Generation of Ultrahigh Peak Capacity LC Separations via Elevated

Sir Alexander Fleming Building, Imperial College London, SW7 2AZ, London, UK. The use of a combination of ultraperformance liquid chromatography at âˆ...
14 downloads 0 Views 219KB Size
Anal. Chem. 2006, 78, 7278-7283

Generation of Ultrahigh Peak Capacity LC Separations via Elevated Temperatures and High Linear Mobile-Phase Velocities Robert S. Plumb,*,† Paul Rainville,‡ Brian W. Smith,‡ Kelly A. Johnson,‡ Jose Castro-Perez,‡ Ian D. Wilson,§ and Jeremy K. Nicholson†

Waters Corporation, Milford, Massachusetts 01757, Department of Drug Metabolism and Pharmacokinetics, AstraZeneca, Mereside, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK, and Biological Chemistry, Division of Biomedical Sciences, Sir Alexander Fleming Building, Imperial College London, SW7 2AZ, London, UK

The use of a combination of ultraperformance liquid chromatography at ∼11 000 psi on sub 2-µm particles combined with reversed-phase gradient chromatography at a temperature of 90 °C is described as applied to the analysis of endogenous and drug metabolites in human and animal urine. By using elevated temperatures, back pressures can be reduced while maintaining high flow rates and chromatographic efficiency, with peaks 1-3 s wide at the base. Application to urine samples provided a peak capacity of ∼700 for a 10-min analysis and greater than ∼1000 in 1 h. Despite the narrow nature of the peaks, good quality mass spectra were also obtained, allowing the identification of typical drug and endogenous metabolites. These ultra-high-resolution chromatograms should be ideal for the analysis of complex samples in, for example, metabolite identification, impurity identification, and metabonomic/metabolomic studies. Applications in natural product analysis and proteomics can also be envisaged. The production of higher resolution chromatographic separations has been the aim of the separation scientist since the advent of the technique to characterize or quantify a specific analyte, in the shortest possible time.1 Chromatographic resolution can be improved by either increasing the column length or reducing the particle size. Of these two options, reducing the particle size is by far the most effective.2 Thus, increasing the column length increases analysis time while reducing particle size reduces it. However, the use of small particles results in higher operating pressures, as the column back pressure is proportional to the square of the particle size (P ≈ dp2). Despite this limitation 1.5µm nonporous stationary phases have been used to separate peptides, vitamins, proteins, and protein fragments.3,4 MacNair et al. demonstrated the use of long capillary columns (in excess of * Corresponding author. E-mail: [email protected]. † Imperial College London. ‡ Waters Corp. § AstraZeneca. (1) Chen, H.; Horvath, C. S. J. Chromatogr. 1995, 705, 3. (2) Halasz, I.; Endele, R.; Asshauser, J. J Chromatgr. 1995, 112, 37. (3) Ohmacht, R.; Kiss, I. Chromatographia 1996, 42, 595. (4) Ohmacht, R.; Boros, B.; Kiss, I.; Jelinek, L. Chromatographia 1999, 50, 75.

