Bioassay screening and bioassay-directed identification · PDF file04/12/2006 1 Bioassay...
Transcript of Bioassay screening and bioassay-directed identification · PDF file04/12/2006 1 Bioassay...
04/12/2006
1
Bioassay screening and bioassay-directed identification of known and unknown hormone
active substances
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might be relevant……but in current residue monitoring: only limited number of target compounds…unable to detect new or outdated ones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might be relevant……but in current residue monitoring: only limited number of target compounds…unable to detect new or outdated ones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
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/QTOF
Other hormone bioassays under development
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
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
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
Bioassay versus GC/MS/MS screening
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
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/QTOF
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
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/QTOF
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
LC / Androgen Bioassay detection
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
LC / Androgen Bioassay detection
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