Chem. Res. Toxicol. 1991,4, 102-106
102
Isolation of Urinary 3-Methyladenine Using Immunoaffinity Columns Prior to Determination by Low-Resolution Gas Chromatography-Mass Spectrometry Marlin D. Friesen, Liliane Garren, Virginie Prevost, and David E. G. Shuker* International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon Cedex 08, France Received August 27, 1990
An ammonium sulfate precipated immunoglobin G (IgG) fraction from rabbit antiserum, prepared by use of novel haptenic derivatives, was used to make immunoaffinity columns for purification of 3-methyladenine (3-MeAde) from human urine. IgG was covalently bound t o protein A-Sepharose, and the resulting affinity gel columns were sufficiently stable for multiple reuse. 3-MeAde (up to 200 ng) was adsorbed a t p H 7.4 and, after extensive washing, eluted with 1 M acetic acid. Recovery of 3-MeAde was typically >go%. For gas chromatography-mass spectrometry analysis, deuterium-labeled (d3)3-MeAde (50 ng per sample) was used as an internal standard. 3-MeAde was determined as the mono-tert-butyldimethylsilylderivative and quantitated by measurement of ions a t m/z 206 (3-MeAde-do) and m/z 209 (3-MeAde-dJ. Repeated analyses of a human urine sample show excellent reproducibility of the method.
Introduction 3-MedA’ is one of the major adducts formed in DNA by methylating carcinogens, including N-methyl-N-nitroso compounds ( I ) . 3-MedA is unstable and spontaneously depurinates to give the corresponding methylated base, 3-MeAde (2). There is also an efficient, specific, DNA glycosylase which catalyzes the depurination of 3-MedA to give 3-MeAde as the first step of a repair mechanism for both bacterial and mammalian DNA (3). In common with many alkylated purine bases, 3-MeAde is not catabolized via purine salvage pathways and is excreted intact and quantitatively in urine (4,5). Its use as a noninvasive marker of DNA methylation has been proposed for a several years (6), but the analytical methodology for its determination has only recently become available. 3-MeAde can be quantified in human and animal urine by GC-MS following cleanup by XAD-2 column chromatography and HPLC fractionation, with subsequent conversion to a volatile tert-butyldimethylsilyl derivative (7, 8). By use of this methodology, both humans and laboratory animals were found to excrete low levels of 3-MeAde when they were not exposed to exogenous methylating agents. In experimental animals, a precursor-product relationship between nitrosatable drugs (such as aminopyrine) and urinary 3-MeAde was shown by using stable isotope labeling where deuterium (d3) labeled methyl groups in the drug were transferred to give 3-MeAde-d3 (9). The development of methods for the monitoring of human populations for exposure to methylating carcinogens by the use of urinary 3-MeAde requires an analytical method that permits the analysis of a large number of samples. This paper describes the preparation of a rabbit antiserum against 3-MeAde and its use in the preparation of efficient, reusable, immunoaffinity columns for purification of 3-MeAde from urine prior to GC-MS. Preliminary accounts of the preparation of the antiserum and immunoaffinity columns have been published (IO, 11).
* To whom
correspondence should be addressed.
0893-228x/91/2704-0102$02.50/0
Scheme I. Synthesis of 3-MeAde-Protein Conjugate
O 3 HCI/EtOH
O 3
I
I1
Experimental Procedures Materials. 3-MeAde-d3and [3H]-3-MeAdewere synthesized as previously described (7). 3-MeAde (Fluka, Buchs, Switzerland), MTBSTFA (Pierce Chemical Co., Rockford, IL), and protein A-Sepharose CL4B (Pharmacia,Uppsala, Sweden) were used as supplied. Polystyrene minicolumns and hydrophobic polyethylene frits (Pierce Chemical Co., Rockford, IL) were used for the preparation of a f f ~ t columns y as described in the manufacturer’s literature. All other solutions and buffers were prepared from standard laboratory reagents of the highest grade of purity available.
