–absorption and laser-induced fluorescence detection, J. Chromatogr. B 808 (2004) 75–82. [19] V. Sanz-Nebot, F. Benavente, E. Hernandez, J. Barbora, Evaluation of the electrophoretic behaviour of opioid peptides separation by capillary electrophoresis–electrospray ionization mass spectrometry, Anal. Chim. Acta 577 (2006) 68–76. [20] V. Solinova, V. Kasicka, P. Sazelova, T. Barth, I. Miksik, Separation and investigation of structure–mobility relationship of gonadotropin-releasing hormones by capillary zone electrophoresis in conventional and isoelectric acidic background electrolytes, J. Chromatogr. A 1155 (2007) 146–153. [21] C.H. Lin, T. Kaneta, On-line sample concentration techniques in capillary electrophoresis: Velocity gradient techniques and sample concentration techniques for biomolecules, Electrophoresis 25 (2004) 4058–4073. [22] M.R.N. Monton, S. Terabe, Sample enrichment techniques in capillary electrophoresis: Focus on peptides and proteins, J. Chromatogr. B 841 (2006) 88–95. [23] Z. Mala, L. Krivankova, P. Gebauer, P. Bocek, Contemporary sample stacking in CE: A sophisticated tool based on simple principles, Electrophoresis 28 (2007) 243–253.
302
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
[24] S.L. Simpson, J.P. Quirino, S. Terabe, On-line sample preconcentration in capillary electrophoresis: Fundamentals and applications, J. Chromatogr. A 1184 (2008) 504– 541. [25] P. Petr, Determination of chlorite in drinking water by on-line coupling of capillary isotachophoresis and capillary zone electrophoresis, Talanta 62 (2004) 977–982. [26] P. Gebauer, P. Bocek, Recent progress in capillary isotachophoresis, Electrophoresis 23 (2002) 3858–3864. [27] P. Gebauer, Z. Mala, P. Bocek, Recent progress in capillary ITP, Electrophoresis 28 (2007) 26–32. [28] H. Okamoto, A.R. Timerbaev, T. Hirokawa, Simultaneous determination of metal ions, amino acids, and other small biogenic molecules in human serum by capillary zone electrophoresis with transient isotachophoretic preconcentration, J. Sep. Sci. 28 (2005) 522– 528.
[29] R.L. Chien, Sample stacking revisited: A personal perspective, Electrophoresis 24 (2003) 486–497. [30] L. Krivánkova, P. Bocek, Isotachophoresis and stacking phenomena, Electrophoresis 21 (2000) 2745. [31] L. Krivankova, P. Pantuckova, P. Bocek. Isotachophoresis in zone electrophoresis, J. Chromatogr. A 838 (1999) 55–70. [32] Z.K. Shihabi, Stacking and discontinuous buffers in capillary zone electrophoresis, Electrophoresis 21 (2000) 2872–2878. [33] Z.K. Shihabi, Transient pseudo-isotachophoresis for sample concentration in capillary electrophoresis, Electrophoresis 23 (2002) 1612–1617. [34] F. Cañada-Cañada, A. Kasselouri, P. Prognon, P. Maillard, D.S. Grierson, S. Descroix, M. Taverna, Enhanced detection of seven glucoconjugated and hydroxylated porphyrins and chlorins by nonaqueous capillary electrophoresis combined with stacking, J. Chromatogr. A 1068 (2005) 123– 130.
