is Best Method for Analysis of Drug Residues in Animal ... · kanamycin, 4 gentamicins, neomycin, apramycin ... ampicillin, cefazolin, penicillin G, oxacillin, cloxacillin ... 5‐Day
Post on 19-May-2018
215 Views
Preview:
Transcript
What is the Current Best Method forMulticlass, Multiresidue Analysis of
Veterinary Drug Residues in Animal Tissues?
Steven J. Lehotay, Alan R. Lightfield, Lucía Geis‐Asteggiante, and Marilyn J. Schneider
Violation Rate on Inspector‐GeneratedBovine Slaughter Classes
1498 violations
See OIG Report: www.usda.gov/oig/webdocs/24601-08-KC.pdf
0.00%
0.20%
0.40%
0.60%
0.80%
1.00%
1.20%
1.40%
1.60%
1.80%
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
%Vi
olat
ions
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
400,000beta-Lactams AminoglycosidesSulfas TetracyclinesMacrolides FlunixinTotal No. Samples
Veterinary Drug Residue Analysis by FSIS
Slide modified from James Withee, FSIS
≈6,300SlaughterHouses
InspectorSamplesn ≈ 150,000/yr≈0.5% of cows
KISTM
orFAST
3 – 6 hrs
–FSIS Lab
St. Louis, MO
+
7‐PlateBioassay
24 – 48 hrs
Shipment24 – 48 hrs
–
> 7 days
–
Analytical Methods*
+
*Confirmation or Determination by Class of Drugs+
QuantitativeBioassay
24 hrs < Tol.
> Tol.
FDA‐CVMEnforcement
Proposed FSIS Residue Monitoring Scheme
≈6,300SlaughterHouses
InspectorSamplesn > 150,000/yr>0.5% of cows
KISTM
orPremi ?
3 hrs
–FSIS LabsSt. Louis, MOAthens, GAAlameda, CA
+
UHPLC‐MS/MS*24 hrs
Shipment24 – 48 hrs
– *Qualitative Screening,Identification, and Semi‐Quantification
QuantitativeBioassay
24 hrs
+
Analytical Confirmation and
Quantification
< 7 days
Bridging
Studies? years
> Tol.
FDA‐CVMEnforcement
4444(+13)(+13) Antibiotics in Initial MMMAntibiotics in Initial MMM(+AMGs)(+AMGs)
Tetracyclines Tetracyclines (3)(3) tetracycline, oxytetracycline, chlortetracycline
AminoglycosidesAminoglycosides (13)(13) spectinomycin, streptomycin, dihydrostreptomycin, amikacin, kanamycin, 4 gentamicins, neomycin, apramycin, hygromycin, tobramycin
MacrolidesMacrolides (8)(8) erythromycin, tilmicosin, lincomycin, tylosin, clindamycin, pirlimycin, tulathromycin, gamithromycin
ββ‐‐LactamsLactams (10)(10) desacetyl cephapirin, amoxicillin, DCCD (cefitofur metabolite), ampicillin, cefazolin, penicillin G, oxacillin, cloxacillin, nafcillin, dicloxacillin
SulfonamidesSulfonamides (16)(16) sulfamethazine, sulfathiazole, sulfachloropyridazine, sulfadoxine, sulfadimethoxine, sulfapyridine, sulfadiazine, sulfamerazine, sulfaquinoxaline, sulfaethoxypyridazine, sulfanilamide, sulfamethoxazole, sulfamethizole, sulfabromomethazine, sulfanitran, sulfamethoxypyridazine
FluoroquinolonesFluoroquinolones (7)(7) ciprofloxacin, danofloxacin, enrofloxacin, sarafloxacin, difloxacin, desethylene ciprofloxacin, norfloxacin
1818 Other Drugs Added to the Target ListOther Drugs Added to the Target List
OtherOther (4)(4) melengestrol acetate, zeranol, carbadox, 2‐quinoxaline carboxylic acid (marker residue for carbadox)
PhenicolsPhenicols (3)(3) chloramphenicol, florfenicol, florfenicol amine
ββ‐‐agonistsagonists (5)(5) cimaterol, clenbuterol, ractopamine, salbutamol, zilpaterol
