WO2018174891A1

WO2018174891A1 - Quantitative targeted metabolomic analysis based on the mixture of isotope-and nonisotope-labeled internal standards

Info

Publication number: WO2018174891A1
Application number: PCT/US2017/023863
Authority: WO
Inventors: Bruce Xuefeng Ling; Limin Chen; Shiying Hao
Original assignee: Mprobe Inc.
Priority date: 2017-03-23
Filing date: 2017-03-23
Publication date: 2018-09-27

Abstract

A mixture of internal standards for the quantitative analysis of targeted metabolites using flow injection chromatography interfaced with electrospray tandem mass spectrometry has been developed and validated. This mixture contains both non-isotope- and isotope-labeled chemicals to be used as internal standards for the correction of extraction recovery, normalization of ionization efficiency, and quantification of endogenous metabolites of interest. Compared to the conventional internal standard mixture where only isotope-labeled chemicals are included, this mixture allows the quantitative targeted metabolomics to be implemented in a more cost-effective and economically favorable manner without compromising the quality of the analysis in terms of accuracy and precision.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention describes a robust, cost-effective, and high-throughput analytical approach for quantification of primary metabolites in the routine clinical samples based on a novel isotope- and non-isotope-labeled internal standard mixture by flow injection liquid chromatography interfaced to electrospray tandem mass spectrometry. This method physically combines the outstanding sensitivity and specificity features from electrospray tandem mass spectrometry, high-throughput capability from the flow injection analysis, and the optimal cost-effectiveness merit from the new internal standard mix into a single standardized analytical platform to allow accurate and precise determination of targeted metabolites. Compared to the conventional pure isotope-labeled internal standard mix, our new mix displays more favorable economic advantage while maintaining the satisfactory performance in terms of quantitative analysis.

2. Description of the Prior Art

Metabolomics is defined as the global profiling of the major metabolic intermediates involved in individual metabolism pathways. Certain group of metabolites originated from specific pathways are biological indicators of current functional states of living organisms, representing functional interaction between gene and environment. Nowadays, despite the existence of huge gap on our current knowledge towards the complete understanding of mechanistic connections between genetics and metabolic outcome, the disturbance on endogenous metabolism has been well evident by substantial studies to be pathologically implicated in the development of various diseases. See Sommer et al., Increased Prevalence of the Metabolic Syndrome in Patients with Moderate and Severe Psoriasis, Arch Dermatol Res (2006), 298:321-328. Thus, quantitative profiling of metabolites from certain metabolic pathways, also known as targeted metabolomics, has become one of the routine clinical methods of great importance to determine the presence of systematic dysfunction in terms of metabolism for the prognosis and diagnosis of disorders before the relevant symptoms are clinically noticed.

As can be exemplified by the case of newborn screening, the presence of inborn errors of metabolism, including amino acid metabolism disorder, organic acid disorder, and fatty acid oxidation disorder, all of which are metabolic disorders caused by genetic defects on certain enzymes in the pathway and accumulation of cytotoxic metabolic intermediates, are clinically diagnosed by quantitatively measuring the levels of a panel of relevant primary metabolites in the blood. This process allows the affected newborns to start the treatment at their earliest age before clinical manifestations are evident.

In last several decades, the emerging of tandem mass spectrometric technology has made significant contributions to advance the clinical detection of metabolic syndromes, giving rise to the confident diagnosis of different metabolic disorders of great medical interest based on levels of a variety of endogenous metabolites. With the established tandem mass spectrometry- based method, a large amount of metabolic disease markers could be simultaneously screened and quantified to clinically diagnose the occurrence of different metabolic syndromes on a routine basis in a high-throughput fashion. The high specificity and sensitivity merits of tandem mass spectrometry enable the diagnostic quality to be consistently and confidently maintained from sample to sample. These advantages make this instrumentational setup soon become the golden standard for the diagnosis of metabolic disorders in the clinical setting. See Chace et al., Neonatal Screening for Inborn Errors of Metabolism by Automated Dynamic Liquid Sewary Ion Tandem Mass Spectrometry New Horizons in Neonatal Screening, 1994. In this method, the incorporation of isotope-labeled internal standards using isotope dilution technique provides the normalizing factor to correct variations of extraction recovery and ionization efficiency as well as a quantitative reference to determine the level of specific metabolite in each sample. The concentrations of individual internal standards labeled with isotopes are optimized based upon their own linear ranges in the sample matrices in terms of signal to noise ratio.

