High-performance liquid chromatographic determination of N1-alkylnicotinamide in urine

ANALYTICAL

BIOCHEMISTRY

147, 108-I 13 (1985)

High-Performance Liquid Chromatographic Determination of A/ ‘-Alkylnicotinamide in Urine TERUHISA HIRAYAMA, KOICHI YOSHIDA, KAZUHIKO UDA,

MOTOSHI NOHARA, AND SHOZO FUKUI Kyoto

Pharmaceutical

University,

S Nakauchi-cho,

Misasagi,

Yamashina-ku,

Kyoto

607, Japan

Received July 18, 1984 A simple high-performance liquid chromatogmphic method has been developed for determining N’-alkylnicotinamides, including C&s alkyl derivatives, in urine. N’-Alkylnicotinamides were reacted with acetophenone in strong alkali medium at O’C and then formic acid was added. The reaction mixture was heated in acidic medium at above 93OC, and the fluorescent product, 1-alkyd-7-phenyl- I .5dihydro-5-oxo- I ,6-naphthyridine, was chromatographed by HPLC, using a Zorbax SCX-300 column with a mixed mobile phase of acetonitrile-0.04 M ammonium phosphate, monobasic. fV’-Alkylnicotinamides can be determined as 1,6-naphthyridine derivatives by a fluorometric detector at a level of 100 pg (signal/noise = 2). Recoveries of N’-alkylnicotinamides in urine were satisfactory. Interfering reaction products from NAD’ and NADP+ were clearly eliminated for determination of N’-alkylnicotinamides without pentyl derivatives. (0 I985 Academic Press, Inc.

KEY WORDS: high-performance liquid chromatography: N’-alkylnicotinamides; compounds; l-alkyl-7-phenyl-l,5-dihydro-5-oxo1.6~naphthyridine: urine.

Nicotimc acid and nicotinamide are involved in many important enzyme systems in the nucleotides NAD’ and NADP’. These nucleotides are coenzymes in numerous dehydrogenases and are involved in glycolysis, hydrogen transport, and ATP formation. N1-Methylnicotinamide is the major urinary excretion product of nicotinic acid and nicotinamide metabolism in mammals. N’Methylnicotinamide was first identified in human urine by Huff and Perlzweig (l), and its formation from nicotinamide decomposition in rat liver slices by Perlzweig et ul. (2). A soluble enzyme system catalyzing the methylation of nicotinamide was prepared by Cantoni (3,4), who subsequently identified ,S-adenosylmethionine as the methylating agent (5). In addition to reflecting nutritional status (6), the metabolites of nicotinamide have been implicated in the regulation of intracellular pyridine nucleotides (7,8), and attempts have been made to correlate the urinary 0003-2697185 $3.00 Copyright 0 I985 by Academic Press. Inc. All rights of reproduction in any form reerved.

fluorescent

levels of these substances with various disease states. It has been reported that burns (9) and the administration of chemical carcinogens (lo- 12) lead to an increased urinary output of N’-methylnicotinamide, whereas schizophrenia ( 13) depression ( 14), and mental retardation ( 1516) are associated with a decreased urinary output. In studies of the nicotinamide metabolites, fluorometric determination of N’-methylnicotinamide in urine has been reported by many workers. The fluorometric methods ( 1,17- 19) are modifications of the original procedure of Najjar and Wood (20) and are based upon conversion of the nonfluorescent N’-methylnicotinamide by treatment with alkali and n-butanol into a fluorescent derivative designated as FZ. FZ possesses a very low fluorescence intensity in aqueous solution. Huff and Perlzweig (2 1) further modified the method to form a highly fluorescent and stable compound by the use of acetone instead of n-butanol. Clark et ul. (22) obtained 108