Synthesis of Haptenic Derivatives of 3-MeAde (Scheme I). (A) Preparation of N6-(Carbamoylmethy1)adenine(11). 6-Chloro-9-(tetrahydro-2-pyranyl)purine (4.72 g, 20 mmol, Sigma Abbreviations: 3-MedA, 3-methyl-2‘-deoxyadenosine; 3-MeAde, 3methyladenine; MTBSTFA, N-methyl-N-(tert-butyldimethylsily1)trifluoroacetamide; DMF, N’,N’-dimethylformamide; DMA, N’,N’-dimethylacetamide; MBSA, methylated bovine serum albumin; EDC, 1[3-(dimethylamino)propy1]-3-ethyldicarbodiimide hydrochloride; KLH, keyhole limpet hemocyanin; IgG, immunoglobulin G; PBS, phosphatebuffered saline; ELISA, enzyme-linked immunosorbent assay; F,, constant region of immunoglobulin molecules.
0 1991 American Chemical Society
Chem. Res. Toxicol., Vol. 4 , No. 1, 1991 103
Urinary 3-Methyladenine Product No. (2-6878) in DMF (40 mL) was treated with glycinamide hydrochloride (4.44 g, 40 "01) in water (40 mL) containing sodium bicarbonate (5.0 g, 60 mmol). The clear solution was heated a t 90 "C for 5 h, by which time the starting material had disappeared (TLC; silica gel using CHCl,/MeOH, 1 0 1 v/v). The mixture was cooled and evaporated to dryness. The residue was stirred with water (150 mL) and the precipitate filtered and washed with large quantities of cold water, which gave, after drying in vacuo, the product 9-(tetrahydropyranyl)-P-(carbamoylmethy1)adenine (I) (3.16 g, 57%). MS m / z 276 (M+), 192 (M+ - tetrahydropyran), 148 (M+ - tetrahydropyran - CONH2). The product was used in the next step without further purification. I (2 g, 7.5 mmol) in absolute ethanol (40 mL) was treated with aqueous HCl(1 M, 10 mL, 1.0 mmol) and the solution stirred a t room temperature for 24 h. Upon cooling a t 4 "C overnight, colorless crystals formed and were filtered and washed with cold ethanol and ether to give I1 as the hydrochloride (1.55 g, go%), which was converted to I1 as follows: the product was dissolved in a minimum volume of water and treated with dilute sodium hydroxide until precipitation occurred. After cooling a t 4 "C, the precipitate was filtered, washed with water, and dried in vacuo to give crude I1 which was recrystallized from water to give I1 (0.8 g, 55%). MS m / z 192 (M+), 148 (M+ - CONH2), 119 (M+ NHCH&ONHZ). (B) P r e p a r a t i o n of N6-(Carboxymethyl)-N-3-methyladenine (111). A suspension of I1 (180 mg, 0.9 mmol) in DMA (3.6 mL) was treated with methyl iodide (360 pL, 5.4 mmol) and heated a t 60 "C in a sealed tube until the solution became clear (6 h). DMA was removed in vacuo and the mixture dissolved in 1 M HCl (4 mL) and heated at 100 "C for 1 h. After careful adjustment to pH 4, the solution was cooled on ice and the precipitate collected. The product was dissolved in a minimum volume of boiling water and treated with activated charcoal. The hot solution was quickly filtered through a plug of diatomaceous earth to give a colorless solution which was cooled to 4 "C. The resulting crystals were filtered and washed with a small quantity of ice-cold water to give I11 (24 mg, 13%). MS m / z 207 (M+), 189 (M+ - HZO), 163 (M+ - COZ), 162 (M+ - COOH), 134 (M+ - CHZCOOH), 119 (M+ - NHCHZCOOH). A,, (e) (pH = 2) 280 (17500), 214 (9800); (pH = 7) 287 (13800), 203 (32300); (pH = 14) 287 (13400), 210 (34200). (C) Preparation of 3-MeAdeProtein Conjugates. I11 (12.4 mg, 60 pmol), in water (1.14 mL) containing sufficient sodium bicarbonate to produce a clear solution, was added, over a period of 3 h, to a solution of MBSA (55 mg, Sigma Product No. A-1009) containing EDC (23.30 mg, 120 pmol). The pH was maintained a t 6.5. On completion of the addition, the solution was stirred at room temperature for a further 20 h. The red reaction mixture was applied to a column of Sephadex G-50 (20 X 2.5 cm) which was eluted with 0.15 M NaC1. The first UV-absorbing peak to elute was collected and concentrated/dialyzed to a volume of 1 mL in a vacuum concentrator (ProDiMem concentrator, Polylabo, Strasbourg, France). The concentrate was lyophilized to give 3-MeAde-BSA conjugate (45 mg). The hapten/carrier ratio was 6.4 mol of 3-MeAde/mol of MBSA, as determined by UV spectroscopy. 