Determination of Phenylbutazone, Tolbutamide and Metabolites in Plasma and Urine Using Chemical Ionization Mass Spectrometry Robert J. Weinkam,? Malcolm Rowlandf and Peter J. MeffiP Department of Pharmaceutical Chemistry, School of Pharmacy, University of California, San Francisco, California 94143, U.S.A..

Quantitative analytical procedures for the analysis of phenylbutazone and tolbutamide levels in plasma have been developed which involve the addition of deuterium labeled internal standards to plasma followed by extraction and direct sample insertion into a mass spectrometer operating under chemical ionization conditions. Peak height ratios used to calculate plasma levels were determined by using either selected ion monitoring or repetitive scan data. The scan approach was used in a related procedure for the simultaneous determination of tolbutamide and two metabolites from urine. The accuracy, precision and sensitivity of the direct sample UlSertiQR opmach to drug level mewreme& kas been determined. Examples are given of data obtained in the course of pharmacokinetic studies in which this analytical approach appears to offer advantages in the analysis of multicomponent mixtures encountered in drug-drug interaction studies.

INTRODUCTION Chemical ionization mass spectrometry (c.i.m.s.) was introduced' as a method of ionizing gas phase sample molecules through reactions in reagent gas ion plasma.2 A reagent gas such as isobutane allows protonation of most organic molecules with little energy transfer so that the probability of fragmentation of the protonated molecular ion is lowered. Because many compounds are ionized with little or no fragmentation, chemical ionization can be used in the analysis of mixtures such as biological fluid extract^.^ Indeed, the incidence of two or more compounds contributing to the ion current at a given mass is reasonably low. The general approach of using c.i.m.s. analysis of biological fluid extracts combined with stable isotope labeled internal standards has been applied to several problems. Milne, Fales and Axenrod demonstrated the c.i.m.s. identification of drugs from stomach contents4 and they successfully applied this technique to distinguish between different barbit~rates.~ Garland, Trager and co-workers quantified quinidine, lidocaine and related compounds in human plasma,6 and then studied the liver microsomal and human metabolism of warfaThis laboratory has studied the stereoselective metabolism of 1-(2,5-dimethoxy-4-methylphenyl)-Zamino C.i.m.s. scans of individual rat brain tissue sections following ion exchange isolation of the amine fraction and derivatization allowed quantification of the endogenous catecholamines dopamine and norepinephrine and the L-Dopa metabolites crmethyldopamine and (Y -methyn~repinephrine.~ In a

t Author to whom correspondence should be addressed. $ Present address, Department of Pharmacy, University of ManChester, Manchester, England. 0 Present address, Cardiology Division, School of Medicine, Stanford University, Stanford, California 94305, U.S.A.

@ Heyden & Son Ltd, 1977

42 BIOMEDICAL MASS SPECTROMETRY, VOL. 4, NO. 1, 1977

preliminary study a direct insertion probe analysis of tolbutamide and its hydroxy and carboxy metabolites plasma levels was achieved followingderivatization with diazomethane and ether extraction. The deuteriodiazomethane derivative of each compound was used as standard in this procedure." Two advantages of direct c.i.m.s. analysis of biological fluid extracts are that several constituents can be analyzed simultaneously and that certain thermally labile compounds which are not amenable to gas chromatography can be analyzed. Both of these features were utilized in this effort to develop an analytical approach to the problems encountered in the course of drug-drug interaction studies in which the presence of one agent will influence the plasma level of a second agent and its metabolites. Tolbutamide, N - butyl - N ' - (4- methylbenzenesulphony1)urea (1) is an orally effective hypoglycemic drug, commonly used to treat diabetes. Numerous reports of complications arising from hypoglycemic episodes have been reported when other drugs are administered with tolbutamide. l 1 Hypoglycemic crises have occurred in patients receiving both tolbutamide and phenylbutazone, 1,2-dipheny1-3,5-dioxo-4-nbutylpyrazolidine (4), an antiflammator a ent used in the treatment of rheumatoid arthritis.", One possible mechanism for the potentiation by phenylbutazone of the hypoglycemic effect of tolbutamide is an inhibition of tolbutamide metabolism. In man, tolbutamide is eliminated by oxidative processes: it is first transformed to hydroxytolbutamide (2), then further oxidized to carboxytolbutamide (3). In order to investigate this possible mechanism, it is necessary to measure plasma concentrations of tolbutamide and phenylbutazone and the urinary excretion of tolbutamide and its two metabolites. Several approaches have been used to measure tolbutamide in biological fluids. Older methods used either ultraviolet absorption14 or colorimetric determination

