See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/8222741

Development and Validation of a Gas Chromatography/Mass Spectrometry

Procedure for Confirmation of Para-Toluenesulfonamide in Edible Fish Fillet
Tissue

Article  in  Journal of AOAC International · September 2004

DOI: 10.1093/jaoac/87.5.1098 · Source: PubMed

CITATIONS READS

0 450

4 authors, including:

Remi Idowu Philip James Kijak

U.S. Food and Drug Administration U.S. Food and Drug Administration
35 PUBLICATIONS   561 CITATIONS    26 PUBLICATIONS   471 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Remi Idowu on 26 September 2018.

The user has requested enhancement of the downloaded file.

1098 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004
DRUGS, COSMETICS, FORENSIC SCIENCES
Development and Validation of a Gas Chromatography/Mass
Spectrometry Procedure for Confirmation of
Para-Toluenesulfonamide in Edible Fish Fillet Tissue
OLUTOSIN R. IDOWU1 and PHILIP J. KIJAK
U.S. Food and Drug Administration, Center for Veterinary Medicine, 8401 Muirkirk Rd, Laurel, MD 20708
JEFFERY R. MEINERTZ and LARRY J. SCHMIDT
U.S. Geological Survey, Biological Resources Division, Upper Midwest Environmental Sciences Center, 2630 Fanta Reed
Rd, La Crosse, WI 54603
Chloramine-T is a disinfectant being developed as
a treatment for bacterial gill disease in cultured
fish. As part of the drug approval process, a
method is required for the confirmation of
chloramine-T residues in edible fish tissue. The
marker residue that will be used to determine the
depletion of chloramine-T residues from the edible
tissue of treated fish is para-toluenesulfonamide
(p-TSA), a metabolite of chloramine-T. The
development and validation of a procedure for the
confirmation of p-TSA is described. Homogenized
fish tissue is dried by mixing with anhydrous
sodium sulfate, and the mixture is extracted with
methylene chloride. The extract is passed through
a silica gel solid-phase extraction column, from
which p-TSA is subsequently eluted with
acetonitrile. The acetonitrile extract is evaporated,
and the oily residue is dissolved in hexane. The
hexane solution is shaken with fresh acetonitrile.
The acetonitrile solution is evaporated and the
residue is redissolved in dilute potassium
hydroxide solution. The aqueous solution is
extracted with methylene chloride to further
remove more of the fat co-extractive. The aqueous
solution is reacted with pentafluorobenzyl bromide
in presence of tetrabutylammonium
hydrogensulfate. The resulting
di-(pentafluorobenzyl) derivative of p-TSA is
analyzed by gas chromatography/mass
spectrometry. This method permits the
confirmation of p-TSA in edible fish tissue at
20 ppb.
B
acterial gill disease (BGD) is a serious disease of
intensively cultured fish, particularly salmonids. In the
United States, BGD has been responsible for
substantial production losses in federal, state, and commercial
Received October 23, 2003. Accepted by JM February 26, 2004.
Corresponding author's e-mail: oidowu@cvm.fda.gov.
hatcheries. Despite the immense salmonid losses from this
disease, no therapeutant has presently been approved in the
United States for the control of BGD.
Chloramine-T (n-sodium-n-chloro-p-toluenesulfonamide)
is a disinfectant that is effective in controlling BGD in
cultured fish (1, 2). State and federal hatcheries are currently
using chloramine-T to control BGD under a compassionate
investigational new animal drug (INAD) application granted
by the U.S. Food and Drug Administration (FDA). Use of
chloramine-T as a therapeutic drug in cultured fish beyond the
INAD requires approval of a New Animal Drug Application
(NADA) by the FDA. The U.S. Geological Survey Upper
Midwest Environmental Sciences Center (UMESC) has been
conducting studies in support of an NADA for use of
chloramine-T in public aquaculture. One requirement for
approval is the availability of a method for the unambiguous
detection and confirmation of chloramine-T residues in fish
fillet tissue. The marker residue to be used to determine the
depletion of chloramine-T residues from the edible tissue of
treated fish for chloramine-T is a metabolite,
para-toluenesulfonamide (p-TSA). The chemical structures of
chloramine-T and p-TSA are shown in Figure 1.
The objective of this study was to follow FDA’s guidelines
to develop and validate a procedure to confirm p-TSA in fillet
tissue from rainbow trout (Oncorhynchus mykiss), channel
catfish (Ictalurus punctatus), yellow perch (Perca flavescens),
and hybrid striped bass (Morone saxatilis x M. chrysops). The
method should be capable of confirming p-TSA at a level of
10 ppm.
Experimental
Apparatus
(a) Extraction column.—19 ´ 300 mm, 250 mL capacity
chromatography column (Labglass, Buena, NJ).
(b) Glass wool.—Deactivated (VWR Scientific Products,
Bridgeport, NJ). The glass wool was washed with methanol and
methylene chloride and dried at room temperature before use.
(c) Solid-phase extraction (SPE) manifold.—Baker
SPE-24G column processor and vacuum chamber (J.T. Baker,
Phillipsburg, NJ).
 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004 1099
Figure 1. Chemical structures of chloramine-T and
p-toluenesulfonamide (p-TSA).
(d) Vacuum pump.—Gast Model DOA-102-AA (VWR
Scientific Products).
(e) SPE columns.—Baker Bond SPER silica gel columns,
6 mL, 1000 mg (VWR Scientific Products).
(f) SPE accessories.—75 mL reservoir, stopcocks,
adaptors, and stainless steel needles (VWR Scientific
Products).
(g) Column frit for 75 mL reservoir.—Polyethylene,
20 mm diameter, 2 mm porosity (Alltech Associates, Inc.,
Deerfield, IL).