3-MeAde-modified KLH was prepared in an identical manner except that, due to problems of solubility, the column fraction was dialyzed against several changes of PBS and made up to a standard solution of 1 mg of protein/mL containing 0.02% sodium azide as preservative. 3-MeAde-KLH contained approximately 100 mol of 3-MeAde/mol of protein as determined by UV spectroscopy. Preparation of Rabbit Serum. Two rabbits were immunized a t multiple subcutaneous sites on the shaved back with an emulsified suspension of 3-MeAde-MBSA (1 mg) in Freund's complete adjuvant (1 mL) per animal. Four and eight weeks later, the rabbits were boosted with the antigen in Freund's incomplete adjuvant. After 10 weeks the animals were bled under local anesthesia from the lateral ear vein and sera were prepared. The properties of the antisera were determined by ELISA using 3MeAde-KLH as the coating antigen, and the results have been described in detail elsewhere (10). For long-term storage, the antiserum was lyophilized and then stored at 4 "C. Preparation of 3-MeAde Immunoaffinity Columns. The procedure for making gel sufficient for 20 X 1 mL columns was
500
n.
W
400
V e
,
v
300
e,
4
.-. j i j i
1
I
4
3
r i j
IIi
1
I
200
100
1 I
0 0
5
10
15
20
Volume of eluate ( m l )
Figure 1. Elution profile of [,H]-3-MeAde in human urine (2 mL) using a 3-MeAde immunoaffinity column. Eluting solvents as described under Experimental Procedures. Solid line: Elution a t 4 OC. Dotted line: Elution a t 4 "C, but with the acetic acid wash a t room temperature. as follows: Lyophilized rabbit serum was reconstituted to its original volume with sterile water. A crude IgG fraction was prepared according to a literature procedure (12). Saturated ammonium sulfate (13.3 mL) was added dropwise to gently stirred serum (20 mL) in a beaker to give a 40% solution. After stirring for a further 30 min, the precipitate was collected by centrifugation (3000g) and the pellet washed twice with 50% saturated ammonium sulfate solution. The pellet was resuspended in PBS (20 mL) and dialyzed overnight against several changes of PBS (3 L). The dialysate was centrifuged a t 3000g to remove suspended matter and used directly in the next step. Protein A-Sepharose CL 4B (20 mL) was washed well with Tris buffer (0.1 M, pH 7.4) and suspended in the same buffer to a total volume of 40 mL. The gel suspension was divided between two 50-mL polypropylene screw-cap centrifuge tubes, and dialyzed IgG fraction (10 mL) was added to each tube. The mixture was stirred end over end for 30 min at room temperature. The gel was recovered by low-speed centrifugation and washed several times with Tris buffer to remove unbound IgG. The gel was then washed with triethanolamine buffer (0.2 M, pH 8.2) and divided between eight centrifuge tubes. Each tube of gel was then treated with dimethyl pimelimidate (Pierce, 50 mL, 20 mM freshly made up in triethanolamine buffer) and mixed end over end for 45 min a t room temperature. The gel was recovered by low-speed centrifugation and treated with aqueous buffered ethanolamine (50 mL per tube, 20 mM in triethanolamine buffer) for 5 min to block unreacted cross-linking agent. The gel was then recovered and washed well with PBS. The gel was poured into polystyrene minicolumns in 1-mL aliquots and maintained in place by use of hydrophobic plastic frits. Prior to use, the columns were washed extensively with PBS/O.O2% azide and stored at 4 "C with a small volume of buffer above the gel. Optimization of Immunoaffinity C l e a n u p f o r U r i n e Samples. [,H]-3-MeAde (100 ng, 700 dpm) in PBS (2 mL) was