Y g

PHENYLBUTAZONE, TOLBUTAMIDE AND METABOLITES

with 2,4-dinitrofluorobenzene.15 These methods have poor sensitivity and selectivity. More recently, a number of gas chromatographic methods have been described for tolbutamide which involve its conversion to a more volatile N-methylsulphonylurea.16-1*

4

Gas chromatographic methods for the analysis of tolbutamide and its metabolites have been d e v e l ~ p e din ~~ which * ~ ~the drug and its hydroxy and carboxy derivatives are selectively extracted and converted to the electron capture responsive N-butyl-2,4dinitroaniline derivative. These methods are sufficiently sensitive to be applied to pharmacokinetic studies, but they are very time consuming and are limited in dynamic range because the selective extraction of the metabolites is incomplete. Several spectrophotometric methods have been reported for the analysis of phenyl6utazone in biological fluids, but the specificity of some of these methods is in doubt.21-2s Gas chromatographic analyses for phenylbutazone have been described which are more sensitive and specific than spectrophotometric analyses.26-29 Since it was necessary to measure levels of tolbutamide and phenylbutazone in plasma as well as tolbutamide and its two metabolites in urine, the possibility of analysis by c.i.m.s. was examined. This approach offered the possibility of analyzing simultaneously for tolbutamide and phenylbutazone in plasma and for tolbutamide in urine. This would avoid the necessity of a separate analysis for each compound.

EXPERIMENTAL

Mass spectrometric analysis The instrument used in this study was an A.E.I. MS-902 high resolution mass spectrometer which has been modified to operate under c.i. condition^.^' Samples were introduced using a ceramic direct insertion probe in which the sample was carried into the ionization chamber by the reagent gas flow. All spectra were obtained using isobutane reagent gas at 0.4 Torr and 200 "C ion chamber temperature. When operated in the scan mode, the instrument was repetitively scanned over the mass region of interest as the sample was warmed by induction from the ion chamber. The relevant protonated molecular ion [MH]+ peak heights of approximately five scans were measured manually and averaged to determine ion abundance. This mode was used primarily for the simultaneous quantification of several components in a single sample such as the urine tolbutamide and metabolite analysis (Fig. ij.

Figure 1. An isobutane c.i.m.s. scan of a urine sample extract in which the sample was taken following a bolus dose of tolbutamide. The sample was prepared through addition of internal standards, ether extraction and diazomethane treatment. The scan shows the presenceof the methylated tolbutamides ([MH]+, 285, 287), hydroxytolbutamides ([MHl+, 301, 304), [2H]2phenylbutazone([MH]+, 325), a urine component ([MH]', 327) and the carboxytolbutamides ([MHl+, 329,331).

When the instrument was operated in the selected ion monitoring mode, two ions of interest could be measured using the peak matching circuitry of the MS-902. This provided accelerating voltage alternation to focus two ions with the auxiliary scan coil disconnected. The signal across the collector meter was monitored using a strip chart recorder as the sample was evaporated into the ion chamber. The height of the two alternating signals was measured near the apex of the evaporation curve as this was found to be more precise than measurement of the area under the evaporation curve (Fig. 2). No deuterium isotope effect was observed on evaporation of a labeled compound from the direct insertion probe. The percent coefficients of variation for peak ratios measured at the beginning, middle and end of evaporation curves were within 0.5%.

Syntheses n-[l,l-2H]2ButyIamide hydrochloride. The compound was prepared by the Amudsen and Nelson31 method, using 3.46 g (0.059 mol) of n-butylnitrile and 2.0 g of lithium aluminum deuteride. The ethereal extract of the reaction mixture was dried over KzC03 and converted -311.173

'iL TlW

Figure 2. Stripchart recorder output showing selected ion intensity of phenylbutazone ([MH]+, 309.160) and [2H]2phenylbutazone([MH]+, 311.173) from a plasma sample ether extract.