(h) Autosampler vials.—Microvial (200 mL), polypropylene
with snap caps (VWR Scientific Products).
(i) Vortex mixer.—S|P Vortex mixer (American Scientific
Products, American Hospital Supply Corp., McGaw Park, IL).
(j) Shaker.—Variable-speed reciprocal shaker 5900
(Eberbach Labtools, Ann Arbor, MI).
(k) Centrifuge.—IEC Centra-8R bench-top centrifuge,
swinging bucket rotor with 15 mL tube holders.
(l) 112 Nitrogen evaporator.—N-Evap™ (Organomation
Associates Inc., Berlin, MA).
(m) Gas chromatography/mass spectrometer (GC/MS)
system.—Hewlett Packard (Palo Alto, CA) 5890 gas
chromatograph with Hewlett Packard 5989A mass selective
detector, Hewlett Packard 7673A automatic sampler, and
Hewlett Packard 7673A autosampler controller.
(n) Column.—30 m ´ 0.25 mm capillary column coated with
DB-5 (film thickness 0.25 mm; (J&W Scientific, Folsom CA).
(o) Injector.—Splitless, quartz 2 mm id, 250 mL,
deactivated liner, Hewlett-Packard No. 5181-8818.
GC/MS operating conditions: The carrier gas flow
(helium) was maintained at 30 cm/s. The injector temperature
was set at 250°C. The oven temperature program consisted of
an initial temperature of 150°C and initial time of 1 min
followed by ramping to 280°C at 5°C/min, and then keeping
at a final temperature of 280°C for 3 min.
Mass spectrometer operating conditions: Spectrometer
was operated in the electron impact (EI) mode with selected
ion monitoring of 4 ions from the pentafluorobenzyl
derivative of p-TSA (m/z 531, 376, 155, and 91). Interface
temperature was set at 280°C, dwell time set at 100 ms, and
electron multiplier voltage set to 2200 V.
System suitability: On each analysis day “Instrument
Check/Standardization” of the mass spectrometer was performed
by automatic tuning in the standard autotune mode using
perfluorotributylamine as the calibration compound. Ions m/z 69,
219, and 502 were monitored. The system was considered
suitable for use if the relative abundance of ion 502 was 5% or
greater and ion 69 was the base peak.
Chemicals
(a) Acetonitrile.—Liquid chromatography (LC) grade,
99.9% assay purity (Fisher Scientific, Fair Lawn, NJ).
(b) Methylene chloride.—LC, GC/MS grade, >99.9%
assay purity (Fisher Scientific).
(c) Hexane.—LC, GC, pesticide residue analysis and
spectrophotometry grade, >99.9% assay purity (Burdick &
Jackson, Muskegon, MI).
(d) Methanol.—LC, GC, pesticide residue analysis and
spectrophotometry grade, >99.9% assay purity (Burdick &
Jackson).
(e) Potassium hydroxide (KOH).—Fisher Scientific.
(f) Sodium sulfate.—Anhydrous powder (Fisher
Scientific).
(g) 2,3,4,5,6-Pentafluorobenzyl bromide (PFB-Br).—>99.9%
purity (Aldrich Chemical Co., Milwaukee, WI).
(h) Tetrabutylammonium hydrogensulfate (TBA-HSO4).—>99%
assay purity (Aldrich Chemical Co.).
(i) p-TSA.—>99% assay purity (Aldrich Chemical Co.).
(j) Perfluorotributylamine.—Agilent Technologies, Inc.
(Wilmington, DE.)
(k) Water.—Deionized to an electrical resistance of at
least 17.8 MW-cm, generated in-house from a Milli-Q UV
Plus system (Millipore Intertech, Marlboro, MA).
Reagent Solutions
(a) Potassium hydroxide solutions.—Prepare ca 1M
solution KOH by dissolving 6.5 g KOH in 100 mL water.
To prepare 0.2M solution, pipet 20 mL 1M solution into
100 mL volumetric flask and adjust volume to 100 mL with
water.
(b) Tetrabutylammonium hydrogensulfate solution.—Dissolve
1.365 g TBA-HSO4 in 40 mL water. Add 10 mL 1M KOH
solution to result in 0.1M TBA-HSO4 in 0.2M KOH solution.
(c) Primary p-TSA stock, 3000 mg/mL.—Weigh 0.0302 g
p-TSA and transfer to 10 mL volumetric flask. Dissolve in
acetonitrile and dilute to 10 mL with acetonitrile. Store all
p-TSA solutions at room temperature. Solution is stable for
9 months.
(d) Fortification solution, 300 mg/mL.—Pipet 1 mL
3000 mg/mL primary stock into 10 mL volumetric flask and
adjust volume to 10 mL with acetonitrile. Solution is stable for
9 months.
(e) Fortification solution, 60 mg/mL.—Pipet 2 mL
300 mg/mL solution into 10 mL volumetric flask, and dilute to
10 mL with acetonitrile. This solution and all subsequent
solutions are stable for 6 months.
1100 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004
(f) Fortification solution, 24 mg/mL.—Pipet 800 mL
300 mg/mL solution into 10 mL volumetric flask, and dilute to
10 mL with acetonitrile.
(g) Fortification solution, 6 mg/mL.—Pipet 1 mL
60 mg/mL solution into 10 mL volumetric flask, and dilute to
10 mL with acetonitrile.
(h) Fortification solution, 2.4 mg/mL.—Pipet 1 mL
24 mg/mL solution into 10 mL volumetric flask, and dilute to
10 mL with acetonitrile.
(i) Fortification solution, 1.2 mg/mL.—Pipet 500 mL
24 mg/mL solution into 10 mL volumetric flask, and dilute to
10 mL with acetonitrile.
Sample Preparation
Fish fillets were prepared as previously described (3).
Briefly, this consists of homogenizing fish fillet to a fine
powder with dry ice in a commercial stainless steel blender.
The fillet-dry ice matrix was initially stored at –20°C in an
open plastic freezer bag to allow the dry ice to sublimate. For
long-term storage, the bag was sealed and kept at –70°C.
Extraction of p-TSA from Fortified or Incurred Fillet
Tissue
A 3 g (±0.05 g) portion of thawed homogenized control
(or incurred) fish tissue was weighed in a tared 250 mL
beaker. To fortify control fish tissue, 50 mL spiking solution
of p-TSA was added to the control tissue. Anhydrous
sodium sulfate, 20 g (±0.1 g), was poured onto the tissue,
ensuring that the tissue was completely covered by sodium
sulfate. The sample was allowed to stand at room
temperature for about 10 min. The tissue/salt matrix was
then thoroughly mixed with a spatula, breaking apart any
