AMI. Ctmm.
i w a , 65, 1723-1727
1723
KO2 Chemical Transformation/Mass Spectrometry Detection of Covalent Damage to the DNA of Cultured Human Lymphocytes Exposed to Benzo[ alpyrenes Kariman Allam: Samy Abdel-Baky; and Roger W.Giese’J Department of Pharmaceutical Sciences, Bouve College of Pharmacy and Health Professions, and Barnett Institute of Chemical Analysis and Materials Science, Northeastern University, Boston, Massachusetts 02115, and Department of Chemistry, Alexandria University, Alexandria, Egypt
Cultured human lymphocytes were exposed to benzo[a]pyrene (€%[alp), and diol epoxide-type DNA adducts arising from this chemical were detected by a method consisting of the following sequence of steps: (1)isolate the DNA; (2) subject the DNA to mild acid hydrolysis to release the polyaromatic moiety as a tetrahydrotetrol; (3) add an internal standard; (4) oxidize the tetrahydrotetrol with potassium superoxide to pyrene2,3-dicarboxylicacid; (5) derivatize the latter with pentafluorobenzyl bromide; (6) purify the diester product on a silica cartridge; and (7) detect this product by gas chromatography electron capture negative ion mass spectrometry. From the dose (1 &mL) of B[a]P applied, five adducts in lo7normal nucleotides were found. Largely because steps 2-5 of the method take place sequentially in a single vial, the procedure is convenient and affords precise results. To demonstrate the potential of the method to detect KOzsusceptible polyaromatic hydrocarbon DNA adducts in general, including unknowns, it was also applied to picomole and femtomole amounts of a standard of chrysene-l,4quinone using scanning and selected ion monitoring conditions, respectively, in the MS. Since standard products can be detected with selected ion monitoring at levels lo4 below those encountered here (prior work), it should be possible in the future to extend the method to samples containing smaller amounts of such adducts. INTRODUCTION Diol epoxide metabolites of polyaromatic hydrocarbons (PAH) are an important class of carcinogens.’ This is due, at least in part, to their electrophilicity, leading to their covalent attachment to DNA. The products can be called “diol epoxide PAH DNA adducts”. For example, benzo[a]pyrene (B[alP) tends to undergo metabolic activation to a corresponding diol epoxide, which then predominantly alky-
* Address reprint requests to this author.
+ Northeastern University.
Alexandria University. Current address: BASF Corp., P.O. Box
C$5CHzOOC,@ C,FSCHzOOC
2
1 F1
3
4
Flguro 1. Structures of compounds: B[a]P adduct (arbitrarily shown as a deoxynucleoside)(1); diester 2 (2); chrysenel,4quinone (a), and 1,2-bis(pentefluoroben~yl)phenanthrene dicarboxylate (4).