NSAIDsNSAIDs (6)(6) β+dexamethasone, flunixin, oxyphenylbutazone, phenylbutazone, prednisone
Int StdsInt Stds / / QCQC (6)(6) sulfamethazine‐13C6, flunixin‐d3, penicillin G‐d7atrazine, methomyl, imidacloprid
62(+13)62(+13) Drugs in Initial Method(s) Drugs in Initial Method(s) +6+6 ISIS//QCQC
Developed, Validated, and TransferredDeveloped, Validated, and TransferredMethod for Method for Aminoglycosides (AMGs)Aminoglycosides (AMGs)in Bovine Kidney, Liver, and Musclein Bovine Kidney, Liver, and Muscle
Mix a 2.0 g sample with 10 mM NH4OAc / 0.4% EDTA / 1% NaCl / 2% TCA buffer in stomacher.Centrifuge and decant supernatant.Adjust pH to 6.5 with NaOH and/or HCl solutions.Sample clean‐up / analyte recovery performed using disposable pipette extraction (loose 50 mg WCX sorbent in 5 mL tip) on manifold extractor apparatus.Elute the analytes using 1.0 mL of 10% formic acid in H2O
4‐fold higher sample throughput thanprevious USDA‐FSIS method
UHPLC‐MS/MS Chromatographic Profile of 13 Aminoglycosides
2.5 min
>10‐fold faster than previous USDA‐FSIS method
HPLC‐MS/MS ChromatographicProfile of Veterinary Drugs in Tissues
20 min
UHPLC‐MS/MS Chromatographic Profile of Veterinary Drugs in Tissues
Mobile Phase: A – 95% water / 5% MeCN / 0.1% formic acidB – 100% MeCN / 0.1% formic acid
7.5min
GoalGoal:: Identify in Kidney at Identify in Kidney at ½½ ““ToleranceTolerance”” LevelsLevels½ “Tol.” (ng/g) Analytes No.
3 ββ‐‐Agonists, zeranol, chloramphenicolAgonists, zeranol, chloramphenicol 7
5 amoxacillin, ampicillin, cloxacillin 3
10 melengestrol acetate 1
12.5 flunixin 1
15 carbadox, 2‐quinoxaline carboxylic acid 2
25 Fluoroquinolones, penicillin G 8
50 Sulfonamides, remaining NSAIDs, Macrolides, ββ‐‐LactamsLactams 31
60 tilmicosin 1
100 florfenicol, tylosin 2
150 florfenicol amine 1
250 pirlimicin 1
500 oxytetracycline, tetracycline 2
1000 chlortetracycline, tulathromycin 2
400 Fixed IS/QCs 6
1)Mol et al. (Rikilt – The Netherlands)2)Martos et al. (U. Guelph – ON, Canada)3)Mastovska et al. (USDA‐ARS – Wyndmoor, PA)4) Leepipatpiboon et al. (Chulalongkorn U., Thailand)5) Stubbings et al. (FERA – York, UK)6) Kaufmann et al. (Switzerland) – Too Tedious!
Comparison of 6 Vet. Drug MMMs
All methods gave similar qualitative MS/MS screening capabilities with nearly all of the drug analytes meeting identification criteria at ½ “tolerance” level in kidney.
Speed, cost, ease of use and ruggedness become the differentiating aspects.
Comparison of 5 Vet. Drug MMMs½x, 1x, and 2x “Tolerance” Levels (n= 6 each) in Kidney, No I.S.
0%
20%
40%
60%
80%
100%
120%
Avg. R
ecoveries
Mol Martos Mastovska Leepipatpiboon Stubbings6
β‐Lactam
s7 Fluoro‐
Quinolones
8 Growth
Prom
oters
7Macrolides
14Sulfonamides
3Tetracyclines
Evaluation of Incurred Samples
• FSIS provided 10 kidneys found in their monitoringprogram to contain drug residues.
•We analyzed the samples in blind fashion(unknown drugs and unknown levels).
• These were each analyzed in duplicate using thedifferent features of Mol, Martos, and Mastovska (et al.) 3 MMMs to compare and assess their performances on real samples.
