J:\pubsdept\oma-17 temp\cd-rom2\995_04.vp

previously charged with Cu ions. Column is washed and AOAC Official Method 995.04
tetracyclines are specifically eluted with buffer containing EDTA.
Multiple Tetracycline Residues in Milk
Eluate is ultrafiltered to remove remaining protein and filtrate is in- Metal Chelate Affinity-Liquid Chromatographic Method
jected onto LC column. Tetracyclines are adsorbed and concentrated First Action 1995
by using 100% aqueous buffer mobile phase, and then eluted with Final Action 1999
phase of 22% acetonitrile and 8% methanol.
(Applicable to determination in raw bovine milk of tetracycline at B. Apparatus
15–80 ng/mL; chlortetracycline, and oxytetracycline at (a) Liquid chromatograph (LC).—With quaternary LC pump
15–60 ng/mL; and as screening assay for demeclocycline, with automatic He solvent sparging [binary LC pump may be used; doxycycline, methacycline, and minocycline at ≥15 ng/mL.) see C(j) for modifications of LC mobile phases and I(a) for LC oper-
See Tables 995.04A and B for the results of the interlaboratory
ating conditions]; UV detector set at 355 nm (Note: Use a full scale study supporting the acceptance of the method.
setting equivalent to 0.02–0.05 absorbance unit. The 150 ng/mLtetracyclines chromatographic solution should give full scale deflec- Caution: Tetracyclines are irritants. Tetracycline itself is a possi- tion); manual injector or autosampler equipped with 2 mL loop; and ble teratogen. Handle tetracycline standards with care.
integrator or computer assisted data capture and analysis system.
(b) LC column. —PLRP-S, 150 × 4.6 mm id, 5 µm particle size,
Use LC grade water throughout the method. Quality of H2O is 100 Å, with guard column containing same packing material. Avail- able from Polymer Laboratories, Inc., Amherst Fields Research A. Principle
Park, 160 Old Farm Rd, Amherst, MA 01002, USA, Part No.
Tetracyclines are specifically removed from milk extracts by 1S12-1S00. Substitutions not recommended.
chelation to Cu ions reversibly bound to iminodiacetic acid ep- (c) Refrigerated centrifuge.—Operating at 10°C, with fixed an-
oxy-activated resin. Test portion is defatted, acidified, and centri- gle rotor holding 18 mm diameter tubes, and with rotor holding dis- fuged. Clear supernate is applied to chelating mini-column Table 995.04A Interlaboratory study results for determination of multiple tetracycline residues in fortified milk by metal chelate
affinity—liquid chromatographya
Known triplicates prepared by collaborators; based on data submitted by 8 laboratories, except as noted.
Table 995.04B Interlaboratory study results for determination of multiple tetracycline residues in milk containing unknown
incurred and fortified tetracycline residues using metal chelate affinity—liquid chromatographya
Blind duplicates; based on data submitted by 8 laboratories, except as noted.
Numbers in parentheses are nominal milk concentrations for fortified tetracycline residues, ng/mL.
(d) Mini-columns.—Disposable, polypropylene, with 10 mL res-
and 0.5M NaCl. Filter solution through 0.2 µm Nylon 66 filter. Store ervoir, containing frit at the bottom, graduated every 0.2 mL for the lower 2 mL. (Bio-Rad, 2000 Alfred Nobel Dr, Hercules, CA 94547, (d) Tetracyclines (TC) analytical standards.—Certified refer-
ence standards of oxytetracycline; and hydrochloride salts of (e) Rack.—To hold mini-columns (optional). Commercial or
minocycline, tetracycline, demeclocycline, chlortetracycline, methacycline, and doxycycline hyclate. Available from U.S. Phar- (f) Centrifugal ultrafilters.—Capacity ca 2.5 mL. Used to re-
macopeia, 12601 Twinbrook Pkwy, Rockville, MD 20857, USA.
move proteins of molecular weight ≥30 000 daltons without signifi- (1) TC stock standard solutions.—100 µg/mL. Place 10 mg each cantly reducing tetracycline concentration. Immediately before use, tetracycline (corrected for potency and HCl content) in individual wash ultrafilters by centrifuging 15 min at 1500 × g with 2 mL H amber or foil wrapped 100 mL volumetric flask. Dilute to volume Shake retentate and filtrate chambers to remove all H with methanol. Shake mixture to dissolve. Stock solutions are stable (g) Syringe filter.—0.2 µm Nylon 66. Alternative to (f). See H,
(2) TC combined working standard solution.—1 µg/mL. Transfer 0.5 mL each TC stock standard solution into one 50 mL amber or foil (h) Filters.—Glass LC solvent filtration apparatus with 0.2 µm
wrapped flask. Dilute to volume with methanol. Store solution (i) Volumetric flasks. —1 L, and 100 and 5 mL. Class A.