However, one of the major pitfalls associated with this methodology is its critical dependence on the presence of stable isotope-labeled standards in the sample, which are spiked into the sample during the preparation phase to serve as both denominators and quantifiers for the analysis of metabolites of interest, leading to the higher cost of the reagents associated with the assay in relative to other routine tests, which limits the popularization of those tandem mass spec-based assays in the global scope.

So, in the present invention, a novel internal standard mix consisting of non-isotope-labeled unnatural structural analogs as well as minimal amounts of isotope-labeled standards has been developed and validated in terms of accuracy and precision in relative to the classical isotope- labeled standard only mix in different sample matrices, including human serum and dried blood spot, within a desired population of samples.

3. Prior Art

U.S. Pat. No. 5, 206, 50, Apr. 27, 1993 describes a tandem mass spectrometric system which can extract tandem mass spectra for each parent ion without separating from other parent ions with different masses. This system would in addition provide the capability to select a specific parent ion prior to excitation.

SUMMARY OF THE INVENTION

The primary objective of this invention is to develop a novel internal standard mixture consisting of both non-isotope and isotope-labeled standards in place of the classical pure isotope-labeled internal standard mixture for the comparably quality of quantitative analysis targeting various primary metabolites, including amino acids, acylcarnitines, urea, and creatinine, in human serum and dried blood spot by using flow injection electrospray tandem mass spectrometric analysis (FIA-ESI-MS/MS). The development of this non-isotope- and isotope-labeled standard-containing internal standard mixture allows the quantitative targeted analysis of primary metabolite to be implemented with more desired cost-effectiveness without compromising the assay performance in terms of accuracy and precision.

Electrospray ionization coupled to tandem mass spectrometry is a well-accepted classic instrumental setup for the quantitative analysis of metabolites. By virtues of its high sensitivity and specificity, individual metabolites of interest can be resolved based on mass to charge ratio by using multiple parent-to-daughter mass transitions, each of which is characteristic and unique to one metabolite. With the addition of stable non-isotope-&isotope-labeled internal standards, the extraction variation and ionization bias are effectively eliminated, and endogenous levels of a wide group of metabolites can be simultaneously and selectively determined with high level of confidence.

The second objective of this invention is to validate the quantitative performance of the non- isotope-&isotope-labeled internal standard mixture in terms of accuracy and precision in comparison with the conventional isotope-labeled mixture for targeted metabolites by implementing parallel analysis with the same set of serum/dried blood spot samples using both mixtures. This validation study allows the global evaluation of the non-isotope-&isotope- labeled internal standard mixture in relative to the isotope-labeled one at the statistic level. The third objective of this invention is to improve the instrumental throughput by injecting the sample extracts into the LC loop with the flow injection method from an autosampler, delivering the injected samples with intended mobile phases, and detecting the signal responses from metabolites of interest with selected reaction monitoring. With this novel instrumental setup, the analytical timeframe will be significantly shortened into a 2-min period to screen through a large panel of metabolites for a single sample.

The fourth objective of this invention is to improve the experimental throughput by implementing the non-derivatization method on the sample preparation to achieve the optimal efficiency of the sample processing. Compared to the derivatization method, the non- derivatization method increases the sample handling throughput by bypassing the esterification step following the analyte extraction, thereby saving the time consumed by the derivatization step and maximizing the sample preparative efficiency.

BRIEF DESCRIPTION OF THE DRAWINGS

The presented method generally consists of a series of principal steps as described in greater detail below:

In the development phase, a set of serum(n=38)/DBS(n=60) samples from human are collected and divided into appropriate aliquots. The aliquots are immediately stored into -80 °C freezer prior to use. Thereafter, two identical sets of aliquots from serum(n=38)/DBS(n=60) are automatically prepared in parallel by extracting with a methanol-based extraction buffer containing known concentrations of non-isotope-&isotope-labeled and isotope-labeled mixtures, respectively, on the 96-well plates through a programmed automatic station. In parallel, quality control samples, encompassing normal, abnormal, and standard controls in duplicate, are prepared by following the identical protocol to ascertain the quality of assay performance on each plate. Afterwards, prepared samples are analyzed by high performance liquid chromatography system in flow injection configuration interfaced to electrospray tandem mass spectrometry in selected reaction monitoring mode for maximal analytical throughput. The quality control samples are matched against their nominal values and thresholds to determine whether the acquired results from unknown samples on that plate will be accepted or rejected. Based on the satisfied QC values, the correction factors are calculated based on differences on quantification results between two internal standard mixtures to normalize the biases in extraction recovery and ionization efficiency. The coefficient variation of each correction factor on individual metabolite being quantified systematically decides whether the obtained factor should be discarded and retained for the validation phase.