lates the N2position of guanine sites on DNA, producing the adduct represented by structure 1 in Figure 1.2 Adducts of this type have been detected in biological samples by several techniques, especially 32P-postlabeling,3.4 immunoB88ay,Sfluorescen~e,6*~ and mass spectrometry (MS)? 32P-Postlabelingis sensitive but does not, by itself, provide much characterization of unknown DNA adducts. Immunoassay also is sensitive, but an authentic adduct must be available f i s t to prepare the antibody, and the subsequent antibody may cross-react with related adducts without defining their nature. Fluorescence is less sensitive, but provides some qualitative information about the adduct. MS as used to date for such adducts also is less sensitive, but provides information about the mass of the adduct and its fragments. This has been achieved by using mild acid hydrolysis to release the PAH moiety as a tetrahydrotetrol from the DNA, conducting a silylation reaction, and detecting the resulting derivative by gas chromatographyJelectron impact MS.’ We are developinga technique that is related to the current MS method for the detection of diol epoxide PAH DNA adducts, with the intent of achieving higher sensitivity. In our technique, the acid-liberated tetrahydrotetrol PAH is
13528,Research Triangle Park, NC 27709-3528. I Abbreviations: B[a]P, benzo[a]pyrene; diester 2,2,3-bis(pentaflu0robenzyl)pyrenedicarboxylate; drdiester 2, [1,2,3,4,5,6,11,12-2H&2,3bis(pentafluorobenzyl)pyrenedic~~~~, drketone, [1,2,3,4,5,6,11,122H~]-9,10-dihydrobenzo[a]pyren-7(8H)-one; GC/ECNI-MS, gas chromatography electron capture negative ion mass spectrometry; MS, mass spectrometry; PAH, polyaromatic hydrocarbon; SIM, selected ion monitoring; tetrol, benzo[alpyrene-7,8,9,10-tetrahydrotetrol. (1)Harvey, R.G., Ed. Polycyclic Hydrocarbons and Carcinogenesis; ACS Symposium Series 283;American Chemical Society: Washington, DC, 1985. 0003-2700/93/0305-1723504.00/0
(2)Singer, B.; Grunberger, D. Molecular Biology of Mutagens and Carcinogens; Plenum Press, New York, 1983;p 151. (3)Fbddy, M.; Randerath, K. Mutat. Res. 1990,241, 37-48. (4)Beach, A. C.; Gupta, R. C. Carcinogenesis 1992, 13, 1053-1074. (5)Santella, R.M.; Weston, A.; Perera, F. P.; Trivers, G. T.; Harris, C. C.; Young, T. L.; Nguyen, D.; Lee, B. M.; Poirier, M. C. Carcinogenesis 1988,9, 1265-1269.
(6)Weston, A.; Bowman, E. D. Carcinogenesis 1991,12, 1445-1449. (7)Weston, A.; Rowe, M. L.; Manchester, D. K.; Farmer, P. B.; Mann, D. L.; Harris, C. C. Carcinogenesis 1989,10, 251-257. 0 1883 Amcnlcen Chemical Soclety
1724
ANALYTICAL CHEMISTRY, VOL. 65, NO. 13, JULY 1, 1993
Table I. Analytical Procedure and Results for the Measurement of Tetrol from B[a]P-Exposed Cells and as a Standard cella calf thymus DNAa standards steps nonexposed B[alP exposed A B C blank tetrol procedure add tetrolb purify/HCl DNA add tetrolb vials (no.) DNA/vial (pp! K02/PFBzBr/Si GC/ECNI-MS results fmol of diester 2/vial
X
X
X
X X
1-3 109
4-6 114
7,s 114
9,lO 113
X
X X
X
X
X
X
11.5 13.3 13.5
49.6 56.3 50.2
12.3 13.1
X X
X
113
13,14
15,16
0
0
X
X X
X X
X X
8.8e 12.7
79.2d 117.1
6.2 10.8
192* 198
11,12
a A, DNA blank; B, spiked; C, spiked later. b Amount of tetrol added 128 pg (390 fmol). e Spiked tetrol is not seen because of the organic extraction step. Also internal standard (106 pg or 380 fmol of da-ketone in 20 p L of methanol) was added to each sample except viala 3, 7, 8,13, and 14. The average yield (86fmol after background correction)correspondsto an absolute yield of 22%. e The average yield (182 fmol after background correction) corresponds to an absolute yield of 47 % .
oxidized to a corresponding PAH dicarboxylic acid by potassium superoxide, followed by derivatization with pentafluorobenzyl bromide and detection of the resulting PAH pentafluorobenzyl diester by gas chromatography/electron capturenegative ion mass spectrometry (GC/ECNI-MS). We have detected a standard of 2,3-bis(pentafluorobenzyl)pyrenedicarboxylate (diester 2) at the low attomole level by GC/ECNI-MSs and used a prior version of this method to determine a standard of the tetrahydrotetrol of B[alP at the midfemtomole level.g Here the method is improved and applied to a biological sample containing a similar level of analyte.
