ToxSci Advance Access originally published online on December 22, 2006
Toxicological Sciences 2007 96(2):285-293; doi:10.1093/toxsci/kfl195
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Effect of Smoking Conditions and Methods of Collection on the Mutagenicity and Cytotoxicity of Cigarette Mainstream Smoke
* Labstat International Inc, Kitchener, Ontario, Canada N2C 1L3
Lauterbach and Associates, LLC, Macon, Georgia 31210-4708, USA
1 To whom correspondence should be addressed. Fax: +519-748-1654. E-mail: wrickert{at}labstat.com.
Received August 17, 2006; accepted December 20, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
There is a history for the use of in vitro bioassays to assess the toxicological properties of mainstream cigarette smoke (MSS). The results described in the literature were, for the most part, obtained with MSS collected under Federal Trade Commission (FTC) or International Organization for Standardization (ISO) conditions. However, numerous studies have shown that smokers smoke their cigarettes more intensely (e.g., they take larger puffs and/or more frequent puffs and/or partially occlude filter ventilation) than they are smoked on smoking machines operated under FTC (or ISO) conditions. It has also been reported that MSS composition changes with changes in smoking conditions. Furthermore, some governmental agencies have adopted regulations that specify more intensive protocols (i.e., Health Canada Intensive, HCI) for the collection of MSS for in vitro toxicological assays. Consequently, the performance of the Ames assay (TA98 + S9, TA100 + S9) and neutral red uptake assay under ISO and HCI protocols was studied with two blended (KR1R4F/KR2R4F, KR1R5F) and one flue-cured (CIM-7) reference cigarettes. The main outcome was when results were reported on a per milligram TPM (that portion of the mainstream smoke which is trapped in the smoke trap, expressed as milligrams per cigarette) basis generated under ISO conditions was more mutagenic and more cytotoxic than was TPM generated under HCI conditions. However, the decrease in biological activity could not be explained only by the increased in the water content of the TPM on going from ISO to HCI smoking conditions, and the results may be influenced by differences in smoke chemistry as a result of differing smoke collection systems.
Key Words: mutagenicity; Ames assay; Salmonella typhimurium TA98/TA100; cytotoxicity; neutral red assay; TPM; cigarette smoke gas-vapor phase; Kentucky reference cigarette; Canadian Industry Monitor cigarette.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
There has been increased interest in assessing the toxicological properties of mainstream cigarette smoke (MSS). Such assessments have used smoke chemistry and one or more bioassays of whole smoke or smoke fractions. In vitro bioassays generally are more practical and cost effective than in vivo ones. In vitro assays include those for genotoxicity (e.g., mutagenicity [Ames Salmonella reverse mutagenicity assay], clastogenicity [in vitro micronucleus assay]), and cytotoxicity [neutral red uptake assay, NRU]. The rationale for using in vitro assays is that they may respond to toxicants and/or combinations of toxicants that would be missed by determination of specific smoke components. This work focuses on use of the Ames and NRU assays and how the results from these assays can be affected by changes in smoking regimen (puff volume, puff frequency, and blocking of filter ventilation).
The Ames assay continues to be used for the toxicological assessment of cigarette mainstream smoke condensate (MSC, also known as TPM, that portion of the mainstream smoke which is trapped in the smoke trap, expressed as milligrams per cigarette.). Several governments have specified its use as part of their tobacco control programs (DIN, 2004
; Health Canada, 2004a). The Ames assay has been used in two studies on the toxicological properties of ingredients added to cigarette tobaccos (Baker et al., 2004; Roemer et al., 2002). Furthermore, the Ames assay has been used as part of the evaluation of cigarettes that were designed to have the potential to reduce smoking-related diseases (Bombick et al., 1997; Tewes et al., 2003).
