You are on page 1of 6

Journal

Vol.

of Pharmaceutical

9, No.

Printed

9, pp. 699-704.

in Great

& Biomedical

0731-7085/91$3.00
+ 0.00
@ 1992 Pergamon Press plc

Analysis

1991

Britain

Determination of hippuric acid and O-, m- and
methylhippuric acids in urine by capillary gas
chromatography
HIROYUKI
MAKITA
Faculty

KATAOKA,*

of Pharmaceutical

KEIKO

Sciences,

MANABE,

Okayama

SHOKO

University,

NAKASE

Tsushima,

Okayama

p-

and MASAMI

700, Japan

Abstract:
A capillary gas chromatographic
(GC) method for the simultaneous
determination
of urinary hippuric acid
(HA) and o-, m- and p-methylhippuric
acids (MHAs), metabolites
of toluene and o-, m- and p-xylenes,
respectively,
is
described.
These metabolites
are converted
into their isopropyl
derivatives
by extractive
alkylation
with tetrahexylammonium
ion as extracting
agent and isopropyl bromide as alkylating reagent in benzene. The derivatives
are
analysed using a chromatograph
equipped with hydrogen flame ionization detector,
split injection system and DB-17
capillary column. Benzoylleucine
is used as an internal standard. The derivatives are well separated within 5 min and no
interfering
peaks are observed. The calibration
curves of HA and MHAs in the range l-50 pg are linear and sufficiently
reproducible
for quantitative
analysis. Urine can be analysed accurately and precisely by this method without prior cleanup of the sample.
Keywords:
Capillary CC; hippuric acid; o-, m- and p-methylhippuric acids; extractive isopropylation;
toluene and o-, m- and p-xylenes; urinary excretion of hippuric acid and methylhippuric acids.

Introduction
Toluene and xylene are organic solvents used
in industry as a replacement
for the carcinogen
benzene,
but these solvents also exhibit considerable
toxicity [l]. Estimation
of the exposure
levels or inhaled
amounts
of these
solvents is important
in monitoring
the health
of workers exposed to such solvents. However
it is difficult
to measure
the atmospheric
concentrations
of these solvents in the workplace. The major metabolites
of toluene and
xylene are hippuric
acid (HA) and methylhippuric acid (MHA),
respectively
[2]. Concentrations
of these metabolites
in urine of
workers
quantitatively
reflect
the solvent
vapour concentrations
in their workplace,
and
have been regarded as indices of exposure to
these solvents [3-91. There are many factories
where both toluene and xylene are used in a
mixture or simultaneously,
and xylene often
occurs as a contaminant
of the toluene used in
industry [lo]. Xylene used in industry consists
of a mixture of o-, m- and p-isomers,
and these
isomers are metabolized
to the corresponding
isomers of MHA, respectively.
If there are
differences
in the toxicity
of toluene
and
*Author

to whom correspondence

should

metabolites of

isomers of xylene, the separate determination
of HA and MHAs is necessary. Therefore,
the
simultaneous
determination
of urinary HA and
MHAs
concentrations
is useful
for occupational health control.
Many methods
for the determination
of
in
these
metabolites
urine
have
been
described,
such as thin-layer
chromatography
[II], gas chromatography
(GC) [12-161, isotachophoresis
[17, 181 and HPLC [l&26].
However,
there is no report describing
complete separation
of these metabolites,
except
some HPLC methods [23-261. These HPLC
methods can be performed
without pretreatment of sample, however the contamination
of
columns with other urinary components
causes
deterioration
of column performance.
In the present work, a simple and reliable
method for the simultaneous
determination
of
HA and o-, m- and p-MHAs by capillary GC
after extractive alkylation is described.

Experimental
Reagents

HA and benzoylleucine
as the internal standard
(i.s.)
were purchased
from Nakarai

be addressed.

