Technical Note pubs.acs.org/ac
Measuring Complete Isotopomer Distribution of Aspartate Using Gas Chromatography/Tandem Mass Spectrometry Jungik Choi, Matthew T. Grossbach, and Maciek R. Antoniewicz* Department of Chemical and Biomolecular Engineering, Metabolic Engineering and Systems Biology Laboratory, University of Delaware, Newark, Delaware 19716, United States S Supporting Information *
ABSTRACT: We have developed a simple and accurate method for determining the complete positional isotopomer distribution of aspartate carbon atoms by gas chromatography/tandem mass spectrometry for 13C-metabolic flux analysis. First, we screened tandem mass spectrometry (MS) spectra of the tert-butyldimethylsilyl (TBDMS) derivative of aspartate for daughter fragments with the necessary carbon atom fragmentations to fully resolve all 16 isotopomers of aspartate. Tandem MS scanning parameters were optimized for each daughter fragment, and the accuracy of tandem MS measurements were evaluated. We selected five accurate fragments that provided a redundant set of 47 labeling measurements to quantify the complete isotopomer distribution of aspartate by least-squares regression. The validity of the approach was demonstrated using six 13C-labeled aspartate standards and natural aspartate. n the past decade, 13C-metabolic flux analysis (13C-MFA) has emerged as a powerful method for measuring intracellular metabolic fluxes in living cells in fields ranging from metabolic engineering to biomedical sciences.1−4 In 13C-MFA, an isotopic tracer (e.g., [1,2-13C]glucose) is introduced to a biological system where it is metabolized by cells resulting in the incorporation of 13C-atoms into intracellular metabolites, macromolecules, and metabolic products.5 The amount of 13Clabeling and positional distribution of 13C-atoms are then assessed using techniques such as gas chromatography/mass spectrometry (GC/MS), liquid chromatography mass spectrometry (LC−MS), and nuclear magnetic resonance (NMR).6−9 Metabolic fluxes are finally calculated from these labeling measurements using a model-based approach that maximizes the fit between the measured and model predicted labeling distributions.10 To determine accurate and precise metabolic fluxes it is important that (i) a proper selection of 13C-tracers is made for the given metabolic network model of interest,11−13 (ii) as much information as possible is obtained about the incorporation of 13C-atoms and the positional distribution of 13 C-atoms;14 (iii) appropriate computational tools are available to simulate isotopomer measurements and fit the labeling data to a given metabolic network model, followed by statistical analysis of confidence intervals of fluxes.15,16 Issues associated with proper selection of isotopic tracers and fitting of labeling data to metabolic network models have been largely resolved in recent years using techniques based on elementary metabolite units (EMU).11,17,18 Currently, the main challenge limiting the estimation of metabolic fluxes at a high resolution stems from the difficulty in measuring complete isotopomer distributions of metabolites. MS and NMR based techniques provide only
I
© 2012 American Chemical Society
partial information on labeling distributions. MS based techniques quantify mass isotopomer distributions, which are useful for determining total enrichment but provide only limited information on the position of 13C-atoms. In contrast, NMR based techniques give useful information regarding local 13 C-labeling patterns at specific carbon atoms; however, because of low sensitivity it may be difficult to quantify isotopomers in low concentration metabolites where 13C-atoms are not adjacent, e.g., [1,4-13C]aspartate. Thus, there is a clear need for new techniques that can provide more detailed labeling information for 13C-MFA. In this work, we introduce for the first time a method for measuring the complete isotopomer distribution of a key intracellular metabolite, aspartate, using tandem mass spectrometry (MS/MS). Aspartate is the proteinogenic amino acid that is derived from oxaloacetate by transamination. Oxaloacetate is centrally located in the metabolism of mammalian and microbial cells at the intersection of several key metabolic pathways, including citric acid cycle, anaplerosis, gluconeogenesis, amino acid metabolism, and glyoxylate shunt.19,20 We have recently demonstrated the potential of using tandem MS measurements of aspartate for 13C-MFA.21 We showed in simulation studies that a single tandem MS fragment of aspartate can improve the precision of estimated metabolic fluxes by about 5-fold compared to MS measurements, e.g., ∼20% flux confidence intervals for MS measurements compared to ∼4% for MS/MS. Additionally, we determined that by measuring the complete isotopomer distribution of Received: March 1, 2012 Accepted: April 17, 2012 Published: April 17, 2012 4628
dx.doi.org/10.1021/ac300611n | Anal. Chem. 2012, 84, 4628−4632
Analytical Chemistry
Technical Note
aspartate the flux resolution could be further improved by 10fold. The goal of this study was to develop this method to fully determine all 16 isotopomers of aspartate carbon atoms using gas chromatography/tandem mass spectrometry (GC/MS/ MS).