CHROMATOGRAPHIC

DETERMINATION

a higher molar relative fluorescence by substituting acetophenone for acetone. Their modification of the Huff procedure results in a further, 20-fold increase in molar relative fluorescence. They noted that NADP+, NAD’, and NMN’ each yield derivatives with a molar fluorescence about l/25 that of N’-methylnicotinamide and the presence in serum of NAD+ results in falsely elevated Zv”-methylnicotinamide levels. Nelis et af. (23) reported the sensitive fluorometric procedure for the determination of aliphatic epoxide by the use of Clark method. They also noted that the sensitivity of the method under physiological conditions makes it attractive for the possible in situ detection of epoxide, formed from alkene precursors. So far, however, attempts have failed to demonstrate any alkylating activity in S9 mixtures incubated with styrene as a substrate. This is attributed to the exceptionally high background, caused by the presence of the required metabolic cofactor NADPH. Oxidized NADP’, being an alkylated nicotinamide itself, could contribute to the interfering fluorescence. The high-performance liquid chromatographic method described here for measurement of IV’-alkylnicotinamides (methyl, ethyl, propyl, butyl, and pentyl derivatives) in urine is based on the Clark et af. reaction procedure (22). Fluorescent derivatives of iv’-alkylnicotinamides were separated from interfering fluorescent substances, which are derived from NAD’ and NADP+, by cationic-exchanged resin column and detected by a fluorometric detector. MATERIALS

AND

OF

Apparatus. HPLC was carried out using a Model LC-2 chromatograph (Shimadzu Seisaku-she, Kyoto, Japan) with a column (15 X 0.46-cm i.d.) packed with Zorbax SCX300 (7-S pm, Du Pont) and an RF-530 fluorescence spectromonitor (Shimadzu Seisaku-sho), at ambient temperature. Synthesis of N’-alkylnicotinamide iodide. To 1.0 g of nicotinamide, dissolved in 5 ml of ethanol, was added 1.5 ml of methyl iodide, and the mixture was heated at 50°C for 5 h. After cooling, the resulting materials gave colorless needles which were recrystallized from methanol and acetone to furnished N’-methylnicotinamide iodide (2.0 g) of mp 2 1 I .5-2 12’C. The other N1-alkylnicotinamide iodides were also synthesized as described above and their melting points, appearance, and yields are listed in Table 1. Standard solution of N’-alkylnicotinamides. Four N’-alkylnicotinamide iodides (100 mg as free N’-alkylnicotinamide; N’-methyl, -propyl, -butyl, and -pentyl derivatives) were accurately weighed, and all standard samples were transferred into l-liter volumetric flasks, diluted to volume with 10m4 M aqueous hydrochloric acid, and stored at 4’C (stock solution). Standard solutions of 2-10 pg/ml Nr-alkylnicotinamides for calibration curves were prepared by dilution of the stock solution with lop4 M hydrochloric acid. N’-Ethylnicotinamide internal standard solution. N’-Ethylnicotinamide iodide ( 184 mg) was transferred into a l-liter volumetric flask, diluted to volume with 10e4 M hydrochloric acid, and stored at 4°C. N’-EthylnicTABLE

METHODS

Chemicals. Nicotinamide, acetophenone, formic acid, and acetonitrile (guaranteed grade) were purchased from Nakarai Chemicals Ltd., Kyoto, Japan. The acetophenone was redistilled and stored in an amber bottle. Reagent-grade absolute ethanol was freed from aldehydes and ketones by refluxing with aluminum and potassium hydroxide before distillation.

109

N’-ALKYLNICOTINAMIDE

I

N'-ALKYLNICOTINAMIDE R Methyl’ Ethyl“ PropyI Butyl Pentyl ’ Ref. (24).

w

PJ

211.5-212.0 205.5-206.5 185.0-186.0 159.0-160.5 126.0-127.0

IODIDE Crystal Needles Needles Needles Plates Plates

Yield 85.4 83.5 81.5 86.3 87.4

(%)

110

HIRAYAMA

otinamide internal standard solution (10 pg/ ml) was prepared by dilution of the above solution. Assay of N’-alkybucotinamides in urine. A half-milliliter of urine was transferred into a 1O-ml glass-stoppered centrifuge tube, and 0.5 ml of N’-ethylnicotinamide internal standard solution, 0.5 ml of 100 rnM acetophenone in ethanol, and 1.0 ml of 6 N sodium hydroxide were added under cooling with a mixture of ice and salt at 0°C. After the mixture cooled for 10 min at O’C, 0.5 ml of formic acid was added at same temperature and the mixture was allowed to stand for 15 min at 0°C. Then the mixture was heated in a boiling-water bath for 3 min at above 93V. Aider cooling to ambient temperature, ethanol was added to a volume (5.0 ml). Chromatography oj” 1-alkyl-7-phenyl-ISdihydro-5-oxo-1,6-naphthyridine. Four microliters of ethanolic reaction mixture was injected into the HPLC system by use of a microsyringe. HPLC conditions were as follows. The mobile phase was 0.04 M ammonium phosphate, monobasic-acetonitrile ( 1: 3) at a flow rate of 1.5 ml/min. The fluorescence detector was operated at an excitation wavelength of 382 nm and an emission wavelength of 440 nm. If the sample peak was not within points on the standard curves, the sample was diluted quantitatively with water as required for reexamination of the reaction procedure. Identification of individual peaks of the fluorescent N1-alkylnicotinamide derivatives was done by comparison of the retention times of the N1-ethylnicotinamide internal standards, as well as by spiking sample with individual standards. Calibration curves and calculation. Calibration curves from the five N’-alkylnicotinamide standard solutions containing 2, 4, 6, 8, and 10 pg/ml of these compounds were prepared, and ratios of N’-alkylnicotinamide peak height to internal standard peak height against micrograms of N’-alkylnicotinamide per milliliter were plotted. From the resulting ratios of N’-alkylnicotinamide peak height