(j) Automatic pipets.—Adjustable, 10–200 µL and 0.05–1.00 mL,
(3) TC chromatographic standard solutions.—Not stable; pre- with >98% accuracy of delivery and <2% RSD.
pare on day of analysis. (i) 150 ng/mL.—Transfer 0.750 mL TC (k) Centrifuge tubes.—15 mL disposable polypropylene with
combined working standard solution to 5.0 mL volumetric flask. Di- lute to volume with McIlvaine–EDTA–NaCl buffer, (c).
C. Reagents
(ii) 100 ng/mL.—Dilute 4 mL 150 ng/mL TC chromatographic stan-dard solution with 2 mL McIlvaine–EDTA–NaCl buffer (2 + 1).
(a) Water.—LC grade. Deionize distilled H2O and then irradiate
(iii) 50 ng/mL.—Dilute 2 mL 100 ng/mL TC chromatographic stan- with UV to remove trace organic impurities.
dard solution with 2 mL McIlvaine–EDTA–NaCl buffer (1 + 1).
(b) Solvents.—Methanol and acetonitrile; LC grade.
(c) McIlvaine–EDTA–NaCl buffer.—Place 12.9 g citric acid
(iv) 20 ng/mL.—Dilute 1 mL 150 ng/mL TC chromatographic stan- dard solution with 4 mL McIlvaine–EDTA–NaCl buffer (1 + 4).
The 20, 50, 100, and 150 ng/mL TC chromatographic standard so- lutions are equivalent to 10, 25, 50, and 75 ng/mL in undiluted milk Prepare McIlvaine–EDTA–NaCl buffer as follows: Add 37.2 g (e) Metal chelate resin.—Iminodiacetic acid covalently bound to
2EDTA⋅2H2O and 29.2 g NaCl to 1 L volumetric flask and dilute to volume with McIlvaine buffer. Solution contains 0.1M EDTA epoxy activated Sepharose 6B Fast Flow in 20% ethanol suspension.
Store refrigerated. Available from Amersham Pharmacia Biotech,SE 75184, Uppsala, Sweden as No. 17-0578-01, or 800 Centennial Ave, PO Box 1387, Piscataway, NJ 08855-1387, USA, 1.0–1.2 mL. Wash resin 3× with 2 mL H2O, and then add 2 mL CuSO4 solution, C(g). Wash mini-column again 2× with 2 mL H2O.
(f) Sodium succinate buffer.—0.1M, pH 4.0. Place 11.8 g
Bed volume should be 1.0–1.2 mL, with ca 0.7 mL blue top from succinic acid in 1 L volumetric flask and dissolve in <1 L H Cu2+ adsorption. One-third bottom of mini-column should remain trate to pH 4.0 with 10M NaOH (1 + 1, w/w). Dilute to volume with white. Mini-column operates using gravity feed. Same column may H2O. Store refrigerated. Discard after 1 month or if any particulate (g) Copper sulfate.—10mM. Dissolve and dilute 0.5 g
G. Extraction
4 5H2O to 200 mL with H2O. Store at room temperature.
(h) Ethanol.—20% (v/v).
(1) Transfer 5.0 ± 0.1 mL test milk (whole raw milk, fresh or pre- (i) LC mobile phases.—(1) Solvent A.—10mM oxalic acid
viously frozen, showing no signs of souring or curdling) into 15 mL (1.26 g oxalic acid⋅2H2O/L). Stable 1 month. (2) Solvent B.—Meth- disposable centrifuge tube and centrifuge 15 min at 1500 × g at 10°C anol. (3) Water. (4) Acetonitrile.
Filter all mobile phases through 0.2 µm Nylon 66 filter.
(2) Transfer lower (skim) layer into clean 15 mL centrifuge tube If binary pumping system is used, solvent B may be varied from using 9 in. Pasteur pipet. Alternatively, while still cold, puncture a c e t o n i t r i l e – m e t h a n o l ( 2 2 + 8 , v / v ) t o 1 0 m M o x a l i c through solid fat layer on opposite sides with Pasteur pipet and de- acid–acetonitrile–methanol (70 + 22 + 8, v/v/v) to meet system suit- cant skim milk through holes. Discard fat. Add 10 mL sodium ability requirements, D.
succinate buffer, C(g), to defatted milk, cap tube, mix contents by in-
D. System Suitability
verting tube several times, and centrifuge 30 min at 1500 × g at 10°C.
(3) Apply clear supernate directly onto mini-column, F. If reser-
Check LC system suitability with every set of test samples by run- voir is not large enough, apply supernate in 2 batches. Let solution ning TC chromatographic standard solutions using gradient as in filter through. Avoid disturbing column bed excessively. After no I(a). LC system must meet the following criteria:
liquid is visible above resin, proceed with next step. Do not let (1) Absence of peaks in region of tetracyclines during runs made with no injection or with only buffer injected.