applied to a column, followed by a further 3 mL of PBS. The column was then washed with water (5 mL). One-milliliter fractions of the column eluate were collected, and radioactivity was measured by liquid scintillation counting. Quantitative elution of [3H]-3-MeAdewas obtained by washing the column with 1M aqueous acetic acid (5 mL), with most of the radioactivity eluting in the first 2 mL (data not shown). With urine samples (2 mL, buffered by addition of 100 pL of 0.5 M phosphate buffer, pH 7.4) spiked with the same amount of t3H]-3-MeAde, some radioactivity was detected in early elution fractions (data not shown). However, quantitative retention of radioactivity was obtained if the columns were preequilibrated at 4 "C. In addition, the volume of wash water was increased to 10 mL in order to minimize the interference of urine residues in the GC-MS analysis (a typical elution profile for a urine sample is shown in Figure 1). If the columns were equilibrated at room temperature prior to acetic acid elution, greater than 90% of the retained [3H]-3-MeAdewas then eluted in 2 mL (Figure 1).
104 Chem. Res. Toxicol., Vol. 4, No. 1, 1991 inn
i
5
.-
w
-
2500
751 i
2
2000
: 1500
\
1
C
0 C
'\ '.-
Friesen et al.
J
P
0 0
200
400
600
800
1000
1000
3-MeAdP a d d e d ( n g )
Figure 2. Evaluation of binding capacity of 3-MeAde immunoaffinity columns. Determination of Column Capacity for 3-MeAde. PBS (2 mL) containing [3H]-3-MeAde(156 pg, lo00 dpm) and 3-MeAde (O-lOOO ng) was applied to immunoaffinity columns at 4 "C. The columns were washed with PBS (3 mL) and water (10 mL). The columns were then equilibrated to ambient temperature and washed with acetic acid (1M, 2 mL), and the eluate was collected directly into scintillation vials. Picofluor (Beckman, 10 mL) was added to each vial and radioactivity determined by scintillation counting. The results were expressed as the percentage of [3H]-3-MeAderetained on the column (Figure 2). Affinity Purification of 3-MeAde. Solutions of 3-MeAde standards in PBS or urine (2 mL, buffered by addition of 100 pL of 0.5 M phosphate buffer, pH 7.4) containing 3-MeAde-d3(50 ng) were applied to the affinity columns at 4 "C (in a cold room) and allowed to elute under gravity [for urine, samples were applied to a small precolumn containing unmodified Sepharose CL-4B (0.5 mL) and allowed to elute directly onto the affinity column]. The columns were washed, successively,with PBS/O.O2% azide (3 mL) and water (10 mL). The compound was eluted with 1M acetic acid (2 mL). The acetic acid eluate was collected into silanized flame seal ampules (2.5-mL capacity) and evaporated to dryness in a centrifugal vacuum evaporator (Speedvac, Savant Instruments). If necessary, the ampules were sealed with aluminum foil and parafilm and stored at 4 "C prior to GC-MS analysis. Otherwise, GC-MS analyses were carried out immediately. The immunoaffinity columns were washed with a further 3 mL of 1M acetic acid and then with PBS/O.O2% azide (10 mL) and stored at 4 "C. After this step, columns were then reusable. GC-MS Analysis of 3-MeAde. MTBSTFA (10 pL) and dry pyridine (10 pL) were added into the ampules containing evaporated acetic acid fractions from the affinity columns, and the ampules were flame sealed and heated at 110 O C for 20 min. The ampules were inverted and centrifuged such that the derivatization mixture was in the cone of the ampule, and then the ampules was carefully opened by scratching with a glass file and breaking at the neck. The reaction mixture was evaporated to dryness in the centrifugal vacuum evaporator and the residue dissolved in dry ethyl acetate (10-40 pL) containing MTBSTFA (1%).Aliquots (0.2-1 ML)of this solution were analyzed by GC-MS (Hewlett Packard 5970A). A typical GC-MS selected ion monitoring (SIM) chromatogram is shown in Figure 3 (for conditions, see legend).
