31
Metabolomics for Food Composition and Food Biomarker Analysis David Wishart, University of Alberta 2017 ILSI North America FNSP Mid-Year Meeting July 18, 2017 Arlington Virginia

Metabolomics for Food Composition and Food Biomarker Analysisilsina.org/wp-content/uploads/sites/6/2017/03/9.-ILSI-2017... · Composition and Food Biomarker Analysis ... determination

Embed Size (px)

Citation preview

Metabolomics for Food

Composition and Food

Biomarker Analysis

David Wishart, University of Alberta

2017 ILSI North America FNSP Mid-Year Meeting

July 18, 2017 Arlington Virginia

Disclosures and Conflicts of Interest

I have no conflict of interest to

report in relation to this

presentation.

Funding

Some Definitions

• Metabolomics – The high throughput

analysis and characterization of the

chemicals constituting the metabolome

• Food Composition Analysis – The

determination of the chemical

(metabolite) components in food

• Food Biomarkers – Food components

or food metabolites found in the

metabolome that are characteristic of

specific foods

The Metabolomics Workflow

1234567ppm

Biological or Tissue Samples Extraction Biofluids or Extracts

Chemical AnalysisData Analysis

Metabolomics Technologies

• UPLC, HPLC

• CE/microfluidics

• GC-MS

• LC-MS

• LC-MS/MS

• ICP-MS

• NMR spectroscopy

• X-ray crystallography

• FTIR

Compound ID by GC/LC-MS

LC/GC-MS total

Ion chromatogram

CH3

Compound ID by NMR

http://bayesil.ca

Technology & Sensitivity

M mM M nM pM fM

# M

eta

bo

lite

s d

ete

cte

d (

Lo

g1

0)

0

1

2

3

4

Sensitivity or LDL

LC-MS or

DI-MS

NMR

GC-MS

Quad

GC-MS

TOF

Known unknowns

Unknown unknowns

Metabolomics for Food Analysis

1234567ppm

Food Samples Extraction Food Extracts

Chemical AnalysisData Analysis

Food Component Analysis

• Protein

• Fat

• Ash

• Minerals

• Carbohydrates

• Calories

• Water content

• Alanine

• Tryptophan

• Methyl-histidine

• PE(18:0/18:2)

• PE(16:0/18:1)

• TG(16:0/16:0/18:0)

• Phosphate

• Calcium

• Zinc

• Fructose

• Glucose

• N-acetylglucsoamine

• Apigenin

• Gallic acid

• Resveratrol

• Epigallocatechin Gallate

• Proline betaine

• …..

Traditional Metabolomics

Advantages of Metabolomics

• Applicable to food composition analysis

and food biomarker analysis

• Uses rapid, high throughput

methodologies (robotics, UPLC, MS/MS)

• Offers exquisite sensitivity (< 1 nM)

• Can be absolutely quantitative

• Can detect 100s to 1000s of compounds

vs. only 10’s for conventional methods

The Rush to Metabolomics

WineScreener & JuiceScreener

Bruker BioSpin ($600,000)

The Alberta Food

Composition Project• A 5 year metabolomics project aimed at

measuring the chemical composition of

~50 Alberta-grown food products (meat,

poultry, dairy, cereals, oils, vegetables

& fruits)

• Measuring 100’s of compounds via

quantitative metabolomics techniques

• Intent is to extend and validate

literature-based composition data and

identify candidate food biomarkers

Type Model

NMR Bruker 700 MHz cryoprobe & autosampler

NMR Varian/Agilent 600 MHz

ICP-MS Perkin Elmer NexION 350

GC-MS 2x Agilent 7890A GC-MS with autosampler

GC-MS Aglient/HP Series 5890 GC-MS

LC-MS Bruker 9.4T FT-ICR MS w. cap HPLC

LC-MS 2x ABI Qtrap 4000 MS w. Turbo ESI

LC-MS Agilent LC-ESI ToF MS

LC-MS 2x Bruker maXis II qTOF

HPLC Agilent w. Fluorescent detector

UPLC 2x Agilent 1290 Infinity UPLC

TMIC Overview

www.metabolomicscentre.ca

Using Multiple Platforms

Typical Results

• ~50 water-soluble metabolites by NMR

• ~80 compounds by DFI/LC-MS/MS via the

Biocrates AbsoluteIDQ™ kit

• ~400 lipids and fatty acids via GC-MS/LC-MS

• 53 trace elements by ICP-MS

• 48 small metabolites via GC-MS

• 17 polyphenols

• 9 water- and lipid-soluble vitamins

• Identification & quantification of ~600 cmpds

Data Summary (Non-lipid

Components)