The Ames assay as applied to TPM is generally not a single assay but a series of assays. Generally, five Salmonella strains with and without metabolic activation (S9) are used, but there is controversy as to which combinations of strains and metabolic activation are responsive (Lauterbach, 2004). However, it is generally accepted that TPM from conventional cigarettes is responsive with TA98 + S9 and TA100 + S9 and that this responsiveness is likely due to nonvolatile, heterocyclic, nitrogen compounds in TPM (DeMarini, 2004). More recently, it was reported that increasing levels of filter ventilation yield increased specific activity when the cigarettes are smoked according to the U.S. FTC protocol (Steele et al., 1995). On the other hand, use of smoking regimens that involve more intensive puffing and/or blocking for filter ventilation reduce specific activity (Appleton et al., 2003; Roemer et al., 2004) Furthermore, values reported for the specific activities of the TPM of several reference cigarettes differed though condensate collection, and assay procedures were similar.
The NRU assay has become the favored assay for determining the cytotoxicity of TPM, the gas-vapor phase (GVP) of smoke and mixtures of both fractions (Baker et al., 2004; Bombick et al., 1997, 1998; DIN, 2004; Health Canada, 2004b; Roemer et al., 2004; Tewes et al., 2003). Cytotoxicity values are typically reported as EC50 values, and the units are micrograms of TPM per milliliter. Under this convention, less cytotoxic smoke has a higher EC50 value than does smoke that is more cytotoxic; but differences in procedure and choice of the cell line used may make comparisons of data obtained by different laboratories problematic (Putnam et al., 2002).
Much less is known about the cigarette blend and design features that affect the results of NRU assay than is known about similar factors and the Ames assay. Bombick et al. reported that when they studied the TPM from the Kentucky reference (KR) cigarettes KR2R1, KR1R4F, and KR1R5F, there was no significant difference in the cytotoxicities (as measured by the EC50 values expressed as micrograms of TPM per milliliter) among the TPM from these three cigarettes even though the blends and designs are considerably different (Bombick et al., 1998; Diana, 1990). There was a directional change in the EC50 values with KR1R4F being most cytotoxic, the KR1R5F the least and the KR2R1 being intermediate. Another finding from that study was that the TPM from upper stalk burley and flue-cured tobaccos were more cytotoxic than TPM from the corresponding lower stalk tobaccos. In addition, Roemer et al. (2004) reported that the KR1R5F showed different trends in cytotoxicity than did the KR1R4F on going from ISO to Massachusetts' smoking.
Because of the apparently disparate results being reported and the fact that there was increasing use of the Ames assay and NRU assays in the regulatory process, we decided it was time to clarify these issues by conducting a systematic study of both types of assays using several reference cigarettes and two smoking conditions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Materials.
The Canadian Industry Monitor cigarette (CIM-7) used in this study was prepared from 100% flue-cured tobacco by the Canadian Tobacco Manufacturers Council (CTMC), Ottawa, Ontario, Canada, and provided free of charge. KR cigarettes 1R4F and 1R5F were purchased from the Kentucky Tobacco Research and Development Center (formerly known as the Kentucky Tobacco and Health Institute), Lexington, KY. Descriptions and analytical data on the 1R4F and 1R5F were published by Diana (1990)
. The KR2R4F cigarette was used in place of the KR1R4F in the cytotoxicity studies. This new reference cigarette has been described and characterized (Chen and Moldoveanu, 2003).
Generation and collection of TPM for the Ames assays.
