Bioassay screening and bioassay-directed identification · PDF file04/12/2006 1 Bioassay screening and bioassay-directed identification of known and unknown hormone active substances

04/12/2006

1

Bioassay screening and bioassay-directed identification of known and unknown hormone

active substances

[email protected]

Outline

Introduction: the hormone residue challenge

Reporter gene estrogen bioassay

Validation data for calf urine and feed samples

Bioassay versus GC/MS/MS screening

Bioassay-directed identification: LC/bioassay/QTOFMS

Reporter gene androgen bioassay

Bioassay-directed identification:LC/bioassay/QTOFMS

Other hormone bioassays under development

Conclusions

04/12/2006

2

Hormone abuse in food production

Abuse of steroids (estrogens, androgens, gestagens, corticosteroids) and beta-agonists as growth promoting agents in food producing animals

EU ban since 1988: …prohibit...substances having hormonal action...and beta-agonists… (96/22/EC)

Thousands of substances might be relevant……but in current residue monitoring: only limited number of target compounds…unable to detect new or outdated ones

Target level: between zero and MRPL (≤ 1-2 ng/g)Unrealistic to enforce EU ban with analyte-list approach !

Hormone abuse in food production

EU ban since 1988: …prohibit...substances having hormonal action...and beta-agonists… (96/22/EC)

Thousands of substances might be relevant……but in current residue monitoring: only limited number of target compounds…unable to detect new or outdated ones

Solution: Yeast screening methods based on hormonal activity !

simple

robust

fast

applicable to urine, feed, illegal preparations, water and environmentalsamples

04/12/2006

3

Outline

Introduction

Reporter gene yeast bioassays





Bioassay-directed identification: LC/bioassay/QTOF


Conclusions

In vivo bioassays have some disadvantages...

04/12/2006

4

...and perhaps in vitro bioassays should be preferred

Not applicable anno 2006 ?

How to design such an in vitro bioassay ?

Biological action of estrogens

04/12/2006

5

Rikilt yeast Estrogen Assay (REA)

Genetically (stable) modified yeast• That expresses the human estrogen alpha receptor: cDNA of the ERalpha behind the strong GPD-promoter (glyceraldehyde-3-phosfate-dehydrogenase)

• That contains a reporter construct that enables the yeast to produce a green fluorescent protein (yEGFP) following binding of an estrogen to the receptor: two consensus ERE-sequences in a truncated CYC1-promoter (not active cytochrome-c oxidase promoter, due to deletion of UAS1 and UAS2: no induction from sugars, oxygen and iron)

T.F.H. Bovee et al., Gene, 325 (2004) 187-200

Rikilt yeast Estrogen Assay (REA)

Genetically (stable) modified yeast• expresses the human estrogen alpha receptor• contains a reporter construct: the yeast produces a green fluorescent protein (yEGFP) following binding of an estrogen to the receptor

High sensitivity• EC50 of 30 picogram estradiol per well

Fast and easy• only 4 or 24 hours • no cell wall disruption, no addition of a substrate

T.F.H. Bovee et al., Gene, 325 (2004) 187-200

04/12/2006

6

REA yeast bioassay

ERE yEGFP

0.001 0.01 0.1 1 10 100 1000 10000 1000001000000

Concentration [ nM]

200

500

800

1100

1400

1700

2000

Fluo

resc

ence

E2b

E2-benz oate

zearalenone

genistein

E1

DES

EE2

estriol

Response of estrogens in the REA bioassay

04/12/2006

7

REP: Relative Estrogenic Potency

0.013.9 x10-3

5.0 x10-4

0. 01genistein8-prenylnaringenin

0.020.03zearalanol0.020.0917α-estradiol0.10.2estrone

1.0 x10-40.1mestranol0.090.6dienestrol0.090.4hexestrol2.01.0diethylstilbestrol1.01.2ethynylestradiol1.01.017β-estradiol

REP (ERβ)REP (ERα)estrogen

T.F.H. Bovee et al., J. Steroids Biochem. & Molec. Biol. 91 (2004) 99-109

With this REA bioassay you see

That all kind of different estrogenic compounds give a response and this response corresponds to the estrogenic potency of that compound.Only estrogenic compounds. Negligible response to testosterone, progesterone, dexamethasone etc.You will detect known and unknown substances that have estrogenic properties.