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Separation and determination of peptide hormones by capillary electrophoresis with laser-induced fluorescence coupled with transient pseudo-isotachophoresis preconcentration Ying Chen a,b, Liangjun Xu b, Lan Zhang a, Guonan Chen a,* a b
Ministry of Education, Key Laboratory of Analysis and Detection Technology for Food Safety, and Department of Chemistry, Fuzhou University, Fuzhou, Fujian 350002, China Analytical and Testing Center, Fuzhou University, Fuzhou, Fujian 350002, China
a r t i c l e
i n f o
Article history: Received 19 April 2008 Available online 5 June 2008 Keywords: Capillary electrophoresis Laser-induced fluorescence Cholecystokinin tetrapeptide Neurotensin Neurotensin hexapeptide Neurokinin B Transient pseudo-isotachophoresis
a b s t r a c t A new method for the determination of the peptide hormones and their fragments by capillary electrophoresis (CE) with laser-induced fluorescence (LIF) detection and transient pseudo-isotachophoresis (pseudo-tITP) preconcentration was established in this study. The LIF detector used an argon ion laser with excitation wavelength at 488 nm and emission wavelength at 535 nm. Fluorescein isothiocyanate (FITC) was used as precolumn derivatization reagent to label cholecystokinin tetrapeptide (CCK-4), neurotensin (NT), neurotensin hexapeptide (NT8–13), and neurokinin B (NKB). Borate (10 mmol/L, pH 9.0) was selected as derivatization medium to get the high efficiency. When the addition of 70% (v/v) methanol and 1% (m/v) sodium chloride (NaCl) to the sample matrix, and with borate buffer (110 mM, pH 9.5) and 20% (v/v) methanol as running buffer, a preconcentration based on the pseudo-tITP afforded 100-fold improvement in peak heights compared with the traditional hydrodynamic injection (2.3% capillary volume). The detection limits (signal/noise = 3) based on peak height were found to be 0.04, 0.1, 0.2, and 0.08 nmol/L for NT8–13, NT, NKB, and CCK-4, respectively. The method was validated and applied to qualitative analysis of NT and NT8–13 in human cerebrospinal fluid sample. Ó 2008 Elsevier Inc. All rights reserved.
Neurotensin (NT)1 and its C-terminal hexapeptide fragment NT8– , neurokinin B (NKB), and cholecystokinin tetrapeptide (CCK-4) are 13 a group of peptide hormones that act as neurotransmitters and neuromodulators in the central nervous system (see Table 1). These peptide hormones are thought to be involved in various biological functions such as sensory information, antipsychotic, food intake control, and regulation of neuroendocrine processing [1–3]. These peptide hormones tend to exist at trace level in biological samples. Their abnormal secretion in the body may lead to severe pathological changes. Despite recent progress in peptide hormone research, many of the body mechanisms and the complete clinical and pharmacological aspects of these peptides remain complex. Therefore, trace amounts of these peptides in biological fluids must be determined
* Corresponding author. Fax: +86 591 83713866. E-mail address: [email protected] (G. Chen). 1 Abbreviations used: NT, neurotensin; NKB, neurokinin B; CCK-4, cholecystokinin tetrapeptide; RIA, radioimmunoassay; ELISA, enzyme-linked immunosorbent assay; HPLC, high-performance liquid chromatography; CE, capillary electrophoresis; LIF, laser-induced fluorescence; MS, mass spectrometry; ITP, isotachophoresis; CZE, capillary zone electrophoresis; pseudo-tITP, transient pseudo-isotachophoresis; CSF, cerebrospinal fluid; CCK-4HCl, cholecystokinin fragment 30–33 amide hydrochloride; FITC, fluorescein isothiocyanate; RSD, relative standard deviation. 0003-2697/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2008.06.002
