Tobacco Documents Online

I I 1_' CHARGE NUMBER: PROGRAM TITLE: PERIOD COVERED: PROJECT LEADER: WRITTEN BY: DATE OF REPORT: 6908 SMOKE CONDENSATE STUDIES July 1-31, 1981 R. N. Ferguson A. Warfield August 7, 1981 A. SALT EFFECTS Gas evolutiona data obtained by R. Fenner14 have shown differences between RL base web controls and KOAc and' KC1 treated samples. Profiles for C02, H20, HOAc, and MeOH revealed differences in transitions for carbohydrate and pectic decomposition. The addition of KOAc resulted in increased ammonila evolution at lower temperatures (240-320°C), followed by a maxilmum at 360°, and a decrease in evolution above 400°. The addition of KC1 shifted the maximumm sliightly, but did not change the profile. Changes were also observed in formic and acetic acid profiles in the 220-340° range. Addition of KOAc caused a decrease in formic acid and'increase in acetic acid evo- lution, presumably related to release of the weaker acid from its salt, and/or decomposition around the melting point of KOAc (292°).1 A low solubles base web has been sprayed with KOAc and' KC1 solutions in 300 g batches along with a control.1 Cigarettes will be smoked and base fractions isolated to facilitate studies of changes in basic components due to salt effects. Attempts will be made to confirm previously observed changes in SaZnaneZZa/microsome activi,ty and in decreased condensate and gas phase del~iveries due to sal~t effects. B. MW 288 L. Decomposition products of a-4,8,13-duvatri.ene-l,3-diol were analyzed by gc/ms (N. Einolf). Several MW 288 isomers were indicated, along with lesser amounts of MW 270 materials. Normal phase HPLC was used to isolate three compounds, one of which was identified by 'H. NMR as (5S)-5-isopropyl-8,12-dimethyl-3E,8E,12E,14-pentadecatetraen- 2-one.2 C. NITROSAMINES Problems with poor resolution of nonvolatile nitrosamines on 3% SP2250 appear to result from a bad lot of packing. A new lot of this packing has been obtained, and is currently being evaluated. Six RL type cigarettes submitted by Project 6900 wereismoked and analyzed'for volatile nitrosamines in mainstream smoke. Nonvolatile nitrosamines will be determined on these samples when satisfactory gc conditions are reestablished.3

Charge Number 6908 2 August 7, 1981 D. PAH PROCEDURE Development work on the PAH cleanup method'was continued.4 A much cleaner prep-scale PAH fraction was obtained from 2R1 IT CSC (MS) using a 35/70 mesh Si-40 column and a small hexane fore-cut. Compared to standards, low MW PAHs eluted early, indicating a deacti- vation of the highly active silica by the CSC hexane extract. Slight adjustments will probably be made in the LH-20 step, and a third purificati'on step remains to be investigated. Substitution of Si-60 for Si'-40 will be tried' also. GC scans of preparative fractions are now quite similar to TLC/LH-20 analytical scans. E. CARBOLINES IN SMOKE (with 6906) Results from the SaZmonetta/microsome assay were obtained5 on silica LC fractions from an acti.ve subfraction of X6D3IM Batch #3, LH-20 co1'. #2, Fr. #7.6 One of the fractions showed high activity, and a single major component was indicated by LC profiling. GC/'MS analysis showed that the component, MW 270, was apparently identical to a compound isolated previously.7 Another fraction was much more active, but consisted of multiple components. GC/MS indicated several MW 235, 244 and 258 compounds.8 Mass spectra were indicative of fused ring systems, some being alkyl substi- tuted. SaimoneZZa/microsome activity profiles5 of small scale LH-20 column fractions from X6D3IM toluene extract showed that later fractions are more active, and that activity can be concentrated' with this method. The 53 g of remaining X6D3IM pot residue was, therefore, carried through the toluene/ 50% methanol partition step, and the tol,uene extract chromatographed on LH-2'0 in three portions.6 When SaZmoneZta/microsome data are complete, the most active fractions will be combined for further fractionation. A portion of the 50% methanol extract was also chromatographed on LH-20 to determine if the SalmnneZZa/tinicrosome activity would'follow a similar pattern. Results to date indicate that the weight distribution and activity patterns for the two extracts on LH-20 are similar. Two of the later LH-20 fractions from the 50% methanol extract had accountabilities of 125 and 175% based on the activity of the extract. Apparently an interaction occurs in the latter, which masks the activity until the interacting components are removed. Reference samples of Glu-P-1 and Glu-P-2 were analyzed by gc/ms,8 anda mass spectral difference presumably due to the presence or absence of a methyl group was observed. Earlier profiles were examined for indications of these compounds, with no success. Analytical and'preparative HPLC conditions were worked out for Glu-P-1 and Glu-P-2.9 Using an Ultrasphere ODS column and a solvent system similar to that employed for the 2AC/H/NH analysis,10 a number of base fractions were.profiled using fluorescence conditions specific for G1u-P-1 and Glu-P-2. X6DOAD and X6D31M were chosen as likely candidates for

