Research article Received: 24 June 2013,
Revised: 11 December 2013,
Accepted: 18 December 2013
Published online in Wiley Online Library: 23 January 2014
(wileyonlinelibrary.com) DOI 10.1002/bmc.3135
Analysis of normal and modified nucleosides in urine samples by high-performance liquid chromatography with different stationary phases Sylwia Studzińska and Bogusław Buszewski* ABSTRACT: The main aim of the present work was to study the retention behavior and quantification of nine nucleosides with the use of octadecyl, alkylamide, cholesterol and alkyl-phosphate stationary phases. The influence of organic solvent and buffer concentration on the separation of these compounds was under investigation. The retention factor had the highest values for the octadecyl and cholesterol packing materials. Complete separation of all the studied nucleosides was achieved in case of cholesterol stationary phase. The optimized separation method was applied for the quantification of nucleosides in the urine samples. Calibration plots showed good linearity (R2 > 0.999) and the limits of detection were in a range of 0.3–0.5 μg/mL, while the limits of quantitation were >0.9 μg/mL. Accuracy was in the range of 5–11%. Copyright © 2014 John Wiley & Sons, Ltd. Keywords: reversed-phase high-performance liquid chromatography; normal and modified nucleosides; stationary phase; separation; urine
Introduction
1140
Nucleosides are the building blocks of nucleic acids. They are composed of a purine or pyrimidine base (adenine, thymine, uracil, guanine or cytosine) and ribose or deoxyribose sugar (Cooperwood, 2002). Naturally occurring nucleoside modifications are an essential attribute of transfer RNA (tRNA) (Kool, 2008). Modified nucleosides are formed at the post-transcriptional stage by chemical and enzymatic modification of standard nucleosides (Cooperwood, 2002; Kool, 2008). (Cooperwood, 2002; Kool, 2008) and are excreted in urine. The level of their concentration in the biological fluids is a biomarker of a tumor (Dudley et al., 2003; Jeng et al., 2009). Their measurement in urine may be a noninvasive diagnostic tool. However, it is important to emphasise that the utilization of nucleosides in the diagnosis of cancer is still confirmatory in nature. Methods for the separation of nucleosides and their modified forms in urine are of great importance (Struck et al., 2011). Numerous methods for analysis of nucleosides have been used and described; however, reversed-phase high-performance liquid chromatography (RP HPLC) is the most popular (Hsu et al., 2009; Li et al., 2009; Struck et al., 2011). The method is simple, precise and selective. Octadecyl (C18) stationary phase is commonly used when RP HPLC is used for the analysis of normal and modified nucleosides (Cho et al., 2009; Hsu et al., 2009; Li et al., 2009; Liebich et al., 2005; Qian et al., 2008). Its application is not suitable for the separation of small and very polar nucleosides, since they are eluted close to the dead volume of the column. The separation of the nucleosides was performed on C18 columns with different dimensions, e.g. length between 250 and 100 mm (Hsu et al., 2009, 2011; Liebich et al., 2005; Zhao et al., 2006). The use of C18 involves a time-consuming separation step (to avoid co-elution of similar compounds) when multiple analyses of nucleosides in urine are performed (Dudley et al., 2000; Xu et al., 2000; Qian et al., 2008). On the other hand there are a variety of stationary phases for RP-HPLC and selectivity of separation
Biomed. Chromatogr. 2014; 28: 1140–1146
may be improved when other columns are applied for the analysis of studied compounds (Kowalska et al., 2005; Studzińska and Buszewski, 2012). Aminopropyl, alkylamide and cholesterol packing materials have been utilized for this purpose (Kowalska et al., 2005). It was shown that the pH of mobile phase buffer influences the retention of nucleosides on these three packing materials, owing to the protonization of aminopropyl groups (Kowalska et al., 2005). However specific packing materials were not utilized in the analysis of urine samples. Owing to the polarity of the studied compounds, the mobile phase usually contains organic solvent (methanol or acetonitrile) and an appropriate buffer (Cho et al., 2009; Hsu et al., 2009; Liebich et al., 2005). The chromatography was performed with the use of phosphate buffer (Liebich et al., 2005; Qian et al., 2008) or ammonium acetate (Cho et al., 2009; Hsu et al., 2011; Takeda et al., 2000) as mobile phase components. The elution is carried out usually with buffer concentration in the range 10–50 mM and pH A > N2mG > 1mG>T > 1mI > G > U > C. The most polar C was eluted near the dead volume of the system when mobile phase contained >10% v/v of methanol. A similar elution order was noticed for SG-AP (Fig. 2B), although above 15% v/v of methanol some changes may be observed (e.g. C is eluted before U and 1mI before G). Furthermore, C and U were retained inside the chromatographic column in the wider range of methanol content (10–20% v/v) in comparison with SG-C18 (Fig. 2B). This is a consequence of the specific hydrophobic–hydrophilic structure of SG-AP (Fig. 1) and interactions taking place during the chromatographic process. Apparently, a mixed structure of packing material favors
1.32 — 1.13 1.03 1.18 1.21 1.26 1.25 1.23
S. Studzińska and B. Buszewski
Figure 2. Dependencies of log k vs the percentage part of methanol in the mobile phase (ϕ) for the stationary phases: (A) SG-C18; (B) SG-AP; (C) SGCHOL; (D) SG-P-C10. The mobile phase consisted of mixtures of methanol and 30 mM K2HPO4–KH2PO4, pH 6.5; the flow rate was equal to 0.5 mL/min.