BIOMEDICAL MASS SPECTROMETRY, VOL. 4, NO. 1, 1977 43

R. J. WEINKAM. M. ROWLAND AND P. J. MEFFIN

to the salt using HCl gas to give 4.95 g (0.044 mol, 75% yield) of the dideuteriobutylamine hydrochloride. Mass spectral analysis showed that deuterium incorporation exceeded 98%. n-[l,l~H]zButytisocyanate. The above amine was converted to the isocyanate by a modification of the n-[l,l-2H]2butylamine method of Staab and hydrochloride, 4.95 g (0.044 mol) was dissolved in 250ml of CHzClz and stirred over anhydrous K2C03 (50 g) for 24 h. The K2C03 was removed by centrifugation and washed with 50 ml of CH2C12.Diimidazolecarbony1 (7.43 g, 0.045 mol) was placed in a 500 ml round bottom flask equipped with a magnetic stirrer and drying tube. The combined CH2C12solutions of amine were added slowly over a period of 15 min. After continued stirring at ambient temperature for 15 min the solvent was removed under vacuum and the residue was distilled to give 1.35g (0.013mo1, 30% yield) of the isocyanate, b.p. 110-114°C. No loss of deuterium was observed in the course of the reaction. N-[1,1-2H]zB~tyl-N-(methylbenzenesulfonyl)urea, [2Hlztotbutamide.[2H]2tolbutamidewas prepared from 1.35g (0.013mol) of the above isocyanate and 4methylbenzenesulfonamide using the method of K i m b r o ~ g h . ~The ~ crude reaction product was crystallized from aqueous ethanol (1 : 1) to yield 2.03 g (0.0075 mol, 54% yield). Mass spectral analysis showed 98% dideuterium, 1.7% monodeuterium incorporation. N-[l,l-2H]zButyl-N-(4- earboxybenzenesulfonyl) urea, [2H]2carboxytolbute. [2H]2carboxytolbutamide was prepared by the biotransformation of 1.0 g (0.003 mol) of [2H]2tolbutamide in man. Urine was collected for 12 h following an oral dose, adjusted to pH 1using 12 N HCl, and extracted with 3x300 ml of ether. The ether was extracted with 25 ml 2 N NaOH and discarded. ~2H]2carboxytolbutamidewas precipitated from the NaOH solution on acidification and crystallized from aqueous ethanol (1 :1)to give 0.25 g (0.0008 mol, 28% yield of [2H]2carboxytolbutamide, m.p. 209-212 "C. Mass spectral analysis showed 97% dideuterium, 2.2% monodeuterium incorporation. N-Butyl-N- [2H]3methyl-N-(4-hydroxymethylbenzenesulfony1)urea. Active hydrogens from 1.O g (0.0035 mol) of hydroxytolbutamide (2)in 150 ml of ether was exchanged by washing 4x2 ml of 2% 'HC1 in 2 HzO. The ethereal solution of 2 was added to an excess of ethereal deuteriodiazomethane (from deuteriodiazald, Aldrich) and 1ml of 2H20with vigorous stirring. After standing for 30 min at ambient temperature, 5 mi of 1N NaOH was added and the ether layer removed, dried over Na2S04, and evaporated to give 0.72g (0.0024moJ 68% yield) of pure hydroxytolbutamide N'-[ H],methyl. Mass spectral analysis showed 9 1YO trideuterium, 6.4% dideuterium and 2.1% monodeuterium incorporation. Diethyl 2-(1,1-[ZH]zbutyl)malonate. This precursor in the phenylbutazone synthesis was prepared from diethyl malonate and 0.66 g (0.0048 mol) of l-bromo-1,lC2HJ2butaneby the method of Adams and Kamm34to give 0.8 g (0.0037 mol, 77% yield) of the malonate, b.p. 235-240 "C. 1,1-[2H]2butan-l-ol was prepared from butyric acid on reduction with lithium aluminum deuteride according to Nystrom and Brown.35 The resulting i,l-[2H]2butan-l-ol was converted to 1-

-

44 BIOMEDICAL MASS S P E C ~ O M E ~ YVOL. , 4, NO. 1, 1977

bromo-1 l-['HI2butane by the method of Kamm and Marvel." 1,2-Diphenyl-3,5-dioxo-4-(l,l-[2H]zbutyl), pyrazotidine, [2H]zphenylbutazone. ['HI2phenylbutazone was prepared from 0.8 g (0.0037 mol) of the above diethylrnalonate and N,N-diphenylhydrazine by the method of Buchi et ~ 1 . Crystallization 3 ~ of the crude product from ethanol gave 0.55g (0.0018 mol, 49% yield) of [2H]2phenylbutazone,m.p. 103-104 "C. Mass spectral analysis showed 96% dideuterium, 3.6% monodeuterium incorporation.

