Tobacco Documents Online

Topical Report NCI/S&HP/ORNL #41 SITE VISIT I TO BATTELLE PACIFIC NORTHWEST LABORATORIES RAT MODEL INHALATION BIOASSAY Richland, Washington, February 23-24, 1977 5-27-77 R. A. Jenkins and B. E. Gill Tobacco Smoke Research Program Bio/Organic Analysis Section Analytical Chemistry Division Oak Ridge National Laboratory Oak Ridge, Tennessee 37830 Interagency Agreement (ERDA-NIH/NCI) No. 40-485-74, Part II Internal (ORNL) Contract•Charge No. 3390-0224 Intended for informal communication with project management only. Confidential until published or released by author.

Site Visit I to Battelle Pacific Northwest Laboratories Rat Model Inhaltion Bioassay R. A. Jenkins and B. E. Gill Summar~y. Data gathered on the first visit to Battelle Pacific Northwest Laboratories (BNW) to monitor the rat model inhalation bioassay suggest that the smoke dose which the animals receive is similar in magnitude - when normalized for body weight - to that which dogs in other NCI sponsored inhalation bioassays are offered. The overall quality of the operation and exposure procedure at BNW is very high. However, comparison of detailed analyses of the on-site smokes with those generated analytically suggest that significant changes in the chemical profile of the Code 27 variant may occur under exposure conditions. This variant has also exhibited large month-to-month variations in total particulate delivery. Introduction. A first site visit was made to Battelle Pacific Northwest Laboratories (BNW) on February 23-24, 1977. The primary purpose of the visit was to °'shake down° monitoring procedures, identify future monitoring needs, and discuss and observe the rat inhalation bioassay being conducted there. Because of the helpful and cooperative attitude of the BNW staff, we were able to obtain nearly 60 samples of smoke of the variants in use at the time of the study. The purpose in obtaining these samples was to estimate the amount of smoke being offered and retained by the animals, and to obtain some measure of the chemical profile of the smoke being generated on-site and offered the animals. Experimental. Details of the exposure protocol can be found in the BNW Annual Progress Report, "Inhalation Bioassay of Cigarette Smoke in Rats, January 1- December 31, 1976." Briefly, eighty (80) animals per variant began full dose exposure in May, 1976. Each animal is exposed to her share (ten animals per chamber) of the smoke from eight cigarettes per day, spaced fairly evenly over the course of 12 hours, seven days per week. Animals smoking the Code 27 variant are exposed to extra puffs, as that particular variant has a lower average number of puffs. To collect smoke particulates generated by the experimental variant, a Cambridge filter pad is installed between the cigarette and the exposure Cl~ GD ~ chamber. Grab samples of smoke inside the exposure chamber are taken by .~ ~ (b ~

withdrawing a 20-m1 sample immediately following introduction of smoke into the chamber (at 3 seconds into the one minute exposure cycle), and immediately prior to the flushing of the smoke from the chamber (at 27 seconds into the exposure cycle). The 3 second sample is designed to provide a measure of the amount of smoke available to the animal for inhalation. The 27-second sample measures the amount of smoke remaining in the chamber. The smoke is drawn through a small filter pad, and the nicotine present is taken as an in- dicator of the amount of smoke sampled. Presumably, the difference in the two samples is a measure of the amount of smoke actually inhaled and retained by the animals. There are some difficulties associated with this method of dose documentation, and these will be discussed later. Smoke samples obtained at BNW were returned to ORNL for analysis. Nico- tine is determined by extracting the filter pad with ethanol and analyzing an aliquot of the extract by gas chromatography. Other analyses (referred to as a multi-aliquot procedure and discussed later) are performed by extracting the filter pad with pyridine, subjecting small aliquots to a tri-methyl silylating procedure and analyzing these derivatived aliquots under varying gas chroma- tographic conditions. The multi-aliquot procedure permits a much more detailed chemical description of the smoke, as several components correlate highly with the presence of entire classes of smoke constituents. General Observations. The overall impression of the ORNL site visit team was that the staff at RNW is doing a very good job of exposing the animals to a reproducible amount of smoke, given the inherent limitations of the exposure device and the animal model. While no cigarette static burn tests were made on this visit (there was a problem with the balance used to weigh cigarettes - see below), we noted that environmental conditions in the exposure wards were as close to standard analytical smoking conditions (76°F, 60iu relative humidity) as any we have observed in a chronic exposure situation. The temperature in both exposure wards was recorded to be about 74°F, with a 65-72% relative humidity. Ventilating air flow across the cigarettes was minimal, yet expo- sure technicians commented that the fresh air supply was more than adequate for comfort. During our observation of the exposures, we noticed that there are a sufficient number of technicians to handle the machines and animals. We believe this to be an integral part of the overall quality of the 5NW exposure.