Results Preparation of a 3-MeAde Rabbit Antiserum. 3MeAde is too small a molecule to elicit an immune response and needed to be bound to a carrier protein. Previous studies on the preparation of adenosine antibodies utilized an NG-(carboxymethyl) functionality to bind the nucleoside to protein amino groups (13). Thus, a synthetic route to W-(carboxymethyl)-N-3-methyladenine (111) was devised and is summarised in Scheme I. In principle, the method has general applicability to other 3-alkyladenines. 111was bound to MBSA by use of a water-soluble carbodiimide procedure. The haptenlcarrier ratio was not
2 09
5 00
2 06 OJQ 7
'
'
.
'
I
'
'
8
Time
.
.
I
9
.
.
-
.
*
10
(min.)
Figure 3. SIM trace of an extract of human urine purified by immunoaffiiity chromatography. 3-MeAde-TBDMS derivative, m/z 206 (M+- 57);3-MeAde-d3-TBDMSderivative,m/z 209 (M+ - 57) (internalstandard). This figure is redram from the original, and the retention time of 3-MeAde-dois actually slightly longer than that of the d3internal standard. GC-MS conditions (Hewlett Packard 5890 GC): on-column injection; helium carrier gas at 0.7 mL/min; column, HP ultra-2 25 m; temperature program, 55-270 "C at 40 "C/min, then to 300 "C at 10 "C/min, and hold for 5 min. Connected via interface (260 "C) to HP 5970A mass selective detector operating at an ionization potential at 70 eV.
exceptionally high, but upon immunization in rabbits, good antibody titers (as determined by ELISA) were obtained. Reasonably good recognition of 3-MeAde compared to other closely related purines was observed (10). Preparation and Characteristics of Reusable 3MeAde Immunoaffinity Columns. A crude ammonium sulfate precipitated IgG fraction from a 3-MeAde rabbit antiserum was covalently bound to protein A-Sepharose CL-4B by using a bifunctional cross-linking agent, dimethyl pimelimidate. The advantage of this approach over other methods using variously activated gels is that protein A binds specifically to the F, region of IgG, thus leaving the antigen binding site maximally exposed (Goding, 1985). The first attempts to make 3-MeAde affinity columns were successful, and no attempt was made to investigate the effect of altering the loading of IgG on the gel. Despite the relatively high cost of protein A-Sepharose gels, their ease of use and the high stability of the resulting affinity columns make them extremely useful. The binding capacity of the columns for 3-MeAde was determined by using a simple saturation assay. Separate solutions each containing high specific activity [3H]-3MeAde (156 pg, 1000 dpm) and increasing amounts of unlabeled 3-MeAde (0-1000ng) in PBS were applied to columns at 4 "C. The standard elution protocol was used, and radioactivity eluting with 1 M acetic acid was measured (Figure 2). Thus, it can be seen that the capacity of the columns was between 150 and 200 ng. Cross-reacting substances present in urine (probably nucleic acid bases such as adenine, which are normal urinary catabolites) resulted in decreased binding of 3MeAde to the column a t room temperature. However, at 4 "C, binding of 3-MeAde is improved, and almost quantitative retention was observed under conditions of the assay (Figure 1). This phenomenon is due to a "nonequilibrium" effect as a result of decreased dissociation of the antigen-antibody complex at low temperature
Chem. Res. Toxicol., Vol. 4, No. 1, 1991 105
Urinary 3-Methyladenine
available methods. Studies of 3-MeAde excretion in different human populations have been undertaken (11). A monoclonal antibody against 3-MeAde has recently been prepared (16),but its use in preparing immunoaffinity columns has not yet been explored. However, the use of the above-described antiserum-based immunoaffinity columns in combination with quantification by monoclonal antibody based ELISA provides a wholly immunochemical rapid method for 3-MeAde determination
BO r
(16).