EXPERIMENTAL SECTION Materials. Potassium superoxide (KO*), 18-crown-6, pentafluorobenzylbromide, and NJV-dimethylformamidewere purchased from Aldrich (Milwaukee,WI). Hydrochloricacid, acetic acid, ethyl acetate, hexane, acetonitrile, isoamyl alcohol, methanol, silanized glass wool, and silica gel particles (40pm) were purchased from J. T. Baker (Phillipsburg, NJ). Sodium lauryl sarcosine, phenol, N-(2-hydroxyethyl)piperazine-N’-(2-ethanesulfonic acid) (HEPES), and calf thymus DNA were from Sigma Chemical Co. (St. Louis, MO). NazEDTA and chloroform were from Fisher Chemical Co. (Fairlawn, NJ). Proteinase K and ribonuclease A (RNase A), DNAase free, were obtained from Boehringer Mannheim Co. (Indianapolis, IN). Triethylamine was purchased from Pierce (Rockford, IL). Distilled water was purified to HPLC grade with a Nanopure/Organic Pure system (Barnstead,Boston,MA) or purchased fromJ. T. Baker. Toluene wae purchased from Burdick & Jackson (Muskegon,MI). Benzo[a]pyrene-7,8,9,10-tetrahydrotetrol (tetrol) was from the NCI Chemical Carcinogen Repository, Midwest Research Institute (KansasCity, MO). [1,2,3,4,5,6,11,12-2H&9,lO-Dihydrobenzo[a]pyren-7(8H)-one(&ketone) was prepared as deecribed.eGlaes conical centrifuge tubes (15 mL) fitted with glass pennyhead stoppers, used for the phenol extractions, were from Scientific Products (Bedford, MA). Glass-stoppered 4.0-mL conical glass vials (all-glassvials) were prepared in-house10 for the sequential hydrolysis, oxidation, and alkylation steps. SilicaCartridge Column. Silica gel particles (100mg) were sandwichedbetween 10-mgsilanizedglase wool plugs in a Pasteur pipet, and the column was washed (gravity flow) with 3 mL each of ethyl acetate and hexane before the sample was applied. Equipment. A Model 5988Amass spectrometer from Hewlett Packard (Palo Alto, CA) was fitted with a chemical ionization detector and connected to a Hewlett-Packard 5890 Series 11gas (8) Abdel-Baky,S.; Giese, R. W. Anal. Chem. 1991, 63, 2986-2989.
(9) Li, W.; Sotiriou, C.; Abdel-Baky, 5.;Fisher, D.; Giese, R. W. J. Chromotogr. 1991,588,273-280. (10) Abdel-Baky,S.; Allam, K.; Gieae, R. W. Anal. Chem. 1992, 64,
2882-2884.
