Technical Note pubs.acs.org/ac
Online Three Dimensional Liquid Chromatography/Mass Spectrometry Method for the Separation of Complex Samples Shuangyuan Wang,†,‡ Xianzhe Shi,*,† and Guowang Xu*,† †
CAS Key Laboratory of Separation Science for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China ‡ University of Chinese Academy of Sciences, Beijing 100049, China S Supporting Information *
ABSTRACT: In this work, a novel online three dimensional liquid chromatography (3D-LC) system was first developed by effectively coupling of preseparation and comprehensive 2DLC using a stop-flow interface, aiming at improving the separation of complex samples. The sample was separated into two or several fractions through the first dimensional separation, and then each fraction was transferred in an orderly way into the following comprehensive 2D-LC part for further analysis. More optimal conditions could be operated in the second and third dimensions according to the properties of each fraction. Thus, the resolution of the 3D-LC system was substantially improved. Analysis of soybean extract was taken as a proof-of-principle to demonstrate the powerful separation of the established 3D-LC system. The amide column was selected as the first dimension column. Weakly polar metabolites (such as lipids, aglycones, etc.) and polar metabolites (such as glycosides, etc.) were separated into different fractions. Fluorophenyl and C18 columns were used in the second and third dimensions of the 3D-LC system for further separation, respectively. There were 83 flavonoids characterized in the soybean extract, including many difficult to separate isomers and low-abundance flavonoids; in total, they were nearly 30% more than those identified in the comparative comprehensive 2D-LC approach. In conclusion, this 3D-LC system is flexible in construction and applicable to complex sample analysis.
M
Although 2D-LC methods are powerful, a good separation of very complex samples is still difficult. Inherently, a huge amount of compounds in these samples with considerably varied concentrations and properties pose challenges in their separation. In comprehensive 2D-LC, the same second dimensional gradient is usually used for all the cuts from the first dimension, which results in a waste of second dimensional analytical time and poor separation. Recently, a shift gradient mode was reported and applied in 2D-LC methods.14−17 The orthogonality was greatly improved especially for RPLC × RPLC, which produced better separation. However, this mode had difficulties in performing universal 2D-LC systems with good orthogonality. Therefore, three dimensional (3D) chromatography methods have become alternative choices toward better separation of very complex samples. Offline 3D-LC methods, which are easy to establish, have been reported.18,19 Valeja et al. established an IEC-hydrophobic interaction chromatography (HIC)-RPLC method to separate intact proteins.18 As a result, 201 nonredundant proteins were characterized in the 3D-LC approach, but only 47 non-
ultidimensional liquid chromatography (MD-LC) methods are particularly attractive and widely used in the separation of complex samples due to extremely high resolution.1−4 As they are easy to implement, two dimensional (2D)-LC methods in comprehensive and heart-cutting modes are widely applied. Generally, comprehensive 2D-LC methods, which can provide twice separation of the entire sample, are suitable for untargeted and complex constituents’ analysis because of the substantially improved peak capacity.5,6 Kalili et al. developed a hydrophilic interaction chromatography (HILIC) × reversed phase (RP) LC method and applied it in grape seed analysis.1 Coupled with a time-of-flight mass spectrometer (TOF MS), this method was powerful in separation and characterization of tannins.1 Besides HILIC and RPLC, normal phase (NP) LC, ion exchange chromatography (IEC), size exclusion chromatography (SEC), etc. can also be used in constructing 2D-LC systems and flexibly combined to meet specific separation requirements.7−9 In contrast with comprehensive mode, heart-cutting 2D-LC methods are beneficial to targeted analyses.10−12 Ma et al. proposed a heart-cutting 2D-LC method for analysis of carbohydrates. Due to online purification, this method was demonstrated to be suitable for milk powder samples.13 © XXXX American Chemical Society
Received: November 9, 2016 Accepted: December 27, 2016
A
DOI: 10.1021/acs.analchem.6b04401 Anal. Chem. XXXX, XXX, XXX−XXX
Technical Note
Analytical Chemistry
Table 1. Gradient Conditions of 3D-LC Method and Comparative 2D-LC Method for Separation of Soybean Extract 3D-LC
fraction 1 fraction 2
2D-LC
a
1st dimension
2nd dimension
3rd dimensiona
0−5.7 min, 100% B1 5.71−30.99 min, stop 31−42 min, 100% A1 42.01−111 min, stop
0−6.2 min, 30% B2 6.21−30.2 min, 30−75% B2 30.21−30.99 min, 95% B2 31−42.6 min, 15% B2 42.61−111 min, 15−40% B2 0−1.1 min, 15% B2 1.11−69.5 min, 15−40% B2 69.51−93.5 min, 40−75% B2
0−0.8 min, 35−70% B3 0.81−1.0 min, 35% B3 0−0.8 min, 10−50% B3 0.81−1.0 min, 10% B3 0−0.8 min, 10−70% B3 0.81−1.0 min, 10% B3
Cycle time of the 3rd dimension is 1 min. Gradient conditions in one cycle are listed here.