accurately so that their physiological effects and therapeutic measures can be better collected and understood. Traditionally, radioimmunoassay (RIA), enzyme-linked immunosorbent assay (ELISA), and immunohistochemical techniques have been the most widely used methods for the measurement of these peptides [4,5]. However, the most obvious disadvantage of RIA is the safety concern of using radioactive species. In addition, RIA involves labor-intensive procedures aside from sample preparation and lengthy reaction times. Although ELISA does not use radioactive reagents, some intrinsic problems involved in immunoassay, such as cross-reaction and nonspecific interference of coexistent compounds with the binding of the analytes, still need to be resolved. High-performance liquid chromatography (HPLC) with immunoassay or mass detection has been used to analyze these peptide hormones [6–8]. Nowadays, compared with HPLC, capillary electrophoresis (CE) has become a more attractive separation technique in peptide analysis due to its many advantages, including high efficiency, low waste production, and fast separation [9– 11]. Two basic approaches can be distinguished to improve detection sensitivity in CE either by using more sensitive detection methods, such as laser-induced fluorescence (LIF) and electrochemical mode, or by increasing analyte mass loading [12]. Recently, CE with different detection methods, such as mass
298
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
Table 1 Three-letter codes of analyzed peptide hormone of brain and intestine Peptides
Molecular mass
Three-letter codes of compound
NKB CCK-4 NT8–13 NT
1210.4 633.2 874.0 1672.0
Asp-Met-His-Asp-Phe-Phe-Val-Gly-Leu-Met-NH2 Trp-Met-Asp-Phe-NH2 Arg-Arg-Pro-Tyr-Ile-Leu pGlu-Leu-Tyr-Glu-Asn-Lys-Pro-Arg-Arg-Pro-Tyr-Ile-Leu
spectrometry (MS) and LIF, has been applied in peptide hormone analysis [13–20]. But to the best of our knowledge, CE–LIF has not been reported for simultaneous analysis of NT, NT8–13, CCK-4, and NKB. Sample stacking is an inherent and exclusive feature of CE that can be used directly in commercially available CE instruments without complicated steps or special equipment [21–24]. The coupling of isotachophoresis (ITP) to capillary zone electrophoresis (CZE) can be superior. ITP can be induced as a transient ITP during CZE by the composition of the sample itself or of the electrolyte system (i.e., the addition of leading/terminating ion). But ITP requires careful planning regarding the pH as well as leading and terminating ions and co-ions [25–31]. Shihabi developed a preconcentration method termed transient pseudo-isotachophoresis (pseudo-tITP) [32–34]. In pseudo-tITP, the inorganic ions in the sample act as leading ions, whereas organic solvents give the high field strength similar to the action of the terminating ions but without the rigid requirements of pH, concentration, or counter ions necessary in ITP. The aim of our work was to develop a sensitive method based on CE–LIF coupled with pseudo-tITP preconcentration for separation and determination of four peptide hormones. To optimize the CE method, the effects of experimental parameters, such as derivatization, methanol content in the buffer, and the amounts of NaCl and methanol in the sample matrix, were studied. Finally, this method was validated in terms of linearity, limit of detection, and repeatability and was applied to the analysis of human cerebrospinal fluid (CSF) samples. Materials and methods Chemicals NT, NT8–13, cholecystokinin fragment 30–33 amide hydrochloride (CCK-4HCl), NKB, and fluorescein isothiocyanate (FITC) were obtained from Sigma (St. Louis, MO, USA). Methanol (HPLC grade), sodium hydroxide, and sodium borate (analytical grade) were obtained from Shanghai Chemical Factory (Shanghai, China). Deionized water was obtained from a Milli-Q water purification system. Instrument A P/ACE MDQ CE instrument equipped with an LIF detector (Beckman, Fullerton, CA, USA) was used for the experiments. The