tissue/salt lumps by mashing them against the side of the
beaker. The tissue/salt mixture was allowed to stand at
room temperature for 2 h, with vigorous stirring at intervals
of 20–25 min. After the 2 h drying period, the tissue/salt
matrix was free flowing. p-TSA in the tissue/salt matrix
was stable at room temperature for 5 days.
An extraction column was prepared by plugging the
effluent end with glass wool and pouring 10 (±0.1 g) sodium
sulfate into the column. The column was clasped to a ring
stand support, and the tissue/salt mixture was poured into the
extraction column. The beaker was dry-rinsed with 10 g
anhydrous sodium sulfate, and the sodium sulfate was
poured on top of the tissue/salt matrix in the extraction
column. The beaker was then rinsed with 30 mL methylene
chloride, and the rinse was transferred into the extraction
column.
A closed SPE stopcock was attached to a 75 mL SPE
reservoir and, after a polyethylene column frit was placed in
the bottom of the reservoir, 10 (±0.1) g anhydrous sodium
sulfate was poured into the reservoir. The reservoir was then
clamped under the extraction column to receive the effluent
from the column. The stopcock of the extraction column was
opened to allow the methylene chloride flow through the
tissue/salt mixture at a rate of ca 1 drop/4 s, while the column
effluent was collected in the SPE reservoir. After the flow of
methylene chloride through the tissue/salt mixture stopped,
residual methylene chloride was gently forced through the
column with compressed air.
SPE
To prepare the SPE system, a stainless steel needle was
attached to an outlet port of the SPE manifold; a stopcock was
attached to the manifold head; and a 6 mL, 1000 mg silica gel
column was attached to the stopcock. The silica gel column
was conditioned by aspirating through the column
ca 2 column volumes acetonitrile followed by 4 column
volumes methylene chloride while the solvent level was
maintained above the column bed.
An SPE adaptor was attached to the silica gel column.
The 75 mL SPE reservoir containing the methylene
chloride extract was removed from under the extraction
column, and the reservoir was attached to the adaptor on top
of the silica gel SPE column with the reservoir’s stopcock.
The extract was aspirated through the SPE silica gel column
at a flow rate <5 mL/min, while the solvent level was
maintained above the column bed. The lumen of the empty
reservoir was rinsed twice with ca 5 mL methylene
chloride. After rinsing, the methylene chloride was
aspirated through the SPE column. The SPE column was
dried by aspirating air through the column for 30 min with a
vacuum of ca 500 mm Hg. The waste container was
removed from the vacuum chamber, and the methylene
chloride was discarded. A 10 mL test tube was placed in the
waste container, and the SPE manifold needle was lowered
into the test tube. p-TSA was eluted from the SPE silica gel
column into the test tube with 3 mL acetonitrile.
Back-Extraction of p-TSA
The acetonitrile eluate from the silica gel column was
evaporated, and the oily residue was dissolved in 3 mL
hexane. The hexane solution was extracted with 3 mL
acetonitrile by mixing on a Vortex mixer for about 1 min. The
upper hexane layer was removed with a Pasteur pipet and
discarded. Another 3 mL hexane was added to the acetonitrile
solution and mixed on a Vortex mixer for about 1 min. The
mixture was centrifuged at 4000 rpm (2840 rcf) for 5 min at
5°C. The hexane layer (upper layer) was removed with a
Pasteur pipet and discarded. The acetonitrile solution was
decanted into a fresh test tube and evaporated under a stream
of air (or nitrogen) to obtain an oily residue. p-TSA in the oily
residue was stable at –20°C for 5 days. 0.2M KOH (2 mL) was
added to the oily residue and briefly mixed on a Vortex mixer.
Methylene chloride (2 mL) was then mixed into the solution
by mixing on a Vortex mixer for about 1 min. The mixture was
centrifuged at 4000 rpm (2849 rcf) for 5 min at 5°C. Using a
Pasteur pipet, the lower, methylene chloride layer was
removed, along with the thin ring of emulsion that formed
between the 2 layers. The alkaline aqueous extract was
transferred to a fresh test tube using a Pasteur pipet to ensure
that the aqueous extract was not contaminated by emulsion
particles.
 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004 1101
Figure 2. Reaction of p-TSA with pentafluorobenzyl
bromide.
Note: It is critical to avoid contamination of the aqueous
extract with emulsion particle. It was found easier to achieve
optimum recovery of the aqueous extract, without
contamination, by first removing the lower, unwanted
methylene chloride layer together with the ring of emulsion
rather than try to remove the upper aqueous layer directly.
Derivatization of Extracted p-TSA with PFB-Br
TBA-HSO4 solution (1 mL) was added to the aqueous
alkaline extract followed by 20 mL PFB-Br and 2 mL
methylene chloride. The test tube was placed in a test tube
rack, and the mixture was shaken on a reciprocating shaker
Figure 3. Full-scan mass spectrum of PFB-Br derivative of p-TSA.
for 45 min. The mixture was centrifuged at 4000 rpm
(2840 rcf) for 5 min at 5°C. The aqueous layer was removed
with a Pasteur pipet, and the extract was dried with a small
portion of anhydrous sodium sulfate. The methylene
chloride extract was decanted into a fresh test tube and
evaporated under a stream of air (or nitrogen). The residue
was dissolved in 200 mL acetonitrile with brief shaking on a
Vortex mixer. The solution was transferred to an
autosampler vial with a Pasteur pipet, and a 1 mL portion