LC Conditions of 3D-LC Method and Comparative 2DLC Method. Two CBM-20A controllers, two DGU-20A5 degassers, one SIL-30AC autosampler, one CTO-30A column oven, and six pumps (specific information is provided in the following) were equipped in the 3D-LC system. All LC instrument modules were purchased from Shimadzu (Kyoto, Japan). Mixer 1 (500 μL) and mixer 2 (100 μL) were bought from Shimadzu. Three two-way multiport valves were purchased from Valco (Houston, TX). For the 3D-LC method, a LC-20AB pump and an Acquity BEH Amide column (100 mm × 2.1 mm, 1.7 μm, Waters, Milford, MA) were used in the first dimension to prefractionate samples. The mobile phase A1 was acetonitrile/water (40/60, v/v) with 10 mM ammonium formate, and B1 was acetonitrile/ water (95/5, v/v) with 10 mM ammonium formate. The flow rate of the first dimension was 0.15 mL/min. In the second dimension, two LC-20AD pumps and an Acquity CSH FluoroPhenyl column (50 mm × 2.1 mm, 3.5 μm, Waters, Milford, MA) were used. The mobile phase A2 was water with 0.1% formic acid, and B2 was acetonitrile with 0.1% formic acid. The flow rate of the second dimension was 0.025 mL/min. In the third dimension, two LC-30AD pumps and a SB-C18 column (50 mm × 2.1 mm, 1.8 μm, Agilent, Santa Clara, CA) were used. The column temperature was set at 90 °C. The mobile phase A3 was water with 0.1% formic acid, and B3 was methanol with 0.1% formic acid. The flow rate of the third dimension was 1.0 mL/min. This flow was split at the ratio of 1:1 before entering into the MS. Gradient conditions in all the three dimensions are listed in Table 1. Two serially coupled Acquity BEH C18 columns (5 mm × 2.1 mm, 1.7 μm, Waters, Milford, MA) were used as trap column 1 to collect the fractions from the first dimension. The makeup flow (water) was introduced by a LC-20AD pump at a flow rate of 2.0 mL/min. The mixed flow was split at a ratio of 1:1 before trap column 1. Trap column 2 and trap column 3 were single Acquity BEH C18 columns (5 mm × 2.1 mm, 1.7 μm, Waters, Milford, MA), and the mixed makeup flow rate was 0.5 mL/min. For analysis of soybean extract, the injection volume was 10 μL. Two fractions, which contain weakly polar compounds, and medium and strong polar compounds, were produced through the first dimensional separation. Detailed information on flow scheme of the 3D-LC system and soybean extract separation is described in the Results and Discussion section. As a comparison, a corresponding comprehensive 2D-LC method was developed. The LC instrument, columns, and mobile phases were the same as those of the second and third dimensions in the 3D-LC method. Gradient conditions of this comprehensive 2D-LC method are also listed in Table 1. A 5 μL portion of soybean extract was injected.