699

the reaction mixture was transferred into a separatory funnel with 15 ml of distilled water and this solution was extracted with 20 ml of chloroform.1 of this solution injected into the GC. the supernatant was evaporated at 5O”C. m. and the residue reconstituted in methanol to give a 10% (w/v) solution. 0. The peak heights of HA. The residue was dissolved in 0. ethyl acetate.w. Tetrahexylammonium hydroxide (THA-OH) was prepared as follows: 5 g of tetrahexylammonium iodide (Nakarai Chemicals) were dissolved in 30 ml of 80% methanol and to this solution was added 8 g of silver oxide (Nakarai Chemicals). calculated for the construction of the calibration curves.p. p. The crystalline residue was recrystallized from n-hexane to yield long colourless needles. Preparation of reference compound A reference sample of isopropyl hippurate was prepared as follows: 100 mg of hippuric acid was dissolved in 2 ml of 15% BFsisopropanol (Tokyo Kasei Kogyo) and the mixture was heated for 5 min in a boiling water bath.5 ml kg-‘.83. The output of the FID was recorded on a HewlettPackard 3392A recording integrator. respectively. After washing with 20 ml of water. was 700 Chemicals (Kyoto. After centrifugation.05 ml ml-’ each of toluene and o-.and p-MHAs) were purchased from Tokyo Kasei Kogyo (Tokyo. 64.1 ml of 0. m. Japan). All other chemicals were of analytical grade.0.). 70 eV. After centrifugation. Immediately after addition of 0. Urine samples Normal human urine samples (24 h) were obtained from laboratory personnel. Urine samples (24 h) were collected under toluene and kept frozen if not analysed immediately.25 km film thickness) coated with crosslinked DB-17 (equivalent to OV-17) (J & W. for C12H. 280 and 23o”C.0 and 40 ml min-‘. Japan). The data for elemental analysis were as follows. Folson.33. 6.s03N: C.14. The temperatures of injector. After cooling. identical to that used for GC analysis. USA) was used. N. The combined organic layers were then heated at 90°C for 10 min and washed with 2 ml of a saturated silver sulphate by shaking for 30 s. solution and 0. ionization voltage. ion-source temperature. Found: C.2 ml of isopropyl bromide and 2 ml of benzene. A fusedsilica capillary column (30 m x 0. After addition of 0.. 6. 6. and a 1 p.s. N. Calc.79. 250°C. A standard solution of HA and MHAs was prepared so as to contain 0. The helium carrier gas and nitrogen make-up gas flowrates were 1. the organic layer was transferred to a round-bottomed flask and the solvent evaporated. detector and column were 250.25 mm i.0 ml min-‘. helium flow rate. H. the total volume was made up to 1 ml with distilled water. 84-85°C.HIROYUKI KATAOKA et al. 6. Results and Discussion The extractive alkylation technique [27] introduced by Ehrsson is simple and selective . were measured and the peak height ratios against the i. m.s.d. The rats were treated with solvent mixture containing 0. 65. CA. respectively. The mixture was shaken for 1 h at room temperature. MHA isomers (o-. Gas chromatography-mass spectrometry The Hewlett-Packard 5890A gas chromatograph was operated in connection with HP 5970B mass detector under the following conditions: GC column. H. the mixture was shaken with a shaker set at 300 rpm (up and down) for 2 min at room temperature..83. 1. Analytical derivatization procedure Urine (50 ~1) was pipetted into a 10 ml Pyrex glass tube with a PTFE-lined screw-cap.1 mg ml-’ of each compound in distilled water.and p-xylenes in mineral oil (0. Isopropyl bromide was obtained from Nakarai Chemicals. Each rat was placed in a metabolic cage which allowed separate collection of urine and faeces. the organic layer was transferred to another tube and evaporated to dryness at 90°C under a stream of dry air.22.1 mg ml-’ i. MHAs and the i. After centrifugation at 3000 rpm for 1 min. Split ratio was 2O:l.2 ml of 10% THA-OH. the organic layer was transferred to another tube and the aqueous layer was reextracted with 1 ml of benzene by shaking for 30 s.1 ml of Gas chromatography GC analyses were carried out by means of a Hewlett-Packard 5890A gas chromatograph equipped with a hydrogen flame ionization detector and a split injection system. b. Control rats received the same dose of mineral oil without solvent. Male Wistar rats weighing 350-450 g were used in experiments.s.