chromatograph via a capillary interface kept at 290 OC. The capillary GC column (25 m length, 0.22 mm i.d., 0.1-pm film thickness, Ultra 1 from Hewlett Packard) was temperature programmed from 140OC immediately after on-column injection to 290 OC at 70 OC/min and held for 13 min. The Multi-Mixer, Model4600, for continuous vortexingand L-C Oven,Model3510, to heat this vortexer were purchased from Lab-Line Instruments (Melrose Park, IL). Exposure of Cells to B[a]P. This work was performed by Charles Crespi at GENTEST Corp. (Woburn, MA). MCL-5 cellsll at 5 x 105 cells/& were aliquoted into 12 500-mL glass Erlenmeyer flasks at 65 mL of cells/flask. Six flasks received 65 pL of dimethyl sulfoxide,and six flasksreceived65pL of dimethyl sulfoxide containing 1mg/mL B[a]P (final concentration, 1pg/ mL). Flasks were incubated at 37 OC for 28 h, and then 55 mL from each flask was centrifuged in a plastic centrifuge tube that had been rinsed first with dimethyl sulfoxide. The medium was discarded, and the cell pellet was resuspended in phosphatebuffered saline and recentrifuged. The supernatant was again discarded and the cell pellet frozen. Analytical Procedure. Cellular DNA. Cells (5.5 X 107 as a thawed pellet) were treated with 800 pL of digestion buffer
(0.01MHEPES,pH7.5,0.lMNaCl,0.025MEDTA,0.1mg/mL proteinase K, 2% sodium lauryl sarcosine) and kept, after vortexing, in a shaking water bath at 50 OC for 18h. The digest was transferred to a glass tube and extracted with 800 pL of alcohol2524 1;the phenol solution (phenol/chloroform/isoamyl phenol had been preextracted with 0.01 M HEPES pH 7.5). The DNA was precipitated from the aqueous layer by adding 80 p L of 3 M sodium acetate followed by 1.6 mL of cold ethanol. The precipitated DNA was washed by vortexing once with 1.0 mL of cold 70% ethanovwater, dried under nitrogen, dissolved in 500 p L of water, and treated with 2.5 pg of DNase-free RNase with shakingfor 30 min at 37 OC. The phenol extraction was repeated, yielding (1AU at 260 nm equaled 50 pg of DNA/&) 760 pg of DNA from exposed cells and 1.08 mg of DNA from nonexposed cells as solutions in 500 pL of water. For each sample, A d A w was 1.9. Aliquots in triplicate were made from each DNA sample (as described in Table I), followed by dilution with water up to 500 pL, in all-glass vials. Calf Thymus DNA. Three tubes were set up containing calf thymus DNA: A and C, 560 pg each, and B, 700 pg. To each was added 800 pL of a mixture of 0.01 M HEPES, 0.1 M NaC1, and 0.025 M EDTA. After the DNA dissolved, 20 pL of methanol containing 128 pg of tetrol was added to B. The samples were phenol extracted, RNase hydrolyzed, and phenol extracted as above, yielding 325 pg of DNA from B and 230 pg each from A and C. Water was added to make each volume 500 pL. Organic Extraction. The cellular and calf thymus DNA samples from the above procedures were extracted once with 500 (11)Crespi,C.L.;Gonzalez,F.J.;Steimel,D.T.;Turner,T.R.;Gelboin,
H. V.;Penman, B. W.; Langenbach,R. Chem.Res. Toricol. 1991,4,666572.
ANALYTICAL CHEMISTRY, VOL. 65, NO. 13, JULY 1, 1993
-
B(a)P Exposed Cells 1. Isolate DNA 2. H+ : Tetrol
release
--
3. Add Internal standard: de-Ketone
4. KO:! : Tetroi Diacid 5. PFBzBr: Diacid Diester 2 6. Si Cartridge: Cleanup 7 . GC-ECNI-MS
Amount of Adduct F w r r 2. Analytical scheme. pL of water-saturated isoamyl alcohol and once with 500 pL of ethyl acetate. The volume of each aqueous sample was reduced to about 100 pL under nitrogen, and each volume was measured accurately by pulling up into a 100-pL syringe. Acid Hydrolysis. Each sample volume was adjusted to 450 pL with water, treated with 50 pL of 1.2 N HC1, and heated at 90 "C for 3 h with occasional vortexing. Sample C was treated with 128 pg of tetrol. The samples were evaporated to dryness at 60 O C under nitrogen, followed by addition of 100 pL of acetonitrile and reevaporation. KO2 Oxidation and Pentafluorobemzylation. To two empty vials (blanks) was added 40 pL of methanol, and to other empty vials were added 20 pL of methanol containing 128 pg of tetrol plus 20 pL of methanol containing 106 pg of &-ketone (internalstandard), followedby evaporation. Similarlydgketone was added to most of the other vials (see Table I) followed by evaporation under Nz at 60 OC. Dimethylformamide, 25 p L containing 260 pg of potassium superoxide and 260 pg of 18 crown-6 (freshly prepared mixture), was added to each of the vials, followed by continuousvortexing at room temperature for 18 h. After the addition of 500 pL of 1% acetic acid followed by evaporation,500 pL of 10% triethylaminein toluene was added. After evaporation, 50 pL of toluene containing 2 pL of pentafluorobenzyl bromide and 2.5 pL of triethylaminewas added. Each vial was vortexed at 50 "C for 5 h and then overnight at room temperature. After evaporation, 100 p L of hexane was added. After vortexing and evaporation, 500 pL of hexane was added followed by vortexing and then loading of each sample onto a silica cartridge column with a pasteur pipet. The column was washed twice with 2 mL of hexane taken through the vial. The product (diester 2) was eluted with 2 X 1 mL of ethyl acetate (taken through the vial), followed by evaporation under Nz at 60 OC. To focus the product at the bottom of the vial, 100 pL of acetonitrile was added followed by vortexing and evaporation. Just before injection of 1pL into the GC/ECNI-MS, 20 p L of acetonitrile was added and the sample was vortexed.