redundant proteins were identified in the traditional 2D-LC method.18 These results demonstrated the superiority of the 3D-LC approach. In addition, several solid phase extraction (SPE)-2D-LC methods have also been studied.20,21 However, offline approaches request manual collection of the effluent components, which needs more time and labor and limits their practicability. Online 3D-LC methods, collecting the components automatically, are mainly centered on targeted analysis of trace compounds in complex matrix.22−24 Stoll et al. developed selective 3D-LC methods based on a valve-switching mode, and four targeted compounds were analyzed in three different matrices.22 The 3D-LC setups with serially coupled columns are easily achieved for profiling analysis.25,26 However, this approach is only suitable for limited analytes with distinct structural features, such as proteins. In addition, several LC coupled with multidimensional gas chromatography (MD-GC) systems have been reported.27,28 However, only volatile compounds can be covered with these systems, and the syringe-based LC−GC interface is unsuitable to thermally labile compounds. Up to now, no universal online 3D-LC system has been reported for untargeted analysis of complex samples. In this study, a novel online 3D-LC system was designed and constructed by coupling of prefractionation in the first dimension with LC × LC in the second and third dimensions, aiming at providing a universal strategy for the analysis of very complex samples. With this 3D-LC system, complex samples were initially fractionized into several fractions through the first dimensional preseparation, and then each fraction was orderly reinjected into the following comprehensive 2D-LC part for further analysis by using a stop-flow interface. As a proof-ofprinciple study, an amide−fluorophenyl × C18 system was established to analyze a soybean extract. To our knowledge, this online 3D-LC system is reported for the first time.
■
EXPERIMENTAL SECTION Reagents. HPLC grade acetonitrile and methanol were purchased from Merck (Darmstadt, Germany). Water was prepared by a Milli-Q system (Millipore, Bedford, MA). Ammonium formate and formic acid were purchased from Sigma-Aldrich (St. Louis, MO) and J&K (Beijing, China), respectively. Sample Preparation. Soybean samples were ground into powder and passed through a 60-mess sieve in advance. For extraction, 60 mg of soybean powder was weighed into a 2 mL Eppendorf tube. Then, 1.5 mL of methanol/water (4:1, v/v) was added, and the mixtures underwent ultrasound extraction for about 1 h. Next, centrifugation at 14 000 rpm and 4 °C was carried out for 15 min. Finally, 1.4 mL of supernatant was lyophilized at vacuum conditions. Before injection, the extract was redissolved using 400 μL of methanol/water (4:1, v/v). B
DOI: 10.1021/acs.analchem.6b04401 Anal. Chem. XXXX, XXX, XXX−XXX
Technical Note
Analytical Chemistry MS Conditions. Ion trap (IT)-TOF MS (Shimadzu, Kyoto, Japan) with electrospray ion (ESI) source was used as the detector. Data were acquired in positive mode. Nitrogen (N2) was used as the nebulizing gas at the flow rate of 1.5 L/min. Both the temperatures of CDL and heat block were set at 220 °C. The flight tube temperature was constant at 40 °C. The interface voltage and detector voltage of TOF analyzer were set at 4.5 and 1.77 kV, respectively. High-purity argon was used as collision gas and ion cooling gas. The MS full scan was started from m/z 150 to 1000. The loop time was 0.1 s. For MS2 and MS3, experiments were adopted with precursor ion selection mode, the CID energy was 50%, and collision gas was 50%. Frequency (q) was set at 0.251 (45.0 kHz).