In the validation phase, another set of serum(n=38)/DBS(n=60) samples from human are aliquoted and stored under identical conditions as described above. Thereafter, two identical sets of aliquots from serum(n=38)/DBS(n=60) are automatically prepared in parallel by non- isotope-&isotope-labeled and isotope-labeled mixtures, respectively, following the same protocol as before. As means of assuring the quality of the analysis, identical sets of quality control samples are included into each plate in duplicate. Following the FIA-ESI-MS/MS, the quality control samples are matched against their nominal values in relative to determined threshold before proceeding to the next step. Based on satisfied QC results, the concentration values obtained by the non-isotope-&isotope-labeled mixture are multiplied by the correction factors corresponding to individual metabolites and placed next to the concentration values obtained by the isotope-labeled mixture for a side-by-side comparison. As means to statistically analyze the deviation of quantification results obtained by the non-isotope- &isotope-labeled mixture in relative to the isotope-labeled one, the z-scores between two different internal standard mixtures are calculated based on the population mean as well as the standard deviation of each metabolite and analyzed by the classical box plotting to determine the presence of statistical significance in terms of quantitative outcomes in the given set of serum and dried blood spot samples.

Fig. 1 is a simplified block diagram showing the overall makeup of the workflow. Four primary steps are inter-correlated from sample collection to final application.

Fig. 2 is a block diagram showing the stepwise makeup of the flowchart in the development phase of the non-isotope-&isotope-labeled internal standard mixture.

Fig. 3 is a block diagram showing the stepwise makeup of the flowchart in the validation phase of the non-isotope-&isotope-labeled internal standard mixture.

Fig. 4 is a block diagram showing the details regarding steps involved in the sample preparation flowchart.

Fig. 5 is a tabular diagram showing the roster of both non-isotope-&isotope-labeled and isotope-labeled internal standard mixtures as well as the individual metabolites to which each component from the mixtures serving as the normalizer and quantifier.

Fig. 6 is a box plot diagram showing the statistical difference on the z-score distributions between quantification results obtained by two different internal standard mixtures in serum. The statistical comparison between two different mixtures on individual metabolites is presented in three separate figures based on the number of analytes being included into the measurement.

Fig. 7 is a box plot diagram showing the statistical difference on the z-score distributions between quantification results obtained by two different internal standard mixtures in DBS. The statistical comparison between two different mixtures on individual metabolites is presented in three separate figures based on the number of analytes being included into the measurement.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

The method will now be described in detail in relation to a preferred embodiment and implementation thereof, which is exemplary in nature and descriptively specific as disclosed. As is customary, it will be understood that no limitation of the scope of the invention is thereby intended. The invention encompasses such alterations and further modifications in the illustrated device, and such further applications of the principles of the invention illustrated herein, as would normally occur to persons skilled in the art to which the invention relates. Fig. 1 presents a general stepwise workflow of obtaining a high quality non-isotope-&isotope- labeled internal standard mixture for the quantitative targeted metabolomics analysis using flow injection electrospray tandem mass spectrometric approach. Four principal steps are involved in the generation of the qualified internal standard mixture containing both non- isotope-labeled unnatural structural analogs and isotope-labeled molecular siblings in terms of quantitative performance.

In the first step, intended sample types (serum/dried blood spot) with desired quantities (n=76/120) are collected for the study, all of which are randomized to generate two different sample sets with identical population size per sample type for the subsequent development and validation phases, respectively.

In the second step, the grouped samples are individually barcoded and scanned to log in the sample inventory prior to their storage in the freezer. An excel template is then generated to associate the barcodes of individual samples with their corresponding locations at which they are placed onto the 96- well plate during the subsequent steps.