and SG-CHOL are different (Fig. 1). SG-CHOL combines the advantages of SG-C18 (in terms of retention strength) and SG-AP (in terms of the selectivity interrelated with small, polar nucleosides). This effect is strictly connected to the structural properties of SGCHOL. The large cholesterol molecule is probably responsible for the hydrophobicity of SG-CHOL; on the other hand aminopropyl groups can influence nucleoside retention by hydrogen bonding and electrostatic interactions with the analyzed compounds (Fig. 1). Consequently, the retention mechanism is of a mixed mode nature and appears to have a beneficial impact on the chromatographic analysis of normal and modified nucleosides. The lowest k-values of analyzed nucleosides were determined for SG-PC10 (Fig. 2D). This stationary phase possesses alkylphosphate ligands bonded to the silica surface (Fig. 1). The structure of SG-PC10 is similar to that of SG-AP, because they have alkyl chains and aminopropyl groups (Fig. 1). However SG-PC10 possesses a phosphate group instead of an amide one. The presence of a charged phosphate group is probably mainly responsible for the low k-values (Fig. 2D). Most of the nucleosides were eluted from the chromatographic column during the first minutes of analysis and consequently their separation was not complete.
Separation of nucleosides
1144
An attempt to separate nine nucleosides was performed for all of the studied stationary phases. The best results obtained for each packing material are presented in Fig. 3. Complete resolution of all analyzed compounds was possible for SG-CHOL (Fig. 3A). As mentioned before, this phenomenon is connected to mixed mode properties of SG-CHOL. SG-AP may also be characterized by similar properties (Fig. 1). However the resolution of nucleosides was poor when SG-AP was used for the separation
wileyonlinelibrary.com/journal/bmc
of normal and modified nucleosides (Fig. 3B). This stationary phase possesses aminopropyl groups, similar to SG-CHOL; however, the size and volume of alkyl chains are lower in comparison to the large cholesterol molecule for SG-CHOL. Consequently nucleosides have greater access to the aminopropyl ligands and k-values are lower in comparison to SG-CHOL (Fig. 1, Table 2). For this reason complete separation of all nucleosides was not achieved (Fig. 3B). Application of SG-C18 failed in the separation of polar C and U (Fig. 3C), as a consequence of the hydrophobic character of octadecyl packing. The times of single analysis in the case of SG-C18 and SG-CHOL were similar (about 50 min). Moreover, utilization of a gradient elution could reduce the total time of analysis. The use of SG-PC10 provided short times of analysis; however, the attempt to separate the nucleosides was unsuccessful (Fig. 3D). It may be concluded that SG-PC10 is not suitable for the separation of normal or modified nucleosides.
Application to the urine samples SG-CHOL was chosen for further determination of normal and modified nucleosides in the urine samples purchased from LGC standards. Urine was fortified with nucleosides before analysis, since it did not contain any of the studied compounds. The gradient elution was applied to reduce the total time of single analysis. The linear gradient of increasing the methanol content in the mobile phase was used between the 15 and 20 min of analysis. Consequently, the separation of nucleosides was reduced to 23 min (Fig. 4). The optimized separation method was next applied for the quantification of nucleoside levels in urine. Figure 4 presents the chromatograms obtained for blank urine and urine samples fortified with nucleosides. Studied compounds were identified by comparison with
Copyright © 2014 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2014; 28: 1140–1146
Analysis of nucleosides in urine by HPLC
Figure 4. Chromatograms of a urine sample fortified with nucleosides. Chromatographic conditions: SG-CHOL stationary phase; mobile phase consisted of methanol and 30 mM K2HPO4–KH2PO4, pH 6.5; gradient elution, 0–15 min 5% v/v MeOH, 15–20 min 45% v/v MeOH, 20–25 min 45% v/v MeOH; flow rate, 0.5 mL/min; injection volume, 1 μL. Notation: 1, C; 2, U; 3, G; 4, T; 5, 1mG; 6, 1mI; 7, N2mG; 8, A; 9, 1 mA.