noteworthy that the number of fractions is dependent on the inherent characteristics and complexity of samples. As shown in Figure 1A, all three valves are switched to the loading position at the beginning. At this position, the fractions are generated through the first dimensional preseparation and collected in trap column 1. In order to refocus the compounds on the top of trap column 1, a makeup flow is introduced by valve 1 and mixed with the first dimensional effluent. During the loading procedure, column equilibration is being carried out in the comprehensive 2D-LC part. After one fraction is completely collected, all valves are switched to the elution position (Figure 1B). At this position, the first dimensional flow rate is stopped, and the whole first dimensional system remains airtight. In the meantime, the fraction collected on trap column 1 is eluted by the second dimensional mobile phases and reinjected into the comprehensive 2D-LC part. Backflush pattern is used in the elution procedure of this 3D-LC system to refocus the components and to narrow injection bands.5 In the comprehensive 2D-LC part, an indirect transfer interface with two trap columns is applied to connect the second and the third dimensions. Similarly, the aforementioned makeup flow is introduced and mixed with the second dimensional effluent at this position to help capture the analytes in each second dimensional cut on the trap column 2 or 3. During this procedure, valve 3 is switched periodically according to the third dimensional cycle time. After the second and the third dimensional separations of one fraction are completely finished, all the valves are switched back to the loading position, and the first dimensional separation is restarted to produce sequentially the next fraction. These loading and elution procedures are alternately conducted until all the first dimensional fractions are thoroughly separated. The resolution and peak capacity of this 3D-LC system can be significantly increased as a result of the first dimensional preseparation and multiple comprehensive 2D-LC analysis. Moreover, various types of columns including HILIC, RPLC, NPLC, SAX, etc. can be chosen in the three dimensions of the system on the basis of the features of complex samples, the orthogonality of separation mechanisms, and the compatibility of mobile phases. Application of the Online 3D-LC System. In the present research, the separation of a soybean extract was taken as a proof-of-principle experiment to further demonstrate the powerful capacity of the online 3D-LC system. The soybean extract contains many compounds with greatly varied polarities, for example, nonpolar lipids, weakly polar aglycones, and polar glycosylated compounds including flavonoids, saponins, oligosaccharides, etc. An amide column was used as the analytical column in the first dimension. Through the first dimensional preseparation, the complicated soybean extract was simplified into two fractions according to the polarities of components. Nonpolar and weakly polar components including lipids, daidzein, genistein, glycitein, etc., which were very weakly retained on the amide column, were first eluted under isocratic conditions (acetonitrile/water, 95/5, v/v) and collected as fraction 1. After the analysis of the first fraction on the second and third dimensions was finished, fraction 2, which contains mainly medium and strong polar components, was eluted under the condition of acetonitrile/water (40/60, v/v). In order to capture these components effectively, three columns with different stationary phases were investigated as trap column 1. Specifically, the fluorophenyl column (50 mm × 2.1 mm, 3.5 μm) was demonstrated to be inappropriate because
■
RESULTS AND DISCUSSION Design of the Novel Online 3D-LC System. For samples of extremely high complexity, the separation with traditional 2D-LC methods may be not enough, and the increment of system dimensionality is necessary for further improving peak capacity and resolution.29 On this basis, we designed a fully automatic 3D-LC system by online coupling of LC preseparation with LC × LC. As shown in Figure 1, this
Figure 1. Flow scheme of the established 3D-LC system. Position A is loading position, and B is elution position.
combination is implemented using a stop-flow interface to overcome the challenges associated with the long analytical time of the comprehensive 2D-LC part. Extremely complex samples are divided into several fractions through the first dimensional preseparation so that the complexity of each fraction can be obviously simplified and the difficulty of the following LC × LC analysis is reduced. Moreover, the conditions of the comprehensive 2D-LC part can be adjusted according to the detailed properties of each fraction; thus, the peak capacity and resolution will be improved remarkably. It is C
DOI: 10.1021/acs.analchem.6b04401 Anal. Chem. XXXX, XXX, XXX−XXX
Technical Note
Analytical Chemistry
Figure 2. TIC chromatograms of soybean extract acquired with 3D-LC method. Parts A and B show TIC chromatograms of fractions 1 and 2, respectively.