All smoke collections were done with an ISO harmonized Borgwaldt RM-20 CS smoking machine. Cigarettes were conditioned per the latest version of ISO 3402 (1999). For the TPM collections done with the Cambridge filter pad (CFP), a 92-mm pad was used. ISO smoking (35/60/2 without blocking of filter ventilation) was carried out in accordance with ISO 4387 (2000). Health Canada Intensive (HCI) smoking (55/30/2 with complete blocking of filter ventilation) was done in accordance with Health Canada method T-115 (Health Canada, 1999). A minimum of 200–300 mg of TPM was collected for each replicate sample. At the conclusion of smoking, pads were extracted with sufficient dimethylsulfoxide (DMSO, CAS# 67-68-5) by shaking for 20 min to give a concentration of approximately 10 mg TPM/ml. This solution was filtered through sterile cheesecloth into 25-ml amber vials. These were stored in a cryogenic freezer (– 80°C) prior to testing. For TPM collections done with the electrostatic smoke precipitator (ESP), the Borgwaldt RM-20 CS smoking machine was fitted with a Borgwaldt central electrostatic smoke trap (ESP unit) using a standard electrode and glass electrostatic precipitation tubes. The smoke trap was connected to a Spellman SL30 power supply set at 17.5 kV with a 0.2-mA current limit. The trapped particulate matter was extracted from the walls of the ESP tube by adding a known volume of DMSO (to achieve 10 mg TPM/ml DMSO) by placing on the end caps and using a Thermolyne Maxi Mix II. Once the walls were clear of visible TPM, the solution was poured through sterile cheesecloth in order to treat the sample the same as for samples collected on the CFP. As with the case of the TPM samples prepared by the CFP procedure, the samples prepared by the ESP procedure were also stored in a cryogenic freezer (– 80°C) prior to testing.
Generation and collection of cigarette smoke fractions for the NRU assays.
Cigarettes were conditioned according to the latest version of ISO 3402 (1999). Cigarettes were smoked under ISO conditions as well as under conditions specified in the Health Canada official Method T-502 (Health Canada, 2004b). A Borgwaldt RM-20/CS smoking machine was used. A sufficient number of cigarettes were smoked to provide minimum 180 mg TPM per 92-mm pad. Pads were extracted in DMSO by shaking for 20 min such that the final concentration of MSC is 10 mg TPM/ml. This solution was then filtered through sterile cheesecloth. Sample preparations were completed within approximately 35 min of sample generation. GVP was collected simultaneously with the smoking for the mainstream TPM. The GVP, which passes through the filter, was bubbled through ice-cold phosphate-buffered saline (PBS) as described by Roemer et al. (2002). Since the number of cigarettes smoked was such that minimum 180 mg of TPM was generated, 15 ml of ice-cold PBS was placed into a 70-ml impinger with extracoarse frit for the collection of GVP. The impinger was kept in an ice bath during smoke collection. Upon completion of smoking and the determination of TPM, the final volume of PBS was adjusted such that the GVP concentration was representative of a 10 mg TPM equivalent per milliliter PBS within approximately 30 min. The measure of dosing in the culture medium referred to the trapped GVP constituents corresponding to the collected TPM. The TPM + GVP sample was prepared by combining the TPM/DMSO fraction with the GVP/PBS fraction in a 1:1 volume ratio within approximately 60 min of smoking completion. Since the concentrations per milliliter solution were equivalent, this represented a 1:1 TPM/ml DMSO to TPM equivalent per milliliter PBS ratio. Thus, the TPM + GVP sample was representative of 5 mg TPM/ml + 5 mg TPM equivalent GVP/ml.
Determination of mainstream smoke analytes.
Mainstream smoke analytes were determined using the Health Canada methods. These methods are available on the Health Canada Web site. The URL for the method index is http://www.hc-sc.gc.ca/hl-vs/tobac-tabac/legislation/reg/indust/method/main-principal/index_e.html.
Ames Salmonella reverse mutation assays.
The Ames assays were performed with the same method that has since been adopted by Health Canada as Health Canada Official Method T-501, Bacterial Reverse Mutation Assay for Mainstream Tobacco Smoke (Health Canada, 2004a). Five TPM samples were prepared per brand/smoking condition/preparation method, and three replicate counts were obtained per dose (eight dose levels ranging from 0 to 500 µg TPM/plate) resulting in a 3 x 2 x 2 factorial experiment with 15 (5 x 3) replicates for all nonzero dose levels. The Ames assays were carried out with strains TA98 and TA100 with S9 activation with 20-min preincubation. The S9 (postmitochondrial supernatant in 0.154M KCl) used for Ames Assay was purchased from Molecular Toxicology Inc (also known as Moltox Inc., Boone, NC). It came from the livers of male Sprague-Dawley rats induced with Aroclor 1254.