7278 Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

40 cm) packed with 1-µm stationary phases to produce over 200 000 theoretical plates in just 6 min,5 and Shen et al. used a capillary 87 cm in length and an internal diameter of 15-75 µm, packed with a 3-µm porous silica stationary phase6 achieving a peak capacity in excess of 1000 (in 5 h) for a cellular digest. The application of a commercially available ultra-high-pressure chromatography system with sub 2-µm porous silica, at pressures of up to 15 000 psi, described by Mazzeo et al,7 producing up to 8-fold improvement in sensitivity, a 1.4-fold increase in resolution, and a 9-fold increase sample in throughput. Since this initial work, there have been a number of applications exploiting this new capability to pharmaceuticals8,9(including impurity profiling8), bioanalysis,10-11 analysis of seized drugs,13 sports doping screening,14 food stuffs,15 drug metabolism,16-19and global metabolite profiles in metabonomic studies.20-22 Thus, in previous work, Ultra Performance LC reduced analysis times by a factor of 10, while (5) MacNair, J. E.; Lewis, K. C.; Jorgenson, J. W. Anal. Chem. 1997, 69, 983. (6) Shen, Y.; Zhao, R.; Berger, S. J.; Anderson, G. A.; Rodriguez, N.; Smith, R. D. Anal. Chem. 2002, 16, 4235-49. (7) Mazzeo, J.; Neue, U.; Kele, M.; Plumb, R. Anal. Chem. 2005, 77, 460A67A. (8) Wren, S. A. J. Pharm. Biomed. Anal. 2005, 15, 38, 337-43. (9) Wren, S. A.; Tchelitcheff, P. J. Pharm. Biomed. Anal. 2006, 24, 40, 57180. (10) Li R.; Dong, L.; Huang, J. Anal. Chem. Acta 2005, 546, 167-73. (11) Shen, J. X.; Wang, H.; Tadros, S.; Haynes, R. N. J. Parma BioMed. Anal. 2006, 78, 689-706. (12) Lurie. I. S. J. Chromatogr., A 2005, 1100, 168-75. (13) Kaufmann, A.; Butcher, P. Rapid Commun. Mass Spectrom. 2005, 30, 3694700. (14) Nielen, M. W. F.; Bovee, T. F. H.; van Engelen, M. C.; Rutgers, P.; Hamers, A. R. M.; van Rhijn, J. A.; Hoogenboom, L. A. P. Anal. Chem. 2005, 78, 424-31. (15) Kaufmann, A.; Butcher, P. Rapid Commun. Mass Spectrom. 2005, 30, 19, 3694-300. (16) Plumb, R. S.; Castro-Perez, J. M.; Granger, J. H.; Beattie, I.; Joncour, K.; Wright, A. Rapid Commun. Mass Spectrom. 2004, 18 (19), 2331-7. (17) Castro-Perez, J. M.; Plumb, R. S.; Granger, J. H.; Beattie, I.; Joncour, K.; Wright, A. Rapid Commun. Mass Spectrom. 2005, 19, 843-8. (18) Wrona, M.; Mauriala, T.; Bateman, K. P.; Mortishire-Smith, R. J.; O’connor, D. Rapid Commun. Mass Spectrom. 2005, 19, 2597-602. (19) Plumb, R. S.; Granger, J. H.; C. L. Stumpf, K. A. Johnson, B. W. Smith, S Gaulitz, I. D. Wilson, Castro-Perez, J. Analyst 2005, 130, 844-9. (20) Wilson, I. D.; Plumb, R.; Granger, J.; Major, H.; Williams, R.; Lenz, E. M. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2005, 817, 67-76. (21) Jonsson, P.; Bruce, S. J.; Moritz, T.; Trygg, J.; Sjostrom, M.; Plumb, R. Granger, J.; Maibaum, E.; Nicholson, J. K.; Holmes, E.; Antti, H. Analyst 2005, 130 (5), 701-7. 10.1021/ac060935j CCC: $33.50