The technicians have sufficient time to actually observe the animals during exposure, such that problems that arise during the smoking process receive immediate attention. The animal containment tubes are kept clean. Careful attention is paid to exposure protocols, such that animals smoking the Code 27 variant receive the proper number and type of extra puffs. Smoke exposure machines are kept clean and in good working order. Communication between the exposure technicians and the professional staff appears to be good. Results and Discussion. Measurement of machine puffing parameters (frequency, duration, volume) were not obtained on this visit. The Maddox-ORNL exposure device does not have a fixed puffing volume. Rather, the pressure inside the exposure chamber is decreased such that a 35 ml puff is drawn through the cigarette placed at the front of the chamber. Of course, this net puff volume can fluctuate with differences in resistance-to-draw of the cigarette or minor leaks in the system. The protocol at BNW is to frequently recheck net puff volume on a routine basis, to insure conformance to smoking standards. During the visit, a problem developed with the analytical balance, such that it would not weigh small masses (such as individual cigarettes) with acceptable accuracy. Thus, contrary to our normal procedures, no weights for cigarettes used in the on-site experiments were obtained. However, it has been our experience that random cigarette selection yields averaqe weights differing little from group averaged weights. The total particulate matter (TPM) and nicotine values of the smoke generated by the variants on the Maddox-ORNL machines at BNW are tabulated in Table I. Table II lists the average values for each variant in comparison with those from the same batch returned to ORNL and smoked under analytical conditions. For the Code 32 and 27 variants, TPM and nicotine levels obtained on the exposure devices are substantially above those obtained under analytical conditions (Nicotine content of the Code 27 smoke is 65% greater than the value obtained analytically). Values obtained analytically from the site visit cigarettes are in good agreement with those obtained from the Tray Quality Control cigarettes (See Topical Report x 40). Thus, smoking conditions in the exposure wards at BNW probably contribute to an increased TPM and nicotine delivery for the two variants (Code 13 values are elevated, but not so seriously We would tend to rule out environmental factors (temperature, humidity, air flow) because they are relatively close to standard analytical conditions.

Any alteration in the puffing performance of the exposure device to increase the proportion of the total cigarette smoke which is collected (longer puff duration, large puff volume) would be expected to be accompanied by a similar reduction in the number of puffs. Such is not the case here. BNW has not observed this phenornenen in their exposure ward tests. A conclusive answer must await further on-site documentation. As part of the monitoring protocol for the BNW bioassay, cigarettes are sampled from those destined for use in the exposure and shipped to ORNL on a monthly basis for analysis. This not only aids in better interlaboratory comparison of bioassays, but also provides a measure of potential variation in smoke dose to the animals resulting from fluctuations in cigarette smoke delivery. Table III summarizes the results of these monthly analyses. The Code 32 cigarettes show the least variation in delivery. For the months of June and October, 1976, the Code 27 tar delivery was considerably below its average value. The data here suggests that even for an inhalation bioassay which is relatively well controlled, potential smoke dose variations due to fluctuations in cigarette delivery may be expected. Table IV summarizes chamber depletion data for the three variants as measured on this visit. Nicotine values are expressed in terms of concen- tration of nicotine (ug/ml) in the atmosphere available to the rats. Given the inherent difficulties of measuring inhomogeneous concentrations of "live" smoke, interpretation of chamber depletion measurements is by no means straightforward. As the smoke remains in the chamber, it tends to plate out on the chamber walls, reducing its effective concentration thorugh a non- lung-deposition mechanism. In addition, nose seals are broken, exposing the chamber contents to "fresh" air in the animal containment tubes, reducing smoke concentration via dilution. Also, some smoke particulates condense on the animals' fur. Presumably, chamber nicotine concentrations determined from samples made only one second after the end of the puff (the 3-second sample) should not have been affected by depletion through in-chamber conden- sation, dilution, or lung deposition, and therefore should be a good measure of the amount of smoke actually getting'into the chamber. Any loss of nicotine presumably would be due to condensation of smoke particulates within the inlet to the exposure chamber. Table V compares per cigarette deliveries of nicotine calculated from the 3 second chamber grab samples and those determined by GD