Analysis of Incurred Kidney (2 g)
Pirlimycin in Incurred Kidney (no IS)
0
50
100
150
200
250
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid5 min shake; w/ acid30 min shake; w/o acid30 min shake w/ acid60 min heat; w/o acid60 min heat; w/ acid
Analysis of Incurred Kidney (2 g)
Sulfamethazine in Incurred Kidney (no IS)
020406080
100120140160180
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid5 min shake; w/ acid30 min shake; w/o acid30 min shake w/ acid60 min heat; w/o acid60 min heat; w/ acid
Beta/Dexamethasone in Incurred Kidney (no IS)
050
100150200250300350400450
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid5 min shake; w/ acid30 min shake; w/o acid30 min shake w/ acid60 min heat; w/o acid60 min heat; w/ acid
Beta/Dexamethasone was missed by FSIS
Analysis of Incurred Kidney (2 g)
Analysis of Incurred Kidney (2 g)
Flunixin in Incurred Kidney (no IS)
0
100
200
300
400
500
600
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid5 min shake; w/ acid30 min shake; w/o acid30 min shake w/ acid60 min heat; w/o acid60 min heat; w/ acid
Flunixin vs. Flunixin-d3 IS in Incurred Kidney
0
100
200
300
400
500
600
700
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid
5 min shake; w/ acid
30 min shake; w/o acid
30 min shake w/ acid
60 min heat; w/o acid
60 min heat; w/ acid
Analysis of Incurred Kidney (2 g)
Penicillin G vs. PenG-d7 IS in Incurred Kidney
0
10
20
30
40
50
60
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid 5 min shake; w/ acid30 min shake; w/o acid 30 min shake w/ acid60 min heat; w/o acid 60 min heat; w/ acid
Spikes were ok, but incurredpenicillins can’t tolerate acids
No need for long shake nor heat,so fast and cool is fine!
Analysis of Incurred Kidney (2 g)
““Goldilocks?Goldilocks?”” MMM for Vet. DrugsMMM for Vet. Drugs
extraction
clean‐up
add 10 mL of 4/1 (v/v) MeCN/watervortex briefly, shake for 5 mincentrifuge for 5 min >3500 rcf
add IS mix (SMZ‐IS; flunixin‐d3; PenG‐d7)
2 g tissue in a 50 mL tube
supernatant + 500 mg C18 + 10 mL hexane sat’d w/MeCN; mix for 30 s, centrifuge for
5 min > 3500 rcf; aspirate hexane to waste
evaporate 5 mL extract to 1 mL final vol.
filter extract with the Mini‐UniPrepTM
UHPLC‐MS/MS analysis
Modified from: K. Mastovska, A.R. Lightfield, J. Chromatogr. A 1202 (2008) 118‐123.
Method LogisticsMethod Logistics
1 chemist was able to process 60 pre1 chemist was able to process 60 pre‐‐homogenized homogenized samples in an 8samples in an 8‐‐hr day for an overnight sequencehr day for an overnight sequence
(longest step was 1 hr to evaporate MeCN)(longest step was 1 hr to evaporate MeCN)
No glassware to be cleaned afterwardsNo glassware to be cleaned afterwards
Cost of materials Cost of materials ≈≈ $3/sample (using bulk C18)$3/sample (using bulk C18)
Waste = 10 mL hexane and 5 mL MeCNWaste = 10 mL hexane and 5 mL MeCN(and two 50 mL and one 15 mL PP tubes)(and two 50 mL and one 15 mL PP tubes)
Streamlined Method ValidationNeedsNeeds::•• Trueness (Recoveries at Trueness (Recoveries at ≥≥3 Levels, n=10)3 Levels, n=10)•• Precision (Repeatability & Reproducibility)Precision (Repeatability & Reproducibility)•• Ruggedness (MultiRuggedness (Multi‐‐day, Multiday, Multi‐‐Analyst, etc.)Analyst, etc.)•• Selectivity (Interferences in Blanks?)Selectivity (Interferences in Blanks?)•• Range (Calibration and Matrix Effects)Range (Calibration and Matrix Effects)•• Detection Limits (LOI < Detection Limits (LOI < ½½ ““Tol.Tol.””?)?)•• Qualitative (False Negatives/Positives)Qualitative (False Negatives/Positives)
Can We Meet All Needs in 3Can We Meet All Needs in 3‐‐5 Days?5 Days?