Table 3. Analytical characteristics of the method for nine normal and modified nucleosides analysis with the use of cholesterol stationary phase Nucleoside
Linear regression data Test range (μg/mL)
R
1.41–28.52 1.30–27.77 1.05–24.21 1.00–29.30 1.11–26.60 1.12–23.48 1.11–27.56 1.19–28.34 1.22–27.29
0.9982 0.9989 0.9994 0.9996 0.9996 0.9993 0.9997 0.9992 0.9995
U C G T 1mI 1mG N2mG 1mA 1mI
Figure 3. The results of separation of normal and modified nucleosides for stationary phases: (A) SG-CHOL (5:95% v/v MeOH–phosphate buffer, pH 6.5); (B) SG-AP (5:95% v/v MeOH–phosphate buffer, pH 6.5); (C) SGC18 (5:95% v/v MeOH–phosphate buffer, pH 6.5); (D) SG-P-C10 (100% v/ v phosphate buffer, pH 6.5). Notation: 1, C; 2, U; 3, G; 4, 1mI; 5, T; 6, 1mG; 7, N2mG; 8, A; 9, 1 mA. For detailed experimental conditions see the Experimental section. 1 mA, 1-methyladenosine; 1mI, 1methylinosine; 1mG, 1-methylguanosine; A, adenosine; C, cytydine; G, 2 guanosine; N2mG, N -methylguanosine; T, thymidine; U, uridine.
Biomed. Chromatogr. 2014; 28: 1140–1146
LOD LOQ (μg/mL) (μg/mL) 0.44 0.41 0.33 0.32 0.35 0.32 0.35 0.37 0.38
1.32 1.23 0.99 0.96 1.05 0.96 1.05 1.11 1.14
measure of the linearity of the method. As it can be seen in Table 3, the calibration plots showed good linearity (R2 > 0.998) within the test ranges. Values of LOD were between 0.3 and 0.5 μg/mL, while LOQs were higher than 0.9 μg/mL (Table 3). Reliability of the method in terms of precision and accuracy was also studied. The inter-day CV ranged from 0.6 to 4.3%. Accuracy varied with the concentration, but was generally in the range of 5–11%. The quantification of all the studied nucleosides with the use of SG-CHOL was of acceptable accuracy, precision and linearity. Therefore this stationary phase may be successfully used for the determination of normal and modified nucleosides in urine samples.
Conclusions Stationary phases with specific surface properties appear to be useful in the analysis of nucleosides. Several different functional groups bonded to the silica surface interact with nucleosides in various ways. Therefore, the retention mechanism of these substances is of mixed mode, with both polar (hydrogen bond, electrostatic interactions) and hydrophobic interactions influencing the final result of analysis. On the basis of the obtained results, it may be concluded that methanol should be
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
1145
chromatograms of standards, retention times and spectra. Table 3 collects all of the data obtained for method validation. The calibration plots were appointed for all of the nucleosides in the urine. A linear relationship was obtained between the concentration of analyzed compounds and peak areas determined at 254 nm for concentration ranging between 0.9 and 28 μg/ mL. The squared coefficients of variation were used as the
2
S. Studzińska and B. Buszewski used as an organic modifier of the mobile phase together with a buffer. Its concentration influences both the retention and peak shape of the nucleosides on each stationary phase used in the investigation. A 30 mM phosphate buffer should be used for the routine analysis of normal and modified nucleosides in the case of the cholesterol, alkylamide or alkyl-phosphate stationary phase. Moreover, the complete separation of nine nucleosides was achieved for the cholesterol stationary phase, while U and C were not separated in the case of the octadecyl one. The cholesterol packing material was successfully applied for the quantification of normal and modified nucleosides in urine fortified with studied compounds. Future studies will be performed to determine nucleosides in the real urine samples taken from people with breast cancer.
Acknowledgments Financial support was provided by the National Science Center (Cracow, Poland) under grant no. 2011/01/D/ST4/04142. The authors are grateful to Ms Magdalena Milewska for technical assistance and to Dr Szymon Bocian for his help in the synthesis of stationary phases.