105 (C6HS--CO) and 119 (CHR---ChHS-CO). II = IS).CAPILLARY GC OF HIPPURIC 701 ACIDS for the derivatization of acidic compounds prior to GC analysis.0503X + 0. and the regression lines of HA and o. y = 0. acids.0177 (r = 0. Typical chromatogram of HA and MHAs was shown in Fig. 3(A). The derivative preparation took within 30 min. In each case. the contents of HA and MHAs in urines of human and rat treated with toluene and xylene isomers were analysed. Key: 0 = . As shown in Fig. The elemental analysis data for HA derivative agreed with the theoretical values calculated for the expected structure.and p-MHAs were y = 0.9995. Other ion peaks useful for structure elucidation were as follows: M+ -59 [OCH(CH. y = 0. various amounts of HA and MHAs ranging from 1 to 50 pg were derivatized in a mixture. Of several alkylammonium ions tested.0017 (r = 0. and aliquots representing 0. 291. In the present work. Each compound gave a single and symmetrical peak on GC with a DB-17 capillary column. The structures of the derivatives of HA and MHAs were confirmed by gas chromatography-mass spectrometry (GC-MS) and by elemental analysis. in which acid compound is first extracted as an ion-pair with alkylammonium cation into organic layer. alkylation of these compounds with isopropyl bromide was accomplished within 10 min at 90°C.25 kg.0043 (r = 0. where y is the peak height ratio and x is the amount (pg) of each compound. m.9997.X + 0.0639~ + 0. The derivatives were found to be very stable under normal laboratory conditions. The minimum detectable amounts of HA and MHAs to give a signal three times that of the average noise level were found to be about 1 ng. In order to demonstrate the applicability of the method to urinary samples.)J and m/e 77 (C. As shown in Fig. n = 18). THAOH proved to be the most satisfactory counterion for the rapid and quantitative extraction of HA and MHAs.0417~ .).H. A = o-methylhippuric (B) on the isopropylation acid. However the problem was solved by washing with aqueous silver sulphate solution [30]. benzene was selected as the most suitable solvent for ion-pair extraction.9994. The extractive alkylation process can be described as a two step reaction. M+ -87 [COOCH(CH. 1. II = 18).9% (n = 3) by comparison with the synthetic reference derivative. In order to test the linearity of the calibration curve.01-0.0. a linear relationship was obtained.]. The mean derivatization yield of HA throughout the procedure established above was determined to be 96. THA bromide formed as a by-product in the reaction and its degradation products formed in the injector caused significant peak distortion. n = 18) and y = 0. n = m-methylhippuric Temperature (‘C) of hippuric acid and methylhippuric acid. As shown in Fig.). a molecular ion peak (M+) was observed for each of the derivatives. x = p-methylhippuric acid. A chromatographic run was completed within 5 min. 3(B-F) HA and Time (min) Figure 1 Effect of time (A) and temperature hippuric acid.0535. respectively.5 kg of each compound were injected. and it has been applied to analysis of a great number of organic acids [28. 91 (CH3-ChHS). and next alkylated with a nucleophilic reagent such as alkyl halide in the organic layer.9999. 2. Each peak represents ca 0.0013 (r = 0. and several samples could be treated simultaneously.

6 97.483 0.77 134 119 OS’ 100 162 L he221 I 200 150 250 .020 97.010 0.012 0.0 97.2 95.8 96.004 0.010 NDt ND ND 0.486 0. 119 1D) Table 1 Recoveries rat urines 176 1 150 100 of (A) hippuric acid. @z 200 250 m/z 119 (cl > . M’:235 176 I 200 150 100 of hippuric .480 0.008 96.702 HIROYUKI loo KATAOKA PI al.E 100 a m 0. 1007 146 I: M’-235 1 200 250 m/z 1 Figure 2 CC-MS spectra obtained for the isopropyl methylhippuric acid and (D) p-methylhippuric derivatives acid.018 k 0.015 rf.471 0. 250 m/z (B) o-methylhuppuric acids added to human (mg ml-‘)* Compound added (0. c . acid and o-.6 Rat urine HA o-MHA m-MHA p-MHA ND ND ND ND 0.317 f 0.485 + 0.479 0.2 96.X 94.488 k t * * 0.801 0.0. 105 (A) 50.0 *Mean f SD (n = 3) tNot detectable. m. (C) m- .5 mg ml-‘) Non-addition Addition Human urine HA o-MHA m-MHA p-MHA 0.018 f 0. and Recovery (%) acid._ M+=235 I . . 176 OuI-rJL al m/z 50 91 -l&4& 0 77 148 105 1.and p-methylhippuric Amount found .