was analyzed by GC/MS. A solvent rinse was injected before
each sample injection to prevent carryover from the previous
injection. The derivatized extract in acetonitrile was stable for
5 days at room temperature and for 15 days at –20°C.
Derivatization of p-TSA Comparison Standards
Comparison standards required to establish confirmation
criteria were prepared as follows: A 50 mL portion of each of
the 5 p-TSA working solutions was placed in 5 separate test
tubes. These correspond, respectively, to fillet tissue fortified
with p-TSA at 20, 40, 100, 400, and 1000 ng/g. Acetonitrile
was evaporated under a stream of air. The residue was
redissolved in 2 mL 0.2M KOH. TBA-HSO4 solution (1 mL)
was added to the aqueous solution followed by 20 mL PFB-Br
and 2 mL methylene chloride. Subsequent derivatization
procedures and GC/MS were as described in the previous
section for extracted p-TSA.
Calculations
(a) Peak areas of the reconstructed ion chromatograms for
ions 531, 376, 155, and 91 were integrated.
(b) The relative abundances from the ion chromatogram
peaks were calculated. Relative abundances were normalized
to the most abundant peak.
(c) The average retention times based on the reconstructed
ion chromatograms from all the analytical standards analyzed
with each batch of samples were calculated.
1102 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004
Figure 4. Mass fragmentation ions of
N,N-di(pentafluorobenzyl)-p-toluenesulfonamide.
(d) The signal-to-noise (S/N) ratio (peak-to-peak) for the
ion chromatogram of ion 531 was calculated.
(e) The reconstructed ion chromatograms for ions m/z
531, 376, 155, and 91 were obtained.
(f) Reconstructed ion chromatograms of samples overlaid
on corresponding chromatograms from control samples
analyzed with each batch of samples were obtained to confirm
lack of deviation of sample retention time from retention time
of standards.
Confirmation Criteria
Confirmation was based on matching unknowns to the
comparison standard according to the following criteria:
(a) The retention time of a suspect peak must fall within
2% of the average retention times of the peak obtained from
analysis of p-TSA standards included with each batch.
(b) The relative abundances of the ions of the p-TSA
derivative were determined from the average of all
comparison standards analyzed with each batch. Relative
abundances for fortified and suspect samples must match
those of the m/z 531, 376, 155, and 91 ions within ±10% of an
arithmetic mean. For example, with an ion’s relative
abundance of 50% in fortified standards, the acceptable range
for that ion in samples is 40–60%, not 45–55%.
(c) The 4 ions, m/z 531, 376, 155, and 91, must be
detected.
(d) The S/N ratio (peak-to-peak) for the ion
chromatogram of ion 531 must be 5 or above.
Results and Discussion
Preliminary studies showed that LC with electrospray
ionization MS (LC-ESI) was not a suitable technique for a
confirmatory method for p-TSA. LC-ESI resulted in poor
fragmentation behavior of p-TSA, with the molecular ion
remaining largely intact under all the ionization conditions
investigated. GC/MS with EI ionization was adopted
because fragmentation ions could be obtained in the EI
mass spectrum of p-TSA. Despite the presence of
fragmentation ions, the unchanged p-TSA exhibited poor
GC behavior with a tailing peak of variable retention time.
The poor GC behavior of sulfonamides is well known, and
GC analysis of antibacterial sulfonamides usually involves
the preparation of derivatives in which the acidic
sulfonamido nitrogen is converted to a neutral derivative,
most often through alkylation. A relevant example of this
approach is described in a previous report on the
confirmation and quantitation of sulfamethazine in swine
muscle and liver by GC/MS involving methylation of
sulfamethazine with diazomethane (4). However, use of
diazomethane for alkylation has since gone out of favor
because of safety concerns. Alkylation of
benzenesulfonamide with PFB-Br for the purpose of GC
has been reported (5). With this information, alkylation of
p-TSA for the purpose of GC/MS, using PFB-Br, was
adopted in the present study.
p-TSA was extracted by modification of the extraction
procedure described earlier for the determination of p-TSA
in fish fillet tissue (3). The original extraction procedure did
not involve cleanup with hexane and back extraction into an
alkaline solution. These steps, as described above, were
necessary to avoid interfering peaks in subsequent
chromatography of the PFB-Br derivative of the extracted
p-TSA.
The reaction of p-TSA with PFB-Br to give a
di-(pentafluorobenzyl) derivative is depicted in Figure 2.
The full-scan mass spectrum of PFB-Br derivative of p-TSA
is shown in Figure 3 and the mass fragmentation of the
di-(pentafluorobenzyl) derivative of p-TSA is depicted in
Figure 4. Total ion chromatograms from GC/MS analysis of
the PFB-Br derivative of the p-TSA standard and p-TSA
extracted from fortified and incurred fish tissue are shown in
Figures 5–7, respectively.
The reconstructed ion chromatograms (RICs) of the
PFB-Br derivative of the p-TSA standard are shown in
Figure 8. The RICs of the PFB-Br derivative of p-TSA
extracted from fortified and incurred fish tissue are shown in
Figures 9 and 10, respectively.
The method was validated for p-TSA in skinless fillet
tissue from channel catfish and skin-on fillet tissue from
hybrid striped bass, rainbow trout, and yellow perch. For
each species, 5 replicates of control tissue and control
tissue fortified with p-TSA at 20, 40, 100, 400, and
 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004 1103

