| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Endocrinology and Metabolism |
1 Department of Laboratory Medicine & Pathology, Mayo Clinic and Foundation, 200 First Street SW, Rochester, MN 55905.
aAuthor for correspondence. Fax 507-284-9758; e-mail singh.ravinder{at}mayo.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: After solid-phase extraction, chromatographic separation of normetanephrine (NMN) and metanephrine (MN) was accomplished by use of a cyano analytical column. NMN, MN, d3-NMN, and d3-MN positive ions were detected in the multiple-reaction monitoring mode using the specific transitions m/z 166
134, 180148, 169137, and 183151, respectively, with an atmospheric pressure chemical ionization source.
Results: Multiple calibration curves exhibited consistent linearity and reproducibility. Interassay imprecision values (CV; n = 20) for NMN at 0.64, 1.9, and 2.7 nmol/L were 6.6%, 7.8%, and 13%, respectively. Interassay CV for MN at 0.60, 1.2, and 2.1 nmol/L (n = 20) were 9.2%, 6.8%, and 9.8%, respectively. The mean recoveries of NMN and MN relative to the internal standard were 100% and 96%, respectively. The assays were linear between 0.20 and 10.0 nmol/L. Deming regression of HPLC and LC-MS/MS results yielded slopes of 0.93 (95% confidence interval, 0.89–0.98) and 0.89 (0.85–0.93) and y-intercepts of -0.16 and 0.03 nmol/L for NMN (n = 132) and MN (n = 92), respectively.
Conclusions: This novel LC-MS/MS approach provides a precise, rapid, and specific alternative method to HPLC for the quantification of the low nanomolar concentrations of free metanephrines in plasma.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
80 UMET and 60 urinary catecholamine tests every day, and this test volume has been constant for the last 4 years. During this period there has been a decrease in the requests for plasma catecholamine testing from almost from 30 to 15 a day. Requests for urinary vanillylmandelic acid have also decreased slightly, with a decrease in the daily average from 55 to 45. Recently we noticed a major change in the practice of clinicians ordering PMET tests in the period from 2000 to 2003. The daily volume of requests for PMET testing has increased from 30 to 140 in our reference laboratory over this period. This is as a result of recent studies showing that plasma metanephrines have better sensitivity for the biochemical diagnosis of pheochromocytoma compared with free catecholamines (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13). PMET analyses are performed with HPLC with electrochemical detection (HPLC-EC) (14)(15). HPLC-EC methods are generally labor-intensive, require extensive sample preparation, and are time-consuming. The chromatographic run time is generally >20 min, which limits the number of assays that can be performed daily on a single instrument. Occasionally, there are drugs/substances that coelute with the analytes, further complicating data interpretation. A faster, easier, and more reliable method of analysis is required to keep up with demand for the plasma metanephrine test. Enzymatic immunoassays based on microtiter plate technology have recently been developed for the rapid determination of urinary metanephrine (MN) and normetanephrine (NMN), but compared with HPLC methods, immunoassays are sometimes difficult to quality control in a routine laboratory and are very susceptible to artifacts caused by nonspecific binding. They may also be subject to cross-reactivity and other analytical interferences. Most importantly, there is often very poor agreement among the results obtained by different immunoassays, sometimes even among immunoassays from the same manufacturer, making patient follow-up over time or between laboratories, as well as longitudinal studies, extremely difficult.
