close

Вход

Забыли?

вход по аккаунту

?

2302185

код для вставки
Determination of Phenobarbital, p - Hydroxyphenobarbital and Phenobarbital-N-glucoside
in Urine by Gas Chromatography Chemical
Ionization Mass Spectrometry
B. K. Tang,? B. Yilmaz and W. Kalow
Department of Pharmacology, University of Toronto, Toronto, Ontario, M5S 1A8, Canada
A sensitive and efficient analytical procedure was developed to determine phenobarbital-N-glucoside,phenobarbital
and p-hydroxyphenobarbital in urine. The stable-isotope labeled internal standards used in this assay were obtained
in urine from a subject who ingested "N-labeled phenobarbital.
INTRODUCTION
An investigation of two healthy subjects after ingestion
of radiolabeled phenobarbital (PB) showed that the drug
was excreted in urine as unchanged PB, phenobarbitalN-glucoside (PBNG), free p-hydroxyphenobarbital
(PBOH) and its glucuronate in man.' The amounts of
PB, PBNG and PBOH accounted for 29%, 27% and
18.5% of the dose, respectively.
While there are many chromatographic methods to
determine PB, only a few are suitable for the simultaneous determination of PB and PBOH.," There is no
quantitative method for PBNG. As a consequence, there
is very little information about this new and quantitatively important drug glucosidation. We, therefore,
report here a simple and sensitive gas chromatographic/mass spectrometric procedure for the determination of PBNG in urine. Basis for this assay is the
production of a stable-isotope labeled internal standard
contained in urine from a subject who ingested "Nlabeled PB. Since thereby all metabolites are produced
in labeled form, the same urine can be used as internal
standard for the simultaneous determination of PB and
PBOH.
cations. After heating the mixture containing N-(2,3,4,6tetraacetyl-P-D-glucopyranosyl) urea (1.5 mmol) and 2ethyl-2-phenylmalonyl chloride (3.0 mmol) at 80 "C for
1 h, the caramel coloured mixture was washed three
times with 1 ml of n-hexane and then two times with
1 ml of water. The residue was dissolved in 2 m1 of ethyl
acetate and applied to a glass column (2.2 cm i.d.) which
had been slurry-packed to a bed height of 25 cm with
silica (70-230 mesh, Merck/BDH Chemicals). The
column was eluted with ether at a flow of 1 ml min-'.
Fractions were collected and analysed by thin-layer
chromatography (TLC) and mass spectrometry and
appropriate fractions were pooled and then taken to
dryness. The. white amorphous solid obtained (310 mg,
m.p. 90-1 15 "C) was pure I -(2,3,4,6-tetraacetyl-p-~glucopyranosy1)phenobarbital. Heating the solid
(310mg) dissolved in 2ml of methanol containing
2.2 mmol of 6 N HCI at 75 "C for 3 h gave phenobarbitalN-glucoside (PBNG). After drying by a stream of
nitrogen and recrystallization from ethyl acetate and
cyclohexane gave pure PBNG (130mg, m.p. 120125 "C). The chemical ionization (methane) mass spectral data were: m / z 395 [MH]+, 233 (loo%), 205, 163,
145, 127.
The labeled urine
EXPERIMENTAL
Materials
P-Glucuronidase (Type H-2) was purchased from Sigma
Chemical Co. (St. Louis, Missouri), and N,O-bis(trimethysily1)trifluoroacetamide (BSTFA) from Pierce
Chemical Co. (Rockford, Illinois). Phenobarbital,
("N,)phenobarbital
(96.4 atom %), and p hydroxyphenobarbital were obtained as previously
described. I Phenobarbital-N-glucosidewas synthesized
according to our previous procedure' with some modifit Author to whom correspondence should be addressed. AbbreviPBOH = p-hydroxyphenobarbital;
ations:
PB = phenobarbital:
PBNG = phenobarbital-N-glucoside;
BSTFA= N,O-bis-(trimethylsilyl)trifluoroacetamide.
A healthy subject ingested 120 mg of (15N2)PBand total
urine was collected for 2 days. The urine was pooled
and kept frozen at -20 "C in small batches. The labeled
urine was used without further treatment. It contained
I5N2-labeledphenobarbital and all labeled metabolites,
i.e. ("N,)PBOH, ("N,)PBNG, etc.