T.F.H. Bovee et al., JSBMB 91 (2004) 99-109

04/12/2006

8

Sample clean-up

Calf urine 2 ml, adjust pH 4.8 and incubate o/n at 37 °C with β-glucuronidase/arylsulfataseAdd 2 ml 0.25 M sodium acetate buffer pH 4.8 and subject to SPE on a C18 column (elute with 4 ml acetonitrile)The eluate was applied to an NH2-column and the eluate thus obtained was evaporated to 2 ml under nitrogen

Sample clean-up

100 µL in triplicate for bioassay, remaining 1700 µL for identification (suspect samples only) by LC/bioassay/MS ;

add yeast suspension ; read fluorescence (485/530 nm) and calculate t24-t0 values ;

report suspect (> CCα) or compliant (negative): " on / off "

04/12/2006

9

Extract the sample using the generic SPE...

…put the urine extract into the well...

04/12/2006

10

...add the yeast cell suspension...

...read the fluorescence at t0 and t24 (or t4)

no cell lysis, no reagents, just wait and determine t24 - t0

04/12/2006

11

OutlineIntroduction








Conclusions

Validation data according to 2002/657/EC (1)

CCβ: 20 different blank calf urines and 5x20 calf urines spiked with 17β-estradiol (1 ng/ml), DES (1 ng/ml), ethynylestradiol (1 ng/ml), zearalanol (50 ng/ml), and mestranol (10 ng/ml): suspect

Specificity/interferences: • urine spiked with 1000 ng/ml testosterone and 1000 ng/ml progesterone: compliant• idem, but also spiked with 17β-estradiol (1 ng/ml): suspect

Robustness:• used in routine screening > 2 years: no cell toxicity, blanks always compliant, 1 ng/ml spiked samples always suspect.

ISO 17025 accreditation

T.F.H. Bovee et al., Anal. Chim. Acta 529 (2005) 57-64

04/12/2006

12

Validation data according to 2002/657/EC (2)

CCβ: 20 different blank feeds and 5x20 feeds spiked with 17β-estradiol (5 ng/g), DES (10 ng/g), ethynylestradiol (5 ng/g), zearalenone (1250 ng/g), equol (200000 ng/g): suspect

Specificity/interferences: • feed spiked with 1000 ng/g testosterone and 1000 ng/g progesterone: compliant• idem, but also spiked with 17β-estradiol (5 ng/g): suspect

Robustness:• used in routine screening > 1 year: no cell toxicity, blanks always compliant, 5 ng/g spiked samples always suspect.

ISO 17025 accreditation pending

T.F.H. Bovee et al., Food Add. and Contam. 23 (2006) 556-568

Longterm performance of the estrogen bioassay on urine

T.F.H. Bovee et al., ACA 529 (2005) 57-64

-200

0

200

400

600

800

1000

1200

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63

decision limit CCalpha spiked reagent blank spiked urine

maximum response yeast reagent blank blank urine

04/12/2006

13

Longterm performance of the estrogen bioassay on animal feed

Feed control chart

-200

0

200

400

600

800

1000

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53

Week number

Fluo

resc

ence

decision limit CCα=77

reagent blank spiked

blank feed spiked

reagent blank

blank feed

T.F.H. Bovee et al., FAC 23 (2006) 556-568

Outline

Introduction

Reporter gene estrogen bioassayValidation data for calf urine and feed samplesBioassay versus GC/MS/MS screeningBioassay-directed identification: LC/bioassay/QTOFMS

Reporter gene androgen bioassayBioassay-directed identification: LC/bioassay/QTOFMS

Other hormone bioassays under developmentConclusions

04/12/2006

14

Routine analysis: 126 calf urine samples

Analysed based on estrogen activity using the bioassay

Analysed for specific steroids (incl. stilbenes) by GC/MS/MS

Results:

• GC/MS/MS: 71 samples compliant (< 1 ng/ml); 55 samples contain 17α-estradiol, a few of them also estrone

• Bioassay: 67 compliant (only 3.2 % "false suspects")

Predicted bioassay performance

Bioassay sensitivity is based on hormonal activity:

if the relative estrogenic potency of 17α-estradiol = 0.09,

if CCα17β-E2 corresponds with 0.22, CCβcalc.,17β-E2 with 0.44, and CCβexp.,17β-E2 < 1.0 ng/ml (initial validation study),

then theoretically the bioassay starts seeing 17α-estradiol from 2.4 ng/ml and the 95% detection capability will be between 4.8 and 10.9 ng/ml...