ET

AL.

to internal standard peak height from the sample chromatogram, the concentrations of N1-alkylnicotinamides in sample were read from calibration curves. RESULTS

AND

DISCUSSION

Synthesis of N’-alkylnicotinamide iodide. N’-Alkylnicotinamide iodides (methyl, ethyl, propyl, butyl, and pentyl derivatives) were synthesized by the Tomita method (24). As can be seen in Table 1, these compounds were furnished in good yields. Optimal assay condition for jluorometric detection. From the results of fluorescence spectra of 1-alkyI-7-phenyl- 1,5-dihydro-5-oxo1,6-naphthyridines in 0.04 M (NHJHzPO,+CH$ZN (1:3) solution (Fig. 1), excitation and emission wavelengths were selected at 382 and 440 nm, respectively, for HPLC determination of naphthyridine derivatives. Optimal conditions for liquid chromatographic separation of 1,6-naphthyridine derivatives. The cationic-exchanged resin column was used for separation of five l-alkyl7-phenyl-l,5-dihydro-5-oxo-l,6-naphthyridines. When 0.04 M ammonium phosphate, monobasic aqueous solution was used as eluting solvent, the elution times of these naphthyridine derivatives from HPLC system were 1 h or more. Addition of a threefold amount of acetonitrile to the aqueous buffer solution was suitable for separation of each

50

4 ll

300(ml,

FIG. 1. Excitation (A) and fluorescence (B) spectra of 1-alkyl-7-phenyl1,5-dihydre5-oxo1,6-naphthyridines in 0.04 M ammonium phosphate, monobasic-CH3CN (1:3) solution. Emission spectrum was obtained with excitation at 382 nm and excitation spectrum was obtained with emission at 440 nm.

CHROMATOGRAPHIC

DETERMINATION

OF N’-ALKYLNICOTINAMIDE

111

derivative, and these N1-alkylnicotinamide derivatives were analyzed within 12 min at a flow rate of 1.5 ml/min. A mixed Nr-alkylnicotinamide standard containing 5 pg of methyl, propyl, butyl, and pentyl and the same amount of N’-ethylnicotinamide internal standard was reacted with acetophenone and assayed by cationic-exchanged resin HPLC with fluorescence detector. A chromatogram illustrating the separation of these compounds is shown in Fig. 2. The retention times were 3.2, 4.0, 5.6, 7.2, and 10.0 min for the pentyl, butyl, propyl,

FIG. 3. Linear relationship between amount of Nialkylnicotinamide standards and ratio of peak height (Ni-alkyinicotinamide/internal standard). Detector sensitivity, high: range, X 16.

FIG. 2. Chromatogram of N’-alkyhricotinamides as IaikyI-7-phenyl- 1,5-dihydro-S-oxo- 1,6-naphthyridines. Standard solution (0.5 ml) containing IO &ml of each N’-alkymicotinamide and N’-ethylnicotinamide internal standard solution (0.5 ml) were reacted, and 4 ~1 of reaction solution was injected. Analytical conditions are described under Materials and Methods. Detector sensitivity, high; range, X 16. a = Pentyl, b = butyi, c = propyl, d = internal standard (ethyl), e = methyl.