Assays based on gas chromatography–mass spectrometry address many of the shortcomings of automated immunoassays and are considered to be the most accurate methodology. However, sensitivity is often less than what can be achieved by sensitive immunoassays, and run times may be longer, limiting throughput. Liquid chromatography with tandem MS (LC-MS/MS) is superior to gas chromatography–mass spectrometry in terms of both sensitivity and sample throughput, and in combination with isotope dilution, is also considered as a reference methodology. Sample preparation is relatively simple because of the high specificity offered by MS/MS. We recently published a high-throughput LC-MS/MS method for urinary metanephrines after acid hydrolysis of conjugated metanephrines (16). The concentrations of conjugated metanephrines are two to three orders of magnitude higher in urine than plasma concentrations of free metanephrines (nanomolar concentrations). Method development and validation for the UMET test was easier than for the PMET test. In this study, we describe a sensitive, specific, rapid LC-MS/MS method with high throughput for the analysis of PMET.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,,ß-d3 was purchased from Medical Isotopes, Inc., and d3-metanephrine-HCl-,,ß-d3 was purchased from Cambridge Isotope Laboratories, Inc. Charcoal-stripped serum (cat. no. HS-230) was obtained from SeraCare, Inc. HPLC-grade acetonitrile and methanol were supplied by EM Science. Formic acid and trifluoroacetic acid were purchased from Pierce. Ammonium acetate was obtained from Sigma, and HPLC-grade water was purchased from Fisher. OasisTM HLB 1-mL (30 mg) extraction cartridges were purchased through Waters Corporation.
instrumentation
The HPLC system consisted of two Perkin-Elmer Series 200 isocratic micro pumps, a Perkin-Elmer 200 autosampler equipped with a 200-µL sample loop, a 225-position sample tray insert, and a Supelco Supelpro 10-port switching valve. The detection system was a SCIEX API 3000 triple-quadrupole mass spectrometer interfaced with the HPLC system by an atmospheric pressure chemical ionization (APCI) source. The instrument was interfaced to a compatible desktop computer. All data were acquired and processed by Analyst, Ver. 1.2, software.
lc-ms/ms conditions
The chromatographic separation of NMN and MN was achieved with a Phenomenex Cyano Security guard column [4 x 3.0 mm (i.d.)] and a Phenomenex LUNA Cyano analytical column [15 cm x 4.6 mm (i.d.); 5-µm particle size] (17). This column was also compared with Varian Monochrom Cyano and Inertsil-5 CN-3 columns and a Supelco Discovery Cyano column. The mobile phase was acetonitrile–water (40:60 by volume) containing 1.5 mmol/L ammonium acetate and 0.6 g/L formic acid. The flow rate was 1.5 mL/min, and the autosampler wash solvent was methanol–water (75:25 by volume).
Ionization of NMN and MN with APCI was performed in positive-ion mode with purified nitrogen as the nebulizing and auxiliary gas. Nitrogen was also used as the curtain and collision gas. The APCI interface settings were nebulizer = 10, curtain gas = 7, and temperature = 500, and detector settings were collision-assisted dissociation = 4; nebulizing current = 3. Analyzer settings were declustering potential = 50, focusing potential = 175, entrance potential = 10, collision energy = 25, collision exit potential = 2, quadrupole 1 (Q1) resolution = unit, quadrupole 3 (Q3) resolution = low, deflector = -300, and channel electron multiplier = 3000. Acquisition was achieved in the multiple-reaction monitoring (MRM) mode. The transitions of the precursor ions to the product ions (m/z 166134, m/z 180148, m/z 169137, and m/z 183151) were monitored for NMN, MN, d3-NMN, and d3-MN, respectively, with a dwell time of 400 ms for each. The total chromatographic run time was 6 min, but data were acquired only for the last 3 min.
To maximize use of the MS/MS instrument, samples were injected alternately every 3 min on identical columns 1 and 2, and isocratic mobile phase was used for pumps A and B. Fig. 1
illustrates the column-switching strategy used to maximize sample throughput. Sample 1 was injected on column 1, and the eluate was diverted to waste for the first 3 min of the run (Fig. 1A
). At 3 min, the switching valve changed configuration (Fig. 1B
), and the column 1 eluate was redirected to the MS/MS detector for data collection. At the same time, sample 2 was injected on column 2. During this period, the eluate from column 2 went to waste. After another 3 min, column 2 eluate was redirected to the MS/MS detector and sample 3 was injected on column 1.