04/12/2006

15


0

10

20

30

40

50

60

70

80

<1 1 2 3 4 5 6 7 8 9 10 >10

GC/MS/MS concentration 17a-E2

# sa

mpl

es

GC/MS/MSBioassay

M.W.F. Nielen et al., Food Add. And Contam. 23 (2006) 556-568

Outline

Introduction

Reporter gene estrogen bioassayValidation data for calf urine and feed samplesBioassay versus GC/MS/MS screeningBioassay-directed identification: LC/bioassay/QTOFMS

Reporter gene androgen bioassayBioassay-directed identification: LC/bioassay/QTOF

Other hormone bioassays under developmentConclusions

04/12/2006

16

Bioassay directed identification of unknowns

Sample pretreatmentand clean-up

Gradient Liquid Chromatography

Bioassay plate

Collection plate

Identification on-lineQTOFMS(/MS)

Identification off-line- UPLC/QTOFMS(/MS) - GC/TOFMS

Bioactivity screening: bioassay as LC detector !

Flowsplit

LC/Bioassay/QTOFMS

2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00Time0

100

%

DES

DES

E1

aE2

bE2

E3

LC/bioassay/QTOF: 1 ng estrogens standard

1 7 1 3 1 9 2 5 3 1 3 7 4 3 4 9

w e l l #

ER biogram

RIC

04/12/2006

17

LC/Bioassay/QTOFMS

Spiked calf urine: 2 ng/ml

2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00Time0

100

%

0

100

%

DESE1

aE2

bE2

E3

TIC, 1400x more intense

...than estrogens !

TIC intensity is 1400X higher…

LC/Bioassay/QTOFMS

(Un)known estrogens in calf urine

0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00Time0

100

%

0

100

%

0

100

%

0

100

%

1 7 1 3 1 9 2 5 3 1 3 7 4 3 4 9

w e l l #

ER

bioactive m/z 241

enterolactone m/z 297

inactive m/z 271

TIC

biogram

04/12/2006

18

LC/Bioassay/QTOFMS

Unknown estrogens in calf urine

80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500m/z0

100

%

241.0797

121.0266

C15H13O3

C7H5O2

CxHyOz database search:4 options; RRT 0.77:

equolOOH

OH

Conclusion

17α-estradiol -natural hormone

equol -phytoestrogen metabolite

Bisphenol-like -endocrine disrupter

nonylphenol -endocrine disrupter

Identified estrogens in calf urines

M.W.F. Nielen et al., Anal. Chem. 76 (2004) 6600-6608

04/12/2006

19

Outline

Introduction







Other bioassays under development

Conclusions

Genetically (stable) modified yeast• expresses the human androgen receptor • contains a reporter construct: the yeast produces a green fluorescent protein (yEGFP) following binding of an androgen to the receptor

Highly sensitive for 17β-testosterone, DHT, boldenone, trenbolone, nortestosterone, THG, etc.

Negligible sensitivity for estradiol, progesterone, dexamethasone etc.