Deconjugation and extraction
To 0.2-1.0 ml of urine, 0.3 ml of the labeled urine was
added. The mixture was incubated with 0.1 ml of pglucuronidase in 0.5 ml of 0.1 M acetate buffer (pH 4.5)
at 37 "C for 2 h. Then, 100 mg of ammonium sulfate was
added and the mixture extracted with 6 m l of ethyl
acetate after vortex-type stirring (20 s). About 5 ml of
ethyl acetate was transferred to a 15 ml conical test tube
CCC-O30&042X/ 84/00 1 1-0462 $02.00
462 BIOMEDICAL MASS SPECTROMETRY, VOL. 11, NO. 9, 1984
@ Wiley Heyden Ltd, 1984
PHENOBARBITAL METABOLITES
'ii400o\
361
Y
x
.-
4-l
VI
C
aJ
3 55
c
C
aJ
.->
50-
-
c
0
d
aJ
Ly
-4
I
319
0
1
1
I,
-2i
I
435
371
-
i - MO$i -
697
681
60%25
L
I
I
I
I
I
l
l
1
I1
I,
725
1
1
1
I
760
,
I.
I
m/z
Figure 1. Mass spectrum of the methylated and silylated derivative of phenobarbital-N-glucoside.
and then dried by a stream of nitrogen. The residue was
dissolved in 0.1 ml of ethanol and then methylated with
excess diazomethane in ether at room temperature for
20 min. The dried mixture was redissolved in 50 pl of
dimethylformamide of which 5 pl was taken out and
diluted with 50 p1 of dimethylformamide for gas
chromatographic/mass spectrometric analysis of PB and
PBOH. The remaining 45 yl was silylated with 50 p1 of
BSTFA at 80 "C for 30 min and 1-2 p1 was taken for gas
chromatographic/mass spectrometric analysis of
PBNG.
Gas chromatographic mass spectral analysis
The gas chromatographic/mass spectrometric system
(Biospect) used was described previously.' A glass
column (2 mm i.d., 0.9 m long) containing 3% SE-30 on
80/100 mesh Chromosorb W was used. The gas
chromatographic conditions were: injector port at
280 "C; helium 20 ml min-I. The mass spectrometer was
operated in chemical ionization mode and methane was
used as reagent gas; ion source pressure 1.0 mm Hg at
200 "C.
Analysis of PB and PBOH
The oven temperature was delay-programmed at 170 "C
for 3 min and then to 220 "C at 20 "C min-I. The retention
was 2.7 min for PB and 4.2 min for PBOH. Four channels
were selected to monitor ions at m / z 261 and 263 for
N,N-dimethyl derivatives of PB and (I5N2)PB respectively, and m / z 291 and 293 for N,N,O-trimethyl derivatives of PBOH and ("N2)PBOH respectively.
Analysis of PBNG
The oven temperature was kept at 270 "C and two channels were selected to monitor ions at m l z 697 and 699
for the N-methyl-tetra-0-trimethylsilylderivative of
PBNG and (ISN2)PBNGrespectively. The retention time
was 3.0 min.
Peak height ratics of I4N to "N were measured for
each of six peaks ax:.? the amounts of PB, PBOH and
PBNG were estimate( from three different calibration
curves.
Calibration curves
Known amounts of PB, PBOH and PBNG ranging
between 0.05 and 2.0 pg were added to 0.5 ml of blank
urine. Then 0.3 ml of the labeled urine was added and
samples were deconjugated, extracted, derivatized and
analysed as described. Peak ratios of I4N to ISN were
plotted against added amounts of PB and its metabolites.
RESULTS AND DISCUSSION
The N-methyl derivative of PBNG, like that of amobarbital-N-glucoside could be formed readily with
diazomethane.* Pyrolytic decomposition of N-methylPBNG at 280 "C gave mainly N-methylphenobarbital.
Unfortunately, interferences at m / z 247 from blank
urine excluded the possibility of determination of PBNG
as N-methylphenobarbital.
Because of the thermal instability of N-methyl-PBNG
further derivatization was necessary. Permethylation of
N-methyl-PBNG by methyl iodide under various conditions caused extensive decomposition of the glucoside.