|
preparation of calibrators and calibration curves
A stock solution containing 1 g/L each of NMN (5.4 mmol/L) and MN (5.0 mmol/L) and a stock internal standard solution containing 1 g/L each of d3-NMN (5.4 mmol/L) and d3-MN (5.0 mmol/L) were prepared in water. We serially diluted these stock solutions in water to produce calibrator solutions II (54 nmol/L NMN and 50 nmol/L MN) and III (5.4 nmol/L NMN and 5.0 nmol/L MN) and a working internal standard solution containing 54 nmol/L d3-NMN and 50 nmol/L d3-MN. Calibrators were prepared by adding the analytical solutions to 1 mL of charcoal-stripped serum. The resulting calibration curves were constructed with six concentrations of each analyte (0.22, 0.54, 1.08, 2.7, 5.4, and 10.8 nmol/L NMN; and 0.20, 0.50, 1.0, 2.5, 5.0, and 10 nmol/L MN), and calibrators were kept frozen until use. Internal standard was added, and the calibrators were extracted in the same manner as the unknown samples. Peak areas were calculated using Analyst software, and the calibration curve was constructed by plotting the peak-area ratio of analyte to internal standard vs analyte concentration.
specimen collection
Approval for these studies was obtained from the Institutional Review Board of the Mayo Clinic. All participants from whom samples were collected for method comparison and reference values were asked to discontinue epinephrine and epinephrine-like drugs at least 1 week before sample collection. They were also asked to avoid smoking for 4 h before collection and to avoid coffee, tea, alcohol, or caffeine-containing beverages or medications for 4 h before collection. With the participant seated, 10 mL of blood was collected into a Vacutainer Tube with EDTA as the anticoagulant. The blood was allowed to clot at room temperature for 30 min and then was centrifuged at 1g for 25 min. Plasma was removed from the top of the tube and was kept frozen until the time of analysis.
sample preparation
Once thawed, the samples were mixed by inversion, and 1 mL of plasma was placed in a 13 x 75 mm tube. After 50 µL of working internal standard solution (54 nmol/L d3-NMN; 50 nmol/L d3-MN) was added to the tube, the sample was vortex-mixed. A vacuum manifold (Supelco) at 127 mm of Hg was used in the solid-phase extraction (SPE) procedure. SPE was performed with Oasis HLB 1-mL solid-phase cartridges. The cartridges were conditioned by sequential washing with 1 mL of methanol and 1 mL of water. The plasma mixture was transferred to the cartridge with a disposable plastic transfer pipette, and the cartridge was washed with an additional 2 mL of water. The analytes were eluted with 1 mL of methanol. The eluate was evaporated and reconstituted in 100 µL of methanol, and 30 µL of the reconstituted eluate was injected into the LC-MS/MS system. Results (nmol/L) were automatically calculated by Analyst software.
basic assay performance characteristics
We determined the critical limit and the detection limit of our assay as suggested by the IUPAC guidelines based on statistical calculations (18). The basic assay performance characteristics are defined as the critical limit, detection limit, limit of quantification (functional sensitivity), and upper limit of the detectable range. Twenty replicate measurements each of blanks were used to determine the critical limits for NMN and MN, defined as the concentrations below which there is >95% certainty that no analyte is present. Similarly, the detection limits were established by performing 20 replicate experiments with low amounts of NMN and MN added to charcoal-stripped serum and were the lowest analyte concentrations that could be distinguished from the absence of analyte with >95% certainty. The limits of quantification (functional assay sensitivities) were, in agreement with common clinical laboratory practice, arbitrarily set as the lowest analyte concentrations that had interassay CV <20% when actual patient samples were analyzed. The upper limit of the detectable range was arbitrarily defined as 10 nmol/L for both NMN and MN, the concentrations of the highest calibrators in the assay.
precision, recovery, and linearity
NMN and MN were added to donor plasma at 0.5, 1.0, and 2.0 nmol/L, and multiple aliquots were placed in vials and frozen. These quality controls were assayed with each batch, and the results were used to evaluate assay performance with regard to accuracy and precision. Recovery, relative to the internal standard, was determined by adding the analytes to plasma samples with at three different concentrations (0.5, 2.5, and 10 nmol/L) and analyzing the resulting mixtures. Absolute recovery was determined by adding internal standard after SPE. Assay linearity was determined by serially diluting patient specimens with high analyte concentrations with water and comparing the results with the expected concentration.