considerable components flowed through and the breakthrough time was very short even if a large ratio of makeup flow was introduced. Carbon column (33 mm × 2.1 mm, 5 μm, United Science Corp.) had a very strong hydrophobic interaction and presented a high trap efficiency. Unfortunately, the elution of the trapped compounds was very difficult when acetonitrile/ water was used as the second dimensional mobile phases. Finally, a C18 column, which had a compromise performance in the collection and elution procedures, was adopted as trap column 1. As aforementioned, the flow rate of the makeup phase was a predominant parameter in the collection procedure. Optimization procedures of this parameter have been reported previously.30 Briefly, the breakthrough points of hesperidin with relatively strong polarity were measured by using a traditional 1D-LC system with the trap column 1 as the analytical column. With the flow rate of the mobile phase B1 fixed at 0.15 mL/min, candidate makeup flow rates ranging from 0.5 to 2.0 mL/min were investigated. Finally, the makeup flow rate was optimized at 2.0 mL/min when the breakthrough time was longer than the optimized collection time. Unfortunately, the total pressure of the first dimensional column and trap column 1 at this flow rate exceeded the pressure limit of the first dimensional LC system. Therefore, a splitter with a ratio of approximately 1:1 was utilized before trap column 1. The RPLC × RPLC system is one of the most commonly used comprehensive 2D-LC systems owing to favorable peak capacity, column efficiency, and mobile phases’ compatibility.31 This system was employed in the comprehensive 2D-LC part of the 3D-LC method. With the consideration that many compounds in the soybean extract contain aromatic groups, a fluorophenyl column was utilized in the second dimension. Besides hydrophobic interactions, hydrogen bond interactions and π−π interactions also contributed to separation. With the C18 column used in the third dimension, the fluorophenyl × C18 system was established in the comprehensive 2D-LC part. Targeted gradient conditions were optimized according to the properties of each fraction. As listed in Table 1, the gradients with a relatively high percentage of organic solvent were applied to the separation of fraction 1, which consisted of mostly weakly polar compounds, whereas the gradients with a relatively low percentage of organic solvent were applied to fraction 2 because mainly strong polar compounds were in this fraction. The total ion current (TIC) chromatograms of
fraction 1 and fraction 2 are shown in Figure 2A,B, respectively. To show the separation more clearly, zoom-in TIC contour plots of 71 min cut (A) and 88 min cut (B) are presented in Figure 3. At least 7 and 5 peaks are apparently distinguished
Figure 3. Zoom-in TIC chromatograms of 71 min (A) and 88 min (B) cuts. Arabic figures labeled in the TIC chromatography of 88 min cut stand for identified flavonoids. Peak 1: daidzein O-hexosidemalonylated IV (compound 30) and glycitein O-hexosidemalonylated IV (compound 50). Peak 2: daidzein O-hexosidemalonylated V (compound 31). Peak 3: glycitein O-hexosidemalonylated V (compound 51).
and well-distributed during the gradient time of 40 s, which demonstrates the excellent separation power of this 3D-LC system. As one of the most important functional components in the soybean extract, flavonoids were characterized on the basis of their molecular weight, MS2 and MS3 information after being well-separated by the 3D-LC method. On the basis of the rules of fragmentation patterns, which have been summarized in the previous research,32,33 83 flavonoids were identified in the soybean extract (Table S1, Supporting Information). It was noteworthy that 30% more flavonoids were identified as compared to those from the comparative comprehensive 2DLC method, and most of them were the flavonoid isomers which could not be separated by the 2D-LC method. As is well-known, separation and recognition of isomers are always hard issues due to their similar structures and chromatographic behaviors. The established 3D-LC method displayed excellent ability to identify flavonoid isomers. As shown in Figure 3B, two pairs of isomers (daidzein Ohexosidemalonylated IV (compound 30, peak 1) and daidzein O-hexosidemalonylated V (compound 31, peak 2), and glycitein O-hexosidemalonylated IV (compound 50, peak 1) and glycitein O-hexosidemalonylated V (compound 51, peak 3)) are well-separated. Besides the 88 min cut, many flavonoid D
DOI: 10.1021/acs.analchem.6b04401 Anal. Chem. XXXX, XXX, XXX−XXX
Technical Note
Analytical Chemistry
untargeted analysis of complex samples to improve separation. Complex samples can be separated into several fractions through the first dimensional preseparation. Then, more accurate LC conditions can be adopted in the comprehensive 2D-LC part based on the properties of compounds in each fraction. As a result, the peak capacity and resolution of the 3DLC method are substantially improved when compared with a traditional comprehensive 2D-LC method. Moreover, as the 3D-LC system is convenient in operation and flexible in construction, it is universal and suitable for many kinds of complex samples, including proteins, plant extracts, biosamples, polymers, etc.