lot 01
0
I
10
20
3-MeAde
30
40
50
60
a d d e d (ng/2ml)
Figure 4. Standard curve for the analysis of 3-MeAde in human urine. (14). 3-MeAde was bound to the column a t pH 7.4, and then the columns were extensively washed with water. This step served to remove unbound urinary organic substances as well as removing inorganic salts. 3-MeAde was eluted from the column with 1 M acetic acid, which was then evaporated under vacuum, leaving a residue suitable for derivatization and GC-MS analysis. A selected ion monitoring trace for a human urine sample is shown in Figure 3. Typically, human urine extracts were free of interfering peaks. Interestingly, a closely related methylpurine, 7-methyladenine (7-MeAde), which has recently been found as a urinary metabolite of methylating agents (15),was not retained a t all on the immunoaffinity columns (Mandel, Friesen, and Shuker, unpublished observations) and is, therefore, unlikely to interfere with the quantification of 3-MeAde. Furthermore, the excellent agreement for the determination of urinary 3-MeAde between the immunoaffinity column GC-MS method and the recently developed wholly immunochemical method (16) suggests that the procedure is specific for 3-MeAde. Validation of the Immunoaffinity Cleanup/GC-MS Method for 3-MeAde. Samples of phosphate-buffered saline and urine (2 mL) containing 3-MeAde (0-50 ng) and 3-MeAde-d, (50 ng) were subjected to the immunoaffinity cleanup/GC-MS analysis, and the results are shown in Figure 4. A linear response was found with the nonzero intercept due to the presence of “background” 3-MeAde in the urine sample. The reproducibility of the method was demonstrated by analyzing 4 X 2-mL aliquots of a human urine sample on four different immunoaffinity columns. A mean value of 80 ng ( f 3 ng)/2 mL was found.
Discussion Immunoaffinity columns, prepared from 3-MeAde rabbit antiserum, effectively replace the chromatographic steps required for purifying urinary 3-MeAde prior to GC-MS analysis (7,8). Urine samples were applied to a small gel precolumn and allowed to pass directly onto the affinity columns without further cleanup. The 3-MeAde-containing fraction collected off the column was used directly for GC-MS after derivatization. The immunoaffinity gel was prepared on a batch basis with reproducible performance of columns. Simultaneous use of a number of columns considerably reduced the time for sample preparation. Furthermore, the columns were reusable after washing, and tests for residual cross-contamination of 3-MeAde between runs of the column were negative (data not shown). The columns could be used up to 30 times without detectable loss of efficiency. This method has enabled the analysis of many hundreds of samples of urine in a much shorter time than previously
Recent results indicate that the majority of urinary 3-MeAde is dietary in origin. However, this potential confounding factor can be easily manipulated in experimental studies, and methylating exposures such as cigarette smoke can be readily detected (Prevost and Shuker, to be published). Despite some limitations, the analysis of urinary 3-MeAde remains a good integrated measure of DNA methylation [especially as methylation of adenosine in RNA gives predominantly 1-methyladenine (611. Acknowledgment. Financial support from the U S . National Cancer Institute (Grant CA48473) and the IARC (Special Training Award for V. Prevost) is gratefully acknowledged. Mass spectra of 3-MeAde haptens were obtained by Dr. P. B. Farmer, MRC Toxicology Unit, Carshalton, Surrey, England. John Green, of the MRC Toxicology Unit, is thanked for his help in preparing the rabbit antisera. Registry No. I, 131011-52-4;11,131011-53-5;III,131011-54-6; 3-MeAde, 5142-23-4; 6-chloro-9-(tetrahydro-2-pyranyl)purine, 7306-68-5; glycinamide hydrochloride, 1668-10-6.
References (1) Margison, G . P., and O’Connor, P. J. (1979) Nucleic acid mod-
ification by N-nitroso compounds. In Chemical carcinogens and DNA (Grover, P. L., Ed.) pp 1:111-159, CRC Press, Boca Raton, FL. (2) Fujii, T., Saito, T., and Nakasaka, T. (1980) Synthesis, ring opening and glycosidic bond cleavage of 3-methyl-2’-deoxyadenosine. Chem. Commun., 758-759. (3) Karran, P., and Lindahl, T. (1985) Cellular defense mechanisms against alkylating agents. Cancer Surv. 4,585-599.
