an optimized acquisition database and lc-ms/ms method ... · mrm peak area cv was 5.4% per...

Robustness

To assess robustness, 200 injections of a yeast extractspiked with 25 metabolite standards were analyzed with theLC/MS method. Data acquisition occurred over a 5-dayperiod without interruption (Fig. 5). The overall averageMRM peak area CV was 5.4% per metabolite. Excellentretention time reproducibility was also observed with anoverall average CV of <0.2% per metabolite.

A major challenge in metabolomics is achievingreproducible, robust chromatographic resolution with asingle analytical LC/MS method for endogenous cellularmetabolites due to their diversity of physiochemicalproperties. A more manageable chromatographic solutionis to develop pathway-targeted methods for smaller sets ofmetabolites. A highly reproducible and robust ion-pairbased reverse phase (IP-RP) chromatographic method wasdeveloped providing efficient separation of over 200 anionicand hydrophobic metabolites. In conjunction with the IP-RPmethod a triple quadrupole mass spectrometer operated indynamic MRM (dMRM) mode enabled sensitive detectionof analytes across a wide dynamic range. The assayleverages a curated dMRM transition database containingdefined retention times and optimized acquisitionparameters based on chemical standards, enabling flexibleand rapid method implementation.

Results and Discussion

The Agilent dMRM Database

The Agilent Metabolomics dMRM Database providesoptimized instrument operating parameters for the targetedMS acquisition of >200 metabolites. The choice ofprecursor and product ion ions, RTs, and RT windows werebased on extensive analysis of authentic chemicalstandards. MassHunter (MH) Optimizer Software was usedto determine optimal collision energies and fragmentorvoltages for each metabolite to maximize sensitivity. Thecombination of these curated, optimized parametersgenerates results of the highest quality.

Compound transitions are directly imported into MHAcquisition software from the Database Browser to easilybuild tailored dMRM methods (Fig. 2). Users can alsooptimize new compounds of interest with MH Optimizer andresults are automatically saved to the database.

Introduction

UHPLC-MS/MS Method

The IP-RP method employs a UHPLC C18 column with theion-pairing agent tributylamine (TBA) added into the LCmobile phases. The LC/MS method leverages the 1290Infinity II UHPLC system coupled to a 6470 TripleQuadrupole LC/MS system equipped with a Jet Streamelectrospray ionization source (Fig. 1). The method alsoimplements a phase-matched guard column and switchingvalve to backflush the guard and analytical column duringeach run. This combination effectively clears matrixcontaminants for added robustness. Transitions and MSconditions were taken from the Agilent MetabolomicsdMRM Database (described in Results) to build a dynamicmultiple reaction monitoring (dMRM) method. Major UHPLCand MS parameters are as follows:

The combined UHPLC-MS/MS method and acquisitiondatabase provides:• A robust IP-RP chromatography optimized for

challenging ionic metabolites and suitable for analysesin complex cellular matrices

• Compound-specific optimized dMRM MS acquisitionparameters

• Pre-defined compound retention times and retentiontime windows

• A customizable database format

For metabolomics researchers needing targeted methodswith many metabolites per method, the database will allowfast startup and provide excellent results without the userneeding to optimize each and every compound.

Differential Analysis and Pathway Mapping

To enable multivariate relative comparisons Agilent’s MassProfiler Professional software (MPP) provides the ability to:• Easily import, compare, and visualize LC/MS data• Find meaningful relationships in complex experiments• Identify affected biological pathways with the Pathway

Architect module

To demonstrate the workflow, metabolite extracts fromyeast cells cultured with acetate or glucose as the solecarbon source were analyzed with the dMRM method.After processing the datasets with MH Quant and MPP,pathway analysis indicated several affected pathwaysincluding the TCA cycle (Fig. 7). Notably, pyruvate wasfound to be reduced in the acetate-grown cells, likely due tothe lack of glucose as a substrate for glycolysis.

ASMS 2016

Poster # MP-483

Results and DiscussionExperimental Results and Discussion

Conclusions

Results and Discussion

An Optimized Acquisition Database and LC-MS/MS Method Targeting Central Carbon Pathway MetabolitesMark Sartain1, Amy Caudy2,3, and Adam Rosebrock2,3

1Agilent Technologies, Inc., Santa Clara, CA, USA; 2Donnelly Centre for Cellular and Biomolecular Research; and 3Department of Molecular Genetics, University of Toronto, Toronto ON, Canada

s1

Agilent UHPLC 1290 Infinity II System

with the 1290 Infinity II Flexible Pump (Quaternary Pump)

Column Agilent ZORBAX Extend C18, RRHD,

2.1x150mm 1.8 um (p/n 759700-902)

Guard Column Agilent ZORBAX Extend C18 Fast Guard

2.1 x 5mM, 1.8 um (p/n 821725-907)

Column Temperature 35 ºC

Mobile Phase A= 97:3 Water/Methanol with 10mM

Tributylamine, 15mM Acetic Acid

B = Isopropanol (not used)

C= Methanol with 10mM Tributylamine, 15mM

Acetic Acid

D = Acetonitrile (backflushing solvent)

Analytical Flow Rate 0.25 ml/min

Gradient, Flow Rate

(Analytical Portion only)

Time (min) A B C D

2.5 100% -- 0% 0%

7.5 80% -- 20% 0%

13.0 55% -- 45% 0%

20.0 1% -- 99% 0%

24.0 1% -- 99% 0%

… backflushing and regeneration phase

Total Run Time 40.0 min

Agilent 6470 Triple Quadrupole Mass Spectrometer

Ion source, Polarity Agilent Jet Stream ESI, Negative

Drying gas (nitrogen), Temp 13 L/min, 150 ºC

Sheath flow (nitrogen, Temp 12 L/min, 325 ºC

Nebulizer gas (nitrogen) 45 psi

Capillary voltage -2000 V

Nozzle voltage -500 V

Scan type Dynamic MRM (dMRM)

Q1/Q2 resolution Unit (0.7 amu)

Figure 1. The Agilent Infinity II UHPLC System

with the Agilent 6470 Triple Quadrupole Mass

Spectrometer

Figure 7. Pathway Architect view of TCA Cycle (KEGG). Matching

compounds are highlighted, and heat strips indicate normalized abundances.

