15N investigation into the effect of a pollutant on the nitrogen metabolism of Tetrahymena pyriformis as a model for environmental medical research.

A pilot study was performed to examine the potential of stable isotope techniques for monitoring the impact of a harmful substance on the cellular nitrogen metabolism in the ciliate species Tetrahymena pyriformis. After identical cultivation periods of control cells and toluene-exposed cells in a defined culture medium enriched with [guanidino-15N2]l-arginine, a number of nitrogen-containing pools were analyzed: 1) quantity and 15N abundance of ammonia as the end product of nitrogen metabolism in the system; 2) pattern and 15N abundances of the protein-bound amino acids in the cells; 3) pattern and 15N abundances of free amino acids in the cells; and 4) pattern and 15N abundances of the amino acids in the culture medium. In addition to 15N emission spectrometry, a new gas chromatography/combustion interface-isotope ratio mass spectrometry/mass spectrometry analytical system was used. The production and 15N content of ammonia were higher in the toluene-exposed system by 30% and 43%, respectively, indicating higher deamination rates and greater arginine consumption. The toluene-exposed cells exhibited increased 15N abundances of protein-bound amino acids in alanine, aspartic acid, glutamic acid, and tyrosine. Furthermore, structural analyses revealed the presence of N[Omega]-acetylarginine and pyrrolidonecarboxylic acid--compounds that had not previously been detected in Tetrahymena pyriformis. Differences in the 15N-enrichment of free amino acids were also evident. This new effect-monitoring system designed to investigate the impact of a pollutant on protein metabolism by using a stable isotope-labeled cell culture is a powerful tool for environmental medical research.

organic compounds. Envirn Heal Prpect 106: 493-497 (1998). [Online 9July 1998] hap://ehpnnl.niehs.ni.gov/docs/1998/106p493497ar/absrar html The impact ofchemical, physical, and biological environmental factors on human health must be addressed by scientific risk assessment, a process that requires knowledge of exposure, dose response, and mechanisms. Suitable cell models are frequently used for basic research into effect monitoring at the cellular level. The cell system used for this study was the ciliated protozoan Tetrahymena pyni-Jormis. Tetrahymena, a typical eukaryotic cell, and mammalian cells have much in common in regard to their nutritional requirements, cell compartmentation, metabolic pathways, and sensitivity to cytotoxic substances (1)(2)(3)(4)(5).
Cell culture has been used frequently as a test system for toxicity assessment in pharmacology (4,6,7) and ecotoxicology (8,9). By observing end points such as growth impairment (10), modification of motility (7,11), and inner and outer morphology (6,12,13), only the end result of changes in metabolism caused by harmful substances can be pinpointed. The purpose of this investigation was to identify metabolic changes in normal cell metabolism (biological effect monitoring), namely, at the level of amino acid and protein metabolism. The Tetrahymena cells generally require 11 essential amino acids, including arginine (14). Because these eukaryotes lack the urea cycle present in mammals, the final product of nitrogen metabolism is ammonia (15,16).
The harmful substances we are interested in are volatile organic compounds (VOCs) with toluene as the representative compound. As air is the main exposure route for most VOCs, effects on target cells such as alveolar macrophages (1]) and bovine bronchioepithelial cells (18) have already been studied. Exposure often occurs repeatedly over a longer period. This may result in accumulation ofVOCs in fatty tissue and, after mobilization, in higher blood concentrations (1), which lead to increased exposure of other organs such as the liver. Although various adverse effects of toluene in vivo are described in the literature , the effect of toluene on amino acid metabolism in humans has not yet been satisfactorily darified (23).
To date only a few articles have been published on 15N measurements of amino acids by gas chromatography/combustion interface-isotope ratio mass spectrometry/ mass spectrometry (GC/C-IRMS/MS) coupling (24)(25)(26)(27)(28). None of these studies was designed to observe the use of an essential amino acid under the impact of a harmful substance using a 15N-labeled tracer. Therefore, the aim of our study was to determine changes in the use of the 5N-labeled amino acid L-arginine by Tetrahymena pyri-Jbrmis under the impact of toluene using 15N emission spectrometry and a novel GC/C-IRMS/MS coupling for 15N/I4N analysis. We intended to analyze the quantity and 15N concentrations of the end product ammonia, as well as 15N distribution in amino acids and related metabolites. Ultimately, such a system characterizes the impact of specific environmental pollutants on specific steps of the protein metabolism and serves as an early-effect monitoring method.