Figure 5. Total ion chromatogram from GC/MS analysis of PFB derivative of standard p-TSA (60 ng).

1000 ng/g were analyzed. All control tissue failed to
confirm and all fortified tissue met confirmation criteria.
Retention time for the PBR-Br derivative of p-TSA was
approximately 21.1 min. The retention time was highly
reproducible with interday coefficient of variation of
0.1% or less. Typical within-day reproducibility of the
retention time is illustrated with the results shown in
Table 1 from the analysis of p-TSA in yellow perch fillet
tissue. The interday reproducibility of the retention time is
also illustrated in Tables 2–5 for the 4 fish species.
Relative abundances for the ions monitored at m/z 531,
376, 155, and 91 were approximately 1.5, 100, 25, and
60%, respectively. The relative ion abundances were also
reproducible. On each analysis day, both the retention
time and the relative ion abundances for the fortified and
incurred samples closely matched the values obtained for
the comparison standards as required by the confirmation
criteria. The relative abundance of the ion 531, which is the
weakest ion, showed the greatest variability. Below 20 ng/g
p-TSA in fish tissue (or 60 ng p-TSA standard), the m/z 531
ion was usually too weak to be detected at the confirmation
criterion of an S/N ratio of 5 or greater. Similarly,
background signals obtained from control samples were
usually lacking the m/z 531 ion or the signal was much below
the ion detection criterion of an S/N ratio <5.
Incurred fillet samples obtained from fish exposed to different
levels of chloramine-T were analyzed. The retention times and
ion abundances from these analyses confirmed the presence of
p-TSA in the incurred fish fillet tissue. The chromatograms in
Figure 10, obtained from one of the incursion studies, show the