First, theoretical TMIDs were calculated for the selected tandem MS fragments and for all 16 carbon atom isotopomers of aspartate. For calculating theoretical TMIDs, the following natural isotope abundances were used: 2H (0.0156 at %), 13C (1.082 at %), 15N (0.366 at %), 17O (0.038 at %), 18O (0.204 at %), 29Si (4.69 at %), 30Si (3.09 at %). To relate measured TMIDs to theoretical TMIDs, the following equation was used:21
■
EXPERIMENTAL SECTION Materials. [1-13C]Aspartate (99%), [2-13C]aspartate (99%), [3-13C]aspartate (99%), and [4-13C]aspartate (99%) were purchased from Isotec (Miamisburg, OH). [1,4-13C]Aspartate (99%) and [U-13C]aspartate (97%, algal amino acid mixture) were purchased from Cambridge Isotope Laboratories (Andover, MA). Stock solutions of aspartate standards were prepared at 25 mM in distilled water. Derivatization of Aspartate. For TBDMS-derivatization of aspartate, 50 μL of aspartate solution was evaporated to dryness under an airflow at 65 °C. The sample was then dissolved in 50 μL of pyridine followed by addition of 50 μL of N-methyl-N-(tert-butyldimethylsilyl)-trifluoroacetamide (MTBSTFA) + 1% tert-butyldimetheylchlorosilane (TBDMCS) (Thermo Scientific, Bellefonte, PA).22 Samples were then incubated for 30 min at 60 °C on a dry heating block to complete the derivatization. Tandem Mass Spectrometry Analysis. GC/MS and GC/MS/MS analyses were performed using an Agilent 7890A GC equipped with a DB-5 ms (30 m × 0.25 mm i.d. × 0.25 μm; Agilent J&W Scientific) capillary column, interfaced with a triple quadruple tandem mass spectrometer (Quattro Micro GC, Milford, MA) operating under ionization by electron impact at 70 eV. The injection port, interface, and ion source temperatures were kept at 250 °C. The injection volume was 1 μL, and samples were injected at 1:10 split ratio. Helium flow was maintained at 1 mL/min. The temperature of the column was started at 80 °C for 2 min, increased to 280 °C at 7 °C/ min, and held for 20 min. For GC/MS analysis, mass isotopomer distributions were recorded in selected ion recording (SIR) mode with a 30 ms dwell time. For GC/ MS/MS analysis, argon gas was used as the collision gas to achieve collision induced dissociation (CID). The collision gas pressure was kept at 3 × 10−6 bar, which was determined to be optimal for all daughter fragments, and the collision energy was optimized between 2 and 40 eV for each daughter fragment individually. Daughter spectra were recorded in multiple reaction monitoring (MRM) mode with 30 ms dwell time. Nomenclature. Aspartate isotopomers are denoted using a subscript notation with ones and zeros. As an example, Asp1100 denotes the isotopomer of aspartate where the first two carbon atoms are labeled (i.e., 13C) and last two carbon atoms are unlabeled (i.e., 12C). Mass isotopomers of parent fragments are denoted by M0, M1, etc.; and mass isotopomers of daughter fragments are denoted by m0, m1, etc. Tandem Mass Isotopomer Distributions (TMIDs). TMIDs were obtained as follows.21 First, parent mass isotopomer distributions were measured for two aspartate fragments at m/z 418 and m/z 390. Next, daughter spectra were measured for individual mass isotopomers from these parent fragments, as indicated in the text. Finally, TMIDs were calculated by multiplying the normalized daughter mass isotopomer distributions with the mass isotopomer fraction of the respective parent ion. Calculation of Complete Isotopomer Distribution of Aspartate. The complete isotopomer distribution of aspartate was determined by least-squares regression of tandem MS data.