method comparison and reference intervals
Surplus EDTA-plasma specimens from different patients submitted to the laboratory for testing of plasma free metanephrines by the HPLC-EC method were also assayed by the LC-MS/MS method, and results were compared. EDTA plasmas from 50 healthy individuals were also assayed to determine the reference intervals.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. In positive mode, NMN (Mr 183) and MN (Mr 197) are protonated to produce molecular ions of m/z 184 and 198, respectively. The spontaneous loss of water from the protonated molecular ions during ionization gives fragments with m/z, 166 and 180, respectively (19). Further loss of a protonated methoxy group (-OCH3) in the collision cell gives the major daughter fragments of m/z 134 and 148 for NMN and MN, respectively. The transitions 166134 and 180148 were used to detect NMN and MN, respectively, in MRM mode. These transitions are consistent with the method used for urine metanephrines (16).
|
In addition to the MRM transitions described, we also evaluated the MRM transitions m/z 184166 and 198180 (attributable to loss of water) for quantification of NMN and MN, respectively, and obtained comparable sensitivities. Because the transitions attributable to loss of -OCH3 are more specific than the loss of water, we preferred the m/z 166134 and 180148 transitions for quantification of NMN and MN, respectively. We also compared electrospray ionization with APCI for all pairs of transitions. The sensitivity for plasma free metanephrines was markedly better for APCI than for electrospray.
Final selection of MRM transitions and ionization source was driven by the need for maximum sensitivity (large signal-to-noise ratios) and specificity for quantification of low concentrations of circulating plasma free MN and NMN. The metanephrines were separated from matrix components and interferences by SPE and LC with a cyano column and acetonitrile–water mobile phase. Typical chromatograms of plasma extracts produced by HPLC-EC and LC-MS/MS are shown in Fig. 3
. The chromatograms demonstrate that the analysis time is considerably shorter for the LC-MS/MS method than for the HPLC-EC method. We had previously reported analysis times of 2.5–3.0 min for urine metanephrines by the LC-MS/MS method (19). For plasma, the analysis time was increased to 6 min to enhance sensitivity by preventing suppression of the NMN and MN ions attributable to competing ions produced by the drugs/substances generally present at higher concentrations in the plasma of patients being screened for pheochromocytoma.
|
calibration, precision, recovery, linearity, and stability
Calibration curves and controls were run with every batch of patient specimens. Calibration curves were reproducible with CV for the slope of 9.3% for NMN and 8.7% for MN. Correlation coefficients of the curves (n = 19) obtained on multiple days were consistently >0.98 for both NMN and MN. Interassay CV (n = 20) were 6.6–13% for NMN and 6.8–9.8% for MN. Intraassay CV (n = 20) were 4.5–7.5% for NMN and 4.9–10% for MN (Table 1
).
|
The mean recoveries, relative to the internal standard, for NMN and MN were 100% and 96%, respectively. Mean absolute recoveries were 60% for NMN and 72% for MN. The results showed excellent linearity across the calibration range of 0.20–10.0 nmol/L (Table 2
). No carryover was evident on the LC-MS/MS system. NMN and MN were stable in human plasma stored up to 5 days at 2–5 °C (n = 5). No changes were observed in plasma that had been subjected to one freeze–thaw cycle (n = 5).
|
Metanephrines in plasma are more stable than catecholamines. Even in the absence of a reducing agent, plasma can be kept at 4 °C for 3 days without appreciable degradation. Storage or shipment of longer duration must be at -20 °C or lower, and it has been recommended that blood should be kept at 4 °C and must be centrifuged within 6 h (20). Sodium heparin and serum produced results comparable to those for EDTA plasma (n = 5).