© 2006 American Chemical Society Published on Web 09/16/2006

maintaining the information density of HPLC, or increased the number of ions detected by a factor of 8 with the same analysis time.23 As well as increasing the back pressure to maintain flow rates, and thus minimize analysis time, raising the column temperature has also been advocated as a means achieving the same aim by decreasing back pressure. It has been suggested by other researchers that an added benefit of performing chromatography at elevated temperature is increased column performance resulting from improved mass transfer, resulting in sharper peaks.24 The effects of column back pressure and operating temperature were outlined by Knox in 1980,25 who noted that, as the column particle size is reduced the optimal linear velocity is increased by the power of two, while as the column temperature is increased the mobile-phase viscosity is reduced and the optimal linear velocity is increased. Thus, increasing the column temperature requires an increase in mobile-phase flow rate, somewhat negating the benefit of temperature on column back pressure. Thus, as the temperature is increased from 40 to 90 °C, the optimal flow, for a 2.1-mm-internal diameter column, increases from 0.6 to 1.1 mL/ min. As discussed by previous researchers,25 the overall chromatographic performance in terms of efficiency or peak capacity is not increased by increasing the temperature unless the separation is kinetically limited, but the number of peaks produced per minute increases. Thus, elevated temperatures enable the maximum performance to be achieved in a reduced time scale. Xiang et al. employed elevated column temperatures, high pressures, and 1-µm nonporous particles to produce very fast separations of barbitals.26 By elevating the temperature from 30 to 80 °C, the analysis time was reduced from 6.5 (689 bar) to 1.8 min; when the pressure was increased to 2400 bar, the separation was reduced to just 30 s with little loss in performance. These early activities in ultra-high-pressure separations employed capillary and nanoscale chromatography columns packed in-house and using specially adapted chromatography systems. An alternative to using these small particles for faster analysis is the use of monolithic column technology.27 The relative merits of these two technologies for the analysis of drug metabolites in biological fluids was described by Johnson and Plumb;28 in this study, the sub 2-µm particles showed significant benefits, reducing analysis times and increasing assay sensitivity. Given the benefits to resolution conferred by small particle size and high operating pressure or small particle size and high temperature, there is an obvious potential benefit from combining both small particles and high pressures and high column temperatures (which might also include the use of more exotic mobile phases, such as ethanol, isopropyl alcohol, and dimethyl sulfoxide, whose use is normally precluded due to their inherent viscosity and hence high back pressure). Here we investigate the benefits (22) Plumb, R. S.; Johnson, K. A.; Rainville, P.; Smith, B. W.; Wilson, I. D.; CastroPerez, J. M.; Nicholson, J. K. Rapid Commun. Mass Spectrom. 2006, 20, 1989-94. (23) I.D.; Wilson, J. K.; Nicholson, J.; Castro-Perez, J. H.; Granger, K. A.; Johnson, B. W.; Smith, R. S.; Plumb. J. Proteome Res. 2005, 4 (2), 591-8. (24) Colin, H.; Diez-masa, J. C. J. Chromatogr. 1988, 435, 1. (25) Knox, J. J. Chromatgr. Sci. 1980, 18, 453. (26) Xiang, Y.; Liu, Y.; Lee, M. L. J. Chromatogr. 2006, 1104, 198-202. (27) Preinerstorfer, B.; Lubda, D.; Lindner, W.; Lammerhofer, M. J. Chromatogr., A 2006, 1106 (1-2), 94-105. (28) Johnson, K. A.; Plumb, R.. I. J. Pharm. Biomed. Anal. 2005, 15, 39 (3-4), 805-10.

of small particle columns operated at pressures in the region of 11 000 psi and temperatures up to 90 °C for the analysis of endogenous and drug metabolites in human and animal urine. EXPERIMENTAL SECTION Chemicals. Optima grade acetonitrile (HPLC grade) was purchased from Fisher Scientific (Hampton, NH), and ammonium formate and formic acid (spectroscopic grade) were purchased from Sigma-Aldrich (St. Louis, MO). Distilled water was purified “in-house” using a MilliQ system (Millipore, Billerica, MA). Leucine-enkephalin was obtained from Sigma-Aldrich. Samples. Control rat urine was obtained from a male Wistarderived rat. Urine was also obtained from a rat for the time period 0-96 h following the administration of hydrazine (2 and 80 mg/ kg). Acetaminophen in human urine was obtained by the oral administration of 800 mg of acetaminophen to a healthy male volunteer. The urine was collected 2 h after dosing and stored frozen prior to analysis. Chromatography. Chromatographic separations were performed on a 2.1 × 150 mm ACQUITY 1.7-µm C18 column (Waters Corp., Milford, MA) using an ACQUITY Ultra Performance Liquid Chromatography system (Waters). The column was maintained at 90 °C and eluted with a linear gradient of 0-95% B, where A is 0.1% formic acid and B is acetonitrile/0.1% formic acid. The gradient duration was either 10 or 60 min with a flow rate of 0.8 mL/min. The column eluent was directed to the mass spectrometer. A 10-µL injection of each sample was made onto the column. Mass Spectrometry. Mass spectrometry was performed on a Waters Micromass LCT Premier (Waters MS Technologies, Manchester, UK) orthogonal acceleration time-of-flight mass spectrometer operating in positive ion mode. The nebulization gas was set to 800 L/h at a temperature of 350 °C, the cone gas set to 10 L/h, and the source temperature set to 120 °C. A capillary voltage and a cone voltage were set to 3200 and 60 V, respectively. The LCT Premier was operated in W optics mode with 12 000 resolution (fwhm). The data acquisition rate was set to 0.095 s, with a 0.005-s interscan delay; data were collected for 12 or 65 min. All analyses were acquired using the lock spray to ensure accuracy and reproducibility; leucine-enkephalin was used as the lock mass (m/z 556.2771) at a concentration of 300 pg/µL and flow rate 30 µL/min. Data were collected in centroid mode from 100 to 1000 m/z with a lock spray frequency of 11s and data averaging over 10 scans. RESULTS AND DISCUSSION Endogenous Metabolite Profiling. Metabonomics has been defined as “Quantitative measurement of time-related multiparametric metabolic responses of multicellular systems to pathophysiological stimuli or genetic modification”.29,30 The global metabolite profiles needed for this type of work have been obtained using analytical techniques such as proton NMR,31,32 GC/ MS,33 and LC/MS.34 Often large sample numbers are involved in (29) Nicholson, J.; Lindon, J.; Holmes, E. Xenobiotica 1999, 29, 1181-9. (30) Nicholson, J.; Connelly, J.; Lindon, J.; Holmes, E. Nat. Rev. Drug Discuss. 2002, 1, 153-61. (31) Nicholson, J.; Wilson, I. Prog. NMR Spectrosc. 1989, 21, 449-501. (32) Robertson, D.; Reilly, M.; Lindon, J.; Holmes, E.; Nicholson, J. K. Comprehensive Toxicol. 2002, XIV, 585-610. (33) Fiehn, O. Plant Mol. Biol. 2002, 48, 155-71.