LIF detector used an argon ion laser for excitation at 488 nm and emission at 535 nm. Data collection, processing, and analysis were performed with system 32 Karat software (Beckman) and recorded on an IBM-compatible personal computer. Electrophoresis was performed in bare fused-silica capillary (60 cm [effective length 50 cm] 75 lm i.d., Rui-feng, Yongnian, Hebei, China). C18–ODS cartridges (500 mg/6 ml) were obtained from Agilent Technologies (Palo Alto, CA, USA). Borate buffer (10 mmol/L, pH 9.0) was obtained by titrating 10 mmol/L borate with 10 mmol/L boric acid to pH 9.0. For FITC derivatization reaction, 5 ll standard peptides mixed solution, 5 ll FITC (in acetone), and 90 ll borate buffer (10 mmol/L, pH 9.0) were blended together. The derivatized solution was kept in the dark for 16 h. The obtained solution was diluted prior to injection. Borate buffer (110 mmol/L, pH 9.5) was obtained by titrating 110 mmol/L borate with 1 mol/L sodium hydroxide to pH 9.5. The buffer solution was filtered through a 0.45-lm pore-size membrane filter and degassed before use. Conditioning of new capillary was performed by first flushing with Milli-Q water for 10 min, followed by 20 min with 1.0 M sodium hydroxide and finally with Milli-Q water for 10 min at high pressure (138 kPa, 20 psi). Between runs,the capillary was rinsed with 0.1 mol/L sodium hydroxide for 2 min, with Milli-Q water for 2 min, followed by the running buffer for 2 min at 138 kPa. Preparation of CSF sample CSF samples were collected from a patient with craniocerebral trauma and subarachnoid hemorrhage (provided by the Department of Neurology, Fuzhou General Hospital, Fuzhou, China). The procedures used for obtaining peptide from CSF sample include deproteination and solid-phase extraction using C18–ODS cartridges. After 2 ml of CSF was collected in a 10-ml polyethylene centrifuge tube, 2 ml of acetonitrile was added to the sample. The sample mixture was centrifuged at 5000g for 5 min. Supernatant was passed through the C18–ODS cartridge. The C18–ODS cartridge was washed with 3 ml of water, and the tested peptides were eluted using 3 ml of 50% (v/v) methanol/1% (v/v) aqueous acetic acid. The eluate of CSF sample was dried by N2 and reconstituted in 20 ll of derivatization buffer and then taken for derivatization.
Results and discussion Optimization of fluorescence derivatization conditions Because the maximum excitation and emission wavelengths of derivants formed by FITC and analytes matched those of the used Ar+ laser (with 488 nm as the excitation wavelength and 535 nm as the emission wavelength), FITC was selected as the precolumn label reagent. Fig. 1 shows the scheme of the derivatization reaction between FITC and amino acid. Several effect
Fig. 1. Scheme of the derivatization reaction between FITC and amino acid.
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
factors on the labeling, such as buffer pH, concentration of buffer, reaction time, and concentration of FITC, were investigated in detail to get the maximum labeling efficiency. The effect of pH on the labeling efficiency was investigated in the range of 7.0 to 10.0, and the results show that the peak heights are increased with the increasing pH up to 9.0; after that, the peaks are decreased. Therefore, sodium borate buffer (pH 9.0) was selected as the derivatization buffer. The influence of the concentrations of sodium borate buffer (5– 50 mmol/L, pH 9.0) on fluorescent intensity were studied, and the results show that the maximum reaction yield was achieved at a borate concentration of 10 mmol/L. Thus, 10 mmol/L borate solution was used for subsequent experiments. By comparing the peak heights of derivatives at different concentrations of FITC, the result show that a high concentration and excessive amount of FITC was required to obtain the high efficiency for labeling, with 2.5 10 4 mol/L FITC being found as optimal. The reaction time for derivatization was investigated by comparing peak heights of derivatives at various time intervals. The results show that the derivatization reaction could be completed to get a maximum peak height in 16 h at room temperature and could remain at this level for approximately 40 h. In subsequent experiments, 24 h was used as the analyte derivatization time. Fig. 2 shows the electropherogram of four peptide hormones under the optimal conditions. Enhancing sensitivity by pseudo-tITP When the volume of sample injected was increased over 3% of the capillary volume, the separation deteriorated rapidly. To allow trace amount determination of these analytes in biological sample, we investigated a stacking mode based on a pseudo-tITP