Figure 6. Total ion chromatogram from GC/MS analysis of PFB derivative of p-TSA extracted from fortified fish tissue
(20 ppb).
1104 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004

Figure 7. Total ion chromatogram from GC/MS analysis of PFB derivative of p-TSA extracted from incurred fish
tissue (50 ppb p-TSA).

Figure 8. Reconstructed ion chromatograms of PFB derivative of standard p-TSA.

 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004 1105

Figure 9. Reconstructed ion chromatograms of PFB derivative of p-TSA extracted from fortified yellow perch fish
tissue.

presence of background peaks for the m/z 376, 155, and 91 ions
in the control tissue. These background peaks were not observed
earlier during method development with a different batch of
control tissue. This suggests that the fish used as control during
the incursion studies may have had prior exposure to
chloramine-T. However these control samples fail to confirm
because they lack the m/z 531 ion. It was based on this
experience, and to ensure against false positives, that the
confirmation criterion was set requiring that the m/z 531 ion must
be present at the relatively high S/N ratio of 5 or greater.
The following list of compounds can be used in
aquaculture without interference: quinolones (nalidixic acid,
flumequine, oxolinic acid), sulfonamides (sulfachloro-
pyridazine, sulfadiazine, sulfamethazine, sulfadimethoxine,
sulfamerazine, sulfaquinoxaline), fluoroquinolones
(difloxacin, enrofloxacin, sarafloxacin), fenicols (chloram-
phenicol, florfenicol, florfenicol amine, thiamphenicol), dyes
(gentian violet, leuco crystal violet, leuco malachite green,
malachite green), tetracycline, and oxytetracycline.

Safety

Tetrabutylammonium hydrogensulfate and p-TSA may be

harmful by inhalation, ingestion, or skin contact. The method
also uses bases and a variety of organic solvents. All of the
extraction procedure was performed in a fume hood to prevent
exposure of laboratory personnel to toxic solvent fumes.
Exposure was further limited by the use of laboratory coats,
gloves, and other personal protection devices.
Figure 10. Reconstructed ion chromatograms (solid
lines) of PFB derivative of p-TSA extracted from incurred
fish tissue and reconstructed ion chromatograms
(broken lines) of PFB derivative of an extract of control
yellow perch tissue.
1106 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004

Table 1. Within-day reproducibility of retention time and ion abundances in the GC/MS confirmation of p-TSA in
yellow perch fish fillet tissue