On the other hand, both tetra-acetyl and tetra-trimethylsilyl derivatives could be made easily. The latter gave
more favorable properties for gas chromatographic/ mass spectrometric analysis. Figure 1 shows the
methane mass spectrum of the silyl derivative with protonated parent ion at m / z 697 [MH]+. The gas chromatographic/mass spectrometric chromatograms showed a
single peak when monitoring ions at m / z 609, 625, 681
[MH- 16]+, 697 [MH]+ and were free of interference
from blank urine. When lower masses were monitored
some interferences were observed. The protonated
parent ion was chosen for single ion monitoring because
of the relatively strong signal. General precautions for
BIOMEDICAL MASS SPECTROMETRY, VOL. 11, NO. 9, 1984 463
B. K. TANG, B. YILMAZ AND W. KALOW
gas chromatographic/mass spectrometric analysis of
trimethylsilyl derivatives of glucoside were adopted.'
Although a 1% SE-30 liquid phase gave a sharp peak
with short retention times in the chromatogram it is not
recommended for routine use because of the relatively
short column life.
The tetra- 0-trimethylsilyl derivative of N-methylPBNG was stable in dimethylformamide for 24 h at room
temperature and at least 3 months at -20 "C. However,
it decomposed with protic solvents including moisture.
A single-step extraction of PBNG by ethyl acetate was
sufficient for this assay; in control experiments, about
50% of I4C-labeled PBNG' was extracted from urine
saturated with ammonium sulfate. Under the same conditions, more than 80% of PB and PBOH was extracted.
Thus it was possible to extract PB, PBOH and PBNG
simultaneously after enzymatic hydrolysis of the PBOH
conjugate. Although the N-C bond of PBNG was partly
cleaved by acid hydrolysis (6 N HC1 at 100 "C for 1 h)
it was virtually untouched by glucuronidase.
Attempts to analyse PB, PBOH and PBNG simultaneously by gas chromatography/rnass spectrometry
(GC/ MS) after methylation and silylation failed because
of extensive interferences of silylation with PB and
PBOH derivatives. Consequently, PB and PBOH had to
be analysed after methylation, and then PBNG after
silylation at a high column temperature. The resultant
gas chromatographic/mass spectrometric chromatograms were free of interferences and optimal retention
times were achieved.
The labeled urine which contained I5N2-1abeledPB
and its metabolites used as internal standards for this
assay, was produced by a subject who ingested (I5Nz)PB.
Quantitation and isolation of each metabolite was not
necessary because the same amount of the labeled urine
was used for the testing of samples and for the construction of calibration curves (Fig. 2). The excess amount
of 14N drug and its metabolite could be estimated from
these calibration curves given a measurable I4N/'*N
ratio. The calibration curves of PB, PBOH and PBNG
(Fig. 2) showed linearity between 0 and 2 pg with correlation coefficients of better than 0.999 for PB and PBOH
and better than 0.995 for PBNG. The ordinate intercepts
for PB and PBOH ranged from 0.04 to 0.06, consistent
with the percentage of I5N in the drug ingested. The
intercepts for PNBG ranged from 0.30 to 0.35 which
was consistent with a strong P + 2 peak for trimethylsilyl
derivatives. The different slopes obtained were a reflection of the varying amounts of labeled substances
excreted by the volunteer.
Recoveries and precisions of the assay for the analysis
of spiked urine samples are shown in Table 1. Recoveries
were almost quantitative for the ranges studied. The
coefficients of variation (CV) tended to increase with
decreasing concentrations. The detection limits of this
assay were about 0.1 ng ml-' for these substances in
urine. Samples with concentrations higher than
3 pg ml-I had to be diluted and reanalysed.
Enzymatic hydrolysis of the PBOH conjugate was
completed within 2 h. The single-step extraction of ethyl
acetate was efficient and exhaustive extraction did not
change any of the ratios. Methylation of barbiturates by
diazomethane was reported to give a mixture of N- and
0-methyl derivatives" of which the former predominate. In this assay, minor contribution of the 0-methyl
464 BIOMEDICAL MASS SPECTROMETRY, VOL. 11. NO. 9, 1984
P
PBoH
Figure 2. Calibration curves of p-hydroxyphenobarbital (PBOH),
phenobarbital (PB) and phenobarbital-N-glucoside (PBNG).
derivatives of PB and PBOH did not affect the quantitation because they had longer retention times. An 0methyl derivative of PBNG was not observed.