isomers were also identified in other cuts using the 3D-LC method as listed in Table S1. Discovery of the low-abundance components in complex samples is another challenge because they were easily disturbed by neighboring high-abundance compounds. In this study, kaempferol O-hexosidemalonylated III (compound 57, m/z 535.1072) and glycitein O-hexosidemalonylated III (compound 49, m/z 533.1274) had a very similar retention time. Extracted ion current (EIC) chromatograms of m/z 535.1072 acquired by the 3D-LC and the corresponding 2D-LC methods are presented in Figure 4. As shown in Figure 4A, the two
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b04401. Details of identified flavonoids with this 3D-LC system (PDF)
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. Phone: 0086-411-84379757. Fax: 0086-411-84379559. *E-mail:
[email protected]. Phone: 0086-411-84379530. Fax: 0086-411-84379559. ORCID
Xianzhe Shi: 0000-0001-9306-0130 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was supported by the foundations (21275141, 21375011, 21435006, and 21575142) from the National Natural Science Foundation of China and National Grand Project of Science and Technology (2016ZX08012002-003).
■
Figure 4. EIC chromatograms of kaempferol O-hexosidemalonylated (m/z 535.1072) acquired by a comprehensive 2D-LC method (A) and an established 3D-LC method (B). Peak 1 is the isotope of glycitein Ohexosidemalonylated III (compound 49). Peak 2 is kaempferol Ohexosidemalonylated III (compound 57).
REFERENCES
(1) Kalili, K. M.; Vestner, J.; Stander, M. A.; de Villiers, A. Anal. Chem. 2013, 85, 9107−9115. (2) Stephan, S.; Hippler, J.; Kohler, T.; Deeb, A. A.; Schmidt, T. C.; Schmitz, O. J. Anal. Bioanal. Chem. 2016, 408, 6545−6555. (3) Bang, D. Y.; Moon, M. H. J. Chromatogr A 2013, 1310, 82−90. (4) Wang, S.; Qiao, L.; Shi, X.; Hu, C.; Kong, H.; Xu, G. Anal. Bioanal. Chem. 2015, 407, 331−341. (5) Stephan, S.; Jakob, C.; Hippler, J.; Schmitz, O. J. Anal. Bioanal. Chem. 2016, 408, 3751−3759. (6) Stoll, D. R.; Carr, P. W. J. Am. Chem. Soc. 2005, 127, 5034−5035. (7) Dugo, P.; Herrero, M.; Giuffrida, D.; Kumm, T.; Dugo, G.; Mondello, L. J. Agric. Food Chem. 2008, 56, 3478−3485. (8) Wang, Y.; Lu, X.; Xu, G. J. Chromatogr A 2008, 1181, 51−59. (9) Vonk, R. J.; Gargano, A. F.; Davydova, E.; Dekker, H. L.; Eeltink, S.; de Koning, L. J.; Schoenmakers, P. J. Anal. Chem. 2015, 87, 5387− 5394. (10) Pursch, M.; Buckenmaier, S. Anal. Chem. 2015, 87, 5310−5317. (11) Qiao, X.; Song, W.; Ji, S.; Li, Y. J.; Wang, Y.; Li, R.; An, R.; Guo, D. A.; Ye, M. J. Chromatogr A 2014, 1362, 157−167. (12) Escrig-Domenech, A.; Beneito-Cambra, M.; Simo-Alfonso, E. F.; Ramis-Ramos, G. J. Chromatogr A 2014, 1361, 108−116. (13) Ma, J.; Hou, X.; Zhang, B.; Wang, Y.; He, L. J. Pharm. Biomed. Anal. 2014, 91, 24−31. (14) Bedani, F.; Kok, W. T.; Janssen, H. G. Anal. Chim. Acta 2009, 654, 77−84.
compounds were completely overlapped and could not be separated even when a comprehensive 2D-LC method was used. In addition, the concentration of compound 49 was much higher than that of compound 57 in the soybean extract. The great difference in peak intensity and the similarity in retention time made the molecular ion of compound 57 be completely overlapped by the isotope ion of compound 49. Fortunately, as shown in Figure 4B, compound 57 (peak 2) and compound 49 (peak 1) were very well-separated and easily identified with the established 3D-LC method because of dedicated gradient conditions in the second and the third dimensions of the 3DLC method. All of these results demonstrated that this 3D-LC system is advantageous and powerful in the analysis of complex samples.
