Gas chromatography-mass spectrometry analysis of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine in urine and feces.

A method has been developed to measure levels of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) excreted in urine and feces. The method involves organic solvent extraction, derivatization to form electron-capturing bis-pentafluorobenzyl derivatives, and analysis by gas chromatography-negative ion chemical ionization mass spectrometry using a deuterium-labeled internal standard. The method can detect PhIP at levels of less than 1 ng/g in rat urine (5 ng/24 hr) and 5 ng/g (wet weight) in rat feces (50 ng/24 hr). Sprague-Dawley rats given a single 50 micrograms dose of PhIP by gavage excreted an average of 0.6% of the dose in the urine and 25% of the dose in the feces as unchanged PhIP, in the first 4 days after treatment. To make this method applicable for the analyses of biological fluids of PhIP-exposed human subjects, it is now being improved by using immunoaffinity chromatography.


Introduction
Epidemiological studies suggest a role for diet in the etiology of many human cancers. Among the heterocyclic amine food mutagens thus far isolated, 2-amino-l-methyl-6-phenylimidazo- [4,5-b]pyridine (PhIP) has been identified in fried beef and fish at levels up to 15 ag/kg (1). The compound is mutagenic in bacteria (1) and has been shown to cause lymphomas when administered in the diet to mice (2) and colon and mammary carcinomas in rats (3). To assess possible human exposure to this compound, we have developed a gas chromatography-mass spectrometry (GC-MS) method for measuring PhIP in urine and feces. This study reports on the application and validation ofthis method in rats.

Internal Standard
A d5-labeled PhIP internal standard was prepared by reaction ofd5-phenylalanine and creatinine, with subsequent purification by high-pressure liquid chromatography. Methane negative ion chemical ionization mass spectra of the bis-pentafluorobenzyl ' (BPFB) derivatives of d5-labeled and unlabeled PhIP are shown in Figure 1.

Treatment of Rats with PhIP
Ten 7-week-old male Sprague-Dawley rats (average weight 210 g) were divided into tvo treatment groups and placed into individual metabolic cages. Rats were given free access to tap water, but were not fed 24 hr before treatment and for the first 48 hr after treatment. One group of five rats received a single dose, by gavage, of50 Ag PhIP (.-250 Ag/kg body weight) dissolved in a mixture containing 0.5 mL H20, 15 AL dimethyl sulfoxide, and 0.2 AL 0.1 N HCl to help solubility. The control group of five animals received only the solvent mixture. Urine and feces were collected at 24, 48, 72, and 96 hr and stored at -80°C.

Preparation of Urine Samples
PhIP was extracted from urine using a modification of a procedure described by Murray et al. (4). After thawing, aliquots representing 0.5 % (for 24-, 48-, and 72-hr samples) or 2 % (for 96-hr samples) of the 24-hr urine volumes were dissolved in 5 mL water. The pH ofeach sample was adjusted to 9-10 with 1 mL of 1 M Na2CO3 and 500 pg of d5-PhIP in 20 AL methanol was added as internal standard.

Preparation of Fecal Samples
Fecalsampleswerelyophilized, groundtoapowder, andmixed to homogeneity. After weighing, 0.2% of the 24-hr and 48-hr samples and 0.8% of the 72-hr and 96-hr samples were dis- After extraction into ethyl acetate (2 x 5 mL), samples were centrifuged, and the combined organic phases were extracted with 2 x 5 mLof0.l N HCI. PhIP was recovered from the acidic solution by addition of 1.5 mL of 1 M Na2CO3 and extraction with ethyl acetate (2 x 10 mL). After centrifugation and decantation to remove any remaining aqueous phase, the samples were taken to dryness at ambient temperature in a Speed-Vac centrifugal evaporator.

Formation of BPFB Derivatives
To obtain wvlatile, electron-capturing derivatives, the residues obtained after extraction from urine or feces were dissolved in a small quantity of methanol and transferred to 2-mL flamesealable vials and taken to dryness in a centrifugal evaporator. After adding 20 yL of a 30% (w/v) solution of pentafluorobenzyl bromide in ethyl acetate and 20 ltL of diisopropylamine, the solutions were vortex mixed. After flame-sealing, the vials were heated for 1 hr at 30°(. After reduction to dryness under a stream of nitrogen at 30°C, the residue was dissolved in 200 AL of 0.1 N HC1, mixed, and extracted with 2 x 750 ,uL hexane. The hexane was discarded. After adding 100 ,uL of 1 M Na2CO3 to bring the pH to 9-10, the aqueous phase was extracted with 2 x 500 AL ethyl acetate. The solution was reduced to dryness by a centrifugal evaporator in a small glass vial. The residue was dissolved in 20 jtL (for urine samples) or 200 4L (for feces samples) of ethyl acetate for analysis by GC-MS.
Gas Chromatography-Mass Spectrometry GC-MS analyses were carried out on a Hewlett Packard (HP) 5890 gas chromatograph coupled through a heated interface (310°C) to a HP 5988A mass spectrometer. For urinary extracts, chromatographic separation was achieved on a HP-i fused silica capillary column (12 m x 0.2 mm id). After splitless injection at 1800C, the column oven was heated to 290°C at 30°C/min and then at 100C/min to 3200C. For fecal extracts, a 25 m HP-1 column was used. After splidess injection at 2000(, the column was heated at 300C/min to 3200(, where it was held for 8 min. The mass spectrometer was operated in the negative ion chemical ionization mode with a methane source pressure ofabout 1 torr and source temperature of2500(. Quantification was by selected ion monitoring of the (M-PFB) -ions of the BPFB derivatives of PhIP (m/z 403) and ds-PhIP (m/z 408) as shown in Figure 2 for a typical sample of rat urine. Figure 3 shows the urinary excretion of PhIP for each of the five rats treated with a single 50-pg oral dose of PhIP. An average of0.6% (0.4-1.3 %) of the total dose was excreted as unchanged PhIP in the first 4 days after treatment. Of this amount, an average of 84% was excreted in the first 24 hr after treatment. Figure 4 shows the fecal excretion ofPhIP for each ofthe five rats. An average of25 % (19-32) ofthe total dose was excreted as unchanged PhIP in the first 4 days after treatment. Of this amount, an average of 84% was excreted in the first 24 hr after treatment.

Discussion
A method has been developed that can detect PhIP at levels of less than 5 ng/24 hr (1 ng/g) in rat urine and 50 ng/24 hr (5 ng/g wet weight) in rat feces. It should-be possible to monitor PhIP excretion after a single dose ofabout 200 ng PhIP per rat. Urinary and fecal extracts were quite clean and the amounts of urine or feces extracted could have been augmented to increase sensitivity. This was not necessary in this case, however, as PhIP levels were relatively high.
Rats excreted an average of 0.6% of an oral dose of 50 yg of PhIP in the urine and an average of 25 % in the feces. These results confirm an earlier report (S) where about 2% of a 600-ttg dose in unstarved Fischer rats was excreted unchanged in the first 24 hr in the urine and 51% unchanged in the feces.
Although the results ofthis study are promising, the detection limit ofthis method may not be sufficient to allow routine quantification of PhIP at the levels expected in humans. Work is in progress to improve the clean-up step using immunoaffinity chromatography and to apply the method to matrices other than urine and feces.