Data Review

With large multi-compound sample batches typicallyacquired with the Metabolomics dMRM method, MHQuantitative Analysis software provides easy and quickmethod creation for data review and peak integration. Thebatch type format offers helpful features such ascompound-at-a-glance data review (Fig. 6). Reports can beautomatically generated within the software and serve asinput for the next phase of statistical analysis.

Figure 6. Compounds-at-a-glance view within Quantitative Analysis

software, showing 5 compounds (columns) across 3 samples (rows).

Figure 3. Overlapped MRM chromatograms of more than 100 metabolite

standards at 5 ng on-column.

The IP-RP method is fully detailed and designed to be easilyreplicated across laboratories with little observed variationfrom the database compound RTs, enabling narrow dMRMRT windows and providing added confidence in assigningMRM signals to the correct metabolite in a complex matrixwith multiple isobaric/iso-transition species (Fig. 4).

Compound Area CV RT CV

L-Citrulline 6.98% 0.27%

Melibiose 3.91% 0.43%

D-Maltose 4.89% 0.37%

O-Phosphorylethanolamine 9.81% 0.26%

Cytidine 8.34% 0.31%

Hypoxanthine 6.68% 0.30%

Inosine 3.96% 0.42%

Guanosine 3.80% 0.57%

D-Gluconic_acid 2.72% 0.16%

Glyoxylic acid 6.75% 0.20%

Lactic acid 2.89% 0.09%

D-Fructose 6-phosphate 5.37% 0.15%

D-Glucose 6-phosphate 5.37% 0.15%

Orotic_acid_mono 2.67% 0.08%

Pyruvic acid 7.89% 0.08%

b-NAD 4.98% 0.08%

Nicotinic acid 4.63% 0.05%

4-Methyl-2-oxovaleric acid 3.42% 0.05%

Itaconic acid 3.20% 0.06%

L-Malic acid 3.30% 0.06%

D-Fructose 1,6-biphosphate 9.80% 0.06%

2-3-Pyridinedicarboxylic acid 10.68% 0.09%

cis-Aconitic acid 3.44% 0.06%

3-Indoleacetic acid 3.33% 0.05%

Phenylpyruvic acid 6.75% 0.06%

Average 5.42% 0.18%

Figure 5. MRM response over time for 6

representative metabolites (below). Peak

Area and RT CVs for 25 compounds

representing major metabolite functional

classes (right).

0.0E+0

2.0E+5

4.0E+5

6.0E+5

8.0E+5

1.0E+6

1.2E+6

1.4E+6

0 20 40 60 80 100 120 140 160 180 200

MR

M R

aw

Peak

Are

a

Number of Injections

O-Phosphorylethanolamine Inosine

b-NAD 4-Methyl-2-oxovaleric acid

Itaconic acid D-Fructose 1,6-biphosphate

Figure 4. To demonstrate method consistency retention times from the

curating laboratory were verified at a second laboratory with a subset ofstandards (n=67). Excellent agreement was observed with an average RT<0.1 minute. Shown in the inset is a histogram of the compound RT

deviations observed between the two laboratories.

0

2

4

6

8

10

12

14

16

18

0 2 4 6 8 10 12 14 16 18

Cura

ting L

abora

tory

RT (

min

)

Second Laboratory Observed RT (min)

Retention Time Agreement

13

10

16

37

< -6

0 se

c

-60 to

-45 se

c

-45 to

-30 se

c

-30 to

-15 se

c

-15 to

-9 se

c

-9 to

-3 se

c

-3 to

+3 se

c

3 to

9 se

c

9 to

15 se

c

15 to

30 se

c

30 to

45 se

c

45 to

60 se

c

> 6

0 se

c

Figure 2. Dynamic Acquisition Workflow. In this scenario all 407 MRM

transitions representing 218 compounds were imported from the

Metabolomics dMRM Database through the Database Browser (top), and a

single dMRM method was automatically built through MH Acquisition

software. The Dynamic MRM Viewer (bottom) displays dMRM statistics.

Import Compounds of Interest

Experimental

Sample Preparation

More than 200 individual authentic chemical standardswere analyzed for database curation. Cultures ofSaccharomyces cerevisiae (s288c background) were grownin synthetic media containing either acetate or glucose (2%w/v) as the sole carbon source. Cultures were grown tomid-log phase (OD600=0.4), vacuum filtered onto 0.2μmnylon membranes, and immediately immersed in -20˚CACN:MeOH:H2O (40:40:20) for quenching and extraction.Clarified metabolite extracts were dried under nitrogen andresuspended in water prior to LC-MS analysis.

The Agilent dMRM Method

An ion-pair based reverse phase (IP-RP) chromatographicmethod was developed that leverages the provenrobustness of Agilent C18 UHPLC columns, with addedfunctionalization of the stationary phase with the ion-pairing agent tributylamine to facilitate retention of polarmetabolites from multiple functional classes (Fig 3.). Thepresence of the ion-pairing ternary amine TBA in thechromatography was found to facilitate retention of acidicmetabolites, while non-acidic metabolites still maintainedconventional reverse-phase selectivity.

Concurrent MRMsvs Retention Time

For Research Use Only. Not for use in diagnostic procedures. This information is subject to change without notice.