This assay was applied to follow urinary excretion of
PB and its metabolites for 20 days after ingestion of
120 mg of PB by three healthy subjects. Table 2 shows
that the recovery of PB and its major metabolites
averaged 69.3% of dose of which PB, PBNG and PBOH
accounted for 26.3%, 25.7% and 11.3%, respectively.
This is consistent with our previous report based on
Table 1. Relative recovery and precision of
measurement of phenobarbital-N-glucoside,
phenobarbital
and
phydroxyphenobarbital added to blank
urine and assayed in triplicate
Amount added
Found by GC/MS
Mean *SD
(ws m1-7
(ws ml-')
Phenobarbital-N-glucoside
0.102
1.02
2.04
cv (%I
0.095i 0.006
1.02* 0.04
2.02f0.02
6.5
3.5
1 .o
0.106i0.009
0.99*0.01
1.98* 0.02
7.7
1 .o
1 .o
Phenobarbital
0.100
1 .oo
2.00
p-Hydroxyphenobarbital
0.109
1.09
2.17
0.106* 0.003
1.13f 0.00
2.15f0.04
2.9
0.0
2.0
PHENOBARBITAL METABOLITES
“Ol
Table 2. Amounts (in ‘/o of dose) of phenobarbital (PB), phydroxyphenobarbital (PBOH) and phenobarbital-A’glucoside (PBNG) excreted in 20-day urine in three
healthy subjects after ingestion of 120 mg of phenobarbital
Subject
1
2
3
Mean fSD
Weight
(kg)
50
64
68
PB
PBOH
PBNG
Total
28
16
24
68
26
70
22
22
29
14
27
70
26.33~3.1 17.353.4 25.73Z1.2 69.3*0.9
PBNG
z
L
0
c
._
L
3
-
0.011
0
I
I
I
I
I
,
, , ,
10
Time ( Days)
b
I
20
I
,
Figure 3. Urinary rate of excretion of PBNG ( 0 3 ) .PB (A-A)
and PBOH (04)
from a healthy subject after ingestion of 120 rng
of PB.
ingestion of I4C-labeled PB.’ Figure 3 shows that urinary
rates of excretion of PB and its metabolites are parallel
and follow first-order kinetics.
The success of this procedure further demonstrated
the efficiency of using biosynthetic production of stableisotope labeled compounds as internal ~ t a n d a r d s . ~
Labeled urine worked well as internal standards also as
part of a recently described method involving acid
hydrolysis and extractive methylation for the simultaneous determination of PB and PBOH by GC/MS.6
However, it failed to extract the main metabolite PBNG
and hence has limited usefulness.
Acknowledgement
T h i s work was supported by Medical Research Council, Ottawa,
Canada, Grant No. MT-4763 and MA-5799 and MA-7303.
REFERENCES
1. 13. K. Tang, W. Kalow and A. A. Grey, Drug Metab. Dispos. 7 ,
315 (1979).
2. N. KAllberg, S. Agurell, 6. Ericsson, E. Bucht, 6. Jalling and L. 0.
Borens. Europ. J. Clin. Pharmacol. 9, 161 (1975).
3. M. P. Whyteand A. D. Dekaban, Drug Metab. Dispos.5.63(1977).
4. R. W. Dykeman and D. J. Ecobichon, J. Chromatogr. 162, 104
(1979).
5. D. Kadar, 8. K. Tang and A. W. Conn, Can. J. Anaesth. SOC.J.
29, 16 (1982).
6. A. Van Langenhove, J. E. Biller, K. Biemann and T. R. Browne,
Biomed. Mass. Spectrom. 9, 207 (1982).
7. B. K. Tang, H. Uchino, T. lnaba and W. Kalow, Biomed. Mass
Spectrom. 9,425 ( 1 982).
8. B. K. Tang, W. Kalow and A. A. Grey, Res. Comm. Biochem.
Path. Pharmacol. 21, 45 (1978).
9. E. M. Martinelli, Europ. J. Mass Spectrom. 1, 33 (1980).
10. D. J. Harvey, J. Nowlin, P. Hickert, C. Butler, 0. Gansow and
M. G. Horning, Biomed. Mass Spectrom. 1, 340 (1974).
Received 27 September 1983; accepted (revised) 27 December 1983
BIOMEDICAL MASS SPECTROMETRY, VOL. 11, NO. 9, 1984 465
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 518 Кб
Теги
1/--страниц
Пожаловаться на содержимое документа