collecting the smoke as it leaves the butt of the cigarette. For the Code 27 and 32 variants, the data suggest that nearly half of the particulate phase of the smoke condenses within the chamber inlet. Some inlet condensation of smoke was observed, but we would estimate this amount to be much less than 50%. Since condensation and lung depositon of smoke are known to be time dependent, the conclusion is that dilution of the smoke with "fresh" air from the animal containment tubes must occur more rapidly than originally anti- cipated. (Documentation of the actual amount of smoke drawn into the inhalation chamber following completion of the puff - and thereby available for inhalation - is important. We are presently developing a method by which the entire chamber contents may be sampled for nicotine and carbon monoxide, and expect to initiate such sampling on the next site visit.) Despite the difficulties with the chamber depletion measurement, it is still possible to estimate, using chamber depletion data, the amount of smoke retained by the animal. Some preliminary dosimetry experiments in the ORNL rat inhalation bioassay showed that about 76% of the smoke depleted from the chamber - as measured by the chamber depletion method - deposited in the respiratory tract of the animal. (This number may vary greatly for different cigarette codes, depending on the "inhability" of the smoke.) Table VI com- pares estimated weekly nicotine doses based on the chamber depletion data on Table IV. (The Code 27 data has been adjusted for two additional puffs per cigarette.) Interestingly, when body weights are normalized, rats smoking the Code 32 variant at BNW receive about the same order of magnitude of smoke as the Hazleton Laboratory dogs being exposed to the same variant (20 mg per week per kg of body weight for the 200 g rats vs ,,14 mg for the 10 kg dogs). One of the important functions of bioassay monitoring is to determine the extent to which (if any) the smoke being offered the animals is an artifact of the method of exposure (i.e., does the smoke being generated on-site differ significantly in chemical composition from that generated under analytical conditions). In an effort to more exactly define the chemical composition of the particulate phase of the smoke being offered the animals, chamber input samples were subjected to a multi-aliquot analysis procedure. Tables VII A-C compare results obtained for the on-site samples with those obtained by smoking cigarettes returned from BNW under analytical conditions. Comparison ~ ~ C 1 J M aD GID

is made between the absolute levels of the given constituent in the smoke, and also between levels which have been normalized to a constant nicotine delivery. For the Code 32 variant, smoke generated on-site shows little difference in composition from that generated analytically. Such is not the case for the other two variants. Code 13 smoke generated by the Maddox-ORNL contains substantially more neophytadiene than "analytical" smoke. The Code 27 °on-site® smoke exhibits a radically different chemical profile than the analyt- ical smoke. The "on-site" smoke appears to contain substantially lower levels of the free fatty acids, and significantly higher levels of neophytadiene, phenol, and the cresols. As suggested by, the wide differences in the nicotine deliveries, the Code 27 cigarettes used at BNW on this visit may not be truly representative. However, the data does indicate that significant changes in smoke chemistry of the Code 27 may occur under exposure conditions at BNW. Confirmation of this will be necessary on future site visits.

TABLE I Total Particulate Matter (TPM) and Nicotine Delivery of Cigarettes Smoked on Selected Maddox-ORNL Smoking Machines in Service at BNW on February 24, 1977 Cigarette Code Platfornz Desjqnati on Sa r~1 l e No. TPM mq/ci c~ Nicotine mg/ci g Nicotine:TPM Rati o _ _ _ __ 32 I 1 36.6 2.22 .0607 I 2 39.4 2.35 .0596 I 3 39.0 2.66 .0682 I 4 32.8 2.24 .0683 G 5 45.1 2.22 .0492 G 6 45.7 2.35 .0514 G 7 38.5 2.55 .0662 G 8 38.8 2.53 .0652 Average 39.5 ± 4.2 2.39 ± 0.16 .0611 ± 0.0074 27 I 1 23.8 1.59 .0668 I 2 21.8 1.47 .0674 I 3 26.8 1.61 .0601 I 4 19.6 1.28 .0653 H 5 19.6 1.24 .0633 H 6 27.0 1.51 .0559 H 7 24.7 1.55 .0628 H 8 24.6 1.53 .0622 Average 23.5 ± 2.9 1.47 ± 0.14 .0630 ± 0.0038 13 1 1 29.6 0.23 .0078 1 2 29.9 0.23 .0077 1 3 31.5 0.25 .0079 1 4' 29.0 0.22 .0076 G 5 31.0 0.22 .0071 G 6 27.5 0.22 .0080 G 7 33.3 0.23 .0069 G 8 28.9 0.22 .0076 Average 30.1 ± 1.8 0.23 ± 0.01 .0076 t.0004