299
method first described by Shihabi [31,32]. Miscible organic solvents, such as methanol, ethanol, isopropanol, and acetonitrile, have low electric conductance. When these organic solvents are added to the sample matrix to provide the high electric field strength necessary for sample zone sharpening, especially in the presence of salts, a stacking similar to transient ITP is expected. In this method, a 110-mM borate solution at pH 9.5 was selected as running buffer; it was known that at the high pH derivatized peptides would migrate as anions. The chloride anion that had high mobility could act as leading ion. Borate ion might act as terminating ion. The leading ion rapidly migrates ahead of the analytes, leaving behind an area of higher field strength. This difference in the field strength accelerates the stacking process before the analytes enter the separation buffer. Our studies show that when the sample was dissolved in a mixture of 70% (v/v) methanol and 1% (m/v) NaCl, a sample volume of 50% of the capillary can be injected without band broadening. As shown in Fig. 3, we obtained an increase of peak height by approximately 100-fold with regard to the traditional injection procedure (Fig. 2A). Optimal conditions, such as the effect of methanol and NaCl concentrations in the sample matrix, must prevail for stacking. Effect of the percentage of methanol in the sample matrix We studied the effect of the percentage of methanol in the sample matrix in the presence of 1% (m/v) NaCl on the peptide separation and on the sensitivity. As demonstrated in Fig. 3, the sensitivity is clearly improved with increasing percentages of methanol. This is because increasing the amount of methanol in the sample matrix may increase the field strength in the sample zone, which in turn will accelerate the peptide stacking and
Fig. 2. Electropherograms of four peptide hormones derivatized with FITC using LIF detection. (A) Four peptides. (B) FITC blank. running buffer: borate buffer (pH 9.5, 110 mM) and 20% (v/v) methanol; separation voltage: 25 kV; injection: 3.45 kPa for 10 s. 1, NT8–13 (5.0 10 9 mol/L); 2, NT (1.0 10 8 mol/L); 3, NKB (5.0 10 9 mol/L); 4, CCK-4 (5.0 10 9 mol/L).
300
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
Fig. 3. Effect of percentage of methanol in the sample matrix. The concentration of NaCl in the sample matrix was held constant at 1% (m/v). Injection: 34.5 kPa for 25 s; separation voltage: 20 kV; running buffer: borate buffer (pH 9.5, 110 mM) and 20% (v/v) methanol. Other conditions are the same as in Fig. 2. RFU, relative fluorescence units.
improve the sensitivity. Based on sensitivity and separation, 70% (v/v) methanol was chosen in the sample matrix. Effect of the percentage of NaCl in the sample matrix From a practical point of view, this type of stacking is dependent on the concentration of both the salt and the buffer ionic strength. The influence of the concentration of NaCl was then studied, and the results show that change in the concentration of NaCl in sample matrix should affect the analyte ion velocities and, consequently, the stacking. Increasing the NaCl concentration in the sample to 0.5% (m/v) would improve the stacking. The stacking was maintained by increasing the NaCl concentration to 1% (m/ v), whereas increasing of NaCl concentration to 3% (m/v) would decrease the stacking. Effect of sample volume Sample solutions containing four peptides (NT8–13 [5.0 10– 9 M], NT [1.0 10 8 M], NKB [5.0 10 9 M], and CCK-4 [5.0 10 9 M]) were injected into the capillary at different time intervals under the injection pressure of 34.5 kPa. The peak heights were proportional to the injection time between 5 and 25 s. The longer injection time meant that it took a longer time to stack the samples. In addition, peak heights would decrease when the injection time was longer than 25 s. Thus, an injection time of 25 s at 34.5 kPa was chosen.