Plasma sample analysis A volume of plasma up to 1ml containing 2-50 pg of phenylbutazone and/or tolbutamide was added to a glass tube containing 1ml of 0.2 N HCl and 5 ml of ether. Internal standards [2H]2phenylbutazone (100 p1, 9.48 pg in 0.05 N NaOH stored at 4 "C) and/or [2H]2tolbutamide (100 pl, 0.50 pg in pH 7.4 phosphate buffer stored at 4 "C]were added and the tube shaken and the aqueous layer was frozen. The ether layer was transferred to a second glass tube containing 0.5ml of 0.5N NaOH, shaken, centrifuged, frozen and the ether layer poured off. The aqueous layer was acidified with 100 pl of 6 N H2S04together with 5 ml of ether. The tube was shaken, centrifuged and frozen. An aliquot of the ether solution was transferred to a conical tube and evaporated to dryness at 20-45 "C using boiling chips. The residue may be analyzed mass spectrometrically for phenylbutazone (Fig. 2). A second aliquot of the ether solution was treated with excess ethereal diazomethane (stored at 4 "C).After 10 min the solution was evaporated to dryness as above. The residue may be analyzed mass spectrometrically for tolbutamide. Tolbutamide is converted to the N-CH, derivative on treatment with diazomethane. Methylation increases the volatility of the sulfonylurea and stabilizes the function toward thermal degradation. Occassionally side reactions during the methylation of phenylbutazone interfered with the [2H]2phenylbutazone measurement so that the simultaneous analysis of the two compounds was not pursued.

Urine sample analysis A 100 p1 urine sample containing tolbutamide and its metabolites was added to a glass tube containing 0.5 ml of 1.5N HC1 and 3ml of ether. Internal standards [2H]ztolbutarnide(100 pl, 9.50 p g in pH 7.4 phosphate buffer stored at 4 "C) and [2H]2carboxytolbutamide (100 pl, 17.84 pg in pH. 7.4 phosphate buffer stored at 4 "C) were added and the tube shaken and frozen. The ether layer was removed and the aqueous solution again extracted with 3ml of ether. The sulfonylureas were extracted from the combined ether solutions with 0.3 ml of 0.5 N NaOH and the ether removed. The aqueous layer was acidified with 100 pl of 1.5 N HCl and the internal standard N'-[2H]2methylhydroxytolbutamide (100 pl, 57.76 p g in ether stored at 4 "C) was added along with 3ml of ether. The aqueous solution was washed twice with ether as above and the ethereal solution treated with excess ethereal diazomethane.

PHENYLBUTAZONE. TOLBUTAMIDE AND METABOLITES

After 10 min the solution was evaporated to dryness at 2 0 4 0 ° C using boiling chips. The residue may be analyzed mass spectrometrically for tolbutamide, carboxytolbutamide and hydroxytolbutamide (Fig. 1).

l

~

I

*

l

'

I

.

1

~

1

.

I

26 22

E

-

RESUL'IS AND DISCUSSION

P f

The approach to the analysis of phenylbutazone and tolbutamide described in this report was based on the observation that there is little interference to the quantitative measurement of the phenylbutazone and tolb u t a i d e protonated molecular ion current in an isobutane c.i. scan Of a plasma or urine extract containing these drugs. This is illustrated in Fig. 3 which shows a high resolution isobutane c.i. scan of Nmethyltolbutamide ( m / e 285.145) and N-methyl[2H;l,tolbutamide (m/e 287.158). Approximately 10 pg of each compound was added to 1 ml of sheep plasma, extracted and the extract scanned while being evaporated into the ionization chamber. Instrument background and plasma extract related ions are present in this scan, but they contribute less than 2% to the total ion current at each nominal mass. The contribution of these ions to the peak height of the drug related ion at low resolution (M/AM 2500) would be less than 1%, or the equivalent of 100ngml- in this scan. No significant ion current was observed at the exact masses of the drug related peaks in a similar scan of a blank plasma extract. The predominance of the drug and standard peaks Over other ions in the plasma extract is the result of a combination of factors. The conditions of isobutane c.i.m.s. are sufficientlymild so that extensive fragmentation of organic molecules is less than electron bombardment or methane c.i.m.s. Therefore, higher molecular weight compounds present in plasma in abundance are less likely to yield fragment ions at the nominal mass of the compound of interest. The concentrations of both drugs as used in this study were in the p g ml-' plasma range and are thus present at much higher levels than are most endogenous substances. Drug levels were calculated directly from peak heights after correcting for isotope contributions from the drug or labeled standard. Curves of calculated vs added levels for phenyibutazone and tolbutamide are shown in Figs. 4 and 5. Tables 1and 2 compare the precision and accuracy of the plasma assay of phenylbutazone and tolbutamide using the selected ion monitoring (s.i.m.) and scan modes of ion current measurement. The s.i.m, mode was used primarily in single compound analysis while the