References Buszewski B, Jezierska-Świtała M, Kaliszan R, Wojtczak A, Albert K, Bochmann S, Matyska MT and Pesek JJ. Selectivity tuning and molecular modeling of new generation packings for RP HPLC. Chromatographia 2001; 53: 204–212. Buszewski B, Krupczyńska K, Gadzała-Kopciuch R, Rychlicki G and Kaliszan R. Evaluation of HPLC columns: a study on surface homogeneity of chemically bonded stationary phases. Journal of Separation Science 2003; 26: 313–321. Cho SH, Choi MH, Lee WY and Chung BC. Evaluation of urinary nucleosides in breast cancer patients before and after tumor removal. Clinical Biochemistry 2009; 42: 540–543. Cooperwood JS, Gumina G, Boudinot FD, Chu CK, in: Chu CK (Ed.). Recent Advances in Nucleosides: Chemistry and Chemotherapy. Elsevier Science 2002; 91. Dudley E, El-Sharkawi S and Games DE. Analysis of urinary nucleosides. I. Optimisation of high performance liquid chromatography/ electrospray mass spectrometry. Rapid Communications in Mass Spectrometry 2000; 14: 1200–1207. Dudley E, Lemiere F, Van Dongen W, Esmans E, El-Sharkawi AM, Games DE, Brenton AG and Newton RP. Urinary modified nucleosides as tumor markers. Nucleosides, Nucleotides and Nucleic Acids 2003; 22: 987–989. Gadzała-Kopciuch R and Buszewski B. Comparative study of hydrophobicity of octadecyl and alkylamide bonded phases using the methylene selectivity. Journal of Separation Science 2003; 26: 1273–1281. Hsu WY, Chen WTL, Lin WD, Tsai FJ, Tsai Y, Lin CT, Lo WY, Jeng LB and Lai CC. Analysis of urinary nucleosides as potential tumor markers in
human colorectal cancer by high performance liquid chromatography/electrospray ionization tandem mass spectrometry. Clinica Chimica Acta 2009; 402: 31–37. Hsu WY, Lin WD, Tsai Y, Lin CT, Wang HC, Jeng LB, Lee CC, Lin YC, Lai CC and Tsai FJ. Analysis of urinary nucleosides as potential tumor markers in human breast cancer by high performance liquid chromatography/electrospray ionization tandem mass spectrometry. Clinica Chimica Acta 2011; 412: 1861–1866. Jeng LB, Lo WY, Hsu WY, Lin WD, Lin CT, Lai CC and Tsai FJ. Analysis of urinary nucleosides as helper tumor markers in hepatocellular carcinoma diagnosis. Rapid Communications in Mass Spectrometry 2009; 23: 1543–1549. Kool ET. Modified Nucleosides: in Biochemistry, Biotechnology and Medicine, Herdewijn P (ed.). John Wiley & Sons: Chichester, 2008; 49. Kowalska S, Krupczyńska K and Buszewski B. The influence of the mobile phase pH and stationary phase type on the selectivity tuning in HPLC nucleosides separation. Journal of Separation Science 2005; 28: 1502–1511. Li Y, Wang SM and Liu HM. Separation and identification of purine nucleosides in the urine of patients with malignant cancer by reverse phase liquid chromatography/electrospray tandem mass spectrometry. Journal of Mass Spectrometry 2009; 44: 641–651. Liebich HM, Muller-Hagedorn S, Bacher M, Scheel-Walter H-G, Lu X, Frickenschmidt A, Kammerer B, Kim K-R and Gerard H. Age-dependence of urinary normal and modified nucleosides in childhood as determined by reversed-phase high-performance liquid chromatography. Journal of Chromatography B 2005; 814: 275–283. Qian ZM, Wan JB, Zhang QW and Li SP. Simultaneous determination of nucleobases, nucleosides and saponins in Panax notoginseng using multiple columns high performance liquid chromatography. Journal of Pharmaceutical and Biomedical Analysis 2008; 48: 1361–1367. Struck W, Waszczuk-Jankowska M, Kaliszan R and Markuszewski MJ. The state-of-the-art determination of urinary nucleosides using chromatographic techniques ‘hyphenated’ with advanced bioinformatic methods. Analytical and Bioanalytical Chemistry 2011; 401: 2039–2050. Studzińska S and Buszewski B. A new way to fast and high resolution determination of modified nucleosides. Journal of Chromatography B 2012; 887: 93–101. Studzińska S and Buszewski B. Effect of mobile phase pH on the retention of nucleotides on different stationary phases for high performance liquid chromatography. Analytical and Bioanalytical Chemistry 2013; 405(5): 1663–1672. Takeda N, Yoshizumi H and Niwa T. Detection and characterization of modified nucleosides in serum and urine of uremic patients using capillary liquid chromatography–frit–fast atom bombardment mass spektrometry. Journal of Chromatography B 2000; 746: 51–62. Xu G, Enderle H, Liebich H and Lu P. Study of normal and modified nucleosides in serum by RP-HPLC. Chromatographia 2000; 52: 152–158. Zhao X, Wang W, Wang J, Yang J and Xu G. Urinary profiling investigation of metabolites with cis-diol structure from cancer patients based on UPLC-MS and HPLC-MS as well as multivariate statistical analysis. Journal of Separation Science 2006; 29: 2444–2451.