theor. theor. TMID = x0000·TMID0000 + x0001· TMID0001 + ... theor. + x1111·TMID1111
(1)
Equation 1 illustrates that a measured TMID can be viewed as a linear combination of theoretical TMIDs, with the isotopomer fractions (x) acting as the weighting factors. Five of such equations can be written, one for each of the five selected tandem MS fragments described in this paper. All of these equations can be conveniently combined into a matrix formulation: theor.frag1 ⎡ TMIDfrag1 ⎤ ⎡ TMID0000 ⎢ ⎥ ⎢ theor.frag2 ⎢ TMIDfrag2 ⎥ = ⎢ TMID0000 ⎢... ⎥ ⎢... ⎢ ⎥ ⎢ ⎣ TMIDfrag5 ⎦ ⎢⎣ TMIDtheor.frag5 0000 ⎡ x0000 ⎤ ⎢x ⎥ ⎢ 0001 ⎥ ⎢... ⎥ ⎢⎣ x1111 ⎥⎦
theor.frag1 theor.frag1 ⎤ TMID0001 ... TMID1111 ⎥ theor.frag2 theor.frag2 ⎥ TMID0001 ... TMID1111 ⎥· ... ... ... ⎥ theor.frag5 theor.frag5 ⎥ TMID0001 ... TMID1111 ⎦
(2)
Or in short notation: (3)
T = N ·x
The matrix N, which contains the theoretical TMIDs for the selected tandem MS fragments, is full-rank (i.e., rank of 16) ; thus, the tandem MS measurements provide the required 16 independent constraints needed to uniquely quantify all 16 isotopomers of aspartate. The matrix N is given in the Supporting Information (Table S-1). To calculate the complete isotopomer distribution of aspartate, the following linear leastsquares regression problem was solved in Matlab R2008b (MathWorks, Inc.): min ∑ (T − T meas.)2 = min ∑ (N ·x − T meas.)2 s. t.
x ≥ 0,
∑x = 1
(4)
To solve the regression problem we used Matlab’s function lsqnonneg.
■
RESULTS AND DISCUSSION Screening of Aspartate Tandem MS Fragments. TBDMS-derivatized aspartate was analyzed by GC/MS/MS to identify daughter fragments that would allow the complete isotopomer distribution of aspartate to be quantified. We analyzed tandem MS spectra for two abundant parent fragments in the GC/EI-MS spectrum of aspartate, m/z 418 that contains C1-C4 of aspartate and m/z 390 that contains C2C4. We obtained tandem MS spectra for natural aspartate (with m/z 418 and 390 as parent ions) and for [1-13C], [2-13C], [3-13C], and [4-13C]aspartate (with m/z 419 and 391 as parent ions). The measured tandem MS spectra are given in the Supporting Information (Figures S-1 and S-2). For each parent fragment, we identified the most abundant daughter fragments and determined which carbon atoms were retained in each 4629
dx.doi.org/10.1021/ac300611n | Anal. Chem. 2012, 84, 4628−4632
Analytical Chemistry
Technical Note
data was used for least-squares regression: mass isotopomer distributions of aspartate parent fragments m/z 418 and m/z 390 (obtained by SIR) and daughter spectra for m/z 418 > 117 and m/z 418 > 346 (with M1 and M2 as parent ions), m/z 418 > 103 and m/z 390 > 346 (with M1, M2, and M3 as parent ions), and m/z 418 > 244 (with M1, M2, M3, and M4 as parent ions; obtained by MRM). The measured tandem mass isotopomer distributions are given in the Supporting Information (Table S-4). Using the regression method described in the Experimental Section, we fitted 47 tandem mass isotopomers to estimate 16 isotopomers of aspartate. As such, our regression method has 31 (= 47 − 16) redundant measurements. The use of redundant measurements has important advantages, e.g., it improves the precision of the estimated isotopomer distributions and helps in identifying potential measurement errors to ensure the high accuracy of our method. To assess the validity of our method for calculating aspartate isotopomers, we determined the complete isotopomer distributions for natural aspartate and six 13C-labeled aspartate standards: [1-13C], [2-13C], [3-13C], [4-13C], [1,4-13C], and [U-13C]aspartate. The estimated isotopomer distributions are shown in Table 3. The estimated values corresponded well with the expected values based on the manufacturers’ specifications. As an example, for [1,4-13C]aspartate, the estimated abundances of Asp1001, Asp0001, and Asp1000 were 97.1%, 1.1%, and 1.4%, respectively, compared to predicted values of 98%, 1%, and 1%, respectively, i.e., assuming the manufacturers’ specification of 99 atom % enrichment. The estimated abundances of aspartate isotopomers Asp1000, Asp0100, Asp0010, and Asp0001 for [1-13C], [2-13C], [3-13C], and [4-13C]aspartate were 97.7%, 97.8%, 97.2%, and 96.4%, respectively, i.e., compared to 99% expected. It is important to note that since we considered natural abundances for all atoms in the theoretical TMIDs (see the Experimental Section), the calculated isotopomer distributions shown in Table 3 are already corrected for all the natural isotope abundances, including the natural isotope abundance of 13 C. This is a useful feature of our method, as it allows easy interpretation of the calculated isotopomer distributions; 13Catoms incorporated from the tracer experiments are easily identified, and there is no need for data corrections before the data is used for 13C-MFA.