RESULTS AND DISCUSSION The analytical procedure that we used to measure acidlabile, diol epoxide B[a]P DNA adducts in a cell culture of human lymphocytes exposed to B[a]P is summarized in Figure 2. In step 1,the DNA is purified from the cells by a procedure involving cell lysis with detergent, enzymatic digestion, phenol extraction, and finally extraction of the dissolved DNA with isoamyl alcohol and ethyl acetate. The purpose of the latter extraction is to complete the removal of any residual, noncovalently bound B[alP and its metabolites. A known procedure7 is then used to release benzoblpyrene7,8,9,10-tetrahydrotetrol (tetrol) from the DNA, and the resulting solution, after addition of an internal standard (deketone), is evaporated and subjeded to oxidation by potassium superoxide followed by pentafluorobenzylation to afford the bis(pentafluorobenzy1)ester of pyrene-2,3-dicarboxylicacid (diester 2) and its deanalogue &diester 2), respectively(steps 2-5). After purification of the resulting sample on a small column packed with silica,quantitation is done by GC/ECNIMS with selected ion monitoring (SIM).
1725
Prior to discussing the results, we wish to point out how the method has been modified relative to an earlier version which we used to determine a standard of the t e t r ~ l In . ~ the previous procedure, a high level of background interferences was encountered in the sample of diester 2 even though no DNA was present and the diester 2 was derived from pure tetrol. This necessitated purification of the final sample by double passage through an HPLC column prior to injection into the GC/ECNI-MS. Since then we have tuned some of the conditions in the method (especially the amounts and use of the acetic acid and triethylamine), substituted toluene for the previous acetonitrile,'Z introduced an all-glass vial,lo and employed continuous vortexing to establish the more convenient procedure along with cleaner final product that is reported here. Now a simple solid-phase extraction is adequate for postderivatization sample cleanup prior to GC/ ECNI-MS, at least at the level of sensitivity tested to date. This is in spite of the fact that biological samples now are being tested. The results of our method applied to B[alP-exposed lymphocytes, including controls (unexposed cells and also calf thymus DNA) and standards (pure tetrol) from a representative experiment are shown aa absolute values in Table I. Representative GC/ECNI-MS chromatograms are shown in Figure 3. As seen in Table I, except for just one of the duplicates (vials 11 and 121, the precision is very high, taking into account the sample complexityand level of analyte. Remarkably, the precision is not improved, or only somewhat, by use of the internal standard (data not shown). Perhaps interferences in the detection step by GC/ECNI-MS are the major sources of error. This includes the possibility of matrix effects in the ion S O W C ~ especially ,~~ because of the slightly earlier elution of de-diester 2 than diester 2. The absolute yield of the method is satisfactory: 47 % (average of 46 and 47%) for the conversion of 128 pgs (390 fmol) of standard tetrol to diester 2 and 22% (average of 17 and 27%; lower precision data from vials 11and 12) when this same amount is first spiked into calf thymus DNA. Corresponding yields of de-diester 2 from the de-ketone are 39% (average of 38 and 40% from standard vials 15 and 16) and 25 f 3.8% (Xf SD for nine values from all the samples containing DNA to which de-ketone was