In the third step, the groups of samples designated for the development phase would undergo sample preparation, flow injection mass spectrometric analysis, and acquired data processing steps to obtain two sets of quantitative results per sample type using non-isotope-&isotope- labeled and isotope-labeled internal standard mixtures, respectively, in a parallel setting. By comparing two different sets of quantitative data, the individual correction factor dedicated to each measured metabolite could be calculated to obtain information regarding the normalization.

In the fourth step, the groups of samples designated for the validation phase would similarly undergo sample preparation, flow injection mass spectrometric analysis, and acquired data processing steps to obtain two sets of quantitative results per sample type using non-isotope- &isotope-labeled and isotope-labeled internal standard mixtures, respectively, in a parallel setting. Based on the concentrations values obtained by two distinctive internal standard mixture, the correction factors measured from the development phase are applied to correct the methodological bias between two sets of quantitative data, and statistical analysis is implemented to evaluate the overall quantitative performance of the new mix in relative to the old one based on accepted criteria. Fig. 2 displays the stepwise flowchart of the study implemented in the development phase. Briefly, two sets of randomized samples (serum/dried blood spot) with intended quantities (n=38/60) are divided into aliquots with preferred amounts before proceeding to the sample preparation step. The sample preparation is implemented in parallel on two identical sets of aliquots per sample type along with another two sets of quality control samples, comprising normal control, abnormal control, and standard control, by extracting the targeted metabolites with different methanolic solutions containing non-isotope-&isotope-labeled and isotope- labeled internal standard mixtures, respectively. Following the sample preparation, the samples are injected into the sample loop by autosampler and delivered to the electrospray tandem mass spectrometer by a constant flow of mobile phase with no column present. When the injected sample arrives at the electrospray ionization source, the tandem mass spectrometer automatically scans over each parent-to-daughter mass pair dedicated for each metabolite of interest to ensure the speed of the analysis. The acquired data from the samples is processed along with quality controls by assigned software to obtain concentration values of individual metabolites generated from the scans of mass spectrometer, and the concentration values are reformatted into a spreadsheet to undergo further inspections on the calculations as a means of ascertaining the performance with high quality. Based on the qualified QC data, by dividing the concentration values obtained by isotope-labeled internal standard mixture over the non- isotope-&isotope-labeled one, the individual correction factor dedicated to each measured metabolite could be calculated to mathematically normalize the distinctive chemical behaviors between non-isotope- and isotope-labeled internal standards during the extraction and ionization of targeted metabolites. Fig. 3 shows the stepwise flowchart of the study implemented in the validation phase. Briefly, two separate sets of randomized samples (serurn/dried blood spot) with intended quantities (n=38/60) are prepared, analyzed, processed, and reformatted along with two sets of quality control samples as described above to generate a spreadsheet containing quantitative results obtained by two different sets of internal standard mixtures. Once the quality of the performance is assured, the quantitative data obtained by non-isotope-&isotope-labeled mixture are subject to multiplying by the correction factors determined from the previous phase to normalize the differential signal intensities on mass spec caused by structural discriminations on the functional groups at the molecular level, allowing the corrected concentration values to be calculated. Based on the values of population mean as well as standard deviation associated with individual targeted metabolite from each data set, the statistic conclusion could be drawn on the degree of fluctuation from the novel non-isotope- &isotope-labeled mixture towards the true values obtained from the pure isotope-labeled mixture in terms of the quality of quantitative performance.

Fig. 4 shows an overview of primary steps involved in the sample preparation procedure.

Following the randomization and grouping of collected samples (serum/dried blood spot) in respect to their practical applications, selected group of samples are documented to associate their barcodes to the locations at which they are placed on the 96-well plate. Depending on various physical forms presented by different sample types (serum/dried blood spot), either 10 uL of serum or 3mm dried blood spot punch is prepared and placed into the designated well on the plate. Afterwards, extraction buffers containing non-isotope-&isotope-labeled and isotope-labeled are prepared, respectively, by diluting the stock solutions with methanolic solution to desired concentrations, which are then added into each of the sample-containing well from corresponding sample set by the programmed liquid handling through an automated laboratory station. After programmed shaking and centrifuge, the supernant is transferred to a new 96-well plate and placed into the autosampler for subsequent injection prior to the analysis. The plate is sealed to avoid any solvent evaporation prior to the injection.