D. Occup. T. J. . MHAs were completely separated from urine constituents. Flek. Okutani. Blaton. Med. Bieniek. Ind. Brit. Y. Am. . H. 34501 234501 . 0 1 234501 703 ACIDS . Takeuchi. 114-121 (1978). T. J. M. Stokholm. J. m. and urinary HA excretions were 323-759 mg 24 h-’ (n = 5). extractive alkylation and subsequent capillary GC analysis is a simple and reliable method for the simultaneous determination of HA and MHAs in urine. (B) normal human urine. 330-334 (1978).1% (n = 3). Takatsuka. Matsumoto. On the other hand. 244246 (1969). the References Stand. Takatsuka. J. ‘Arch. J. Ind. The present study indicates that this method becomes a useful tool for occupational health control. Work Environ.’ Int.and p-methylhippuric acids). . Ueno. Tomokuni and Y. Ind. Gustafsson.P. Health 37. . Ikeda and H. Japan J. MHAs were not detected in the normal human urine [Fig. I 012 . V. m. In conclusion. Stand. b. .I. 129-134 (1967). [91 K. N. J. Ind. Health 24. 5 . Tomokuni and Y. 382-385 (1971). (A) Standard solution containing 25 ug each of four acids (hippuric acid and o. K. and their relative standard deviations were 0. 219-232 (1976). Environ. 30. Iwata. Med. Brit. Fukaya. 2nd.0. (min) 2 3 4 5 (min) Figure 3 Capillary gas chromatograms of hippuric acid and methylhippuric acids obtained from standard solution and urine samples. J. 0. (41 M. 3 = benzoylleucine (i. Ind. Lauwerys. Buchet and R. . Peaks: 1 = hippuric acid. Hisanaga. 35. 305-312 (1982). 28. 0 1 2345 . Pfaffli and V. D. Riihimaki. Mizuno. PI T. Engtrom. 28. GCIMS analysis of the peaks of HA. P. 131 L.5-4. Sedivec and J.s. Environ. Ogata. [101 Y. 27. 5 = p-methylhippuric acid. Ind. indicating that this method is accurate and precise. the recovery rates of HA and MHAs added to urine samples were 94.s. [‘I K. Ohtsuji. CC conditions: see Experimental. (C) sample B plus for acid standards.. Med. .and pxylenes in mineral oil. Brit. PI M. Brit. Y. 125-128 (1973). Health 4. Husman. 3(F)].).R. Cohr and J. (D) control rat urine.M. Ass. from urine samples confirmed that each peak was almost uniform. 26. K. but the solvent-treated rats excreted the corresponding metabolites of each solvent [Fig. As shown in Table 1. 3(B)]. Ogata. Seki and S. A. not only in cases of exposure to one solvent but also in cases of exposure to a mixture of solvents. 43-50 (1970). Health 5. Pagnott and L. J. Work PI R. [121 J. Work Environ. van Kerckhoven.6%.2-97. 2 = omethylhippuric acid. Wilczok and G. 4 = m-methylhippuric acid. Stand. Acknowledgements The authors wish to thank Yamashita of Yokogawa Electric Corporation for capillary CC-MS measurements. . 2 3 4 .H. J.5 ml kg-‘. K. 71-90 (1979). 100. K. Peeters. [61 M. Goto. Brit. (E) sample D plus four acid standards: (F) solvent-treated rat urine (the rat was treated with solvent mixture containing toluene and o-. . 1-18 (1980). MHAs and i. I . Toftgard and J.). Lieberman. Med. Vandamme and 1111G. Ono.w. Health 6. 3(D)]. Chromatogr. Med. 215-217 (1974). the control rat excreted neither HA nor MHAs [Fig.A. p. Hyg. [71 V. M.CAPILLARY CC OF HIPPURIC 1 . H.

fnf. Heulth 50. H. PI R. PI G. 182-188 (1983). Chambon and N. Sekiya. 275-311 (1981). 346-349 (1981). Chromafogr. 145. Husman and J. Chem. U. 34. :I. Inr. Environ.L. Environ. Acta Pha>m. Taeuchi. Chromatogr. HIROYUKI KATAOKA eial. 113-I 18 (1971). J.C. Med. Brodeur and G. Niinuma. Guisiani. J. Chambon. Ehrsson.D. J. Med. Occup.Environ. Y. &e&a 8. R. Arch. 281-286 (1978). Morin. Lowrv. Baldesten. P. Arch Occup. Ind. Roseboom and A. Sugihara Environ. Palagi. 220. 469-476 (1977). Anal. 9.R. Hulshoff and A. Imamura. Ar&. Sollenberg. M. (191 M. Loi. 25-31 (1982). J. J. 305-309 (1977) L1. Caperos and J. S. Oeata and T. (231_ T. Engstrom. Sakai. Plaa. Chromafogr. 229-233 (1977). matogr. and M. 121-12971986). Kira. H. H. Health 41. A. R. Yanagihara _ 1241 . Chromafogr. 65-74 (1979). 313-316 (1989). Baschieri. Dazzi. Paggiaro. Chromutoar. and K. Ehrsson. Health 39. J. [I71 J. Bichet. Poggi. 132. F.L. Sugihara and S. J. A. 210. J. 419-423 (1985). Matsui and T. revised manuscript received 23 July 19911 . 46. OCCUD. [I41 S. R. Arch. 276. Sollenberg and A. Heal& 58. Ogata. Kasao and S. M. Siclari and L. Chromatogr. fm. Environ. 13. 343. J. Kira.M.K. 34. Ini. Occup. Hulshoff.704 1131K. . C. Matsui. 231-236 (1978). Chromatogr.G. c‘hro(211 H. [181 J. Rantanen. Forth. H. Arch. Int.I. P. F. Int. Occup. M. Anal. 922-924 (1974). Brit. Brit. Phipps. 199-206 (1977). J. 496. Ushio. Ogata. [I’31 M. Fernandez. Health 36. Tardif. 1 [25] [26] (271 [2X] [29] [30] [Received for review 8 April 1991. Stringer and L.51J. 189193 (1989). 153-160 (1976). 173. Toxicol. K.