1146 wileyonlinelibrary.com/journal/bmc
Copyright © 2014 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2014; 28: 1140–1146
Revised: 11 December 2013,
Accepted: 18 December 2013
Published online in Wiley Online Library: 23 January 2014
(wileyonlinelibrary.com) DOI 10.1002/bmc.3135
Analysis of normal and modified nucleosides in urine samples by high-performance liquid chromatography with different stationary phases Sylwia Studzińska and Bogusław Buszewski* ABSTRACT: The main aim of the present work was to study the retention behavior and quantification of nine nucleosides with the use of octadecyl, alkylamide, cholesterol and alkyl-phosphate stationary phases. The influence of organic solvent and buffer concentration on the separation of these compounds was under investigation. The retention factor had the highest values for the octadecyl and cholesterol packing materials. Complete separation of all the studied nucleosides was achieved in case of cholesterol stationary phase. The optimized separation method was applied for the quantification of nucleosides in the urine samples. Calibration plots showed good linearity (R2 > 0.999) and the limits of detection were in a range of 0.3–0.5 μg/mL, while the limits of quantitation were >0.9 μg/mL. Accuracy was in the range of 5–11%. Copyright © 2014 John Wiley & Sons, Ltd. Keywords: reversed-phase high-performance liquid chromatography; normal and modified nucleosides; stationary phase; separation; urine
Introduction
1140
Nucleosides are the building blocks of nucleic acids. They are composed of a purine or pyrimidine base (adenine, thymine, uracil, guanine or cytosine) and ribose or deoxyribose sugar (Cooperwood, 2002). Naturally occurring nucleoside modifications are an essential attribute of transfer RNA (tRNA) (Kool, 2008). Modified nucleosides are formed at the post-transcriptional stage by chemical and enzymatic modification of standard nucleosides (Cooperwood, 2002; Kool, 2008). (Cooperwood, 2002; Kool, 2008) and are excreted in urine. The level of their concentration in the biological fluids is a biomarker of a tumor (Dudley et al., 2003; Jeng et al., 2009). Their measurement in urine may be a noninvasive diagnostic tool. However, it is important to emphasise that the utilization of nucleosides in the diagnosis of cancer is still confirmatory in nature. Methods for the separation of nucleosides and their modified forms in urine are of great importance (Struck et al., 2011). Numerous methods for analysis of nucleosides have been used and described; however, reversed-phase high-performance liquid chromatography (RP HPLC) is the most popular (Hsu et al., 2009; Li et al., 2009; Struck et al., 2011). The method is simple, precise and selective. Octadecyl (C18) stationary phase is commonly used when RP HPLC is used for the analysis of normal and modified nucleosides (Cho et al., 2009; Hsu et al., 2009; Li et al., 2009; Liebich et al., 2005; Qian et al., 2008). Its application is not suitable for the separation of small and very polar nucleosides, since they are eluted close to the dead volume of the column. The separation of the nucleosides was performed on C18 columns with different dimensions, e.g. length between 250 and 100 mm (Hsu et al., 2009, 2011; Liebich et al., 2005; Zhao et al., 2006). The use of C18 involves a time-consuming separation step (to avoid co-elution of similar compounds) when multiple analyses of nucleosides in urine are performed (Dudley et al., 2000; Xu et al., 2000; Qian et al., 2008). On the other hand there are a variety of stationary phases for RP-HPLC and selectivity of separation
Biomed. Chromatogr. 2014; 28: 1140–1146
may be improved when other columns are applied for the analysis of studied compounds (Kowalska et al., 2005; Studzińska and Buszewski, 2012). Aminopropyl, alkylamide and cholesterol packing materials have been utilized for this purpose (Kowalska et al., 2005). It was shown that the pH of mobile phase buffer influences the retention of nucleosides on these three packing materials, owing to the protonization of aminopropyl groups (Kowalska et al., 2005). However specific packing materials were not utilized in the analysis of urine samples. Owing to the polarity of the studied compounds, the mobile phase usually contains organic solvent (methanol or acetonitrile) and an appropriate buffer (Cho et al., 2009; Hsu et al., 2009; Liebich et al., 2005). The chromatography was performed with the use of phosphate buffer (Liebich et al., 2005; Qian et al., 2008) or ammonium acetate (Cho et al., 2009; Hsu et al., 2011; Takeda et al., 2000) as mobile phase components. The elution is carried out usually with buffer concentration in the range 10–50 mM and pH A > N2mG > 1mG>T > 1mI > G > U > C. The most polar C was eluted near the dead volume of the system when mobile phase contained >10% v/v of methanol. A similar elution order was noticed for SG-AP (Fig. 2B), although above 15% v/v of methanol some changes may be observed (e.g. C is eluted before U and 1mI before G). Furthermore, C and U were retained inside the chromatographic column in the wider range of methanol content (10–20% v/v) in comparison with SG-C18 (Fig. 2B). This is a consequence of the specific hydrophobic–hydrophilic structure of SG-AP (Fig. 1) and interactions taking place during the chromatographic process. Apparently, a mixed structure of packing material favors
1.32 — 1.13 1.03 1.18 1.21 1.26 1.25 1.23
S. Studzińska and B. Buszewski
Figure 2. Dependencies of log k vs the percentage part of methanol in the mobile phase (ϕ) for the stationary phases: (A) SG-C18; (B) SG-AP; (C) SGCHOL; (D) SG-P-C10. The mobile phase consisted of mixtures of methanol and 30 mM K2HPO4–KH2PO4, pH 6.5; the flow rate was equal to 0.5 mL/min.