Optimization of separation conditions Sodium borate solution (110 mM, pH 9.5) was selected as running buffer and was found to be a good compromise among relatively low joule heat, short analysis time, and high selectivity. When pseudo-tITP was used, the effect of the content of methanol in the running buffer on the separation was studied. Without methanol, CCK-4 and NKB cannot be separated from the big background peaks of FITC. As a result, only NT and NT8–13 can be detected without methanol in the running buffer. Baseline separation of the four peptides was obtained with 20% (v/v) methanol in the running buffer. Further increases in the content of methanol in the running buffer will lead to increased analysis time and decreased sensitivity. The optimal conditions for separation and preconcentration of four analytes were a running buffer consisting of 110 mM sodium
borate (pH 9.5)–methanol (80:20, v/v), sample matrix methanol (70%, v/v), and NaCl (1%, m/v). A hydrodynamic injection of 25 s and 34.5 kPa was selected, corresponding to an injection volume of 50% the total capillary. Due to the large injection volume and the mechanism of pseudo-tITP, the four peptides were baseline separated and the peak heights were enhanced approximately 100-fold. Method validation The linearity of the method was assessed by preparing standards of different concentrations. Each standard was analyzed in triplicate under the experimental conditions mentioned above. The linear response range and the detection limits for NT8–13, NT, NKB, and CCK-4 under the optimal conditions are listed in Table 2. The detection limit was defined as the minimum analyte concentration yielding a signal/noise ratio equal to 3. The precision of the method was determined by measuring the repeatability of migration times and peak heights for each analyte. The precisions for the peak heights and migration times of NT8–13, NT, NKB, and CCK-4 were evaluated by the relative standard deviation (RSD) of replicate experiments (n = 6) with the concentrations of 5.0
Table 2 Regression equations, linearities, and detection limits Peptide
Regression equationa
R2
Linear range (nmol/L)
Detection limit (nmol/L)
NT8–13 NT NKB CCK-4
y = 238855x 48646 y = 67587x 13525 y = 63808x 18370 y = 85934x 31931
0.9945 0.9972 0.9998 0.9974
0.25–10 0.25–10 0.30–10 0.40–10
0.04 0.10 0.20 0.08
a
The variable y is peak height and x is concentration (in nmol/L).
Table 3 Determination of NT and NT8–13 and recovery of four peptides in human CSF sample Specie
Peptide
Concentration (nmol/L ± SD) (n = 3)
Added (nmol/L)
Found (nmol/L)
Recovery (%)
RSD (%) (n = 3)
CSF
NT NT8–13 NKB CCK
2.48 ± 0.03 0.57 ± 0.04
6.00 6.80 1.00 1.00
5.84 6.08 0.88 0.84
97.4 89.2 87.9 84.4
3.3 4.5 3.2 3.0
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
301
Fig. 4. Electropherogram of the human CSF sample. Sample matrix and CE conditions are the same as in Fig. 3.
10 9 mol/L for NT8–13, 1.0 10 8 mol/L for NT, 5.0 10 9 mol/L for NKB, and 5.0 10 9 mol/L for CCK-4. The results show that the RSD of migration times was less than 2%. The precisions for the peak heights were in the range of 3 to 6%. The results indicate that the proposed method has excellent precision. Application to human CSF sample The human CSF sample was collected from the patient with craniocerebral trauma subarachnoid hemorrhage. The human CSF sample was pretreated as described in Materials and Methods. Because the concentrations of NKB and CCK-4 in CSF are beyond their detection limits in our method, they cannot be detected. Table 3 shows the determination of NT and NT8–13 and the recovery of four peptides in the human CSF sample. A typical electropherogram of NT and NT8–13 in the CSF sample is shown in Fig. 4. The identity of the peaks was confirmed by adding the standard derivatized peptides to the sample. The multiple peaks near NT may be the metabolites of NT or the impurities in the CSF sample. We also experimented with the recoveries of the four peptides in the CSF sample spiked with different concentrations of NT, NT8–13, NKB, and CCK-4, and the results show that the recoveries of the four peptides ranged from 84.4 to 97.4%. Conclusions The CE–LIF combined with pseudo-tITP preconcentration allows the four peptide hormones (NT, NT8–13, NKB, and CCK) to be baseline separated and an improvement of peak heights of 100-fold with regard to the traditional hydrodynamic injection (2.3% capillary volume). We used this pseudo-tITP with LIF detection for the determination of NT and NT8–13 in a human CSF sample with satisfactory results. Acknowledgments This project was financially supported by the National Nature Sciences Foundation of China (20735002, 20575011) and the Key Natural Sciences Foundation of Fujian Province, China (D0520001). References [1] M.D. Davis, C.B. Nemeroff, A.K. Sen, T. Lee (Eds.), Neurotensin, dopamine, and schizophrenia, Receptors and Ligands in Psychiatry, Cambridge University Press, Cambridge, UK, 1988, pp. 167–186. [2] C. Hultsch, B. Pawelke, R. Bergmann, F. Wuest, Synthesis and evaluation of novel multimeric neurotensin8-13 analogues, Bioorg. Med. Chem. 14 (2006) 5913–5920.