In Fig. 5, the 1^st column to the left includes the roster of individual components presented in the non-isotope-&isotope-labeled internal standard mixture, some of which are standards labeled with stable isotopes with given information on the molecular location as well as the types of isotope substituents, including ²D, ¹³C, and ¹⁵N, whereas the rest of the internal standards are unlabeled analogs sharing certain degree of structural homology with the targeted metabolites in terms of molecular backbone. The preparation of non-isotope-&isotope-labeled internal standard mixture-containing extraction buffer is implemented by diluting stock solutions from 11 standards, encompassing 1 unlabeled amino acid derivative, 1 unlabeled acylcamitme derivative, 6 isotope-labeled amino acids, and 3 other isotope-labeled metabolites, with methanolic solution at a ratio of 1 :400 (v/v) to obtain the daily working solution with desired concentrations as list in the 2^nd column to the left. This extraction buffer plays multiple functions as extraction medium, experimental variation normalizer, and endogenous concentration quantifier. The 3^rd column to the left shows the roster of individual standards labeled with stable isotopes, which details not only the molecular position of the stable isotope but also the type of isotope substituent presented at a specific molecular spot. The isotope- labeled internal standard mixture-containing extraction buffer is prepared through serial dilution of 5 stock mixes, comprising encompassing 15 amino acids, 13 carnitine/acylcarnitine, and 3 other metabolites, with methanolic solution at a ratio of 1 :200 (v/v) to obtain the daily working solution with desired concentrations as provided in the 4^th column to the left. Due to the limitation on obtaining all isotope-labeled standards for individual metabolites of interest, a single internal standard might serve as both normalizer and quantifier for multiple metabolites based on structural similarity. Greater details in respect to the group of metabolites to which each internal standard is normalizing and quantifying are stated in the 5^th column. Fig. 6 shows the difference on z-score-based distributions between non-isotope-&isotope- labeled and isotope-labeled internal standard mixtures in terms of quantitative performance at the statistical level in the serum measurement. Upon the calculation of quantitative results from serum in the last few step of the validation phase, both sets of concentration values obtained by two different internal standard mixtures are mathematically transformed into z-score based on their population mean and standard deviation, which are later plotted against each other to determine the presence of statistical significance on the distribution of measured results between two different standard mixes as a function of the size of overlapping area in terms of individual metabolites. The p-value is calculated by student t-test based on the area of distribution overlapping for each targeted metabolite to statistically quantify the chances of getting equivalently level of results while two mixtures are both engaged for the measurements of identical set of samples in parallel.

Fig. 7 shows the difference on z-score-based distributions between non-isotope-&isotope- labeled and isotope-labeled internal standard mixtures in terms of quantitative performance at the statistical level in the dried blood spot measurement. The statistical analysis step is implemented following the same procedure as described above, and the p-value is also calculated by student t-test to evaluate the statistical likelihood of acquiring comparable levels of concentration value between two internal standard mixtures for each metabolite of interest.

Claims

Claim:

1. A method of generating a novel non-isotope-&isotope-labeled internal standard mixture for quantitative targeted-metabolomics analysis, comprising:

a. Collecting the samples in intended sample type with desired quantity;

b. Randomizing the samples into two group with comparable size;

c. Developing the internal standard mixture using one group of samples;

d. Validating the developed internal standard mixture using another group.

2. The method of claim 1 wherein said sample type comprises serum and dried blood spot.

3. The method of claim 1 wherein said quantity comprises 76 serum and 120 dried blood spot samples.

4. The method of claim 1 wherein said population comprises 38 serum and 60 dried blood spot samples.

5. The method of claim 1 wherein said section c further comprising steps of:

a. Dividing the selected group of samples into aliquots;

b. Preparing sample aliquots by extraction buffer 1 and 2, respectively;

c. Analyzing the prepared samples by FIA-ESI-MS/MS;

d. Processing the raw data with assigned software to generate quantitative values for both mixtures;

e. Evaluating quality controls samples by matching to their nominal values and determined threshold;

f. Calculating correction factors associated with individual targeted metabolite;

g. Retaining the qualified calculated factors based upon the coefficient of variation across measured samples in relation to metabolites of interest.