Materials and Methods
Cultivation. Tetrahymenapyriformis strain W (1630/1W) was obtained from the Culture Collection of Algae and Protozoa (CCAP; Ambleside, UK). The cells were grown in a chemically defined medium with salts, trace salts, and vitamins in accordance with Szablewski et al. (29). The medium contained the 11 essential amino acids (6), albeit at a fourfold concentration to ensure short generation times and high cell numbers to meet the requirements for isotope analyses. Glucose was added as a carbon source after separate autodaving to a final concentration of 1%. Cells were transferred twice weekly to fresh medium, taking into account a surfaceto-volume ratio of 2:1. The generation time was determined to be 7.5 hr. Stock cultures were maintained in conformity with CCAP information.
In the '5N-labeling experiments, we used 24-hr cultures (late exponential growth phase). The labeled defined medium contained 40 mg/1 [guanidino-15N2]L-arginine ( Fig. 1) and 1,768 mg/1 unlabeled arginine. The 15N enrichment of arginine, calculated by isotope dilution formula, is 2.4 at%. The unit at% expresses the relative 15N-abundance in a sample, which is the ratio of the amount of 15N (mol)/total amount of N (mol). The resulting 15N enrichment of freeze-dried medium was determined to be 0.765 at% by means of elemental analyzeremission spectrometric coupling (FAN Fischer Analysen Instrumente, Leipzig, Germany). The cells were grown in 500-ml screw-capped glass bottles containing a total volume of 30 ml medium, glucose solution, and inoculated cell suspension. Toluene concentration in toluene-exposed cultures was calculated as 989 pM, but the detected exposure concentration was lower. By means of headspace GC (Perkin Elmer, Ueberlingen, Germany), we detected 270 1iM toluene at the beginning of the experiment and 150 pM after 24 hr of cultivation in the nutritional medium. The bottles were rotated. After the 24 hr cultivation period, the cultures were centrifuged at 1,000g at 4°C for 10 min. The growth temperature was 28°C. Cells were counted with a Coulter Counter (model ZM; Coulter Electronics GmbH, Krefeld, Germany). Depending on cell number, the sample was diluted with Isoton II (Coulter Euro Diagnostics GmbH, Krefeld, Germany).
Chemicals. All chemicals were of analytical grade and were obtained from Serva (Heidelberg, Germany) and Merck (Darmstadt, Germany). Trifluoroacetic anhydride (TFAA) was purchased from Merck and [guanidino-15N2]L-arginine (95 at%) from Berlin Chemie (Germany). Determination oftotal ammonia and its 15N enrichment. Ammonia was isolated from the supernatant by microdiffusion (30Q. To perform quantitation, the isolated ammonium sulfate was titrated with 0.01 N HCl and evaporated to dryness. An aliquot of 15 pg nitrogen dissolved in 30 pl doubly distilled water was converted to N2 using the Rittenberg procedure for emission spectrometric 15N isotope analysis (31) (NOI-6PC; FAN Fischer Analysen Instrumente).
Isolation of nitrogen-containing pools. Cells were harvested by centrifugation for 10 min at 800g at 4°C after the addition of ice-cold 10% perchloric acid (PCA) to a final concentration of 1% (v/v), washed twice with cold 0.6% PCA (v/v) and with doubly distilled water, and frozen at -800C. Then frozen cells and frozen medium were freeze-dried (Lyovac GT2; FINNAQUA, Huerth, Germany). For separation into protein and nonprotein nitrogen (NPN), the freeze-dried cells were resuspended in 20 ml of 10% trichloracetic acid (TCA; w/v), heated in a boiling water bath for 10 min, and allowed to stand overnight at 40C. The sample was centrifuged for 30 min at 10,000g, and the sediment (total protein)  [4 = cell enumeration at the beginning of the experiment (to); Zx = cell enumeration at t, + x hours (10)]; nonexposed, cultivation without toluene; exposed, cultivation in presence of toluene; (B) total ammonia formation (106 cells/mI); (C) 15N ammonia formation (106 cells/mI); (D) ratios of 15N ammonia/total ammonia (values of nonexposed and exposed cells differ significantly according to Wilcoxon test, 0.01> a >0.001).