TABLE II Comparison of Cigarette Deliveries Input to Maddox-ORNL SmoLi ng Machi nes at BNW vs Analyti cal Smoki ng Machi ne Maddox-ORNL Analytical* Cigarett Code e # Puffs TPM mg/cig Nicotine mg/cig # Puffs TPM mg/cig Nicotine mg/cig 32 10.3 ± 0.5 39.5 ± 4.2 2.39 ± 0.16 9.8 ± 0.5 34.1 ± 2.1 1.92 ± 0.13 2 7 7.6 ± 0.5 23.5 ± 2.9 1.47 ± 0.14 8.0 ± 0.2 17.7 ± 1.2 0.89 ± 0.06 13 9.1 ± 0.6 30.1 ± 1.8 0.23 ± 0.01 9.6 ± 0.5 27.9 ± 1.6 0.20 ± 0.02 *Average results for 8 cigarettes per code smoked on the Phipps and Bird smoking machine at ORNL at 1 puff/minute, 1 cigarette per pad.

TABLE III TPM, Nicotine, and Tar Deliveries of Cigarettes Sampled Monthly from BNW Rat Inhalation Bioassay Assayed Under Analytical Conditions at ORNLa DATE amponent Cigare~te Code May ' 76 June July August September October November Dec. '76 Jan. '77 Feb. '77 PM mg/cig 32 33.1+0.8 32.0±0.9 29.3±0.8 31.7-+0.9 31.7±1.8 31.9±2.7 32.9=0.7 32.7±1.5 33.5_1.8 27 1'5.4_0.2 ll.l±1.i 17.5+0.1 15.3±0.4 15.3±0.4 10.7±1.6 16.1±1.4 15.2±1.7 16.9±1.9 18.0±1 .2 13 28.4±0.2 21.5±0.6 27.5±1.3 25.7±0.9 25.7±0.9 24.5±5.8 33.7±8.3 27.7~~0.9 28.9±0.8 28.4± 0.1 icotine mg/cig 32 1.84±0.09 1.79+0.04 1.66±0.10 1.81T0.06 1.77±0.03 1.66±0.17 1.67±0.02 1.58±0.13 1.93±0.10 27 0.85±0.01 0.93±0.06 0.98±0.01 0.75±0.02 0.90±0.01 0.624--0.02 0.86±0.04 0.74±0.04 0.71±0 .04 1.04±0.01 13 0.1910.01 0.13+-0.01 0.17±0.01 0.15±0.01 0.19+-0.01 0.16±0.01 0.19±0.01 0.18±0.01 0.19±0.0 1 0.18±0.01 c ar m91ci9 32 27.9+0.7 26.3±0.8 24.6±0.8 26.2±0.08 26.3±1.6 26.7±2.8 27.6_0.5 27.7±1.1 27.6±1.6 27 11.6±0.2 7.3±1.2 12.8±0.4 11.7±0.2 10.6±1.8 8.3±1.4 12.3±1.0 11.9+1.2 13.9+1.9 13.2+0.9 13 24.7=0.1 18.9±0.5 23.3±0.7 22.3±0.7 21.5±5.6 30.0±2.2 22.9±2.2 24.1+0.7 25.0±0.08 24.7±0 .1 a 8 cigarettes per code for Codes 13 and 27, 16 cigarettes for Code 32. b Code "32" is used to represent all of the SEV IV Variants (04,14,29,32). C Tar = TPM less water and nicotine. z694C4se _: . ~ .~. .. ., ~.,