55‐‐Day Validation ExperimentDay Validation ExperimentDay 1Day 1::
•• Analyst 1 in hot lab, Reagents A, 10 matrix blanks Analyst 1 in hot lab, Reagents A, 10 matrix blanks from different sources, 6 spikes at 3 levels each in from different sources, 6 spikes at 3 levels each in 6 matrices + 4 spikes each at same levels in mixed 6 matrices + 4 spikes each at same levels in mixed matrices (1 in glass tubes); 5matrices (1 in glass tubes); 5‐‐point calibration point calibration each in mixed matrix and reagenteach in mixed matrix and reagent‐‐only stds; only stds; reagent blk = 0reagent blk = 0‐‐Std injStd inj’’d after high std to check for d after high std to check for carrycarry‐‐over over
Days 2 & 3 (repeated on Days 4 & 5)Days 2 & 3 (repeated on Days 4 & 5)::
•• Analysts 2 & 3 in cooler labs repeat using Analysts 2 & 3 in cooler labs repeat using Reagents B & C with different sources of matricesReagents B & C with different sources of matrices
QuantitativeQuantitative Method ValidationMethod Validation
A) Trueness: Recoveries of Spiked Samples
B) Precision: Repeatability and Reproducibility
‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐The MMM cannot be used for enforcement actions if
there is a tolerance value unless the method is bridged to the FDA‐approved method (or it becomes an AOAC Official Method). The method may be
determinative if there is no tolerance value for the analyte, or an enforcement action is not being taken.
Recoveries of Tylosin (100Recoveries of Tylosin (100‐‐400 ng/g)400 ng/g)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Day 1 Day 2 Day 3 Day 4 Day 5 Average
Rec
over
y
Stev
e
Luci
a
Ala
n
Mar
ilyn
Chr
istin
e
n = 30 each dayn = 30 each day(10 reps x 3 levels)(10 reps x 3 levels)
n =
150
Avg. Repeatability = 12% RSD
Reproducibility = 21% RSD
Average Recoveries of the 62 DrugsAverage Recoveries of the 62 Drugs
0
10
20
30
40
50
60
70‐100 50‐69 <50
%Recovery
No. of D
rug Ana
lytes
MatrixMatrix‐‐matched stdsmatched stdsw/o internal stdsw/o internal stdsn = 150 per drug n = 150 per drug
(10 reps x 3 levels x 5 days)(10 reps x 3 levels x 5 days)20 different kidneys20 different kidneys
Precision in the Analysis of the 62 DrugsPrecision in the Analysis of the 62 Drugs
Using matrixUsing matrix‐‐matched stds w/o normalization to internal stdsmatched stds w/o normalization to internal stds
0
10
20
30
40
50
60
≤15 16‐20 21‐25 >25%RSD
No. of D
rug Ana
lytes Reproducibility Repeatability
n = 150 per drug n = 150 per drug (10 reps x 3 levels x 5 days)(10 reps x 3 levels x 5 days)
20 different kidneys20 different kidneys
QualitativeQualitative Method ValidationMethod Validation
A) Screening: Capability to replace the 7‐plate bioassay in FSIS
B) Identification: Capability to replace the single class LC/MS‐MS “confirmation” methods
‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐Then, the NADA (FDA‐approved enforcement method)
is used for both quantitative determination and qualitative confirmation of drug violations.
ScreeningScreening Criteria and ResultsCriteria and Results
USDA‐FSIS Criteria:Presence of at least one fragment ionSignal/Noise > 3Retention time match within ±5%
54/62 analytes met screening criteria at ½ “Tol.” levelChloramphenicol met screening criteria at 2x “Tol.” level
8 drugs to be addressed in future work:desacetyl cephapirin, florfenicol amine, clenbuterol,phenylbutazone, oxyphenylbutazone, tulathromycin,cimaterol, amoxacillin
1. Retention time (tR) is within ±4 SD of average tRand peak shape matches that of reference std
2. tR and peak shape of qual. ion(s) matches those of the quantification ion
3. 2 qual. ions ≤|20%| or 1 qual. Ion ≤|10%| of avg ion ratio from contemporaneous reference stds
4. Absence of positive findings in blanks
5. Signal > ½ “Tol.” calibration stds in matrix
Our LC‐MS/MS Identification Criteria
Ret. Time Consistency for IdentificationRet. Time Consistency for Identification
0
2
4
6
8
10
12
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00
Avg tR (min)
4 x
Std
Dev
(s)
Combined, n=250Solvent Stds, n=50Matrix Stds, n=50
Desacetyl cephapirin and Florfenicol amine need sorted out
Tulathromycin A ?
Clenbuterol
Ion Ratios in Spks, n=50 each
0%
20%
40%
60%
80%
100%
120%
5 10 20Spk Conc. (ng/g)
Ion
Rat
io
1/2x Spk (2/1 ions) 1x Spk (3/1 ions) 2x Spk (3/2 ions)
Ion Ratios for Ampicillin in MatrixIon Ratios for Ampicillin in Matrix
Ion ratio variability decreases vs. Ion ratio variability decreases vs. ↑↑ conc.conc.