daughter fragment. The assignment of carbon atoms followed directly from the shift in m/z for [1-13C], [2-13C], [3-13C], and [4-13C]aspartate compared to natural aspartate. As an example, the daughter spectrum of natural aspartate contained an ion at m/z 117, while [1-13C], [2-13C], [3-13C], and [4-13C]aspartate produced peaks at m/z 117, 117, 118, and 118, respectively; thus, the m/z 117 daughter fragment retained carbon atoms C3-C4 of aspartate. On the basis of these assignments, we postulated chemical formulas for each daughter fragment, calculated theoretical TMIDs, and compared these to measured TMIDs. Fragments for which the measured and predicted TMIDs deviated more than 5 mol % were considered inaccurate and were not analyzed further. In total, we evaluated 10 daughter fragments for the m/z 418 parent fragment and 3 daughter fragments for the m/z 390 parent fragment (Supporting Information, Tables S-2 and S-3). We then selected the five most accurate fragments that provided sufficient information to quantify the complete positional isotopomer distribution of aspartate. The selected daughter fragments were m/z 103 (C1, C3H7O2Si), m/z 117 (C3-C4, C4H9O2Si), m/z 244 (C1-C2, C10H22O2NSi2), and m/z 346 (C2-C3, C15H36O2NSi3) for parent fragment m/z 418 and the daughter fragment at m/z 346 for parent fragment m/z 390 (Table 1). The collision energy was then optimized for each Table 1. Selected Tandem MS Fragments of Aspartatea tandem MS fragmentation (m/z) 418 418 418 418 390
> > > > >
103 117 244 346 346
carbon atoms
optimal collision energy (eV)
[1-2-3-4] > [1] [1-2-3-4] > [3-4] [1-2-3-4] > [1-2] [1-2-3-4] > [2-3] [2-3-4] > [2-3]
20 12 6 12 6
a
Fragments of TBDMS-derivatized aspartate: m/z 418 (C18H40O4NSi3), m/z 390 (C17H40O3NSi3), m/z 103 (C3H7O2Si), m/z 117 (C4H9O2Si), m/z 244 (C10H22O2NSi2), and m/z 346 (C15H36O2NSi3).