added). Thus the yield is lower when DNA is present, probably because the acid-released tetrol is not separated from the DNA prior to oxidation by KOz. Of the 16samples in the experiment, 7 were blanks in which no diester 2 should be found. Nevertheless, a consistent amount of an interfering peak corresponding to diester 2 (ranging from 6.2 to 13.5 fmol/vial as seen in Table I) was found in these Samples. We have previously discussed this interference including data suggesting that it is true diester 2 and establishing the oxidation reaction as its origin.10 The relatively small, constant value of this background peak, along with its independence of type or presence of DNA, allows us to subtract its value from that for the diester 2 in the other samples. A task for our future work is to fully remove this interference. Taking into account the above considerations concerning blank correction and yield (22%) of diester 2, we calculate that the DNA of the exposed cells contains 177 fmol of acidlabile diol epoxide B b I P adducts per 114 pg of DNA. Using 330 as the molecular weight of an average nucleotide in DNA, this adduct level correspondsto five adducts in l o 7 nucleotides. This level of sensitivity is adequate for the method to be applied to some human samples, such as lung samples from smokers.6 (12) Allam, K.; Abdel-Baky, S.; Gieee, R. W . Anal. Chem. 1992,64, 238-239. (13) Tondeur, Y.; Albro, P. W.; Hass, J. R.; Hanran, D. J.; Schroeder, J. L. Anal. Chem. 1984,56,1344-1347.
1726
ANALYTICAL CHEMISTRY, VOL. 65, NO. 13,JULY 1, 1993 I o n 469.00
22000 20000
amu.
k-1
1
f r o m
DRTR:CONTROLI.D
NONEXPOSED CELLS
18000
Ion 469.00 a m u . 1 e000j 16000
f r o m DRTR:BIOLOG2.D
1.
f
Jvy
140001
EXPOSED CELLS
16000 14000 12000
12000j
i
10000
10000 0 0
c
u
8000
6000
6000 Ion 477.00 a m u .
from
DRTf3:CONTROLl.D
!
Ion 477.00
am".
f r o m
DRTR:BIOLOG2.D
I e000j
160004 140004
10000
e0007 6000i
12.0 Time
14.0
16.0
,
,
,
,
,v,i
12.0
18.0
Tlme
(mln.)
16.0
14.0
18.0
(mln.)
Ftguro 5. Selected Ion QWECNI-MS chromatogramsfrom nonexposed and B[e]P+xposed cells: (A) rnlz 469 to detect diester 2 (peak 1)and apparent dlester 2 (peak 1'); (B) mlz 477 to detect ddlester 2 (peak 2). A long-range goal is to use this method to detect other acid-labile diol epoxide PAH DNA adducts, including unknowns. For an unknown, the MS would be operated first in a "spectrum mode" (done by repetitive fast scanning on our instrument), to search for distinctive ion peaks not present in a control sample. Such peaks would represent candidate adducts of this type. While they would remain as unknowns at this stage (and scanning is less sensitive than SIM; see below), at least the peaks could be detected subsequently with high sensitivity by SIM. To begin to explore the extension of our method in this way to other KOZ-susceptible PAH DNA adducts, we tested it on a standard chrysene-1,4-quinone(31, at the low-picomole level (725pg). We have already observed that this compound is oxidized by KO2 to 1,2-dicarboxyphenanthrene.14 Here the final sample WBB spiked with a similar amount of standard diester 2, and an aliquot (1:lO) was then injected into the GC/ECNI-MS under scanning conditions. This gave the reconstructed total ion chromatogram shown in Figure 4. The mass spectra of the more prominent peaks were displayed (data not shown, but more details are provided in the caption for Figure 4). Peaks g and n correspond to compound 4 and diester 2, respectively. The inset shows the detection of these latter compounds by