Comprehensive Metabolome Analysis of Small Volume Samples by Two Complementary UPLC-MS Methods 1David  Fischer,  2Giancarlo  Marra,  1Ralph  Schlapbach,  1Endre  Laczko  

¹Functional Genomics Center Zurich, University/ETH Zurich; 2Institute for Molecular Cancer Research, University of Zurich  

Introduction Application examples

Methods

Conclusions

References

Acknowledgments

Metabolome analysis, thought as a complement to genome wide proteomics and transcriptomics, is still at need for a comprehensive as well as efficient identification and quantification of metabolites and lipids in any biological sample, but especially in small volume samples. Here we like to present 1) a complementary set of two capillary scale UPLC-ESI-HRMS methods for the comprehensive analysis of the metabolome and lipidome and 2) their successful application in small volume samples.

General aspects As the result of our UPLC-MS method development efforts in the frame of various metabolomics projects over several years, we have identified two standard methods for the analysis of endogenous metabolites and lipids with high coverage comparable to the coverage reported for proteomics and transcriptomics. Common to both methods is the use of custom or commercial capillary columns with inner diameters of 0.15 to 0.20mm and 50 to 150mm length. In both methods we apply ramped flows in the range of 2 to 6uL/min and we use a nanoESI sources to couple UPLC and MS systems. In any case we use LC packings with particle sizes below 2micron and we inject 1uL samples on this capillary columns. The methods were set up with similar analytical performance on various nanoUPLC-MS systems, including UPLCs from Eksigent, Waters and Thermo and MS systems of types Q-TOF/OT to acquire HRAM data and QqQ to acquire mSRM data. HILIC-UPLC for anionic and polar metabolites •  Waters BEH Amide, 1.7µm •  A water and B acetonitrile, both with 10mM NaHCO3,

adjusted with NH4OH to pH 9 •  Gradient from 10% A to 50% A in 10min •  Reconditioning of the column within 4 to 15min as

indicated by carry over monitoring •  Minimal run time of 14min •  Samples in solvent corresponding to initial LC

condition but increased ionic strength

RP-UPLC for lipids, aromatic and non-ionic metabolites •  Waters HSS T3, 1.8µm •  A 0 to 40% acetonitrile in water and B 10%

acetonitrile in isopropanol, both with 10mM ammonium acetate at pH 7

•  Gradient from 10 to 100% B in 10min, hold 10min •  Reconditioning of the column within 5 to 15min as

indicated by carry over monitoring •  Minimal run time of 25min •  Samples in methanol/water 80:20 (v/v)

Hunting substrates of orphan transporters

•  Downscaling of UPLC methods to the capillary and nanoESI scale increases the metabolome coverage to levels known from proteomics and transcriptomics, thus turning metabolomics to a systems biology tool

•  sub nM LODs enables the analysis of small samples •  Added benefit of short run times and lower costs

The presented work was enabled by various grants of the SNF and the URPP Functional Genomics of the University Zürich

1.Abplanalp, J et al. (2013) The cataract and glucosuria associated monocarboxylate transporter MCT12 is a new creatine transporter, Human Molecular Genetics, in press (DOI 10.1093/hmg/ddt175). 2. Junmin Hu et al. (2013) Alterations of serum free fatty acid and phospholipid levels in feline diabetes using capUPLC-nanoESI-TOF-MS, manuscript in preparation. 3. Ziellonka, J et al.(2008) Detection of 2-hydroxyethidinium in cellular systems: a unique marker product of superoxide and hydroethidine, Nature Protocols 3:8-21.

d = 0.2

H01H_IIIneg H02H_IIIneg

H03H_IIIneg H04H_IIIneg

D02A_IIIneg D03A_IIIneg

D04A_IIIneg D05A_IIIneg D06A_IIIneg H05H_IIIneg

H06H_IIIneg

D07A_IIIneg

H07H_IIIneg

H08H_IIIneg

D01B_IIIneg D02B_IIIneg D03B_IIIneg

D04B_IIIneg

H09H_IIIneg

H10H_IIIneg

H11H_IIIneg

H12H_IIIneg

D05B_IIIneg

D06B_IIIneg

D07B_IIIneg

D08B_IIIneg

D09B_IIIneg

D10B_IIIneg

H13H_IIIneg

H14H_IIIneg

H15H_IIIneg

H16H_IIIneg H17H_IIIneg D11B_IIIneg

D12B_IIIneg

D13B_IIIneg

D14B_IIIneg

D15B_IIIneg

d = 0.2

A

B H

d = 0.2

mz_89.0241_0.9302

mz_116.9724_0.9538 mz_141.0163_0.9777

mz_179.0557_0.9441 mz_215.0323_0.8994

mz_223.0276_1.1394

mz_229.0535_0.985

mz_233.1542_1.3709

mz_241.2169_1.6737

mz_245.0484_0.9716

mz_275.0591_0.9979 mz_301.1111_0.9971

mz_301.2167_1.5033

mz_303.2324_1.6353 mz_305.0223_0.9795

mz_327.0518_0.9703

mz_327.2325_1.574 mz_329.248_1.6756

mz_362.2365_1.37

mz_373.0101_0.9867

mz_387.0257_0.9842

mz_390.2675_1.5641

mz_476.2704_0.999

mz_476.2777_1.5316 mz_554.3454_1.7209 mz_563.5035_1.8176

mz_568.3611_1.8099 mz_578.3011_1.4876 mz_602.3455_1.4871

mz_612.33_1.4536

mz_830.5907_12.2987

A

B H

Eigenvalues

Observed m/z

RT (min)

Mass tolerance (#mDa)

i-FIT value

Elemental composition

Potential biomarkers

Trend

307.2638 9.49 0.1 30.2 C20H36O2 FFA C20:2 !

309.2796 10.12 0.2 32.3 C20H38O2 FFA C20:1 !

327.2322 8.24 -0.2 29.6 C22H32O2 FFA C22:6# !

329.2479 8.65 -0.2 29.9 C22H34O2 FFA C22:5 !

331.2635 9.17 -0.2 24.6 C22H36O2 FFA C22:4 ! 498.2887 5.09 -0.2 25.8 C26H45NO6S TCDCA# "

514.2834 4.89 -0.5 31.7 C26H45NO7S taurine conjugated bile acids

"

Serum lipid profiles of diabetic cats

ROS detection in cell cultures and small organisms

Spotting serum lipid markers of diabetes from a list of hundreds of quantified and annotated serum lipids. [2]

mSRM based capUPLC-MS lowers the LODs to the sub nM level.

0.16nM

0.80nM

4.0nM

20nM

Improved ROS stress determination in small model organisms.

Identification and quantification of bile acids, free fatty acids, lyso-PL, PL, DAG and TAG in full scan MS/MSMS data in the nM range (sera concentrations).

Comprehensive profiling of water soluble metabolites inside and outside Xenopus leavis oocytes bearing a recombinant and membrane localized orphan transporter protein, resulted in a ranked short list of potential substrates out of 554 relatively quantified and annotated compounds. One of the 2 top ranked metabolites, creatine, was confirmed by detailed transport studies using 14C-creatine to be a substrate. [1]

Published LC-MS methods quantify ROS probes at µM level (see [3]).

As Core4Life member we offer training and courses for capUPLC-nanoESI-MS based metabolomics

• • • www.fgcz.ch/applications/metabolomics

• • • metabolomics@fgcz.ethz.ch