basic assay performance characteristics
For the LC-MS/MS method, the critical limit for NMN was 0.09 nmol/L, and the detection limit was 0.13 nmol/L, with a detection range of 0.13–10.0 nmol/L. The CV for the replicates of NMN added to charcoal-stripped serum was 14% at the detection limit. The functional assay sensitivity for LC-MS/MS analysis of NMN was 0.20 nmol/L with a CV of 11%. The critical limit for MN was 0.10 nmol/L, the detection limit was 0.13 nmol/L with a CV of 13%, the detection range was 0.13–10.0 nmol/L, and the functional sensitivity was 0.20 nmol/L with CV of 9.5%. The lower limit of quantification was 0.20 nmol/L for both NMN and MN based on CV of 11% and 9.5%, respectively (Table 1
).
comparison with hplc-ec method
We compared the results obtained by the LC-MS/MS and the HPLC-EC methods for patient specimens routinely screened for pheochromocytoma (Fig. 4
). Deming regression for LC-MS/MS and HPLC results gave slopes of 0.93 (95% confidence interval, 0.89–0.98) and 0.89 (0.85–0.93) and y-intercepts of -0.16 and 0.03 nmol/L for NMN (n = 132) and MN (n = 92), respectively (Fig. 4, A and C
). Bland–Altman plots for the difference between the LC-MS/MS and HPLC-EC methods showed negative differences with mean differences of 0.3 and 0.1 nmol/L for NMN and MN, respectively (Fig. 4
, B and D). With the exception of a few outliers, the results were generally within 2 SD of the difference. During this study we did not have any clinical information on these patient specimens and thus were unable to calculate the sensitivity and specificity of the current LC-MS/MS method for the diagnosis of pheochromocytoma; this will be a subject of our next study. On the basis of the correlation parameters and similarities with the HPLC-EC method, we anticipate that the LC-MS/MS method, which in general is more specific than EC for detection, will have similar sensitivity but improved specificity.
|
reference intervals
We obtained plasma from 58 healthy volunteers (age range, 23–77 years) for determining reference values for NMN and MN in human plasma. The reference intervals were 0.05–0.47 nmol/L for MN and 0.12–1.1 nmol/L for NMN by the LC-MS/MS method (Fig. 5
). The MN and NMN concentrations were similar in males and females, and we observed no gender differences. However, NMN concentrations increased with age, with a slight correlation (0.24; Fig. 5A
). The mean values for healthy individuals were 0.4 and 0.2 nmol/L, whereas the means plus 2 SD were 0.9 and 0.4 nmol/L for NMN and MN, respectively. The data confirmed that the upper limits of the reference intervals were 0.90 nmol/L for NMN and 0.50 nmol/L for MN, which were established with the HPLC-EC method from samples collected in an ambulatory position. These limits are slightly higher than published upper reference limits (0.60 nmol/L for NMN and 0.30 nmol/L for MN) determined by the HPLC-EC method (3). The samples in the published study were collected from patients in a supine position with a catheter placed in the arm, a procedure commonly used for catecholamines (3). The difference in the upper reference limits for plasma free MN and NMN may be attributable to variations in the specimen collection procedures (11).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HPLC-EC has been the methodology of choice for estimation of NMN and MN in plasma of patients being screened for pheochromocytoma (1)(2). Sample preparation for this method is labor-intensive and time-consuming. The HPLC-EC method is also prone to drug interferences. For quantification of MN and NMN by the HPLC-EC method, careful analysis of the chromatograms for peak shape and retention times is essential. This is a time-consuming and meticulous process. Because the chromatographic time is also very long for the HPLC-EC method, only 30 samples can be analyzed in 1 day with one instrument. With the LC-MS/MS method, a single technologist can easily manage the assay with a test volume of >100 samples/day, including the time needed for sample preparation, data analysis, and reporting of the results. Parallel column injection is an effective means of maximizing instrument throughput. Although the initial set up is cumbersome, it saves instrument time, and results are reliable, producing identical results on either column. Both columns are used to collect calibration data, which are linear, as well as quality-control data. The MRM pairs used in the LC-MS/MS method for detection are unique to the molecules, which increases the specificity and reduces interferences. With the MS/MS method there were sporadic peaks attributable to compounds with the same MRM transitions, but there was no effect of these peaks on the quantification of MN or NMN. The interfering peaks were resolved chromatographically.