NRU cytotoxicity assays.
These assays were performed with the same method that has since been adopted by Health Canada as Health Canada Official Method T-502, Neutral Red Uptake Assay for Mainstream Tobacco Smoke (Health Canada, 2004b). There were three independent smoke collections for each brand style.
Chinese hamster ovary cells were used and were exposed to the smoke fractions for 24 h prior to treatment with the neutral red dye solution. The smoke fractions were not treated with S9 prior to assay.
|
STATISTICAL ANALYSES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Factorial Analysis—Ames Assay
The observed colony count at a specific TPM dose was considered the response variable, and the factor/level significance was evaluated by means of a multifactor ANOVA analysis (STATGRAPHICS PLUS). The same approach was used in the evaluation of the significance of brand type (three), smoking condition (two), and method of Cigarette Smoke Condensate (CSC) preparation (two) as determinants of mutagenicity.
Dose-Response Curves—Ames Assay
Based on the results from the analysis of variance, homogeneous data sets were pooled, and the specific activity was determined based on the slope of the linear portion of the dose-response curve. This involved fitting the data using polynomial regression and then eliminating dosages that contributed significant curvature to the plot (p < .05) until a straight line was obtained (Bernstein et al., 1982). In most cases, this approach resulted in the elimination of the two highest doses (250 and 500 µg/plate) leaving five concentrations for slope determinations (regressions were not forced through the origin). The standard error for the slope was taken as the measure of uncertainty in estimates for specific activities.
Dose-Response Curves—NRU
The corrected absorbance values were plotted against concentration (micrograms of TPM per milliliter) for each smoke fraction per cigarette sample. EC50 values were estimated from the logistic regression function fitted to each dose-response curve as described in the Health Canada method. GRAPHPAD PRISM software was used fitting the logistic regression functions to the dose-response curves.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Factorial Analyses—Ames Assay
With respect to TA98 + S9, smoking condition was found to have a highly significant effect (p value < .005) up to a dose of 250 µg/plate. Cigarette type was a highly significant factor at all dose levels. With respect to cigarette type, the CIM-7 was significantly different from KR1R4F and KR1R5F, but the difference between KR1R4F and KR1R5F was not significant. Method of TPM collection (CFP vs. ESP) was not a significant factor at any of the dose levels utilized in this experiment. With respect to TA100 + S9, smoking condition and cigarette type were significant factors but only for doses between 75 and 500 µg/plate and 100 to 500 µg/plate, respectively. In contrast to the TA98 results, method of condensate collection was found to be significant when colony counts at doses of 250 and 500 µg TPM/plate were examined. However, data from those dose levels were not used in the calculation of specific activities.
Specific Activities—Ames Assay
The specific activity of each of the TPM preparations was taken to be the slope of the linear portion of the dose-response curve determined as described earlier. Tables 1 and 2 contain a summary of the results (uncorrected for water content) obtained for all three brands, two methods of TPM collection, and two smoking regimens.
|
|
With respect to both TA98 and TA100, method of TPM collection did not have a significant effect on TPM activity with the exception of KR1R5F (TA100, HCI). Consequently, a decision was made to pool the results from the two methods of preparing providing 10 replicates (3 counts per replicate) for subsequent analyses (see Tables 3 and 4).