1 false neg @ 5 ng/g (2%); no false positives, n=501 false neg @ 5 ng/g (2%); no false positives, n=50
(10 x 5 days)
Ion Ratio Criteria in 2002/657/EC (EU)Ion Ratio Criteria in 2002/657/EC (EU)
Rel. Abundance Rel. Abundance Acceptable Diff. vs. Ref.Acceptable Diff. vs. Ref.vsvs. Base Peak. Base Peak APIAPI‐‐MSMS
>50%>50% ±±20% RSD20% RSD>20>20‐‐50% 50% ±±25% RSD25% RSD>10>10‐‐20%20% ±±30% RSD 30% RSD ≤≤110%0% ±±50% RSD50% RSD
Ref. RatioRef. Ratio EU RangeEU Range** FSIS (1 ion)FSIS (1 ion) (2 ions)(2 ions)70%70% 56% – 84% 60% – 80% 50% – 90%24% 18% – 30% 14% – 34% 4% – 44%12% 8.4% – 15.6% 3% – 23% >0% – 33%4% 2% – 6% >0% – 14% >0% – 24%
* 2 ion transitions needed to achieve 3 ident. points in MS/MS
Ion Ratios of Spks, n=30 each
0%
20%
40%
60%
80%
100%
120%
140%
160%
Ions 2/1 Ions 3/1 Ions 3/2
Ion
Rat
io
Day 1 Day 2 Day 3 Day 4 Day 5
Ion Ratios for Ciprofloxacin in KidneyIon Ratios for Ciprofloxacin in Kidney
Conc.,
ng/g EU FSIS
25 (n=50) 54% 2%
50 (n=50) 40% 0%
100 (n=50) 26% 0%
False Negatives
No False Positives, n=50 (EU or FSIS)
(10 x 3 levels)
Ion Ratios for Lincomycin in KidneyIon Ratios for Lincomycin in KidneyIon Ratios of Spks, n=30 each
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
Ions 2/1 Ions 3/1 Ions 3/2
Ion
Rat
io
Day 1 Day 2 Day 3 Day 4 Day 5
Conc.,
ng/g EU FSIS
25 (n=50) 26% 0%
50 (n=50) 8% 0%
100 (n=50) 2% 0%
False Negatives
No False Positives, n=50 (EU or FSIS)
(10 x 3 levels)
Ion Ratios in Stds, n=10 each
0%
100%
200%
300%
400%
500%
1.5 3.0 6.0 12 18Conc. (ng/g)
Ion
Rat
ioSolvent 2/1 Matrix 2/1Solvent 3/1 Matrix 3/1Solvent 3/2 Matrix 3/2
Ion Ratios for CimaterolIon Ratios for Cimaterol
Quant. ion gave 78Quant. ion gave 78±±14 14 %recoveries, n=150%recoveries, n=150
Kidney matrix interferencesKidney matrix interferences<20 ng/g for ions 2 & 3<20 ng/g for ions 2 & 3
1 false positive in1 false positive in50 matrix blanks50 matrix blanks
Summary of Identification ResultsSummary of Identification Results
41/61 drug analytes met ident. criteria at all levels
3 analytes met ident. criteria at the 1x “Tol.” level:desethylene ciprofloxacin, cefazolin, florfenicol
5 analytes met ident. criteria at the 2x “Tol.” level:sulfaniliamide, sarafloxacin, sulfanitran, cloxacillin, melengesterol acetate
12 drugs did not meet ident. criteria at the spk levels
An UHPLC‐MS/MS instrument upgrade would lowerlimits of identification through dilution of extracts.
Veterinary Drug Residues ConclusionsVeterinary Drug Residues Conclusions
•• The streamlined method has met validation The streamlined method has met validation criteria for criteria for ≥≥49/61 drugs in a 549/61 drugs in a 5‐‐day validation for day validation for qualitative identification screening purposes.qualitative identification screening purposes.
•• Sample throughput is 60 samples/day by 1 chemist Sample throughput is 60 samples/day by 1 chemist for UHPLCfor UHPLC‐‐MS/MS analysis.MS/MS analysis.
•• The method is being implemented for routine The method is being implemented for routine monitoring of cattle (so far) by the USDA labs.monitoring of cattle (so far) by the USDA labs.
•• Quantification is acceptable for Quantification is acceptable for ≈≈60% of the drugs, 60% of the drugs, but enforcement still requires the NADA method. but enforcement still requires the NADA method.