6. The method of claim 1 wherein said section d further comprising steps of: a. Dividing the selected group of samples into aliquots;

b. Preparing sample aliquots by extraction buffer 1 and 2, respectively;

c. Analyzing the prepared samples containing distinctive mixtures by FIA-ESI-MS/MS; d. Processing the raw data with assigned software to generate quantitative values;

f. Applying the developed correction factors to generate corrected quantitative values; g. Determining the presence of statistical significance on data distributions between results using different extraction buffers.

7. The method of claim 5 wherein said section b further comprising steps of:

a. Preparing extraction buffer containing internal standard mixtures;

b. Extracting the metabolites of interest with 9 volumes of extraction buffer on 96-well plate;

c. Vortexing the extracted samples to reach homogenous state;

d. Centrifuging the homogenous samples at high speed;

e. Transferring the supernant to another 96-well plate for the analysis.

8. The method of claim 5 wherein said extraction buffer 1 comprises unlabeled internal standard, labeled internal standard, hydrazine hydrate, 2-mercaptoethanol, methanol, and water.

9. The method of claim 8 wherein said unlabeled internal standard comprising L- Valine Methyl Ester and O-Dodecanoyl-L-Carnitine Methyl Ester.

10. The method of claim 8 wherein said labeled internal standard comprising ²D₃-L- Alanine, 2Dg-L-Valine, ²D₃-L-Leucine, ²D₃-L-Methionine, ¹³C5-Succinylacetone, ¹⁵N2-Urea, ²D₃- Creatinine, ¹³Ce, ¹⁵N4-L-Argininosuccinic Acid, ²D₄-L-Homocysteine, and ¹⁵N2-Uric Acid.

11. The method of claim 5 wherein said extraction buffer 2 comprises labeled internal standard, hydrazine hydrate, 2-mercaptoethanol, methanol, and water.

12. The method of claim 11 wherein said labeled internal standard comprising ²D₄-L-Alanine, 2Dg-L-Valine, ²D₃-L-Leucine, ²D₃-L-Methionine, ¹³C, ¹⁵N-Glycine, ¹³C6-L-Phenylalanine, 1³C6-L-Tyrosine, ²D₃-L-Aspartate, ²D₃-DL-Glutamate, ²D2-L-Ornithine, ²D2-L-Citrulline, 1³C, ²D₄-L-Arginine, ¹³C₅, ¹⁵N-L-Proline, ²D₉-L-Camitine, ²D₃-0-Acetyl-L-Carnitine, ²D₃- O-Propionyl-L-Carnitine, ²D₃-0-Butyryl-L-Carnitine, ²D9-0-Isovaleryl-L-Carnitine, ²D₃-

O-Glutaryl-L-Carnitine, ²D₃-3-Hydroxyisovaleryl-L-Carnitine, ²D₃-0-Octanoyl-L- Carnitine, ²D9-0-Dodecanoyl-L-Carnitine, ²D9-0-Myristoyl-L-Camitine, ²D₃-0- Palmitoyl-L-Carnitine, ²D₃-0-DL-Hydroxypalmitoyl-L-Carnitine, ²D₃-0-Octadecanoyl- L-Carnitine, ¹³C5-Succinylacetone, ¹⁵N2-Urea, ²D₃-Creatinine, ¹³C₆, ¹⁵N₄-L- Argininosuccinic Acid, ²D₄-L-Homocysteine, and ¹⁵N2-Uric Acid.

13. The method of claim 5 wherein said section c further comprising steps of:

a. Placing the sample-containing 96-well plate into the autosampler;

b. Injecting the samples by autosampler through the sample loop with no column attached; c. Eluting the injected sample from the sample loop to the ionization source by a mixture of water/acetonitrile/formic acid;

d. Ionizing the molecules from the samples by electrospray ionization;

e. Detecting the signal intensity by selected reaction monitoring function of tandem mass spectrometer.

14. The method of claim 5 wherein said section d further comprising steps of: a. Acquiring the peak areas from both analytes and internal standards;

b. Calculating the peak area ratio for each metabolite by dividing analyte's peak area over internal standard's peak area;

c. Quantifying the concentration values of analyte by multiplying the concentration values of internal standard in the extraction buffer.