|
|
It is clear, from the summary provided in Tables 3 and 4, that the specific activity of TPM prepared under ISO conditions is greater than that prepared under HCI conditions (p value< .05). As expected from the Roemer's work (Roemer et al., 2004
), the TPM from the CIM-7 (an all flue-cured cigarette) was less mutagenic than that obtained from KR1R4F; a blended cigarette typical of those sold in the United States (Steele et al., 1995). Under ISO conditions, the specific activity (TA98 + S9 and TA100 + S9) of the low-yield product, KR1R5F, was greater than that of KR1R4F; but the difference was not statistically significant. A significant reduction in KR1R5F activity (TA98 + S9 and TA100 + S9) relative to KR1R4F was noted under HCI conditions. A tendency toward increasing activity with decreasing tar has been noted in one large-scale study of commercially available American brands (Steele et al., 1995). It is difficult to find comparable results reported by others due to the relatively recent usage of intense smoking conditions; most results reported thus far have been obtained under ISO conditions. Table 5 shows our results and those of reported by others for the specific activity of the TPM (TA98 + S9, TA100 + S9) for the KR1R4F, and these indicates that the results reported here are "reasonable." The widespread in reported values may be related to differences in the method used to determine the linear portion of the dose-response curves (Roemer et al., 2002; Steele et al., 1995). The differences in results may also be due to differences in laboratory procedures as well as normal variations in the bacterial strains and the potency of the S9 preparations.
|
EC50 Values—NRU Assay
The EC50 values for the TPM fractions, the GVP fractions, and the TPM + GVP fractions are shown in Tables 6–8, respectively. The 95% confidence lower and upper bounds for the EC50 values are also shown. Cytotoxicity is likely driven in part by the most cytotoxic components of smoke (Kensler, 1969
). Since the cytotoxicity results presented here have been calculated on a TPM basis (micrograms of TPM per milliliter of solvent), one might expect change in EC50 values to follow changes in the concentrations of cytotoxic agents in the TPM. The relevant data on some analytes thought responsible for the cytotoxicity of smoke is shown in Table 9.
|
|
|
The data in Table 6 show that the cytotoxicity of TPM from all three cigarettes decreased on going from ISO to HCI conditions. What is interesting is that the difference was greatest for the KR1R5F and least for the CIM-7. These findings are similar to those reported by Foy et al. (2004)
. If the decrease were strictly due to the increase in the relative amount of particulate-phase water on going from ISO to HCI conditions, one would expect the relative difference between ISO and HCI to be about the same for the KR2R4F and CIM-7 and somewhat greater for the KR1R5F. However, there was quite a difference in the responses between the KR2R4F and the CIM-7 on going from ISO to HCI conditions, and this difference does not appear to be reflected in the smoke chemistry data shown in Table 9.
The data in Table 7 show the results for the NRU assays on the GVP fractions. Based on the work with carbon filters (Gori, 1976; Kensler, 1969), it is reasonable to expect that the main drivers of GVP cytotoxicity are acrolein and vapor-phase hydrogen cyanide (HCN). The CIM-7 did not give a reduction in GVP cytotoxicity in going from ISO to HCI conditions, but such a reduction was shown by both blended products. The smoke chemistry data for acrolein and vapor-phase HCN do not follow that trend.
|
The data in Table 8 show the results of the NRU assays for the mixtures of the TPM and GVP fractions. The cytotoxicities of these mixtures may be representative of the cytotoxicities of the whole MSS from which they are derived. In some cases, the EC50 values for the TPM + GVP fraction are close to the arithmetic mean of the EC50 values for the TPM and GVP fractions. In other cases, it is not. The Health Canada method specifies time limits between the end of smoke collection and the start of the NRU assay. These are 60 min for the TPM, GVP, and TPM + GVP fractions. These time limits were established after it was noted that the EC50 values were different when replicate smoke collections were held different lengths of time. The differences in response between the TPM + GVP fractions from blended product and those from all flue-cured products may be the result of differing ratios of HCN to acrolein as well as the formation of cyanohydrins (see "Discussion" section).
Table 9 shows data for HCN and several carbonyl compounds that were determined in the MSS of these cigarettes. It also shows the data expressed as a percentage of TPM as that is in effect how the smoke solutions used in the NRU assay are based. The data in Table 9 show that for the KR1R5F, the concentration HCN in the TPM rose on going from ISO to HCI smoking conditions, but it fell for the KR2R4F and the CIM-7. For acetaldehyde, acetone, and formaldehyde, the percentage decrease in the concentrations of these compounds relative to TPM on going from ISO to HCI smoking conditions is more for the KR1R5F than it is for the other two cigarettes. However, for acrolein, the concentration decreases for all three cigarettes are about the same.