d

10

-

2- I I

.

I

2

6

I

,

.

10

I

14

,

I

.

1

18

.

22

1

26

Expected ips m T ' f

Figure 4. Phenylbutazone standard curve from plasma.

10

--L

8

-n , 6 P tj x

4 2 0

2

4

6

8

1

0

Expected (pg m(-')

Figure 5. Tolbutamide standard curve from plasma.

scan mode was used in the analysis of several components present in a single extract. For purposes of comparison, all standard assays were measured relative to 9.5 pg of 2H2-internalreference per sheep plasma sample. The tables incorporate points from standard curves taken on different dates to reflect day-to-day variations in the assay. These data indicate the validity of computing plasma levels from peak height ratios less than 5 :1 (2-25 pg ml-'). The use of smaller amounts of internal Table 1. Phenylbutazone standard analysis Calculated from mass spectral peak heights using 9.48 pg ml-' sheep plasma of [2H12phenylbutazone as internal reference Selected ion recording mode Mean

n

Average % Average deviation deviation from mean from mean

Added b g m1-l)

b g ml-'1

2.0 5.0 10.0 25.0

1.97 4.85 10.40 25.60

3 3 5 2

0.05 0.05 0.04 0.40

0.90 1.75 5.23 10.10 22.80

2 3 3 3 3

0.10 0.05 0.21 0.46 0.52

obs.

Average error

%Average error

2.5 1.0 0.4 1.6

0.06 0.16 0.47 0.72

3.0 3.9 4.5 2.8

11.1 2.9 4.0 4.5 2.3

0.10 0.25 0.31 0.47 2.70

11.1 14.3 6.2 4.7 11.8

Scan mode

-J! 287.158

4 286.148

'

285.145

A

L

284.173

Figure 3. A high resolution isobutane c.i. scan of Nmethyltolbutamide and Kmethyl-[*HI2tolbutamide from a plasma ether extract treated with cfiazometthane.

1.0 2.0 5.0 10.0 25.0

~~

BIOMEDICAL MASS SPECTROSCOPY, VOL. 4, NO. 1, 1977 45

R. J. WEINKAM, M. ROWLAND AND P. J. MEFFIN

Table 2. Tolbutamide standard analysis calculated from mass spectral peak heights using 9.6 p g ml-' sheep plasma of [ZH]ztolbutamideas internal reference

1400

°

Selected ion recording mode Mean

obs.

n

Average % Average deviation deviation from mean from mean

Added (pg m1-l)

Ips m1-l)

1.0 2.0 2.5 5.0 10.0 27.5

0.97 1.87 2.50 4.98 10.30 16.80

4 3 3 5 5 4

0.14 0.12 0.00 0.45 0.78 0.88

1.83 5.00 10.20 33.60

4 3 3 2

0.08 0.10 0.38 1.35

Average error

% Average

14.0 6.4 0.0 9.0 7.5 3.3

0.15 0.18 0.00 0.46 0.83 1.72

15.0 9.6 0.0 9.2 8.1 6.4

4.3 2.0 3.7 4.0

0.19 0.10 0.41 3.55

10.4 2.0 4.0 10.6

error

0

Scan mode

2.0 5.0 10.0 30.0

't

I

1

i

100

200

I

I

300

500

400

600

700

800

Time (min)

reference permits each assay to be carried to at least 0.5pg ml-' plasma. Plasma sample sizes as small as 50 pl may be analyzed without loss of accuracy. Measurement of drug concentrations exceeding 25 pg ml-' were carried out by using the same aliquot of internal reference, but by reducing the size of the plasma sample. The c.i.m.s. analytical approach has been shown to be as accurate as the gas chromatographic analysis of tolb~tamide.~'

Figure 8. Total phenylbutazone ( 0 )and tolbutamide (0) levels in sheep plasma following a 0.9 g i.v. bolus of phenylbutazone administered during the continuous infusion of tolbutamide (124.7 mg h-' ).