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

7279

Figure 1. TIC of a reversed-phase gradient UPLC separation of rat urine on a 2.1 × 150 mm, 1.7-µm Acquity BEH C18 column over 10 min at 90 °C.

metabonomic studies, and where chromatographic assays are performed, there is always a need to balance the need for high throughput versus long analysis times to recover the maximum amount of information from each sample. Reversed-phase gradient UPLC/MS has been used on plasma and urine samples to generate peak capacities in the region of 300 in 10 min,23 representing a considerable improvement on conventional HPLC. To examine the effects of analysis at elevated temperature, a sample of rat urine, from a rat administered with hydrazine at 80 mg/kg (day 2), was analyzed using a 2.1 × 150 mm 1.7-µm porous alkyl-bonded chromatography column maintained at 90°C and eluted with a linear organoaqueous gradient over 10 min at a flow rate of 800 µL/min as described in the Experimental Section. These conditions resulted in a starting back pressure of 11 000 psi. The chromatogram obtained by this procedure is shown in Figure 1 and shows myriad very sharp peaks. A close inspection of the peaks showed that the average peak width, measured at the base, was just 1.2 s (averaged across the whole chromatogram time domain), resulting in a peak capacity of 720 for the 10-min analysis. The extremely sharp nature of these peaks required the use of a very high data acquisition rate in the mass spectrometer, and on this occasion, an acquisition time of 95 milliseconds was employed with a 5-ms interscan delay, providing ∼9-10 points across each peak. The combined extracted ion chromatogram was produced for the common endogenous metabolites, hippuric acid (m/z 180.0661), kynurenic acid (m/z 190.0504), xanthenuric acid (m/z 206.0453), (34) Granger, J.; Plumb, R.; Castro-Perez, J. I. D. Chromatographia, 61 2005, 7/8, 375-80.

7280

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

and pantothenic acid (m/z 220.1185) as shown in Figure 2. The mass spectra of these compounds are shown in Figure 3. From these data, we can see that the mass accuracy for each of these four compounds was within 3.3 ppm. This demonstrates that it is possible to operate the mass spectrometer at a sufficiently high data capture rate to give a true representation of the LC performance and still obtain acceptable mass accuracy data. The retention times of these ions were determined to be 0.75, 0.93, 1.19, and 2.09 min for xanthenuric acid, kynurenic acid, hippuric acid, and pantothenic acid, respectively. These retention time data can be compared to those obtained for the same ions using the same gradient profile on with a 2.1 × 100 mm UPLC column operated at 40 °C (at 600 µL/min generating a back pressure of 11 000 psi), where the retention times were 1.13, 1.83, 1.68, and 4.72 min, respectively, for the same analytes (data not shown). Thus, one obvious consequence of using elevated temperatures is that analytes are less retained than at lower temperatures. Care must therefore be taken when developing methods as there is clearly the potential for some highly polar analytes to be unretained on the column, and thus, potentially important information could be lost. It should also be noted that that the flow rate employed was 81% of the optimal flow rate. However, this flow rate was employed as the highest practical flow rate that could be employed with the electrospray MS setup used here. By using the (higher) optimal flow rate, it should thus be possible to generate even higher peak capacity separations than this. When the run time was increased from 10 to 60 min, the results were as shown in the chromatogram illustrated in Figure 4. This system produced a high-resolution chromatogram, with peak

Figure 2. Extracted ion chromatogram of xanthuric acid (m/z 206), kyurenic acid (m/z 190), hippuric acid (m/z 180), and pantothenic acid (m/z 220) obtained from the UPLC separation shown in Figure 1.