(4) Shuker, D. E. G., Bailey, E., and Farmer, P. B. (1987) Excretion of methylated nucleic acid bases as an indicator of exposure to nitrosatable drugs. In The relevance of N-nitroso compounds to human cancer: exposures and mechanisms (Bartsch, H., Hemminki, K., and O’Neill, I. K., Eds.) pp 407-410, IARC Scientific Publication 84, IARC, Lyon, France. (5) Hanski, C., and Lawley, P. D. (1985) Urinary excretion of 3methyladenine and 1-methylnicotinamide by rats following administration of [methyl-14C]-methylmethanesulfonate and comparison with administration of [14C]-methionine or formate. Chem.-Biol. Interact. 55, 225-234. (6) Lawley, P. D. (1976) Methylation of DNA by carcinogens: some applications of chemical analytical methods. In Screening tests in chemical carcinogenesis (Montesano, R., Bartsch, H., Tomatis, L., and David, W., Eds.) pp 181-208, IARC Scientific Publication 12, IARC, Lyon, France: (7) . , Shuker. D. E. G.. Bailey. E. Parrv. A.. Lamb. J.. and Farmer. P. B. (1987) The determinition of &;nary 3-methyladenine in humans as a potential monitor of exposure to methylating agents. Carcinogenesis 8, 959-962. (8) Stillwell, S., Xu, H. X., Adkins, J. A., Wishnok, J. S., and Tannenbaum, s. R. (1989) Analysis of methylated and oxidized purines in urine bv caDillarv eas chromatoeraDhv-mass mectrometry. Chem. Red. T&icol 5,94-99. (9) . . Farmer. P. B.. Shuker. D. E. G.. and Bird. I. (1986) DNA and protein adducts as indicators of i n uivo methylation’by nitrosatable drugs. Carcinogenesis 7 , 49-52. (10) Shuker, D. E. G., and Farmer, P. B. (1988) Urinary excretion of 3-methyladenine in humans as a marker of nucleic acid methylation. In Methods for detecting DNA damaging agents in humans: applications in cancer epidemiology and prevention (Bartsch, H., Hemminki, K., and O’Neill, I. K. Eds.) pp 92-96,
IARC Scientific Publication 89, IARC, Lyon, France.
106 Chem. Res. Toxicol., Vol. 4, No. I , 1991 (11) Shuker, D. E. G., Friesen, M. D., Garren, L., and Prevost, V.
(1990) A rapid GCMS method for the determination of urinary 3-methyladenine: applications to human studies. In Relevance t o h u m a n cancer of N-nitroso compounds, tobacco smoke and mycotoxins (O’Neill,I. K., Chen, J. S., Lu, S. H., and Bartsch, H., Eds.) IARC Scientific Publication 105, IARC, Lyon, France (in press). (12) Goding, J. W. (1986) Monoclonal Antibodies: Principles and Practice, Academic Press, London. (13) Bredehorst, R., Wielckens, K., Kupper, E. W., Schnabel, W., and Hilz, H. (1983) Quantification without purification of blood and tissue adenosine by radioimmunoassay. Anal. Biochem. 135, 156-164.
Friesen et al. (14) Zettner, A., and Duly, P. E. (1974) Principles of competitive binding assays (saturation assays). 11. Sequential saturation. Clin. Chem. 20, 5-14. (15) Mandel, H. G., Shaw, J. A., Farmer, P. B., and Martin, J. (1989) Chromatographic resolution of 7-methyladenine in urine of rata administered N-methylnitrosourea; a potential marker for monitoring exposure to methylating agents. Carcinogenesis 10, 757-762. (16) Prevost, V., Shuker, D. E. G., Bartach, H., Pastorelli, R., Stillwell, W. G., Trudel, L. J., and Tannenbaum, S. R. (1990) The determination of urinary 3-methyladenine by immunoaffinity chromatography-monoclonal antibody based ELISA: use in human biomonitoring studies. Carcinogenesis l l , 1747-1751.