(4) Wash mini-column sequentially with 2 mL sodium succinate (2) Presence of peak (signal/noise >10) of 10 ng chlortetracycline (Note: Next 2 steps are critical for good recoveries. Use gravity (3) Baseline or near baseline resolution of all 7 tetracyclines; reso- lution between oxytetracycline and tetracycline should be >90%(height of valley to peak height) and between methacycline and (5) Carefully apply 0.70 ± 0.05 mL McIlvaine-EDTA-NaCl buffer, C(c), onto mini-column. Drip buffer around sides of column
(4) Retention times of tetracycline residues must be stable, with without disturbing column bed. Discard clear flowthrough.
<0.1 min change between TC chromatographic standard solutions (6) Elute tetracyclines from column using additional 2.5 ± 0.05 mL McIlvaine-EDTA-NaCl buffer. Collect eluted solu- (5) Linear standard curves should be obtained. Values calculated tion (should be blue) in test tube and refrigerate until analysis or col- for calibrants should agree with actual values within 10%.
lect it directly in upper (retentate) chamber of centrifugal ultrafilters, E. Preparation of Controls
B(f), and perform ultrafiltration as in H. Mini-column should be
For controls use whole raw bovine milk (fresh or previously frozen, showing no signs of souring or curdling).
( 7 ) C l e a n m i n i - c o l u m n w i t h a d d i t i o n a l 2 – 3 m L (a) Negative control.—Milk free of tetracycline.
McIlvaine-EDTA-NaCl buffer. Wash column 3× with 2 mL H2O (b) Positive control.—Fortify milk to 15, 30, and 60 ng TC/mL
and then with 5–10 mL ethanol, C(i). Cap mini-columns with excess
by adding, respectively, 75, 150, and 300 µL TC combined working of 20% ethanol and store in refrigerator. Before next use, mix con- stock solution, C(d)(2), to 5 mL aliquots of control milk.
tents of column on Vortex mixer or invert column several times to Analyze negative and positive controls initially as part of famil- resuspend metal chelate resin thoroughly. When re-using iarization with method. Analyze, single negative and positive con- mini-column, open top of column and start from step F(3). Do not
trols on day of analysis as part of method quality assurance.
re-use columns that have been exposed to sour milk or excessive F. Preparation of Mini-Column
amounts (>5 µg) of tetracyclines. Columns are reusable at least Mini-columns can be prepared simultaneously in groups; 10–14 test solutions can be extracted in 8 h. Prepare mini-column as H. Ultrafiltration
(1) Swirl bottle containing metal chelate resin, C(e), to obtain
[Note: Eluates collected in G(6) are not stable and develop precip-
itate, which can clog and effectively destroy LC column. Therefore (2) Transfer 2 aliquots of 0.7 mL metal chelate resin to deproteinize test solutions further prior to LC analysis.] mini-column, B(d), using automatic pipet, B(j), with 1 mL tip of
Cap and invert ultrafilter containing eluate, G(6), several times to
which lower 2–3 mm was removed with sharp razor to increase boresize.
ensure homogeneity of solution. Centrifuge 30–90 min at 5000 × g (3) Open bottom outlet of mini-column and let shipping buffer in fixed angle rotor. Stop centrifugation when ≥1 mL filtrate is in drain out. If necessary, add or remove resin so that bed volume is Table 995.04C LC operating conditions for determination of tetracycline residues in milk
0 min: Equilibration at 100% solvent A at 1 mL/min. Solvent A equilibration is slow. On the day of analysis equilibrate LC column that has been stored in 50% acetonitrile solution ≥1 h before first run. Perform first run using buffer blank to remove artifact peaks coming fromprolonged running of aqueous buffers through LC column. Equilibrate LC column between runs a few minutes.
0–1 min: 100% solvent A at 1 mL/min.
1–6 min: Linear ramp to solvent A–acetonitrile–methanol (70 + 22 +8).
6–17 min: Solvent A–acetonitrile–methanol (70 + 22 +8) at 1 mL/min. Time may be extended if necessary to elute all tetracyclines.
17–19 min: Linear ramp to 100% solvent A with flow ramped linearly from 1 to 1.5 mL/min.
19–20 min: 100% solvent A with linear ramp in flow from 1.5 to 1 mL/min.
20–24 min: 100% solvent A at 1 mL/min.
If solvent B contains acetonitrile–methanol (22 + 8, v/v) start at 100% solvent A and ramp over 5 min to 70% solvent A and 30% solvent B, If solvent B contains 10mM oxalic acid–acetonitrile–methanol (70 + 22 + 8, v/v) start at 100% solvent A and ramp over 5 min to 100% [Note: This step may be omitted if extract from G(6) collected in
test tube is refrigerated until LC analysis and then filtered through 0.2 µm syringe filter, B(g), immediately before injection onto LC
column.]