Direct insertion of a plasma extract into the c.i. source offers a rapid method for drug quantification. The mass spectral time interval between sample analysis is less than 10 min, most of which is spent cleaning and loading the sample probe. This approach offers a potential advantage for multiple drug analysis because minimal changes of instrumental parameters are required to change from one assay to another. The data obtainable is

- ,0°~ 400

1

100,

8

0

0

0

0

0 0 Time (min)

Figure 6. Total phenylbutazonelevels in sheep plasmafollowing a 1.0 g i.v. bolus.

0.6 El 0.4-

.+ X

w I

"

0

0.20.1

0

0 0

0 0

4t -2

0

1 100

200

300

400

500

Time (min)

Figure 7. Total tolbutamide levels in sheep plasma following a 1.O g i.v. bolus.

46 BIOMEDICAL MASS SPECTROMETRY, VOL. 4,

NO. 1, 1977

I

I

200

300

I

100

I

Time (min)

Figure 9. Urine excretion rates for tolbutamide (TI, hydroxytolbutamide (H) and carboxytolbutamide (C) following a900 mg i.v. bolus of tolbutamide to sheep.

PHENYLBUTAZONE, TOLBUTAMIDE AND METABOLITES

of sufficient quality to permit a detailed analysis of pharmacokinetic parameters. Plasma concentration-time curves for phenylbutazone and tolbutamide are shown in Figs. 6 and 7.39 Each figure shows the decline in total plasma drug levels following a single i.v. bolus injection of 1.0 g to sheep. An area of growing importance in pharmaceutical research involves the study of drug-drug interactions. The quantitative evaluation of the effects of one drug on another requires the analyses of several components in a single sample. The interactive effects of a 900mg phenylbutazone i.v. bolus with a steady-state tolbutamide plasma level resulting from continuous infusion at 124.7 mg h-' is shown in Fig. 8. An isobutane c.i. scan of a biological fluid extract can be used to quantify several compounds if the appropriate internal standards are added to the fluid. This approach has been used to measure the relative excretion rates for tolbutamide and its metabolites, hydroxytolbutamide and carboxytolbutamide in urine. Figure 9 shows the measured urinary excretion rates at closetime intervals for these compounds immediately following a 900 mg tolbutamide i.v. bolus. The time intervals over which these samples were collected were close relative to the duration of tolbutamide and metabolite excretion so that the average excretion rate is near the

maximum for these compounds. The measured rates for tolbutamide 1.8 1.1 mg h-' ; hydroxytolbutamide, 42.2 *7.9 mg h-'; and carboxytolbutamide, 3.7* 0.9 mg h-' confirm that the hydroxy compound is the major metabolite in sheep (>88%) and are comparable to those measured using other technique^.^' The accuracy of the urine assay using the scan mode as shown in Fig. 1was within the same limits as the plasma assays of Tables 1 and 2. [2H]zTolbutamide and [2H]2carboxytolbutamidewere used as reference compounds and handled in the manner described for the plasma analyses. The reference compound for hydroxytolbutamide was conveniently prepared as the N-C2H3 derivative which is added to the urine extract prior to treatment with diazomethane. Therefore, the reference compound is an analog of the hydroxytolbutamide derivative. It is feasible to use this approach as the hydroxytolbutamide extractions and methylation reactions are quantitative."

*

Acknowledgement This work was supported in part by NIH grant GM 16496 and Research Career Development Award K4-GM-00007 (R.J.W.).