15. The method of claim 5 wherein said Point f further comprising steps of:

a. Obtaining the concentration values of targeted metabolites for individual samples; b. Classifying the sample information in terms of the type of extraction buffer used for sample preparation;

c. Dividing the concentrations of metabolites measured by extraction buffer 1 over extraction buffer 2 in a sample-to-sample fashion to obtain individual correction factors; d. Calculating the population mean and standard deviation of the obtained correction factors.

16. The method of claim 6 wherein said section b further comprising steps of:

a. Preparing extraction buffer containing internal standard mixtures;

c. Vortexing the extracted samples to reach homogenous state;

d. Centrifuging the homogenous samples at high speed;

e. Transferring the supernant to another 96-well plate for the analysis.

17. The method of claim 6 wherein said extraction buffer 1 comprises unlabeled internal standard, labeled internal standard, hydrazine hydrate, 2-mercaptoethanol, methanol, and water.

18. The method of claim 17 wherein said unlabeled internal standard comprising L-Valine Methyl Ester and O-Dodecanoyl-L-Carnitine Methyl Ester.

19. The method of claim 18 wherein said labeled internal standard comprising ²D₃-L- Alanine, 2D₈-L- Valine, ²D₃-L-Leucine, ²D₃-L-Methionine, ¹³C5-Succinylacetone, ¹⁵N₂-Urea, ²D₃- Creatinine, °C6, ¹⁵N₄-L-Argininosuccinic Acid, ²D₄-L-Homocysteine, and ¹⁵N₂-Uric Acid.

20. The method of claim 6 wherein said extraction buffer 2 comprises labeled internal standard, hydrazine hydrate, 2-mercaptoethanol, methanol, and water.

21. The method of claim 20 wherein said labeled internal standard comprising ²D₄-L-Alanine, 2D₈-L-Valine, ²D₃-L-Leucine, ²D₃-L-Methionine, ¹³C, ¹⁵N-Glycine, ¹³C₆-L-Phenylalanine, 2D₃-L-Aspartate, ²D₃-DL-Glutamate, ²D₂-L-Ornithme, ²D₂-L-Citrulline, ¹³C, ²D₄-L- Arginine, ¹³C₅, ¹⁵N-L-Proline, ²D9-L-Carnitine, ²D₃-0-Acetyl-L-Carnitine, ²D₃-0- Propionyl-L-Carnitine, ²D₃-0-Butyryl-L-Carnitine, ²D9-0-Isovaleryl-L-Carnitine, ²D₃-0- Glutaryl-L-Carnitine, ²D₃-3-Hydroxyisovaleryl-L-Carnitine, ²D₃-0-Octanoyl-L-Camitine, 2D9-0-Dodecanoyl-L-Carnitine, ²D9-0-Myristoyl-L-Carnitine, ²D₃-0-Palmitoyl-L- Carnitine, ²D₃-0-DL-Hydroxypalmitoyl-L-Carnitine, ²D₃-0-Octadecanoyl-L-Carnitine, 1³C5-Succinylacetone, ¹⁵N₂-Urea, ²D₃-Creatinine, ¹³C₆, ¹⁵N₄-L-Argininosuccinic Acid, 2D₄-L-Homocysteine, and ¹⁵N₂-Uric Acid.

22. The method of claim 6 wherein said section c further comprising steps of:

a. Placing the sample-containing 96-well plate into the autosampler;

d. Ionizing the molecules from the samples by electrospray ionization;

e. Detecting the signal intensity by selected reaction monitoring function of tandem mass spectrometer;

23. The method of claim 6 wherein said section d further comprising steps of:

a. Acquiring the peak areas from both analytes and internal standards;

24. The method of claim 6 wherein said Point f further comprising steps of:

a. Grouping the quantitative data in terms of the types of extraction buffer used for sample preparation;

b. Multiplying the correction factors for individual targeted metabolites to correct the concentration values obtained by extraction buffer 1 from sample to sample;

c. Calculating the population means and standard deviations based on quantitative results from samples measured by extraction buffer 1 and 2, respectively.

25. The method of claim 6 wherein said section g further comprising steps of:

a. Calculating the z-score for the concentration of each metabolite based on population means and standard deviations from samples measured by extraction buffer 1 and 2, respectively; b. Analyzing the distribution of quantitative data between concentrations measured by two different extraction buffers in terms of z-score;

c. Calculating p-value to determine the statistical likelihood of obtaining comparable levels of concentrations between two different extraction buffers from measurements of identical samples.