•• Is the new MMM Is the new MMM ““just rightjust right”” to deserve the name to deserve the name ““GoldilocksGoldilocks””??
Acknowledgments
USDA FSIS:Terry DutkoLouis BluhmChilton Ng
Pat McCaskeyet al.
US‐Israel Binational Agricultural Research
and Development Grant US‐4273‐09
*Currently at CovanceGreenfield, Indiana
KaterinaMastovska*
ChristineHaines
Tanks Berry Mulch!Tanks Berry Mulch!
+ +
+ + ! !
Thanks very much!Thanks very much!
Contacts:
steven.lehotay@ars.usda.gov
marilyn.schneider@ars.usda.gov
Outline of the TalkOutline of the Talk
I. Current Situation and Proposed ChangesA. Feasibility StudyB. Choice of Target Drugs and LevelsC. Aminoglycosides Method
II. Comparison of MMMs and OptimizationA. Spiked SamplesB. Incurred Samples
III. Validation (Quantitative and Qualitative)IV. Conclusions
Ion Ratios for SulfapyridineIon Ratios for SulfapyridineIon Ratios of Spks, n=30 each
0%
50%
100%
150%
200%
250%
300%
Ions 2/1 Ions 3/1 Ions 3/2
Ion
Rat
io
Day 1 Day 2 Day 3 Day 4 Day 5
Ion source was cleaned Ion source was cleaned between Days 2 & 3 between Days 2 & 3 ––ion ratios changed for ion ratios changed for
some analytessome analytes
DefinitionsDefinitions
•• IndicationIndication = = result of a screening methodresult of a screening method((i.e.i.e. ““presumedpresumed”” positive or negative)positive or negative)
•• DeterminationDetermination = = result from an analytical determinative result from an analytical determinative method (method (e.g.e.g. GC/PFPD, LC/UV)GC/PFPD, LC/UV)
•• IdentificationIdentification = = qualitative result from a highly selective qualitative result from a highly selective method (method (e.g.e.g. GCGC‐‐MS, LCMS, LC‐‐MSMSnn))
•• ConfirmationConfirmation = = result from 2 or more independent result from 2 or more independent analyses in agreement (ideally, one of which uses a analyses in agreement (ideally, one of which uses a
different chemical mechanism or approach)different chemical mechanism or approach)
Syringeless FiltersSyringeless FiltersMini-UniPrepTM (Whatman)
1) Place unfiltered sample (max. 0.5 mL) in chamber.
2) Compress filter plunger into sample chamber. Clean filtrate fills reservoir bottom up.
3) Place the Mini-UniPrepTM vial in an autosampler.
aqueous samples: PVDF(polyvinylidenefluoride) filter
LC-MS/MS InstrumentationWaters Acquity TQD = UPLC-MS/MS system
UPLC (Ultra-Performance Liquid Chromatography)Pressures up to 15,000 psiLow extra-column volumesCompatible with sub-2 μm particle columns
TQD - Fast triple quadrupole MSRapid data acquisition (5 ms dwell time)Rapid positive/negative ion mode switching (20 ms)
• Fast analysis with good resolution• Lower solvent consumption, less waste• Narrow peaks – fast data acquisition needed
Analysis of Incurred Kidney (2 g)
Penicillin G in Incurred Kidney (no IS)
0
10
20
30
40
50
60
1 2 3 4 5 6 7 8 9 10
Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid5 min shake; w/ acid30 min shake; w/o acid30 min shake w/ acid60 min heat; w/o acid60 min heat; w/ acid
Analysis of Incurred Kidney (2 g)
Sulfadimethoxine in Incurred Kidney (no IS)
0
50
100
150
200
250
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid5 min shake; w/ acid30 min shake; w/o acid30 min shake w/ acid60 min heat; w/o acid60 min heat; w/ acid
Analysis of Incurred Kidney (2 g)Tilmicosin vs. SMZ-d6 IS in Incurred Kidney
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid
5 min shake; w/ acid