daughter fragment to maximize the intensity of the daughter fragment and the agreement between the measured and predicted TMIDs. Table 2 compares the measured and theoretical TMIDs for natural aspartate. The agreement was excellent for all five selected fragments, with less than 1 mol % deviation. Calculating Isotopomer Distribution of Aspartate. To calculate isotopomer distributions of aspartate, the following
■
CONCLUSIONS Accurate assessment of 13C-labeling is critical for determining intracellular metabolic fluxes using 13C-MFA. Thus far, most
Table 2. Comparison of Measured and Theoretical Tandem Mass Isotopomer Distributions (Molar Percent Abundances, mol %) tandem MS fragmentation (m/z)
418 > 103 418 > 117 418 > 244 418 > 346 390 > 346
parent
M0
M1
M1
M2
M2
M2
M3
M3
M3
M3
daughter
m0
m0
m1
m0
m1
m2
m0
m1
m2
m3
measured theory measured theory measured theory measured theory measured theory
63.2 63.3 63.2 63.3 63.2 63.3 63.2 63.3 64.2 64.1
17.4 17.4 16.9 16.7 8.8 9.0 1.7 2.2 1.4 1.5
5.2 5.4 5.7 6.1 13.8 13.8 20.9 20.7 20.8 20.9
7.0 6.7 7.6 6.6 3.0 2.9 0.6 0.3 0.8 0.1
1.4 1.5 1.2 1.6 2.0 2.0 0.5 0.7 0.3 0.5
2.4 2.5 2.0 2.6 5.9 5.8 9.7 9.7 9.3 9.8
1.3 1.2 1.4 1.1 0.5 0.3 0.4 0.0 0.6 0.0
0.5 0.6 0.6 0.6 0.6 0.6 0.5 0.1 0.1 0.0
0.7 0.7 0.4 0.7 0.8 0.8 0.2 0.3 0.1 0.2
0.1 0.1 0.1 0.1 0.8 0.8 1.5 2.1 1.6 2.1
4630
dx.doi.org/10.1021/ac300611n | Anal. Chem. 2012, 84, 4628−4632
Analytical Chemistry
Technical Note
Table 3. Isotopomer Distribution of Aspartate Carbon Atoms Determined for Natural Aspartate and Six 13C-Labeled Aspartate Standards (Molar Percent Abundances, mol %)a
a
aspartate isotopomer
natural aspartate
[1-13C] aspartate (99%)a
[2-13C] aspartate (99%)a
[3-13C] aspartate (99%)a
[4-13C] aspartate (99%)a
[1,4-13C] aspartate (99%)a
[U-13C] aspartate (97%)a
0000 0001 0010 0011 0100 0101 0110 0111 1000 1001 1010 1011 1100 1101 1110 1111
99.8 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0
0.7 0.5 0.1 0.2 0.6 0.1 0.0 0.1 97.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0
1.3 0.0 0.6 0.0 97.8 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0
1.3 0.0 97.2 0.0 0.9 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0
1.3 96.4 1.1 0.0 0.0 0.0 0.2 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.2 0.0
0.0 1.1 0.0 0.0 0.0 0.0 0.5 0.0 1.4 97.1 0.0 0.0 0.0 0.0 0.0 0.0
0.2 0.1 0.1 0.3 0.0 0.0 0.3 2.3 0.1 0.5 0.1 1.5 0.3 2.7 2.3 89.2
Numbers in parentheses denote manufacturers’ specification of isotopic enrichment.
■
studies have relied on GC/MS, LC−MS, and NMR measurements for 13C-MFA applications. To our knowledge, there has been only one report on the use of tandem MS for 13C-MFA. Jeffrey et al.23 compared three techniques to measure 13Clabeling of glutamate from tissue extracts of hearts supplied with 13C-tracers: 13C NMR, GC/MS, and GC/MS/MS. For tandem MS analysis, Jeffrey et al. measured two daughter fragments of glutamate, which provided 13 tandem mass isotopomers. Although this was not enough to compute all 32 (= 25) isotopomers of glutamate, the flux results demonstrated the potential of using tandem MS for 13C-MFA. The experimental results indicated that tandem MS produced flux results that were similar, or better, than those obtained by 13C NMR and significantly better than those obtained using traditional full spectrum GC/MS analysis.23 In this work, we focused on establishing an accurate method for measuring the complete isotopomer distribution of aspartate, which is derived from the key intracellular metabolite oxaloacetate. First, we screened daughter spectra of TBDMSderivatized aspartate and identified five daughter fragments that provided a redundant