When comparing cytotoxicity data with chemistry data, one should remember that there are important differences between the collection of smoke for the determination of HCN and aldehydes and the collection of smoke for the determination of cytotoxicity. The typical determination of smoke aldehydes does not use a Cambridge pad because past work showed some of these compounds are retained on the pad (Sakuma et al., 1978). The smoke is simply bubbled through impinger filled with an acidified solution of 2,4-dinitophenylhydrazine. For the determination of HCN, a Cambridge pad is used and the HCN collected on the pad is extracted with an alkaline solution and assayed separately from the HCN that passes through the pad and is collected in an impinger filled with a dilute solution of sodium hydroxide. On the other hand, for the determination of cytotoxicity, the TPM collected on the Cambridge pad is extracted with DMSO, while the volatiles in the GVP passing through the Cambridge pad are collected in physiologically buffered saline. Thus, there could be unexpected changes in the concentrations of cytotoxic compounds on going from ISO to HCI smoking regimes.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Numerous studies over the past decade have shown that some smokers take larger puffs and/or more frequent puffs than the FTC or ISO method specify and/or partially occlude filter ventilation (if present) (Borgerding and Klus, 2005
). Consequently, machine-smoking regimens such as those promulgated by the Massachusetts and Canada have been used to estimate the yields of smoke toxicants smokers may receive if they puff more intensively than called for in the FTC or ISO methods and/or occlude filter ventilation (Dixon and Borgerding, 2006). Studies such as those reported by Counts et al. (2005) have documented how mainstream smoke yields change with cigarette design and smoking regimen. However, it has only been recently that bioassays using these more intensive smoking conditions have been studied (Appleton et al., 2003; Rickert et al., 2002, 2003; Roemer et al., 2004). The results of research reported to date have shown that while intensive smoking conditions give higher TPM yields on a per cigarette basis, both the cytotoxicity and mutagenicity of the TPM is reduced when the results are reported on a per unit TPM basis. Our explanations of these phenomena follow.
Mutagenicity of TPM under ISO and HCI Conditions
In the present investigation, the most striking observation was the decrease in specific activity under "intensive" smoking. Consequently, it was of some interest to determine if the decrease could be related to changes in composition. From a mutagenicity standpoint, this relative increase may be related to the higher specific activity of condensates collected under ISO smoking conditions as noted earlier (see Tables 3 and 4; ISO vs. HCI). This same pattern was noted when the chemical compositions of dry particulate matter (DPM) from KR1R4F and the Virginia flue-cured product (i.e., the Canadian Monitor cigarette, CIM-7) were compared. Particulate-phase constituent concentrations (yield/mg DPM) were increased under ISO conditions relative to the Canadian Monitor, as were the specific activities. For example, using the data of Counts et al. for the KR1R4F cigarette, the concentration of benzo[a]pyrene B[a]P in DPM is about 0.72 ng/mg at ISO conditions, but it drops to about 0.47 ng/mg under HCI conditions (Counts et al., 2005).
Cytotoxicity of TPM, GVP, and TPM + GVP under ISO and HCI Conditions
While no study has been apparently published that identifies the major drivers of TPM cytotoxicity, it is not unreasonable to assume that formaldehyde along with the proportions of the HCN and acrolein remaining in the particulate phase are likely responsible for much of the cytotoxicity. It has been reported that the cytotoxicities of the TPM from the KR1R4F and the KR1R5F were not significantly different from the cytotoxicities of the TPM from many cigarettes on the U.S. market (Putnam et al., 2001). Putnam et al. found that the mean EC50 values were 44.8 ± 12.3 µg TPM/ml for full flavor products, 51.7 ± 11.7 µg TPM/ml for lights products, and 38.5 ± 7.3 µg TPM/ml for the ultralight products. Those findings imply that the cytotoxicity of TPM collected under FTC conditions does not appear to be influenced by the differences in blends and designs used in most U.S. cigarettes. However, as shown in Tables 6 and 7, there are unexpected changes when an all flue-cured cigarette is assayed or when the KR1R5F is assayed under HCI conditions. In this regard, it is important to remember that the MSS carbonyl determinations are done without the Cambridge pad because past work showed some of these compounds are retained on the pad (Sakuma et al., 1978). Conversely, the HCN determinations and the NRU assays are done with the Cambridge pad. Are there reactions on the Cambridge pad that may be affecting the results of the NRU assay?