ethyl, and methyl derivatives, respectively. The limit of detection (peak height, signal/ noise = 2) using this system was approximately 100 pg for the methyl derivative. Modijication of the Clark method for measurement of N’-alkylnicotinamides in wine. Nakamura and Tamura (25) pointed out that the temperature should be kept under O°C in the first step of the reaction, since an amide group in N’-alkylnicotinamide was decomposed at temperatures higher than O’C with the evolution of ammonia. We followed their manner. Furthermore, when we used 0.1 ml each of 10 rnM acetophenone and 6 N sodium hydroxide and 0.5 ml of 7 M formic acid to 1 ml of sample solution (the dosage proposed by Clark et al.), the recoveries were not satisfactory. The results indicate that urinary

112

HIRAYAMA

components such as salt, urea, and other nitrogen-containing substances may interfere with the reaction procedure. Finally, optimal conditions for determination of N’-alkylnicotinamides in urine were decided upon as described under Materials and Methods, since interference was eliminated with an increase in the reagents. To determine the linearity of the fluorescent response as measured by peak height and calculated from the relative intensity against internal standard, standard curves were obtained. The results shown in Fig. 3 indicate linearity between 0.8 and 4 ng per injection of N’-alkylnicotinamide derivatives (r = 0.9968, 0.9957, 0.9994. and 0.9998 for N1-pentyl-, -butyl-, -propyl-, and -methylnicotinamide, respectively). Recoveries of N1alkylnicotinamides from urine (subject C) at two fortification levels are shown in Table 2. Recoveries from urine were satisfactory. Determination of N’-methylnicotinamide in urine. Table 3 shows the amounts of N’methylnicotinamide in single excreted urine from some normal adult males. N1-Methylnicotinamide in urine varied in concentrate TABLE

2

RECOVERESOFI'V"-ALKYLNICOTINAMIDES FROM URINE (SUBJECTC)“

iv”-Alkyl Methyl

Propyl

Butyl

Pentyl

Added (pg)

Recorder responseb

Found

@I

0.270 0.327 0.386

k 0.014 k 0.027 k 0.022

4.0 4.85 5.12

85 86

0 I 2

0.112 0.209

0 * 0.017 k 0.022

0 0.90 1.68

90 84

0 I 2

0.134 0.225

0 & 0.009 * 0.017

0 0.93 I .56

93 78

0 0.168 k 0.017 0.3 17 & 0.035

0 0.89 1.68

-

’ Detector sensitivity, high; range, X 16. b Ratio of peak height (IV’-aIkylmcotinamide/intemal standard (10 *g/ml)); average ? SD (n = 5).

AL. TABLE

3

DETERMINATION OFJP-METHYLNICOTINAMIDE IN LJRINE~ Subject A B C D E

Age

Recorder

response

56 40 29 24 21

3.44 0.594 0.270 0.884 0.702

k 0.16’ 2 0.032 k 0.014 + 0.056d z!z 0.023d

b

Concentration h/W 51.1 8.8 4.0 13.1 10.4

’ Detector sensitivity, high; range, X 16. b Ratio of peak height (iv’-methylnicotinamide/intemal standard (IO pg/ml)); average + SD (n = 5). c lo-fold dilution. d 2-fold dilution,

from 4.0 to 5 1.1 pg/ml urine. In the HPLC chromatograms (Fig. 4), peak e at Rt = 10.0 min, which was obtained from all urine samples, was identified with the reaction product of N1-methylnicotinamide. Urine from subject A contained a peak b at Rt = 4.0 min, which may be assumed to a 1,6naphthyridine derivative of N’-butylnicotinamide because of its eluting behavior, but now identification of this peak is under investigation. A small peak f in urine from subject B and some peaks at solvent front in all urine samples were observed, but the compounds in these peaks are still unidentified.

Recovery

b.4

0 1 2

0 I 2

ET

89 84

d b “k J!

-8J

FIG. 4. Chromatograms of IV’-alkylnicotinamides in human urine. (A) Subject A; (B) subject B; (C) subject D, (D) subject C. b = Butyl, d = internal standard, e = methyl, f and uk = unknown peaks. Detector sensitivity, high; range, X8.

CHROMATOGRAPHIC

DETERMINATION

ill- -

FIG. 5. Chromatograms of NADP+, N’-alkylnicotinamides, and mixture. (A) NADP+ (50 &ml); (C) N’aIkylnicotinamides (each 5 kg/ml); (B) mixture. a = Pentyl, b = butyl, c = propyl, d = internal standard, e = methyl, and g = derivative from NADP’.