REFERENCES 1. M. S.B. Munson and F. H. Field, J. Am. Chem. Soc.88,2621 (1966). 2. G. W. A. Milne and M. J. Lacey, Crit. Rev. Anal. Chem. 4,45 (1974). 3. R. J. Weinkam, in RecentAdvances in Mass Spectrometryin Biochemistry and Medicine, edited by Frigerio, A., Castagnoli, N. p. 83,Spectrum Press, New York, N.Y. (1975). 4. G. W. A. Milne, H. M. Fales and T. Axenrod, Anal. Chem. 43,

1815 (1971). 5. H. M. Fales, G. W. A. Milne and T.Axenrod, Anal. Chem. 42, 114 (1970). 6. W. A. Garland, W. F. Trager and S. D. Nelson, Biomed. Mass Spectrom. 1, 124 (1974). 7. L. R. Pohl, S. D. Nelson, W. A. Garland and W. F. Trager, Biomed. Mass Spectrom. 2,23(1975). 8. R. J. Weinkarn, J. Gal, P. Callery and N. Castagnoli, Anal. Chem. 48,203 (1976). 9. C. R. Freed, R. J. Weinkam, K. L. Melmon and N. Castagnoli, Anal. Biochem. in press.

10. R. J. Weinkarn and S. B. Matin, American Society for Mass Spectrometry, 21st Annual Conference on Mass Spectrom. and Allied Topics, San Francisco, California (1973).Paper

L13. 11. D. A. Hussar, J. Am. Pharm. Assoc. NSIO,619 (1970). 12. L. K. Christensen, J. M. Hansen and M. Kristensen, fancet2, 1298 (1963). 13. F. Kaindl, A. Kretschy, H. Pukandle and J. Wutte. Wein. Klin. Wochenschr. 73,79 (1961). 14. A. A. Forist, W. L. Miller and W. A. Crake, Proc. SOC.Exp. Bid. Med. 96,180(1957). 15. H. Spingler, Klin. Wochenschr. 35,533 (1957). 16. K. Sabih and K. Sabih, J. Pharm. Sci. 59,792(1970). 17. J. Shibasaki, Chem. Pharm. Bull. 21,1747 (1973). 18. L. F. Prescottand D. R. Redman, J. Pharmacol.24,713 (1972). 19. S. B. biatin and M. Rowland, .I. Pharm. Pharmacnl. 25, 186 (1973). 20. S.B. Matin and M. Rowland, Anal. Lett. 6,865(1973). 21. J. J. Burns, R. K. Rose, T. Chenkin, A. Goldman, A. Schulist and B. B. Brodie, J. Pharmacol. Exp. Ther. 109,346(1953).

22. R. Pulver, Schweiz. Med. Wochenschr. 80,308(1957). 23. B. Horman, Med. Exp. 1, 170 (1960). 24. G. R. Von Petten, H. Feng, R. J. Withey and H. F. Letten, J. Clin. Pharmacol. 11,177 (1971). 25. J. E. Wallace, J. Pharm. Sci. 57,2053 (1968). 26. R. Perego, E. Martinelli and P. C. Vanoni, J. Chromatgr. 54, 280 (1971). 27. R. B. Bruce, W. R. Maynard and L. K. Dunning, J. Pharm. Sci. 63,446(1974). 28. K. Sabih, Biomed. Mass Spectrom. 1,252(1974). 29. K. K. Midha, 1. J. McGilveray and C. J. Charette, J. Pharm. Sci. 63,1234(1974). 30. W. A. Garland, R. J. Weinkam and W. F. Trager, Chem. Instrum. 5,271 (1973). 31. L. H. Admunsen and L. S. Nelson, ./.Am. Chem. SOC.73,242 (1951). 32. H. A. Staab and W. Benz, Ann. Chem. 648,72(1961). 33. R. D. Kimbrough, J. Med. Chem. 15,409 (1972). 34. R. Adams and R. M. Kamm, Org. Synth. Soll. Vol. 7 250 (1941). 35. R. F. Nystrom and W. G. Brown, J. Am. Chem. SOC.69,2548 ( 1 9471. 36. 0.KammandC.S. Marvel,Org.Synth.Coll. Vol. 1250(1941). 37. J. Buchi, J. Ammann, R. Leiberherr and E. Eichenberger, Helv. Chim. Acta 36,75 (1953). 38. S.B. Matin and J. B. Knight, Biomed. Mass Spectrom 1,323 (1974). 39. Details of the animal experimental procedures and pharmacokinetic parameters will be published elsewhere.

40. Jake J. Thiesen, Ph.D. Dissertation, Department of Pharmacy, University of California, San Francisco, California

(1975).

Received 22 April 1976 @ Heyden &Son Ltd, 1977

BIOMEDICAL MASS SPECTROMETRY, VOL. 4, NO. 1, 1977 47