[3] S. Khan, I. Liberzon, J.L. Abelson, Effect of repeat exposure on neuroendocrine and symptom responses to pentagastrin, Psychiatry Res. 126 (2004) 189–195. [4] J.E. Rivier, H.L. Lawrence, M.H. Perrin, M.R. Brown, Neurotensin analogues: Structure–activity relationships, J. Med. Chem. 20 (1977) 1409–1412. [5] J.F. Rehfeld, Accurate measurement of cholecystokinin in plasma, Clin. Chem. 44 (1998) 991–1001. [6] J. Gobom, K.O. Kraeuter, R. Persson, H. Steen, P. Roepstorff, R. Ekman, Detection and quantification of neurotensin in human brain tissue by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, Anal. Chem. 72 (2000) 3320–3326. [7] S. Merani, R.M. Palmour, J. Bradwejn, I. Berezowska, F.J. Vaccarino, J. Gutkowska, Development of a sensitive and specific assay system for cholecystokinin tetrapeptide, Peptides 18 (1997) 869–875. [8] G.A. Qureshi, I. Bednar, Q. Min, P. Södersten, J. Silberring, F. Nyberg, M. Thörnwall, Quantitation and identification of two cholecystokinin peptides, CCK-4 and CCK-8, in rat brain by HPLC and fast atom bombardment mass spectrometry, Biomed. Chromatogr. 7 (1993) 251–255. [9] P.G. Righetti, Capillary electrophoretic analysis of proteins and peptides of biomedical and pharmacological interest, Biopharm. Drug Dispos. 22 (2001) 337–351. [10] V. Kasicka, Recent advances in capillary electrophoresis and capillary electrochromatography of peptides, Electrophoresis 24 (2003) 4013–4046. [11] V. Kasicka, Recent developments in capillary electrophoresis and capillary electrochromatography of peptides, Electrophoresis 27 (2006) 142–175. [12] M. Urbanek, L. Krivankova, P. Bocek, Stacking phenomena in electromigration: From basic principles to practical procedures, Electrophoresis 24 (2003) 466–485. [13] E.M.J. Hoyes, U. Bondesson, D. Westerlund, P.E. Andren, Simultaneous analysis of endogenous neurotransmitters and neuropeptides in brain tissue using capillary electrophoresis–microelectrospray–tandem mass spectrometry, Electrophoresis 20 (1999) 1527–1532. [14] S. Xia, L. Zhang, P. Tong, M. Lu, W. Liu, G. Chen, Determination of peptide hormones of brain and intestine by CE with ESI–MS detection, Electrophoresis 28 (2007) 3268–3276. [15] P.E. Andren, R.M. Caprioli, Determination of extracellular release of neurotensin in discrete rat brain regions, Brain Res. 845 (1999) 123–129. [16] F. Benavente, E. Balaguer, J. Barbosa, V. Sanz-Nebot, Modelling migration behavior of peptide hormones in capillary electrophoresis–electrospray mass spectrometry, J. Chromatogr. A 1117 (2006) 94–102. [17] V. Kasicka, Recent developments in CE and CEC of peptides, Electrophoresis 29 (2008) 179–206. [18] V. Solinova, V. Kasicka, D. Koval, T. Barth, A. Ciencialova, L. Zakova, Analysis of synthetic derivatives of peptide hormones by capillary zone electrophoresis and micellar electrokinetic chromatography with ultraviolet–absorption and laser-induced fluorescence detection, J. Chromatogr. B 