Figure 3. Spectra of kyrurenic acid, hippuric acid, pantothenic acid, and xanthurate from Figure 2.

widths in the order of 3-4 s, measured at the base, with an average of 3.5 s/peak giving a potential peak capacity of over 1000 for the 1-h separation. This kind of high resolution is ideal for the challenging task of metabolic profiling and compares favorably with the data generated by Shen et al.,6 giving a similar peak capacity in a much reduced time scale.

Drug Metabolite Detection and Identification. The identification of drug metabolites remains a demanding analytical challenge especially for very potent, low-dose compounds and their metabolites present in complex matrixes such as urine, plasma, bile, or feces. Previously we applied UPLC at 40 °C to advantage for the analysis of acetaminophen and its metabolites in human Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

7281

Figure 4. TIC of a reversed-phase gradient separation of rat urine on a 2.1 × 150 mm, 1.7-µm Acquity BEH C18 column over 60 min at 90 °C.

Figure 5. TIC of a reversed-phase gradient UPLC separation of acetominophen metabolites in human urine separated on a 2.1 × 150 mm, 1.7-µm Acquity BEH C18 column over 10 min at 90 °C.

urine.28 In Figure 5, the TIC chromatogram, for the analysis of a human urine sample following acetaminophen administration, obtained for a 10-min separation at 90 °C is shown. As seen with the earlier example for endogenous metabolite profiling, a highresolution separation was achieved with a peak capacity in excess of 700. The extracted ion chromatograms for the glucuronide (m/z ) 328.1035) is shown in Figure 6. From these data we can see that the peaks are extremely sharp; there are several peaks displayed two of which, eluting at ∼1 and ∼1.25 min, respectively, are acetaminophen related, as they give rise to a product ion of m/z of 152.071, which corresponds to the aglycon. These results 7282

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

are similar to our previously reported experiments.28 A mass spectrum obtained from the major glucuronide peak eluting at 1.25 min is displayed in Figure 6. The exact mass analysis of both of the glucuronide peaks gave a value of 328.1041, which gave an elemental composition of C14H18NO8 with a mass error of 2.6 ppm. Similarly, we were able to detect the unchanged parent compound, several sulfate conjugates, and other metabolites as seen in our previous work28 (data not shown). The detection of the same metabolites in this high-temperature run as seen at lower temperature, in the same peak ratios provides confidence that in this instance that there was no metabolite degradation during the

Figure 6. Extracted ion chromatogram of acetaminophen-glucuronide metabolites in human urine from Figure 5.

analysis (although analyte degradation has been observed for some compounds at high temperature (e.g., see ref 35), and care must clearly be taken when attempting to analyze sensitive compounds). As indicated for the endogenous metabolite profiling example above, the optimal linear velocity for this 10-min separation was not used for reasons associated with the mass spectrometry and it may be possible to obtain even sharper peaks and more resolution still. The use of 1-mm-internal diameter columns would alleviate this issue, and the application of these smaller diameter columns will be addressed in subsequent work. One of the major issues when operating at these higher temperatures is the lifetime of the analytical columns, where the reduction in column performance is mainly due to stationary hydrolysis. The application of a hybrid organosilica stationary phase increases the column stability at higher pH’s while the use of a difunctional bonded ligand also increases column stability. In this work and others performed by the authors, column lifetime of the bridged ethyl (35) Smith, R. M.; Chienthavorn, O.; Saha, S.; Wilson, I.; Wright, B, D.; Taylor, S. D. J. Chromatogr., A 2000, 886, 289-95.

hybrid (BEH) ACQUITY column has not been an issue at these temperatures of 90 °C and above. Indeed, in a recent study, over 1700 injections of precipitated plasma were made onto an Acquity BEH column operated at 90 °C with no loss in performance or retention (unpublished data). CONCLUSIONS As these applications show, the combination of narrow-bore LC, packed with a sub 2-µm porous stationary phase, operated at both high pressure and high temperature (90 °C) can produce extremely high-resolution chromatograms. Such high-resolution separations are ideal when applied to complex biological samples and many applications in, for example, xenobiotic metabolite identification, drug impurity profiling, natural product, and metabonomic/metabolomic and proteomics can be envisaged.

Received for review May 19, 2006. Accepted August 16, 2006. AC060935J

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

7283