Relative ion abundance, %

Retention time, Signal-to-noise ratio

p-TSA level, ppb min 531 376 155 91 (S/N), for m/z 531 Confirm

Fortified samples

20 21.06 1.5 100 21.7 59.2 9.3 Yes

40 21.07 1.3 100 23.1 61.6 25.0 Yes
100 21.07 1.4 100 22.4 59.5 54.0 Yes
400 21.08 1.4 100 23.0 61.4 176.2 Yes
1000 21.10 1.5 100 22.0 59.5 796.5 Yes
a
Control 21.06 0.9 100 19.9 58.0 0.9 No
Standards, ng

60 21.07 1.4 100 22.2 59.5 14.9

120 21.07 2.1 100 20.1 55.3 56.8
300 21.08 1.4 100 22.5 59.6 74.5
Standard’s mean 21.07 1.6 100 21.6 58.1
Sample’s mean 21.08 1.4 100 22.4 60.2
Sample’s CV, % 0.07 6 0 3 2
Maximum allowable 21.49 100 31.6 68.1
Minimum allowable 20.65 100 11.6 48.1 5

a
Ion m/z 531 below detection criterion of S/N ³ 5.

Table 2. Interday reproducibility of retention time and ion abundances in the GC/MS confirmation of p-TSA in
channel catfish fish fillet tissue
Relative ion abundance, % (mean CV, %)

Signal-to-noise ratio
Retention time, min (S/N) for m/z 531
p-TSA level, ppb (mean CV, %) 531 376 155 91 (mean CV, %)

Fortified tissue

20 21.08 (0.2) 1.4 (10) 100 29.9 (26) 65.9 (22) 10.4 (25)
40 21.08 (0.2) 1.3 (12) 100 30.0 (29) 67.8 (25) 19.9 (26)
100 21.08 (0.2) 1.4 (13) 100 30.5 (28) 68.0 (25) 43.6 (24)
400 21.08 (0.1) 1.6 (21) 100 29.5 (32) 66.2 (27) 202.4 (29)
1000 21.10 (0.1) 1.7 (20) 100 28.1 (31) 64.1 (26) 504.3 (32)
Incurred tissue (50 ppb) 21.08 (0.08) 1.6 (3) 100 22.6 (3) 55.9 (2) 30.1 (5)
Comparison standard, ng

60 21.11 (0.1) 1.6 (35) 100 28.7 (30) 64.9 (27) 14.3 (42)
120 21.10 (0.2) 2.2 (12) 100 27.7 (21) 62.3 (19) 34.2 (20)
300 21.07 (0.1) 1.6 (25) 100 29.2 (30) 65.3 (26) 66.5 (55)
a
Control tissue 21.07 (0.2) 0.42 (139) 100 32.1 (28) 66.8 (23) 1.3 (19)

a
Ion m/z 531 below detection criterion of S/N ³ 5.
 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004 1107

Table 3. Interday reproducibility of retention time and ion abundances in the GC/MS confirmation of p-TSA in hybrid
striped bass fish fillet tissue

Relative ion abundance, % (mean CV, %)

Signal-to-noise ratio
Retention time, min (S/N), for m/z 531
p-TSA level, ppb (mean CV, %) 531 376 155 91 (mean CV, %)

Fortified tissue

20 21.2 (0.6) 1.9 (23) 100 18.2 (12) 51.8 (13) 19.6 (20)
40 21.2 (0.6) 1.9 (12) 100 18.4 (10) 51.3 (11) 39.0 (41)
100 21.20 (0.6) 2.0 (20) 100 18.4 (12) 51.2 (13) 99.8 (37)
400 21.20 (0.6) 2.2 (16) 100 18.0 (10) 50.2 (11) 329.3 (48)
1000 21.21 (0.5) 2.1 (10) 100 18.2 (5) 51.0 (6) 932.7 (15)
Incurred tissue

50 21.10 (0.05) 1.4 (7) 100 21.6 (4) 57.9 (5) 54.8 (39)
Comparison standard, ng

60 21.20 (0.6) 2.0 (17) 100 18.4 (11) 51.5 (11) 24.6 (35)
120 21.20 (0.6) 2.7 (12) 100 17.4 (9.3) 49.0 (10) 65.0 (25)
300 21.20 (0.6) 2.1 (13) 100 17.8 (9) 49.7 (10) 163.1 (36)
Control tissuea 21.20 (0.6) 1.5 (62) 100 19.0 (16) 52.2 (15) 3.0 (43)

a
Ion m/z 531 below detection criterion of S/N ³5.