OF N’-ALKYLNlCOTINAMIDE

ment of N’-alkylnicotinamides, at least the Cl to Cd derivatives. Therefore, the fact that the procedure is free from the interference of NAD’ and NADP+ suggests that the present method for determination of N’-alkylnicotinamides can be useful not only in urine but also in blood and tissues. REFERENCES 1. Huff. J. W., and Perlzweig, W. A. (1943) J. BioL Chem 150, 395-400. 2. Perlzweig, W. A., Bemheim, M. L. C., and Bemheim, F. (I 943) J. Biol. Chem. 150, 401-406. 3. Cantoni, G. L. (1951) J. Biol Chem. 189, 203-216. 4. Cantoni, G. L. (1951) J. Biol. Chem. 189, 745-754. 5. Cantoni, G. L. (1953) J. Biol Chem. 204,403-416. 6. Goldsmith, G. A. (1964) in Nutrition (Beaton, G. H., and McHenry, E. W., eds.), Vol. 2. pp. 109-206. Academic Press, New York. 7. Gholson, R. K. (1968) J. Viruminol. 14, 114-121. 8. Dietrich, L. S. (1971) Amer. .I. Clin. Nufr. 24, 800804. 9.

In the present study, N’-ethylnicotinamide was selected for the internal standard, since no peaks around the N’-ethyl derivative area were observed in chromatograms of normal urine samples. Influence of coexisting NAD’ and NADP+. Nelis et al. (23) reported that NADP+ and NADPH yield fluorescent derivatives similar to N’-methylnicotinamide and could contribute to the interfering fluorescence. Figure 5 shows the chromatograms of the reaction products of NADP+ (50 &ml) and N’alkylnicotinamides (each 5 pg/ml) and their cochromatogram. The reaction product of NADP’ was eluted at the position between the N’-pentyl and N’-butyl derivatives, and NADP’ yielded a derivative with a molar fluorescence about l/30-1/40 that of N’alkylnicotinamides under the present conditions. The ratio is lower than that reported by Nehs et al. in their nonseparated fluorometric determination. The reaction product of NAD+ also exhibited the same behavior as NADP’ during HPLC (data not shown). The above results suggest that NAD’ and NADP’ did not interfere with the measure-

113

IO.

I I. 12. 13. 14. 15.

16. 17.

Barlow, G. B., Sutton, J. L., and Wilkinson, A. W. (1977) Chin. Chim. Acta 15, 337-342. Chu. B. C. F., and Lawley, P. D. (1973) Chem. Bioi. Imeract. I, 65-13. Chu, B. C. F., and Lawley, P. D. (1975) Chem. Biol. Inferucf. 10, 333-338. Ohkubo, M., Shimizu, M., Kubo, A.,‘and Fujimura, S. (1977) Chem. Bio/. Interarf. 18, 101-I IO. Brown, F. C., White, J. B., Jr., and Kennedy, J. K. (1960) Amer. .J. Psychiat 117, 63-65. Cazzullo. C. L., Sacchetti, E., and Smeraldi, E. (1976) Psychoi. Med. 6, 265-270. Lis. E. W., Bijan, R.. and Lis. A. W. (1966) J. Cell Biof. 31, 15OA. Lis, E. W., Lis, A. W., and deHackbei1, K. F. (1970) Clin. Chem. 16, 714-721. Najjar, V. A. (1944) Bull. Johns Hopkins Hosp. 14, 392-399.

18. Coulson, R. A., Ellinger, P., and Holden, M. (1944) Biochem. J. 38, 150-154. 19. Hochberg, M., Melnick, D., and Oser, B. L. (1945) .I. Bioi. Chem. 158, 265-278. 20. Najjar, V. A., and Wood, R. W. (1940) Proc. Sot. Exp. Bioi. Med. 44, 386-390. 2 I. Huff, J. W., and Perlzweig, W. A. (1947) 1 Bio/. Chem. 167, 157-167. 22. Clark, B. R., Halpern, R. M., and Smith, R. A. (1957) Anal. Biochem. 68, 54-61. 23. Nelis, Hans, J. C. F., and Sinsheimer, J. E. (1981) Anal. Biochem. 115, 151-157. 24. Tomita, K. (195 I) Yakugaku Zasshi 71, 220-224, 25. Nakamura, H., and Tamura, Z. (1978) Awl. Chem. 50,2047-205 I.