■
CONCLUSIONS By coupling of the first dimensional preseparation and comprehensive 2D-LC part using a stop-flow interface, an online 3D-LC system was successfully established and used in E
DOI: 10.1021/acs.analchem.6b04401 Anal. Chem. XXXX, XXX, XXX−XXX
Technical Note
Analytical Chemistry (15) Li, D.; Schmitz, O. J. Anal. Bioanal. Chem. 2013, 405, 6511− 6517. (16) Jandera, P.; Hajek, T.; Cesla, P. J. Sep. Sci. 2010, 33, 1382−1397. (17) Cacciola, F.; Jandera, P.; Hajdu, Z.; Cesla, P.; Mondello, L. J. Chromatogr A 2007, 1149, 73−87. (18) Valeja, S. G.; Xiu, L.; Gregorich, Z. R.; Guner, H.; Jin, S.; Ge, Y. Anal. Chem. 2015, 87, 5363−5371. (19) Pernikarova, V.; Sedlacek, V.; Potesil, D.; Prochazkova, I.; Zdrahal, Z.; Bouchal, P.; Kucera, I. J. Proteomics 2015, 125, 68−75. (20) Xu, B.; Wang, F.; Song, C.; Sun, Z.; Cheng, K.; Tan, Y.; Wang, H.; Zou, H. J. Proteome Res. 2014, 13, 3645−3654. (21) Vertommen, D.; Van Roy, J.; Szikora, J. P.; Rider, M. H.; Michels, P. A.; Opperdoes, F. R. Mol. Biochem. Parasitol. 2008, 158, 189−201. (22) Simpkins, S. W.; Bedard, J. W.; Groskreutz, S. R.; Swenson, M. M.; Liskutin, T. E.; Stoll, D. R. J. Chromatogr A 2010, 1217, 7648− 7660. (23) Bailey, H. P.; Rutan, S. C.; Stoll, D. R. J. Sep Sci. 2012, 35, 1837−1843. (24) Moretton, C.; Nigay, H. Anal. Methods 2016, 8, 6573−6579. (25) Shen, Z.; Want, E. J.; Chen, W.; Keating, W.; Nussbaumer, W.; Moore, R.; Gentle, T. M.; Siuzdak, G. J. Proteome Res. 2006, 5, 3154− 3160. (26) Wei, J.; Sun, J.; Yu, W.; Jones, A.; Oeller, P.; Keller, M.; Woodnutt, G.; Short, J. M. J. Proteome Res. 2005, 4, 801−808. (27) Zoccali, M.; Tranchida, P. Q.; Mondello, L. Anal. Chem. 2015, 87, 1911−1918. (28) Sciarrone, D.; Panto, S.; Tranchida, P. Q.; Dugo, P.; Mondello, L. Anal. Chem. 2014, 86, 4295−4301. (29) Giddings, J. C. J. Chromatogr A 1995, 703, 3−15. (30) Wang, S.; Li, J.; Shi, X.; Qiao, L.; Lu, X.; Xu, G. J. Chromatogr A 2013, 1321, 65−72. (31) Stoll, D. R.; Li, X. P.; Wang, X. O.; Carr, P. W.; Porter, S. E. G.; Rutan, S. C. J. Chromatogr A 2007, 1168, 3−43. (32) Vukics, V.; Guttman, A. Mass Spectrom Rev. 2010, 29, 1−16. (33) Chang, Y.; Zhao, C.; Wu, Z.; Zhou, J.; Zhao, S.; Lu, X.; Xu, G. Electrophoresis 2012, 33, 2399−2406.
F
DOI: 10.1021/acs.analchem.6b04401 Anal. Chem. XXXX, XXX, XXX−XXX