set of 47 tandem mass isotopomers to determine all 16 (= 24) isotopomers of aspartate. We optimized tandem MS scanning parameters and evaluated the accuracy of our method using natural aspartate and six specifically labeled aspartate standards. The method we present here determines the complete isotopomer distribution of aspartate with an accuracy and precision of 0.3 mol % or better (based on the number of redundant measurements and the measurement accuracy). Given the increasing use of tandem MS in metabolomics studies, we anticipate that our method will find widespread usage for determining aspartate 13C-labeling for metabolic flux analysis studies and beyond.24−26
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Phone: 302-831-8960. Fax: 302831-1048. Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS We acknowledge support by an NSF CAREER Award (Grant CBET-1054120). REFERENCES
(1) Stephanopoulos, G. Metab. Eng. 1999, 1, 1−11. (2) Ahn, W. S.; Antoniewicz, M. R. Biotechnol. J. 2012, 7, 61−74. (3) Zamboni, N. Curr. Opin. Biotechnol. 2010, 22, 103−108. (4) Large, V.; Brunengraber, H.; Odeon, M.; Beylot, M. Am. J. Physiol. 1997, 272, E51−58. (5) Niklas, J.; Schneider, K.; Heinzle, E. Curr. Opin. Biotechnol. 2010, 21, 63−69. (6) Antoniewicz, M. R.; Kelleher, J. K.; Stephanopoulos, G. Anal. Chem. 2007, 79, 7554−7559. (7) Antoniewicz, M. R.; Kelleher, J. K.; Stephanopoulos, G. Anal. Chem. 2011, 83, 3211−3216. (8) Szyperski, T. Q. Rev. Biophys. 1998, 31, 41−106. (9) Szyperski, T.; Glaser, R. W.; Hochuli, M.; Fiaux, J.; Sauer, U.; Bailey, J. E.; Wuthrich, K. Metab. Eng. 1999, 1, 189−197. (10) Reed, J. L.; Senger, R. S.; Antoniewicz, M. R.; Young, J. D. J. Biomed. Biotechnol. 2010, 2010, 207414. (11) Crown, S. B.; Antoniewicz, M. R. Metab. Eng. 2012, 14, 150− 161. (12) Metallo, C. M.; Walther, J. L.; Stephanopoulos, G. J. Biotechnol. 2009, 144, 167−174. (13) Walther, J. L.; Metallo, C. M.; Zhang, J.; Stephanopoulos, G. Metab. Eng. 2012, 14, 162−171. (14) Wittmann, C.; Heinzle, E. Biotechnol. Bioeng. 1999, 62, 739− 750. (15) Leighty, R. W.; Antoniewicz, M. R. Metab. Eng. 2011, 13, 745− 755. (16) Antoniewicz, M. R.; Kelleher, J. K.; Stephanopoulos, G. Metab. Eng. 2006, 8, 324−337. (17) Antoniewicz, M. R.; Kelleher, J. K.; Stephanopoulos, G. Metab. Eng. 2007, 9, 68−86.
ASSOCIATED CONTENT
S Supporting Information *
Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. 4631
dx.doi.org/10.1021/ac300611n | Anal. Chem. 2012, 84, 4628−4632
Analytical Chemistry
Technical Note
(18) Young, J. D.; Walther, J. L.; Antoniewicz, M. R.; Yoo, H.; Stephanopoulos, G. Biotechnol. Bioeng. 2008, 99, 686−699. (19) Ahn, W. S.; Antoniewicz, M. R. Metab. Eng. 2011, 13, 598−609. (20) Crown, S. B.; Indurthi, D. C.; Ahn, W. S.; Choi, J.; Papoutsakis, E. T.; Antoniewicz, M. R. Biotechnol. J. 2011, 6, 300−305. (21) Choi, J.; Antoniewicz, M. R. Metab. Eng. 2011, 13, 225−233. (22) Yoo, H.; Antoniewicz, M. R.; Stephanopoulos, G.; Kelleher, J. K. J. Biol. Chem. 2008, 283, 20621−20627. (23) Jeffrey, F. M.; Roach, J. S.; Storey, C. J.; Sherry, A. D.; Malloy, C. R. Anal. Biochem. 2002, 300, 192−205. (24) Moxley, J. F.; Jewett, M. C.; Antoniewicz, M. R.; Villas-Boas, S. G.; Alper, H.; Wheeler, R. T.; Tong, L.; Hinnebusch, A. G.; Ideker, T.; Nielsen, J.; Stephanopoulos, G. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 6477−6482. (25) Stephanopoulos, G.; Alper, H.; Moxley, J. Nat. Biotechnol. 2004, 22, 1261−1267. (26) Styczynski, M. P.; Moxley, J. F.; Tong, L. V.; Walther, J. L.; Jensen, K. L.; Stephanopoulos, G. N. Anal. Chem. 2007, 79, 966−973.
4632
dx.doi.org/10.1021/ac300611n | Anal. Chem. 2012, 84, 4628−4632