and SG-CHOL are different (Fig. 1). SG-CHOL combines the advantages of SG-C18 (in terms of retention strength) and SG-AP (in terms of the selectivity interrelated with small, polar nucleosides). This effect is strictly connected to the structural properties of SGCHOL. The large cholesterol molecule is probably responsible for the hydrophobicity of SG-CHOL; on the other hand aminopropyl groups can influence nucleoside retention by hydrogen bonding and electrostatic interactions with the analyzed compounds (Fig. 1). Consequently, the retention mechanism is of a mixed mode nature and appears to have a beneficial impact on the chromatographic analysis of normal and modified nucleosides. The lowest k-values of analyzed nucleosides were determined for SG-PC10 (Fig. 2D). This stationary phase possesses alkylphosphate ligands bonded to the silica surface (Fig. 1). The structure of SG-PC10 is similar to that of SG-AP, because they have alkyl chains and aminopropyl groups (Fig. 1). However SG-PC10 possesses a phosphate group instead of an amide one. The presence of a charged phosphate group is probably mainly responsible for the low k-values (Fig. 2D). Most of the nucleosides were eluted from the chromatographic column during the first minutes of analysis and consequently their separation was not complete.
Separation of nucleosides
1144
An attempt to separate nine nucleosides was performed for all of the studied stationary phases. The best results obtained for each packing material are presented in Fig. 3. Complete resolution of all analyzed compounds was possible for SG-CHOL (Fig. 3A). As mentioned before, this phenomenon is connected to mixed mode properties of SG-CHOL. SG-AP may also be characterized by similar properties (Fig. 1). However the resolution of nucleosides was poor when SG-AP was used for the separation
wileyonlinelibrary.com/journal/bmc
of normal and modified nucleosides (Fig. 3B). This stationary phase possesses aminopropyl groups, similar to SG-CHOL; however, the size and volume of alkyl chains are lower in comparison to the large cholesterol molecule for SG-CHOL. Consequently nucleosides have greater access to the aminopropyl ligands and k-values are lower in comparison to SG-CHOL (Fig. 1, Table 2). For this reason complete separation of all nucleosides was not achieved (Fig. 3B). Application of SG-C18 failed in the separation of polar C and U (Fig. 3C), as a consequence of the hydrophobic character of octadecyl packing. The times of single analysis in the case of SG-C18 and SG-CHOL were similar (about 50 min). Moreover, utilization of a gradient elution could reduce the total time of analysis. The use of SG-PC10 provided short times of analysis; however, the attempt to separate the nucleosides was unsuccessful (Fig. 3D). It may be concluded that SG-PC10 is not suitable for the separation of normal or modified nucleosides.
Application to the urine samples SG-CHOL was chosen for further determination of normal and modified nucleosides in the urine samples purchased from LGC standards. Urine was fortified with nucleosides before analysis, since it did not contain any of the studied compounds. The gradient elution was applied to reduce the total time of single analysis. The linear gradient of increasing the methanol content in the mobile phase was used between the 15 and 20 min of analysis. Consequently, the separation of nucleosides was reduced to 23 min (Fig. 4). The optimized separation method was next applied for the quantification of nucleoside levels in urine. Figure 4 presents the chromatograms obtained for blank urine and urine samples fortified with nucleosides. Studied compounds were identified by comparison with
Copyright © 2014 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2014; 28: 1140–1146
Analysis of nucleosides in urine by HPLC
Figure 4. Chromatograms of a urine sample fortified with nucleosides. Chromatographic conditions: SG-CHOL stationary phase; mobile phase consisted of methanol and 30 mM K2HPO4–KH2PO4, pH 6.5; gradient elution, 0–15 min 5% v/v MeOH, 15–20 min 45% v/v MeOH, 20–25 min 45% v/v MeOH; flow rate, 0.5 mL/min; injection volume, 1 μL. Notation: 1, C; 2, U; 3, G; 4, T; 5, 1mG; 6, 1mI; 7, N2mG; 8, A; 9, 1 mA.