Charge Number 6908 3 August 7, 1981 I I I Ii preparing enriched fractions so that the presence of these compounds can be confirmed. Analytical HPLC of spiked X6D0AD and X6D3IM IT CSC base fractions was used to verify the retention times of Glu-P-1 and Glu-P-2 peaks previously observed. After cleanup on a C-18 Sep-pak cartridge, preparative HPLC was carried out on unspiked samples of X6DOAD and X6D3IM base fractions using a M-9 ODS-3 column. Fractions corresponding to the retention volumes of Glu-P-l and Glu-P-2 were collected. These were further separated by TLC on silica gel GF (250 u) using CHC13/acetone/MeOH, 80:20:5. The G1u-P-1 and Glu-P-2 bands were eluted and profiled by HPLC. The large number of components still present indicated a need for further cleanup prior to gc/ms. A normal phase HPLC step is being investigated currently. Samples of Trp-P-1 and Trp-P-2 received recently from Japan were analyzed by gc/ms using a glass 3% SP-2250DB coiumn.11 The retention time and spectrum for Trp-P-2 were comparable to those obtained earlier on a samplie synthesized in-house. In the continuing attempt to find Trp-P-l and Trp-P-2 iin CSC, TLC fractions of cellulose column cuts, obtained by chromatography of X6D3IM toluene extract, were subjected to analytical HPLC, and peaks at the retention times of Tr1p-P-1 and Trp-P-2 were scanned using the PE- 650-10-S spectrofluorimeter. 2 The spectra were complex, and positive confirmation of the presence of these compounds could not be obtained. However, an emission band 'in the Trp-P-1 peak found iin the NaC1/ethanol/ H20 fraction was in the correct position for Trp-P-1. SIM gcfms will be attempted' on this sample and others in this series to determine whether these compounds can be detected. Recovery of SaZmoneZZa,/mi'crosome activity5 i'n the cellulose chromatography of X6D3IM toluene extract was very low (n.14%), and the distribution of activity was very different than reported for broi1ed'fish, in which Trp-P-2 was found using this procedure.13 Therefore, other cleanup methods will be considered if further attempts are made to find Trp-P-1 and Trp-P-2 in CSC. Two major components of the NaC1/ethanol/H20 cellulose column fraction, after preparative TLC, were isolated by preparative HPLC12 and tested i'n the SaZrnaneZZa/microsome assay by 6906 personnel.5 No activity was detected at the doses employed, indicating that these compounds are less active than 2AC. The fluorescence spectra were nearly identical, though the compounds were well separated by HPLC.

Charge Number 6908 4 August 7, 1981 F. REFERENCES 1. Gager, F. Notebook No. 7375, p. 196. 2. Katz, T. Notebook No. 7679, p. 9. 3. Baker, G. Notebook No. 7545, p. 90. 4. Levins, R. Notebook No. 7351, p. 200. 5. Drew, S.; McCoy, W. Notebook No. 7596 (14). 6. Tafur, S. Notebook No. 7581, p. 82. 7. Tafur, S. Notebook No. 7423, p. l81'. 8. Kinser, R. Notebook No. 7476, p. 121. 9. Ellis, C. Notebook No. 7592, p. 147. 10. Ellis, C.; Tafur, S.; Warfield, A. Development of an HPLC/fluorescence quantitative assay for 2-amino-a- carboiine, norharman, and harman in cigarette smoke condensate. Special Report 81-1,25; 1981 June 8. 11. McKay, C. Notebook No. 7477, p. 117. 12. Warfield, A. Notebook No. 7650, p. 28. 13. Yamaizumi, Z.; Shiomi, T.; Kasai, H.; Nishimura, S.; Takahashi, Y.; Nagao, M.; Sugimura, T. Detection of potent mutagens, Trp-P-1 and Trp-P-2, in broiled fish. Cancer Letters 9:75-83, 1980. 14. eF nner,-ff-.-K.-- Effect of potassium salt addition on the pyrol:lytic decomposition of RL bass web. Memo to F. Gager; 1981 July 7. nwp