Table 4. Interday reproducibility of retention time and ion abundances in the GC/MS confirmation of p-TSA in
rainbow trout fish fillet tissue
Relative ion abundance, % (mean CV, %)

Signal-to-noise ratio
Retention time, min (S/N), for m/z 531
p-TSA level, ppb (mean CV, %) 531 376 155 91 (mean CV, %)

Fortified tissue

20 21.07 (0.07) 1.5 (8) 100 24.6 (6) 55.9 (5) 13.9 (41)
40 21.07 (0.09) 1.5 (9) 100 25.8 (6) 59.1 (7) 22.5 (37)
100 21.07 (0.09) 1.4 (6) 100 26.8 (8) 60.7 (8) 58.9 (37)
400 21.07 (0.09) 1.5 (5) 100 26.8 (8) 60.82 (9) 169.2 (18)
1000 21.08 (0.1) 1.6 (3) 100 25.9 (6) 59.6 (5) 593.9 (21)
Incurred tissuea

(Level unknown) 21.08 (0.06) 1.7 (31) 100 31.7 (6) 68.2 (5) 55.8 (25)
Comparison standard, ng

60 21.07 (0.1) 1.4 (14) 100 25.5 (7) 58.5 (5) 8.7 (21)
120 21.09 (0.9) 2.3 (16) 100 22.8 (7) 52.5 (5) 26.3 (25)
300 21.07 (0.1) 1.5 (6) 100 26.9 (6) 60.2 (5) 59.6 (45)
b
Control tissue 21.07 (0.08) 1.5 (54) 100 25.9 (12) 58.2 (8) 2.0 (41)

a
Fish exposed to 20 mg/L chloramine-T.
b
Ion m/z 531 below detection criterion of S/N ³ 5.
1108 IDOWU ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 87, NO. 5, 2004

Table 5. Interday reproducibility of retention time and ion abundances in the GC/MS confirmation of p-TSA in yellow
perch fish fillet tissue

Relative ion abundance, % (mean CV, %)

Signal-to-noise ratio
Retention time, min (S/N), for m/z 531
p-TSA level, ppb (Mean CV, %) 531 376 155 91 (mean CV, %)

Fortified tissue

20 21.06 (0.1) 1.5 (16) 100 23.9 (15) 60.5 (10) 14 (50)
40 21.06 (6) 1.4 (17) 100 24.5 (15) 62.7 (9) 28.8 (31)
100 21.06 (0.1) 1.4 (14) 100 25.0 (16) 63.4 (11) 60.0 (36)
400 21.07 (0.1) 1.6 (22) 100 24.1 (20) 61.8 (14) 158.9 (10)
1000 21.09 (0.1) 1.6 (18) 100 23.5 (20) 61.0 (14) 668.2 (34)
Incurred tissue, 50 ppb 21.07 (0.1) 1.3 (5) 100 30.7 (4) 67.5 (4) 28.0 (30)
Comparison standard, ng

60 21.05 (0.1) 1.4 (18) 100 23.9 (18) 66.8 (15) 13.5 (24)
120 21.06 (0.09) 2.2 (9) 100 21.8 (14) 56.7 (8) 41.4 (27)
300 21.06 (0.1) 1.5 (11) 100 24.5 (17) 62.4 (11) 61.9 (16)
a
Control tissue 21.05 (0.09) 0.28 (146) 100 29.5 (27) 64.2 (15) 0.9 (21)

a
Ion m/z 531 below detection criterion of S/N ³ 5.

Conclusions Acknowledgments

A GC/MS method for the confirmation of p-TSA
residue in fish fillet tissue from rainbow trout
(Oncorhynchus mykiss), channel catfish (Ictalurus
punctatus), yellow perch (Perca flavescens), and hybrid
striped bass (Morone saxatilis x M. chrysops) was
successfully developed and validated. The method can
establish the presence of p-TSA in fish tissue down to a
level of 20 ppb, which is well below the targeted
confirmation level of 10 ppm.
The technical assistance of David N. Heller and Cristina B.
Nochetto (U.S. Food and Drug Administration, Laurel, MD)
is gratefully acknowledged.
References
(1) From, J. (1980) Prog. Fish Cult. 42, 85–86
(2) Bullock, G.L., Herman, R.L., & Waggy, C. (1991) J. Aquat.
Anim. Health 3, 48–50
(3) Meinertz, J.R., Schmidt, L.J., Stehly, G.R., & Gingerich,
W.H. (1999) J. AOAC Int. 82, 1064–1070
(4) Carignan, G., & Carrier, K. (1991) J. AOAC Int. 74, 479–482
(5) Gyllenhaal, O., & Ehrsson, H. (1975) J. Chromatogr. 16,
327–333

View publication stats