reconstructed selected ion monitoring. As seen in Figure 4, scanning was done from rnlz 350 to 550. This range was selected since it would encompass two to five ring PAHs analogous to diester 2. For this range of scanning, it was found that the response (peak height) was about 100-fold lower for the detection of 2 and 4 (asa combined sample of standards in a separate experiment) relative to their detection by SIM with dual ion monitoring. We also subjected 100 pg of chrysene-1,4-quinone, in duplicate, to steps 4-7 of Figure 2, and also 100 pg of 1,2dicarboxyphenanthrene similarly to steps 5-7. The yields of 4 from these two experiments were 71 and 77% (duplicates) and 68 and 87 7% , respectively, using 2 as the calibrant and assuming, arbitrarily, equal molar responses for 2 and 4. The method is successful, then, as designed, in measuring related
Feasibility has been demonstrated for the ability of a new, general method to detect diol epoxide PAH DNA adductsfn a biological sample. The long-term goal is to detect unknown adducts of this type with high sensitivity. Ultimately this might be done best using MS equipment which detects multiple and single ions with equivalent sensitivity, such as one equipped with an array detector.15
(14) Sotiriou, C.;Li, W.; Giese, R. W. J. Org. Chern. 1990,55, 21592164.
(15) Hill, J. A.;Biller, J. E.; Martin, S. A.; Biemann, K.; Yoehidome, K.;Sato, K.Znt. J. Mass Spectrorn. Zon Processes 1989, 92,211-230.
0.0cr0-
, B
, 9
,
,
10
I 1
, I 2 l i m e
.
13
----
!4
:5
:6
17
IB
< m , n o
Figure 4. Reconstructed total ion GClECNI-MS chromatogram (scannlng mode, rnlz 350-550) obtained as follows: (1)subject 725 pg of 3 to steps 4-6 of Figure 2;(2)add 824 pg of diester 2; and (3) inject 1 pL from a total volume of 10 pL into the GWECNI-MS. Inset: reconstructed selected ion chromatogramsat rnlz 445 (top)and 469 (bottom). Peak g corresponds to 4 (and a second compound; see below) and peak n to diester 2. The mass spectra of the lndlcated peaks each consist of essentially a single ion (alongwith correspondlng isotopic abundance Ions) except for peak g, whlch consists of two ions, as follows (peak letter; mlz): a pel),b (381),c (397),d (395), e (4091,f (425),g (399,4454,h (423),i (439),j (398),k (437),I (453), m (451),n (469). adducts, since all of the steps prior to final detection by GCI ECNI-MS process such adducts as a group.
CONCLUSION
ANALYTICAL CHEMISTRY, VOL. 85,
We can detect a standard of diester 2 a t the low-attomole level by GC/ECNI-MS in the SIM mode (13.8 am01 was detected a t S/N = 19): and the amount of tetrol encountered here is 177 OOO amol/ll4 pg of DNA. Thus we plan to extend the method in the future to DNA samples containing a smaller amount of this and related adducts, which will require continued effort to minimize losses and interferences.
ACKNOWLEDGMENT We thank Alan Jeffrey and Charles Crespi for valuable discusaions. The research described in this article was conducted in part under a contract to the Health Effects
NO. 13, JULY 1, 1993 1727
Institute (HEI), an organization jointly funded by theunited States Environmental Protection Agency (EPA) (Assistance Agreement X-812059) and automotive manufacturers. The specific grant was HE1 Research Agreement 86-82. The contents of this article do not necessarily reflect the views of the HE1 nor do they necessarily reflect the policies of the EPA or automotive manufacturers. Contribution No. 574 from the Barnett Institute.
RECEIVED for review December 14, 1992. Accepted March 9, 1993.