If we refer to the extensive data set of smoke analytes published recently by Philip Morris scientists (Counts et al., 2005), we find that the relative proportion of HCN found in the TPM decreases on going from ISO to HCI conditions. This is opposite to what would be predicted by gas-particle partitioning theory (Lauterbach, 2000). What may be happening is that part of the HCN is reacting with aldehydes such as acetaldehyde and acetone to yield cyanohydrins (i.e., lactonitrile) as discussed earlier. Furthermore, Dube and Green (1982) noted that cyanohydrin formation increased as smoke pH increased. The cytotoxicities of these cyanohydrins in the NRU assay apparently have not been reported in the literature; however, they were tested in the KB cell cytotoxicity assay and found to be very toxic (Kensler, 1969). The most prevalent cyanohydrin in TPM is likely acetaldehyde cyanohydrin, which is also known as 2-hydroxypropanenitrile (or lactonitrile). It has been shown to be acutely toxic in several bioassays (Johannsen and Levinskas, 1986; OECD SIDS, 1994). Another factor that may be leading to confusing information is that lactonitrile may not be completely hydrolyzed to cyanide during the assay procedure because it has some stability at pH 9 (OECD SIDS, 1994). Thus, the decrease in KR1R5F TPM cytotoxicity on going from ISO to HCI smoking conditions may be due to a decrease in the concentrations of other cytotoxic substances (e.g., cyanohydrins) that outweigh the increase in the HCN concentration.
It is thought that much of the cytotoxic activity of the GVP is due mainly to HCN, acrolein, and related compounds as noted earlier. While these compounds can be found in the GVP of the MSS aerosol, they also exist in the TPM. The presence of HCN in the TPM has been known for sometime (Nall, 1966). Likewise, the presence of volatile carbonyl compounds in the TPM on the Cambridge pad had been established a decade later (Sakuma et al., 1978). It is also known that acetaldehyde and other aldehydes in smoke can react with HCN to form cyanohydrins, one of which is lactonitrile (Dube and Green, 1982). Lactonitrile, which has a boiling point over 200°C, can be formed by the reaction of HCN and acetaldehyde when smoke is collected on media such as activated carbon (Alford and Johnson, 1966).