Table 3. Analytical characteristics of the method for nine normal and modified nucleosides analysis with the use of cholesterol stationary phase Nucleoside
Linear regression data Test range (μg/mL)
R
1.41–28.52 1.30–27.77 1.05–24.21 1.00–29.30 1.11–26.60 1.12–23.48 1.11–27.56 1.19–28.34 1.22–27.29
0.9982 0.9989 0.9994 0.9996 0.9996 0.9993 0.9997 0.9992 0.9995
U C G T 1mI 1mG N2mG 1mA 1mI
Figure 3. The results of separation of normal and modified nucleosides for stationary phases: (A) SG-CHOL (5:95% v/v MeOH–phosphate buffer, pH 6.5); (B) SG-AP (5:95% v/v MeOH–phosphate buffer, pH 6.5); (C) SGC18 (5:95% v/v MeOH–phosphate buffer, pH 6.5); (D) SG-P-C10 (100% v/ v phosphate buffer, pH 6.5). Notation: 1, C; 2, U; 3, G; 4, 1mI; 5, T; 6, 1mG; 7, N2mG; 8, A; 9, 1 mA. For detailed experimental conditions see the Experimental section. 1 mA, 1-methyladenosine; 1mI, 1methylinosine; 1mG, 1-methylguanosine; A, adenosine; C, cytydine; G, 2 guanosine; N2mG, N -methylguanosine; T, thymidine; U, uridine.
Biomed. Chromatogr. 2014; 28: 1140–1146
LOD LOQ (μg/mL) (μg/mL) 0.44 0.41 0.33 0.32 0.35 0.32 0.35 0.37 0.38
1.32 1.23 0.99 0.96 1.05 0.96 1.05 1.11 1.14
measure of the linearity of the method. As it can be seen in Table 3, the calibration plots showed good linearity (R2 > 0.998) within the test ranges. Values of LOD were between 0.3 and 0.5 μg/mL, while LOQs were higher than 0.9 μg/mL (Table 3). Reliability of the method in terms of precision and accuracy was also studied. The inter-day CV ranged from 0.6 to 4.3%. Accuracy varied with the concentration, but was generally in the range of 5–11%. The quantification of all the studied nucleosides with the use of SG-CHOL was of acceptable accuracy, precision and linearity. Therefore this stationary phase may be successfully used for the determination of normal and modified nucleosides in urine samples.
Conclusions Stationary phases with specific surface properties appear to be useful in the analysis of nucleosides. Several different functional groups bonded to the silica surface interact with nucleosides in various ways. Therefore, the retention mechanism of these substances is of mixed mode, with both polar (hydrogen bond, electrostatic interactions) and hydrophobic interactions influencing the final result of analysis. On the basis of the obtained results, it may be concluded that methanol should be
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/bmc
1145
chromatograms of standards, retention times and spectra. Table 3 collects all of the data obtained for method validation. The calibration plots were appointed for all of the nucleosides in the urine. A linear relationship was obtained between the concentration of analyzed compounds and peak areas determined at 254 nm for concentration ranging between 0.9 and 28 μg/ mL. The squared coefficients of variation were used as the
2
S. Studzińska and B. Buszewski used as an organic modifier of the mobile phase together with a buffer. Its concentration influences both the retention and peak shape of the nucleosides on each stationary phase used in the investigation. A 30 mM phosphate buffer should be used for the routine analysis of normal and modified nucleosides in the case of the cholesterol, alkylamide or alkyl-phosphate stationary phase. Moreover, the complete separation of nine nucleosides was achieved for the cholesterol stationary phase, while U and C were not separated in the case of the octadecyl one. The cholesterol packing material was successfully applied for the quantification of normal and modified nucleosides in urine fortified with studied compounds. Future studies will be performed to determine nucleosides in the real urine samples taken from people with breast cancer.
Acknowledgments Financial support was provided by the National Science Center (Cracow, Poland) under grant no. 2011/01/D/ST4/04142. The authors are grateful to Ms Magdalena Milewska for technical assistance and to Dr Szymon Bocian for his help in the synthesis of stationary phases.