494
Volume 106, Number E was washed three times with doubly distilled water and freeze dried.
The supernatant (NPN) was filtered through a glass-fiber filter, and the clear solution was cleaned up by cation exchange (DOWEX 50Wx8, 60-170 mesh, Merck, Germany) (34). The freeze-dried medium was dissolved in 0.1 N HCI and purified in the same way. The eluates containing the amino acids were evaporated on a rotary evaporator and dried first under a gentle stream ofhelium in a water bath at 600C and then with methylene chloride. The NPN fraction yielded 1-2 mg as dry weight for analysis. The proteins (10 mg) had to undergo acidic hydrolysis (34 before further preparation.
Derivatization. The isolated amino acids were esterified with acidic isopropanol for 1 hr at 1 100C, and the residue was dried after evaporation of the solvent with CH2Cl2.
The residue was derivatized with trifluoracetic anhydride overnight at room temperature using a modified procedure (33). It had been previously demonstrated that this technique does not cause any isotope fractionation (D. Hofmann, personal communication), enabling the isotopic measurement of arginine. Finally, the sample was concentrated under a gentle stream of helium in an ice bath. The amides glutamine and asparagine were hydrolyzed under the acidic conditions of derivatization to glutamic and aspartic acid. The cooling of the autosampler to 5°C improved the reproducibility of the isotopic analysis of the derivatives.
Isotope and organic mass spectromety of amino acidA. We analyzed the nitrogen-specific isotopes (15N/I4N)  to a standard ratio, which is 3676.5 ± 8.1 for air, in terms of parts per thousand (per mil, 0/0o). The standard generally used is atmospheric air, defined as /oo. GC/C-IRMSIMS measurements. The amino acid derivatives were separated on a capillary GC (HP 5890; column: SGE BPX 5, 50 m X 0.32 mm x 0.5 pm). The following temperature program was used: 50°C, held 1 min; ramp in 10 min to 100°C, held 10 min; ramp in 3 min to 175°C, held 5 min; ramp in 7 min to 300°C, held 15 min; injector: 280°C. The GC is connected to a combustion interface (type II, Finnigan MAT; oxidation reactor 980°C, reduction reactor 600°C) coupled to an IRMS (MAT 252) for isotopic analysis. Another part of the GC eluate was applied to an organic mass spectrometer (GCQ, MAT 252, Finnigan MAT).

Results
Both total ammonia and its 15N content were enhanced in the toluene-exposed cultures by 30% and 43%, respectively (Fig.  2). The amounts reflect a cultivation period of 24 hr. All detectable amino acids of the cell proteins (except threonine and lysine) showed an increase in 15N enrichment in both the control and toluene-exposed cultures. However, in the toluene-exposed cells the amino acids alanine, glutamic acid, aspartic acid, and tyrosine were additionally enriched by 10-25 delta units (Fig. 3). In a number of chromatograms, additional peaks of nonproteinogenic amino acids such as N0-acetylarginine and pyrrolidonecarboxylic acid were observed. The structure of N"-acetylarginine was elucidated using MS-MS technology (data not shown). The mass spectrum of pyrrolidonecarboxylic acid is shown in Figure 4.
In the NPN pool, which contains the sample after cleanup, mainly the free amino acids of the cells, differences were detected in the 15N-enrichment in various amino acids, but so far these differences are not reproducible because of the very low amino acid concentrations. The presence of the metabolites ornithine and aminoadipinic acid was established using organic MS. Medium samples examined at the end of each cultivation period indicated no significant differences in most amino acid compositions and 15N enrichment (Fig. 5). There were detectable amounts of high enriched glutamic acid in the medium of the toluene-exposed culture. Ornithine was released into the nutritional medium by both control and toluene-exposed cultures.