30 min shake; w/o acid
30 min shake w/ acid
60 min heat; w/o acid
60 min heat; w/ acid
Analysis of Incurred Kidney (2 g)
Tetracyclines (Sum) in Incurred Kidney (no IS)
0
1000
2000
3000
4000
5000
6000
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid5 min shake; w/ acid30 min shake; w/o acid30 min shake w/ acid60 min heat; w/o acid60 min heat; w/ acid
Uncertainty (U) Limitation in the MethodUncertainty (U) Limitation in the Method• n = 200 extractions among 20 kidneys over 5 days (n = 5)
Method Step QC‐StdAvg. %
RecoveryAvg.%RSD % of U
UHPLC‐MS/MS Atrazine 90* 11 100
Cleanup Methomyl 92 9 0
Extraction Flunixin‐d3 94 10 0
Homogenization Imidacloprid 79† 10 0
* Should be 100% (‐10% bias in spks vs. stds?)† ≈11‐15% probably lost during homogenization
Ion Ratios for Tulathromycin AIon Ratios for Tulathromycin A
Ion Ratios in Stds, n=10 each
0%
20%
40%
60%
80%
100%
120%
140%
160%
500 1000 2000 4000 6000Conc. (ng/g)
Ion
Rat
ioSolvent 2/1 Matrix 2/1Solvent 3/1 Matrix 3/1Solvent 3/2 Matrix 3/2
(UHP)LC(UHP)LC‐‐MS(/MS) CapabilitiesMS(/MS) Capabilities
Simultaneous identification and quantificationSimultaneous identification and quantification
Sensitive determination in complex matrices Sensitive determination in complex matrices
Wide scope (analyte, matrices) Wide scope (analyte, matrices)
Multiclass, multiresidue methods (MMMs)Multiclass, multiresidue methods (MMMs)
Identification/Confirmation
Lab‐Based Screening
Fast, wide-scope sample preparation is the key!
Slide modified from Kate Mastovska
Tier 1: Screening Test Performed by the USDA‐FSIS Inspectors in the Slaughterhouse.
Currently with Microbial Inhibition Tests
Tier 2: Presumptive Positive Samples Sent to the USDA‐FSIS Laboratory for Quantitative and Qualitative Analysis.
Currently with 7‐plate bioassay, but we propose to use UHPLC‐MS/MS at least for lab‐based screening and antibiotic identifications
Two‐Tiered Approach for Antibiotics
Major Classes of Antibiotics Major Classes of Antibiotics
AminoglycosidesAminoglycosides
SulfonamidesSulfonamides TetracyclinesTetracyclinesββ‐‐LactamsLactams
MacrolidesMacrolides QuinolonesQuinolones
Penicillin G Sulfadimethoxine Tetracycline
Gentamicin C1 EnrofloxacinErythromycin
Currently, 219 vet. drugs (including 94 antibiotics)Currently, 219 vet. drugs (including 94 antibiotics) are on our MMM list are on our MMM list
FieldField‐‐Based Microbial Inhibition AssaysBased Microbial Inhibition Assays
FAST Premi® KISTM
Pictures by Paul Pierlott
FAST, PremiFAST, Premi®® & KIS& KISTMTM ComparisonComparisonSpiked Samples of Kidney (Liquid) Spiked Samples of Kidney (Liquid)
Threshold concentration (µg/mL)
Antibiotic FAST Premi® KISTM
U.S. tolerance(µg/g)
Penicillin G 0.4 0.005 0.005‐0.01 0.05
SDMX 1.4 0.12 0.01‐0.1 0.1
OTC 1.5 2.1 0.5‐1.5 12
Tylosin 1.2 0.09 0.1‐0.2 0.2
Danofloxacin 1.7 4.5 4‐>6 0.2 (L)
Streptomycin 1.4 9.0 4‐10 2.0
Neomycin 0.04 1.2 0.4‐0.6 7.2
Spectinomycin 66 3.5 5‐6 4.0
M.J. Schneider, S.J. Lehotay, Anal. Bioanal. Chem. 390 (2008) 1775‐1179
vs. SMZ‐d6
0%
20%
40%
60%
80%
100%
120%
140%
160%
180%
200%
beta
‐lacta
m +
Ceph
alosp
orin
Fluor
oquin
olon
Grow
th pr
omot
ors
Mac
rolid
es +
Linco
sami
des
Sulfa
s
Tetra
cycli
ne
Mol
Martos
ARS
Thai
Stubbings
IS
Comparison of 6 Vet. Drug MMMs
Old School Veterinary Drug AnalysisOld School Veterinary Drug Analysis
Single analyteresidue methods
Single‐classresidue methods
Selective multi‐class residue methods