To achieve maximum sensitivity and specificity it is necessary to retain the analytes chromatographically to separate them from the bulk of the plasma matrix and drug components. This was achieved with the cyano column, but this column has limited throughput in terms of maintaining good chromatography for more than few hundred injections. We also compared cyano columns of the same dimensions from other manufacturers but were unable to separate NMN and MN from the matrix and other interferences.
Among various interfering compounds for the LC-MS/MS method, one obvious one is epinephrine, which is an isomer of NMN (same molecular weight) and also coelutes with NMN under the chromatographic conditions for the method. When an epinephrine calibrator was injected into the LC-MS/MS system, we observed only 5% interference for NMN quantification. This is attributable to the different fragmentation mechanism for epinephrine and generation of a daughter ion of m/z 107 compared with a daughter ion of m/z 134 for NMN; the two MRM transitions are therefore very different. This is clearly an example of the superior specificity offered by MS/MS compared with single-quadrupole MS. The epinephrine interference was totally eliminated when plasma samples with different concentrations of epinephrine added (0.2–10 nmol/L) were extracted and then injected. We also investigated ephedrine and synephrine as possible interfering substances, but they produced no appreciable interference. The lack of perfect correlation between the HPLC-EC and the LC-MS/MS methods in the dynamic range is probably attributable to interferences from drugs that are difficult to separate from polar metanephrines and the hydrophilic components of the complex plasma matrix.
The analytical precision and accuracy of the LC-MS/MS instrumentation are adequate and comparable to the HPLC method, but improved sensitivity and specificity would still be highly desirable. Additional sensitivity and better precision may be obtained with the next generation of MS/MS instruments, and a selective anion-exchange extraction process could improve specificity. The literature has consistently shown that testing of plasma free metanephrines for the biochemical diagnosis of pheochromocytoma has a sensitivity >96% but relatively lower specificity (85%) with the HPLC method (4)(5)(6). The described LC-MS/MS method will maintain the same high sensitivity of the HPLC-EC method, but may have better specificity for the diagnosis of pheochromocytoma.
In conclusion, we have developed a novel LC-MS/MS method for the quantification of NMN and MN at nanomolar concentrations in human plasma with minimal sample preparation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  A. Algeciras-Schimnich, C. M. Preissner, W. F. Young Jr., R. J. Singh, and S. K. G. Grebe Plasma Chromogranin A or Urine Fractionated Metanephrines Follow-Up Testing Improves the Diagnostic Accuracy of Plasma Fractionated Metanephrines for Pheochromocytoma J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 91 - 95. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Annesley Methanol-Associated Matrix Effects in Electrospray Ionization Tandem Mass Spectrometry Clin. Chem., October 1, 2007; 53(10): 1827 - 1834. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Singh and G. Eisenhofer High-Throughput, Automated, and Accurate Biochemical Screening for Pheochromocytoma: Are We There Yet? Clin. Chem., September 1, 2007; 53(9): 1565 - 1567. [Full Text] [PDF]
|
|
|
|
|
|
W. H.A. de Jong, K. S. Graham, J. C. van der Molen, T. P. Links, M. R. Morris, H. A. Ross, E. G.E. de Vries, and I. P. Kema Plasma Free Metanephrine Measurement Using Automated Online Solid-Phase Extraction HPLC Tandem Mass Spectrometry Clin. Chem., September 1, 2007; 53(9): 1684 - 1693. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. d'Herbomez, G. Forzy, C. Bauters, C. Tierny, P. Pigny, B. Carnaille, F. Pattou, J.-L. Wemeau, and N. Rouaix An analysis of the biochemical diagnosis of 66 pheochromocytomas Eur. J. Endocrinol., May 1, 2007; 156(5): 569 - 575. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Taylor, S. K. Grebe, and R. J. Singh Quantitative, Highly Sensitive Liquid Chromatography-Tandem Mass Spectrometry Method for Detection of Synthetic Corticosteroids Clin. Chem., December 1, 2004; 50(12): 2345 - 2352. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