Discussion
This is the first stable isotope (nonradioactive) study into changes induced by pollutants in the amino acid utilization of a cell system related to mammalian cells. With Tetrahymena pyriformis, conventional toxicity tests are usual and are performed using certain complex media (6,(10)(11)(12)(13). In this investigation a combination of the chemically defined media (29) regarding salts, trace salts, and vitamins and of certain amino acid composition (14) was chosen to ensure defined conditions for reproducible isotopic and mass spectrometric analyses. Our working hypothesis was that the impact of a pollutant should be reflected in changes in both the amount and the 15N abundance ofnitrogen in the various pools.
The growth of Tetrahymena pyriformis is accompanied by the secretion of ammonia as the end product of nitrogen metabolism (which includes many metabolic processes). An increase in ammonia production may be caused by gluconeogenesis and the formation of glycogen from amino acids (34,36), or alternatively by protein degradation under conditions of oxygen deficiency (35). In our case the ammonia production under toluene exposure indicates intracellular metabolic changes with increased deamination activity and increased use of arginine. L-Arginine is an essential amino acid for Tetrahymena pyriformis and is a precursor for proline synthesis (37). In the mammalian liver cells, the guanidino group is split off from arginine, as urea and amino acids normally play a minor role in energy production. But arginine plays a key role in various physiological processes (38). In Tetrahymena pyriformis, however, the two-step hydrolytic cleavage results in ammonia formation and provides energy (14,39). The higher arginine consumption is probably due to higher energy requirements under toluene stress. Because of its cell toxicity, the ammonia is fixed intracellularly by the formation of amino acids. Therefore, changes in the amino acid pools are ofparticular interest. According to 15N analysis, the ammonia originating from [I5N]L-arginine is mainly transferred to form certain amino acids via transamination reactions. This is confirmed by the high 15N enrichment in alanine, aspartic acid, glutamic acid, and tyrosine in the proteins. Under toluene stress, the 15N enrichment in these amino acids is higher by [10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25] delta units equal to an increase by 15%. The results are plausibly explained by the well-known reactions and enzymes of amino acid metabolism: in Tetrahymena, ammonia can be fixed to a-ketoglutarate by glutamate dehydrogenase (GDH) and then transferred by aspartate aminotransferase (AAT) via oxalacetate to aspartic acid (40,41). On the other hand, the a-amino nitrogen of glutamic acid is transferred by an alanine aminotransferase (ALAT) via pyruvate to alanine. As mentioned above, we were not able to detect the 15N abundance in amide-nitrogen of glutamine and asparagine because of hydrolysis under acidic derivatization conditions. This problem can only be solved by changing the derivatization method. There is also a tyrosine aminotransferase (41) whose activity seems to increase under the influence of toluene owing to the more greatly enriched tyrosine. The amounts of amino acids in the protein hydrolysate of nonexposed and exposed cells seem to be equal. This will have to be confirmed by quantitative analyses with an internal standard.
Attempts have already been made in occupational medical research (23,42) to identify specific effects on amino acid and protein metabolism resulting from workplace exposures by measuring plasma amino acid concentrations. Due to the lack of isotopic markers, this has only been partly successful because amino acids are involved in many metabolic processes.
To our knowledge, the presence of the metabolites pyrrolidonecarboxylic acid and of N0-acetylarginine has not yet been reported in publications dealing with Tetrahymena pyriformis. Pyrrolidonecar-boxylic acid presents the ring condensation product of glutamic acid and has been proposed as a protective protein end group against proteolytic degradation (43). We assume that this results from intracellular reactions because of the noticeably higher amounts of pyrrolidonecarboxylic acid compared to its formation in standard mixtures. In the arginine pathway of both mammalian cells and Tetrahymena pyriformis, pyrrolinecarboxylic acid is normally formed from glutamic semialdehyde. The existence of the keto group was established by calculations from mass spectrometric data.
Our MS fragmentation study permits the conclusion that arginine is acetylated in the guanidino group. So far only Na-acetylated amino acids are well known compounds in biochemical reactions and have specific functions in metabolism; the most frequently observed acetylated residues are amino acids other than arginine (44). NOL acetylated arginine can be detected in a rare hereditary disorder of the urea cycle, which causes hyperargininemia (45).
The aminoadipic acid detected in the pool of free amino acids (NPN) is a known intermediate of the lysine pathway and ornithine of the arginine pathway. The formation of the latter is caused by physiological activity, as during the derivatization procedure of arginine alone, only traces of ornithine developed. The formation of unlabeled ornithine can take place in arginine catabolism or the formation oflabeled ornithine can take place by transamination ofglutamic semialdehyde.
The amino acids in the nutritional medium showed no differences. Ornithine and glutamic acid were detected, which conforms with the findings of other authors (14,46). The high 15N enrichment of glutamic acid may point out that this compound is a secretion whose purpose is to release ammonia from the cell. But when there are also detectable amounts of glutamic acid in the control culture, it may be that this indicates 15N abundances similar to those in the toluene-exposed culture. Due to the high concentrations of added amino acids, it is difficult to detect the low concentrations of amino acids secreted by the cells. This problem will have to be addressed in further experiments.

Conclusions
The impact of toluene is reflected in metabolic changes in the use of arginine. Qualitative analysis of amino acids must be augmented by quantitative analysis to assess changes in amino acid synthesis rates. Because of the possibility that arginine serves as an energy source in Tetrahymena pyriformis, examining the use of other essential amino acids would be of interest. In addition, the effect of other harmful substances (including mixtures) could be studied.
Environmental medicine requires effectmonitoring methods to detect effects of exposure to harmful substances in addition to determining internal exposure by analyzing biotransformation products. This pilot study is a first step. We believe that the use of the 15N tracer technique in combination with the sensitive GC/C-IRMS/MS coupling provides a powerful tool for the development of such diagnostic methods.