References Buszewski B, Jezierska-Świtała M, Kaliszan R, Wojtczak A, Albert K, Bochmann S, Matyska MT and Pesek JJ. Selectivity tuning and molecular modeling of new generation packings for RP HPLC. Chromatographia 2001; 53: 204–212. Buszewski B, Krupczyńska K, Gadzała-Kopciuch R, Rychlicki G and Kaliszan R. Evaluation of HPLC columns: a study on surface homogeneity of chemically bonded stationary phases. Journal of Separation Science 2003; 26: 313–321. Cho SH, Choi MH, Lee WY and Chung BC. Evaluation of urinary nucleosides in breast cancer patients before and after tumor removal. Clinical Biochemistry 2009; 42: 540–543. Cooperwood JS, Gumina G, Boudinot FD, Chu CK, in: Chu CK (Ed.). Recent Advances in Nucleosides: Chemistry and Chemotherapy. Elsevier Science 2002; 91. Dudley E, El-Sharkawi S and Games DE. Analysis of urinary nucleosides. I. Optimisation of high performance liquid chromatography/ electrospray mass spectrometry. Rapid Communications in Mass Spectrometry 2000; 14: 1200–1207. Dudley E, Lemiere F, Van Dongen W, Esmans E, El-Sharkawi AM, Games DE, Brenton AG and Newton RP. Urinary modified nucleosides as tumor markers. Nucleosides, Nucleotides and Nucleic Acids 2003; 22: 987–989. Gadzała-Kopciuch R and Buszewski B. Comparative study of hydrophobicity of octadecyl and alkylamide bonded phases using the methylene selectivity. Journal of Separation Science 2003; 26: 1273–1281. Hsu WY, Chen WTL, Lin WD, Tsai FJ, Tsai Y, Lin CT, Lo WY, Jeng LB and Lai CC. Analysis of urinary nucleosides as potential tumor markers in
human colorectal cancer by high performance liquid chromatography/electrospray ionization tandem mass spectrometry. Clinica Chimica Acta 2009; 402: 31–37. Hsu WY, Lin WD, Tsai Y, Lin CT, Wang HC, Jeng LB, Lee CC, Lin YC, Lai CC and Tsai FJ. Analysis of urinary nucleosides as potential tumor markers in human breast cancer by high performance liquid chromatography/electrospray ionization tandem mass spectrometry. Clinica Chimica Acta 2011; 412: 1861–1866. Jeng LB, Lo WY, Hsu WY, Lin WD, Lin CT, Lai CC and Tsai FJ. Analysis of urinary nucleosides as helper tumor markers in hepatocellular carcinoma diagnosis. Rapid Communications in Mass Spectrometry 2009; 23: 1543–1549. Kool ET. Modified Nucleosides: in Biochemistry, Biotechnology and Medicine, Herdewijn P (ed.). John Wiley & Sons: Chichester, 2008; 49. Kowalska S, Krupczyńska K and Buszewski B. The influence of the mobile phase pH and stationary phase type on the selectivity tuning in HPLC nucleosides separation. Journal of Separation Science 2005; 28: 1502–1511. Li Y, Wang SM and Liu HM. Separation and identification of purine nucleosides in the urine of patients with malignant cancer by reverse phase liquid chromatography/electrospray tandem mass spectrometry. Journal of Mass Spectrometry 2009; 44: 641–651. Liebich HM, Muller-Hagedorn S, Bacher M, Scheel-Walter H-G, Lu X, Frickenschmidt A, Kammerer B, Kim K-R and Gerard H. Age-dependence of urinary normal and modified nucleosides in childhood as determined by reversed-phase high-performance liquid chromatography. Journal of Chromatography B 2005; 814: 275–283. Qian ZM, Wan JB, Zhang QW and Li SP. Simultaneous determination of nucleobases, nucleosides and saponins in Panax notoginseng using multiple columns high performance liquid chromatography. Journal of Pharmaceutical and Biomedical Analysis 2008; 48: 1361–1367. Struck W, Waszczuk-Jankowska M, Kaliszan R and Markuszewski MJ. The state-of-the-art determination of urinary nucleosides using chromatographic techniques ‘hyphenated’ with advanced bioinformatic methods. Analytical and Bioanalytical Chemistry 2011; 401: 2039–2050. Studzińska S and Buszewski B. A new way to fast and high resolution determination of modified nucleosides. Journal of Chromatography B 2012; 887: 93–101. Studzińska S and Buszewski B. Effect of mobile phase pH on the retention of nucleotides on different stationary phases for high performance liquid chromatography. Analytical and Bioanalytical Chemistry 2013; 405(5): 1663–1672. Takeda N, Yoshizumi H and Niwa T. Detection and characterization of modified nucleosides in serum and urine of uremic patients using capillary liquid chromatography–frit–fast atom bombardment mass spektrometry. Journal of Chromatography B 2000; 746: 51–62. Xu G, Enderle H, Liebich H and Lu P. Study of normal and modified nucleosides in serum by RP-HPLC. Chromatographia 2000; 52: 152–158. Zhao X, Wang W, Wang J, Yang J and Xu G. Urinary profiling investigation of metabolites with cis-diol structure from cancer patients based on UPLC-MS and HPLC-MS as well as multivariate statistical analysis. Journal of Separation Science 2006; 29: 2444–2451.
1146 wileyonlinelibrary.com/journal/bmc
Copyright © 2014 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 2014; 28: 1140–1146