Metabolomics of Beet Root

HPLC NMR GCMS GCMS ICPMS DIMS Lipidomics(volatiles)

Po

lyp

he

nols

, vita

min

s,

ch

loro

phylls

, a

nth

ocya

nin

s

Org

an

ic a

cid

s, a

min

o a

cid

s,

am

ine

s, su

ga

rs, p

oly

ols

Org

an

ic a

cid

s, a

min

o,

acid

s, fa

tty a

cid

s

Vo

latile

org

an

ics, th

iols

,

ke

ton

es, e

ste

rs

Me

tals

Ca

rnitin

es, a

min

o a

cid

s,

ph

osp

ho

lipid

s, a

min

es

Ne

utr

al lip

ids, ch

ole

ste

rols

Ph

yto

ste

rols

, fa

tty a

cid

s

Num

bers

of Q

uantified C

om

pounds

METHOD

34 37

26

4138

85

~400

Alberta Food Composition Database

www.afcdb.ca

The Food Database (FooDB)

• 26,619 compounds, 25,579

structures with 24,843

descriptions

• 171,359 synonyms

• ~700,000 concentration values

• 31,791 references

• 1376 cmpds with health effects

• 2692 cmpds with flavour data

• Content data on 907 raw or

processed foods

• Supports structure & text

searches

• >100 data fields/compound

• Full data downloadswww.foodb.ca

Food & Food Biomarkers

• Data on food composition provides

important hints for potential food

biomarkers

• Food biomarkers need to be identified

to improve or corroborate findings

from food frequency questionnaires, 24

hr dietary recalls or food diaries

• Most food biomarkers are metabolites

of food components

Food & Food Biomarkers

Metabolomics for Food

Biomarkers

1234567ppm

Metabolism Biofluids

Chemical AnalysisData Analysis

Food Samples

Food

Metabolites/Biomarkers• Tea Epicatechin Benzoic acid Hippuric acid

• Meat Carnosine 1-Methylhistidine

• Eggs Lecithin Phosphocholine TMA TMAO

• Wine Ethanol Ethyl glucuronide

• Soy Daidzein Equol

• Onions Quercetin m-Hydroxyphenylacetate

• Anchovies Purines Uric acid

• Coffee Caffeine Theophylline Methylurate

• Grains Alkylresorcinols 3,5 Dihydroxybenzoate

Food Biomarker Alliance

http://foodmetabolome.org

The FoodBall Project

• A 4 year project (ends in 2018) to use

metabolomics to systematically identify and

validate food biomarkers

• 22 different partners in Holland, Belgium,

Ireland, Norway, Spain, Switzerland, France,

Denmark, Germany, Italy, Canada & ILSI

Europe working on diet intervention studies,

literature reviews, large cohort studies

• Funded by the EU & Canada (JPI-BioNH, CIHR)

• Goal is to make all the findings (and most

compounds) freely available

The Human Metabolome

Database (HMDB)• A web-accessible resource

containing detailed

information on 74,462

“quantified”, “detected” and

“expected” metabolites

• Includes many food

components & metabolites

• 1000’s of food additives

• >10,000 reference spectra

• Supports sequence, spectral,

structure and text searches as

well as compound browsing

• Major update for 2017http://www.hmdb.ca

Phenol-Explorer

• >900 peer-reviewed

papers, 37,000 data entries

• Content data on 502

polyphenols in 459

different plant foods

• >300 polyphenol

metabolites

• Data on food, food variety,

commercial origin,

extraction protocol,

analytical methods &

references

• Searchable by food name,

compound name or

polyphenol class

www.phenol-explorer.eu

Exposome-Explorer

• A database of dietary

and pollutant biomarkers

• 142 food/diet biomarkers

• 346 pollutant biomarkers

• >10,500 concentration

values in blood & urine

• From 480 publications

• Correlation values

• Supports text, structure

and concentration

searches

http://exposome-explorer.iarc.fr

Conclusions & Takeaways

• Metabolomics is transforming food

composition analysis

• Food composition databases have grown

considerably in depth and breadth thanks to

metabolomics

• Quantitative metabolomics is leading to the

identification of dozens of new and useful

food biomarkers

• Many of these findings are now freely

available on the web

Acknowledgements

• Augustin Scalbert

• Claudine Manach

• Rosa Vazquez Fresno

• Shima Borzouie

• Rupa Mandal

• Allison Pon

• Tanvir Sajed

• Reza Jafari

• Naama Karu

• Ana Marcu

• An Chi Guo

• Edison Dong

• FoodBall Team