Fast and easy• only 4 or 24 hours • no cell wall disruption, no addition of a substrate

Androgen Bioassay design: like Estrogen Assay

T.F.H. Bovee et al., 2006 submitted

04/12/2006

20

Response of androgens in the RAA bioassay

0.1 1 10 100 1000 10000 100000 1000000

Concentration [ nM]

-100

100

300

500

700

900

1100

Fluo

resc

ence

17ß-T DHT 17a-T PDexa 17ß-E2 THG Bold

T.F.H. Bovee et al., 2006 submitted

Outline

Introduction







Other bioassays under development

Conclusions

04/12/2006

21

Androgens in human urine

Generic screening procedure

2 ml urine -samples, -blanks, -controls;

enzymatic hydrolysis (Helix Pomatia);

SPE C18/NH2, acetonitrile/water eluate, concentrate to 2 ml;

200 µL in triplicate for bioassay, remaining 1400 µL for identification (suspect samples only) by LC/bioassay/MS;

add yeast suspension

Three male and three female volunteers• urine samples as such • urine samples spiked with 5-15 ng/ml THG

Direct Bioassay screening• Result: all urines suspect for androgen activity

LC/bioassay for androgen activity detection• biograms confirm natural androgens via well numbers• biograms indicate additional bioactive well for THG

Example: androgens in human urine

04/12/2006

22

Spiked reagent blank

LC / Androgen Bioassay detection

Biogram

190

290

390

490

590

690

790

890

990

1 11 21 31 41 51 61

well#

AR re

spon

se

βBol

βT

THG

Male urine


Biogram

1 9 03 9 05 9 07 9 09 9 0

1 1 9 01 3 9 01 5 9 01 7 9 01 9 9 02 1 9 0

1 1 1 2 1 3 1 4 1 5 1 6 1

w e l l#

AR

resp

onse

04/12/2006

23

Male urine, spiked with THG


Biogram

1 8 0

6 8 0

1 1 8 0

1 6 8 0

2 1 8 0

1 1 1 2 1 3 1 4 1 5 1 6 1

w e l l#

AR

resp

onse

LC/Bioassay/QTOFMS

LC/TOFMS human urine sampleAndrogenic well #34: C21H28O2 database search

39 commercial, 17 steroids

1626SciFinder

9 steroids,gestagens

44Sigma-Aldrich

gestagens8Steraloids

gestagens5Merck Index

commentsoptionssearch engine

M.W.F. Nielen et al., Anal. Chem. 78 (2006) 424-431

04/12/2006

24

Progesterone

Glucocorticosteroids

Other yeast bioassays in the pipeline

ConclusionA robust bioassay has been developed, validated and

accreditated for estrogens in calf urine and feeds; the androgen version is on track for achievement

Bioassay screening is addressing the 96/22/EC ban on substances having hormonal action

Substances having weaker bioactivity are less sensitive and might not comply with a chemical MRPL (for example zeranol)

Only suspect bioassay screening results must be identified: either by conventional confirmatory GC/MS methods, or usingLC/bioassay/QTOFMS approaches

Conclusions

04/12/2006

25

ConclusionThe RIKILT estrogen bioassay has been given to veterinary control laboratories in the UK (Queens University-Belfast), Italy (University of Turin) and France (Laberca-Nantes) and to several environmental laboratories active in endocrine disruptors (Germany: GSF and TU-Dresden, Netherlands: Aquasense, Belgium: VITO)

The latest laboratories are:

•KFRI - Korean Food Research Institute

•IRAS-UU – University Utrecht – Institute for Risk Assessment Sciences

•Royal Chulabhorn Research Institute in Bangkok

•Gent University, FFW – Faculty Pharmaceutical Sciences

You can try it also and use it, on a co-operation basis.

Invitation

Acknowledgements

Michel Nielen -RIKILT LC-bioassay-MS

Ron Hoogenboom -RIKILT bioassays

Richard Helsdingen -RIKILT bioassays

Gerrit Bor -RIKILT bioassays

Astrid Hamers -RIKILT bioassays

Elsa Antunes Fernandes -RIKILT bioassays

Henri Heskamp -RIKILT LC-bioassay-MS

Eric van Bennekom -RIKILT LC-bioassay-MS

Paula Balzer-Rutgers -RIKILT LC-bioassay-MS

Dutch Ministry of Agriculture, Nature and Food Quality financial support

UK DEFRA (Hormone Radar project) financial support

World anti-doping agency (WADA) financial support

04/12/2006

26

Thank you for your attention

Documents

Bioassay screening and bioassay-directed identification · PDF file04/12/2006 1 Bioassay screening and bioassay-directed identification of known and unknown hormone active substances