Slide by Kate Mastovska
New School Veterinary Drug AnalysisNew School Veterinary Drug Analysis
Multiclass, MultiresidueMethods (MMMs)
(UHP)LC(UHP)LC‐‐MS(/MS)MS(/MS)
Slide modified from Kate Mastovska
Analysis of Incurred Kidney (2 g)
Sulfamethazine vs. SMZ-d6 IS in Incurred Kidney
0
50
100
150
200
250
1 2 3 4 5 6 7 8 9 10Kidney Sample
Con
cent
ratio
n (n
g/g)
5 min shake; w/o acid
5 min shake; w/ acid
30 min shake; w/o acid
30 min shake w/ acid
60 min heat; w/o acid
60 min heat; w/ acid
IS
7‐Plate Bioassay (Lab Screen)
Pictures from FSIS
Microbial Inhibition AssaysMicrobial Inhibition Assays
Relatively inexpensive and easyRelatively inexpensive and easy
2424‐‐hr sample turnaround timehr sample turnaround time
Not sensitive for all antibiotic classes Not sensitive for all antibiotic classes
Cannot distinguish particular analytes or mixtures Cannot distinguish particular analytes or mixtures
Prone to unidentifiable microbial inhibitions (Prone to unidentifiable microbial inhibitions (≈≈3%)3%)
>60% of 7‐plate samples in FSIS are positives and need to be confirmed by chemical‐based
(LC‐MS/MS) methods anyway
>60% of 7‐plate samples in FSIS are positives and need to be confirmed by chemical‐based
(LC‐MS/MS) methods anyway
Slide modified from Kate Mastovska
Antibiotics – with US tolerance Number
• Dihydrostreptomycin ‐ 6‐14,000 ng/mL 7
• Streptomycin ‐ 75 ng/mL 1
• DCCD ‐ 2‐28 ng/mL 10
• Florfenicol amine ‐ 33 ng/mL 1
• Oxytetracycline ‐ 3‐57 ng/mL 13
• Penicillin G ‐ 2‐3 ng/mL 3
• Pirlimycin ‐ 2‐29 ng/mL 5
• Sulfamethazine ‐ 2‐140 ng/mL 2
Drug Residues Identified inDrug Residues Identified in235 Kidney (Exudate) Samples235 Kidney (Exudate) Samples
Feasibility StudyFeasibility Study:: Vet. Drug Residues in Vet. Drug Residues in Bovine Serum and Kidney (Exudate)Bovine Serum and Kidney (Exudate)
Analysis of 121 drug residues, including 65 antibiotics
Two sample preparation and LC‐MS/MS methods1) aminoglycosides (ion‐pairing LC)2) MMM for veterinary drugs (reversed‐phase LC)
Screened / analyzed 235 serum and kidney (exudate) culled dairy cow samples from a slaughterhouse
M.J. Schneider, K. Mastovska, S.J. Lehotay, A.R. Lightfield, B. Kinsella, C.E. Shultz, Anal. Chim. Acta 637 (2009) 290‐297
Addition of Quality Control Spikes, tooAddition of Quality Control Spikes, too
1)1) Imidacloprid added during sample homogenizationImidacloprid added during sample homogenization
2)2) I.S. compounds added prior to extraction I.S. compounds added prior to extraction ‐‐sulfamethazinesulfamethazine‐d6; flunixin‐d3; Pen.G‐d7
3) Methomyl added prior to the cleanup step
4) Atrazine added to final extracts
RSDoverall = √(RSD12 + RSD2
2 + RSD32 + RSD4
2)
…And Then There Were Three…Leepipatpiboon et al. – TCA & Precipitation!Stubbings et al. – Use of PSA Lowered RecoveriesKaufmann et al. – WAY too Long and Intensive
• Mol et al. – 30 min Shake w/ Acid• Martos et al. – 60 min Heat w/ Acid Addition• Mastovska et al. – 5 min Shake w/o Acid
MMM
Antibiotics – no US tolerance No. of CowsGentamicin (serum) 8 ng/mL 1Kanamycin (serum) 1 ng/mL 1Lincomycin (kidney) 4 ng/mL 1
Other vet drugs in kidney (exudate)Flunixin, >300 (≥1,000) ng/mL 23 (17)Beta/dexamethasone, 20‐75 ng/mL 7 Prednisone, <10 ng/mL 3
Drug mixtures (not counting flunixin) 12
Drug Residues Found by LCDrug Residues Found by LC‐‐MS/MS inMS/MS inAnalysis of 235 Culled Dairy CowsAnalysis of 235 Culled Dairy Cows
top related