I. LC Determination
Inject equal volumes of the 4 TC chromatographic standard solu- tions and filtered test solutions onto LC. Monitor UV absorbance at355 nm. Follow procedure outlined below: (a) Mobile phase gradient.—See Table 995.04Cfor specific op -
erating conditions. Inject filtered test solution while mobile phase is100% solvent A at 1 mL/min flow rate. After 1 min, linearly changemobile phase over 5 min to solvent A–methanol–acetonitrile(70 + 8 + 22). Maintain this composition 11 min at 1 mL/min flowrate, before returning linearly over 2 min at 1.5 mL/min to 100% sol-vent A.
Re-equilibrate LC column ≥4 min (to stabilize retention times) at initial conditions before injecting next test solution.
Systems without automatic He sparging may experience de-gassing problems, which typically appear as large, broad peaks
around elution time of oxytetracycline. If de-gassing problems oc-
cur during gradient chromatography, add small amounts of
acetonitrile (2–5%) to solvent A and increase equilibration time be-
tween runs. It may be necessary to prepare combined TC working
standard solution, C(d)(2), with H2O instead of methanol.
Tetracyclines in aqueous solutions are not stable; therefore, aque-
ous TC combined standard solution must be prepared on day of
analysis.
Store LC column in H2O–acetonitrile solution (1 + 1). Before and after storage, flush LC column and chromatographic systemwith LC grade H2O (ca 10–15 min at 1 mL/min) to prevent precip-itation of oxalic acid caused by high concentrations of organicsolvent.
Figure 995.04—Test sample chromatograms. (A) Ex-
(b) Test solution injection.—Use 2.0 mL injection loop. Injection
tract of control milk fortified to 30 ng each TC/mL:
size may vary. Depending on detector sensitivity, inject 0.5–1.0 mL (1) minocycline, (2) oxytetracycline, (3) tetracycline,
filtered test solution. For accurate quantitation, inject identical vol- (4) demeclocycline, (5) chlortetracycline,
(6) methacycline, and (7) doxycycline. (B) Extract of
2O to flush autosampler to prevent salt precip- itation. Inject a TC chromatographic standard solution every milk from cow treated with tetracycline:
5–10 test injections as a check of the retention time.
(8) epitetracycline. (C) Extract of control milk.
Make 1–2 blank gradient runs each day prior to injecting TC chro- 24 h. Degradation products tend to elute earlier than parent com- matographic standard solution or test solutions to ensure absence of pound and usually migrate with oxytetracycline. To avoid this prob- potential background interferences near the retention times of lem, perform all centrifugation steps at 10°C. Refrigerate extracts or analyze them within 4 h of preparation. Mini-column eluate solu- (c) Peak identification.—Tetracyclines elute in the following or-
tions may be refrigerated ≤2 days or frozen ≤1 week with only slight der: minocycline, oxytetracycline, tetracycline, demeclocycline, changes in tetracycline concentrations.
chlortetracycline, methacycline, and doxycycline. (Note: Retention J. Calculations
times tend to shift slightly with increases of age of column and num-ber of injections.) Prepare standard curve for each tetracycline from standard Define very tight windows for peak identification, since tetracyclines elute closely together. All postulated residue peaks Calculate concentrations of tetracyclines using linear regression should have retention times within 0.05 min of retention times ob- Some metabolites may interfere with analysis of other parent compounds. Oral administration of tetracycline to cows may result where C = tetracycline concentration of injected extract, ng/mL; and in appearance of both tetracycline and earlier peaks (e.g., P = tetracycline peak area or peak height.
epitetracycline), which elute very near oxytetracycline.
Each TC standard curve should be linear.
Occasionally, endogenous peak may appear between oxytetracy- Calculate tetracycline concentration in original milk by dividing cline and tetracycline. This may be caused by high concentrations of concentration determined for injected test solution by 2 (dilution riboflavin in milk, the retention time of which is ca 0.1 min earlier factor used in mini-column procedure, which reduces volume of test than that of tetracycline. To reduce interference of riboflavin, double volume of sodium succinate buffer wash of mini-column in extrac- Care should be taken with integration. Baseline determined by au- tion procedure (this only slightly decreases recovery of oxytetracy- tomated data systems should be checked for each chromatogram (see Figure 995.04 for appropriate baseline construction).
(d) Extract stability.—Tetracyclines are not stable at room tem-
References: J. AOAC Int. 76, 329(1993);
perature under acidic conditions (i.e., in McIlvaine-EDTA-NaClbuffer). Tetracycline and chlortetracycline degrade ≥50% within

Source: http://down.40777.cn/stardard/8/33.2.62%20AOAC%20Official%20Method%20995.04%20Multiple%20Tetracycline%20Residues.pdf