808 (2004) 75–82. [19] V. Sanz-Nebot, F. Benavente, E. Hernandez, J. Barbora, Evaluation of the electrophoretic behaviour of opioid peptides separation by capillary electrophoresis–electrospray ionization mass spectrometry, Anal. Chim. Acta 577 (2006) 68–76. [20] V. Solinova, V. Kasicka, P. Sazelova, T. Barth, I. Miksik, Separation and investigation of structure–mobility relationship of gonadotropin-releasing hormones by capillary zone electrophoresis in conventional and isoelectric acidic background electrolytes, J. Chromatogr. A 1155 (2007) 146–153. [21] C.H. Lin, T. Kaneta, On-line sample concentration techniques in capillary electrophoresis: Velocity gradient techniques and sample concentration techniques for biomolecules, Electrophoresis 25 (2004) 4058–4073. [22] M.R.N. Monton, S. Terabe, Sample enrichment techniques in capillary electrophoresis: Focus on peptides and proteins, J. Chromatogr. B 841 (2006) 88–95. [23] Z. Mala, L. Krivankova, P. Gebauer, P. Bocek, Contemporary sample stacking in CE: A sophisticated tool based on simple principles, Electrophoresis 28 (2007) 243–253.
302
Separation and determination of peptide hormones / Y. Chen et al. / Anal. Biochem. 380 (2008) 297–302
[24] S.L. Simpson, J.P. Quirino, S. Terabe, On-line sample preconcentration in capillary electrophoresis: Fundamentals and applications, J. Chromatogr. A 1184 (2008) 504– 541. [25] P. Petr, Determination of chlorite in drinking water by on-line coupling of capillary isotachophoresis and capillary zone electrophoresis, Talanta 62 (2004) 977–982. [26] P. Gebauer, P. Bocek, Recent progress in capillary isotachophoresis, Electrophoresis 23 (2002) 3858–3864. [27] P. Gebauer, Z. Mala, P. Bocek, Recent progress in capillary ITP, Electrophoresis 28 (2007) 26–32. [28] H. Okamoto, A.R. Timerbaev, T. Hirokawa, Simultaneous determination of metal ions, amino acids, and other small biogenic molecules in human serum by capillary zone electrophoresis with transient isotachophoretic preconcentration, J. Sep. Sci. 28 (2005) 522– 528.
[29] R.L. Chien, Sample stacking revisited: A personal perspective, Electrophoresis 24 (2003) 486–497. [30] L. Krivánkova, P. Bocek, Isotachophoresis and stacking phenomena, Electrophoresis 21 (2000) 2745. [31] L. Krivankova, P. Pantuckova, P. Bocek. Isotachophoresis in zone electrophoresis, J. Chromatogr. A 838 (1999) 55–70. [32] Z.K. Shihabi, Stacking and discontinuous buffers in capillary zone electrophoresis, Electrophoresis 21 (2000) 2872–2878. [33] Z.K. Shihabi, Transient pseudo-isotachophoresis for sample concentration in capillary electrophoresis, Electrophoresis 23 (2002) 1612–1617. [34] F. Cañada-Cañada, A. Kasselouri, P. Prognon, P. Maillard, D.S. Grierson, S. Descroix, M. Taverna, Enhanced detection of seven glucoconjugated and hydroxylated porphyrins and chlorins by nonaqueous capillary electrophoresis combined with stacking, J. Chromatogr. A 1068 (2005) 123– 130.