Very recently, a new type of puff-by-puff analysis was used on the MSS of the KR2R4F reference cigarette, and it was found that 40% of MSS acetaldehyde and acetone were found (by calculation) in the MSS particulate phase (Adam et al., 2006). On the other hand, a lesser percentage of isoprene was found in the particulate phase. Thus, one might hypothesize that TPM cytotoxicity is either due to the cyanohydrins or to HCN and uncomplexed aldehydes (acrolein and formaldehyde, in particular) retained on the Cambridge pad. Indeed, there is evidence that points to cyanohydrins as a likely cause of the cytotoxicity. First, the cyanohydrins from acetaldehyde and acetone are much more toxic than other aliphatic nitriles (Johannsen and Levinskas, 1986). Second, the cytotoxicity of these cyanohydrins has been compared with other MSS components thought to be associated with cytotoxicity (Kensler, 1969). Based on both potency and likely amount present in smoke, Kensler estimated that they could be major contributors to smoke cytotoxicity. Taken together, these factors suggest that use of the Cambridge pad as specified in the Health Canada NRU method may be creating artifacts. The GVP fraction may also be subject to artifacts such as cyanohydrin formation. The GVP is trapped in aqueous media, and the trapped GVP could undergo cyanohydrin formation as reported by Dube and Green (1982) for the aqueous trapping of whole smoke. These factors also suggest the need for more understanding of the cytotoxic agents and their concentrations in the smoke fractions used for the NRU assay.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
In both the Ames assay for TPM mutagenicity and the NRU assays for TPM, GVP, and TPM + GVP cytotoxicities, results are generally expressed as activity per unit of TPM. In general, such a method of expressing the results of those bioassays results in higher biological activity being reported for MSS generated under ISO conditions that generated under HCI conditions. However, these decreases in biological activity are not simply caused by an increased dilution of the particulate matter by water on going from ISO to HCI smoking conditions. There are underlying effects related to differences in the relative amounts of pyrolysis products between the two smoking conditions and differences in gas-particle partitioning of semivolatile compounds that are not biologically relevant in the Ames and NRU assays, the influence of smoke collection techniques for the NRU, but not the Ames, assays. Thus, the use of intensive smoking regimens results in reductions in biological activity when the results are reported on the basis of unit TPM values and may, in the case of the Health Canada NRU method, also yield results which are influenced by differences in smoke chemistry as a result of differing smoke collection systems.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
|
ACKNOWLEDGMENTS |
|---|
Conflicts of Interest: Labstat International Inc and Lauterbach & Associates, LLC, (L&A LLC) provide services to the tobacco industry. L&A LLC is a member of the U.S. TAG to ISO TC126 on Tobacco and Tobacco Products. Both companies are members of Cooperation Centre for Scientific Research Relative to Tobacco.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSTATISTICAL ANALYSESRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Adam T, Mitschke S, Streibel T, Baker RR, Zimmermann R. (2006) Quantitative puff-by-puff resolved characterization of selected toxic compounds in cigarette mainstream smoke. Chem. Res. Toxicol. 19:511–520.[CrossRef][Web of Science][Medline]
Alford ED and Johnson RR. (1966) Cyanohydrins in carbon-trapped vapor phase of cigarette smoke. Report 66-30R. Brown & Williamson Tobacco Corp., October 10, 1966. Available at: http://legacy.library.ucsf.edu/tid/epq00f00. Accessed on August 16, 2006.
Appleton S, Krautter GR, Lauterbach JH, St Charles FK. (2003) Influence of puff volume, puff frequency, and vent-blocking on the mutagenicity of mainstream cigarette smoke condensate. Program Booklet and Abstracts 57th Tobacco Science Research Conference, Vol. 57, p. 46 (Abstract).
Baker RR, Massey ED, Smith G. (2004) An overview of the effects of tobacco ingredients on smoke chemistry and toxicity. Food Chem. Toxicol. 42:Suppl., S53–S83.
Bernstein L, Kaldor J, McCann J, Pike MC. (1982) An empirical approach to the statistical analysis of mutagenesis data from the Salmonella test. Mutat. Res. 97:267–281.[Web of Science][Medline]
Bombick BR, Murli H, Avalos JT, Bombick DW, Morgan WT, Putnam KP, Doolittle DJ. (1997) Chemical and biological studies of a new cigarette that primarily heat tobacco. Part 2. In vitro toxicology of mainstream smoke condensate. Food Chem. Toxicol. 36:183–190.
Bombick DW, Putnam K, Doolittle DJ. (1998) Comparative cytotoxicity studies of smoke condensates from different types of cigarettes and tobaccos. Toxicol. In Vitro 12:241–249.
Borgerding M and Klus H. (2005) Analysis of complex mixtures—Cigarette smoke. Exp. Toxicol. Pathol. 57:Suppl. 1, 43–73.
Chen PX and Moldoveanu SC. (2003) Mainstream smoke chemical analyses for 2R4F Kentucky reference cigarette. Beitr. Tabakforsch. Int. 20:448–458.
Counts ME, Morton MJ, Laffoon SW, Cox RH, Lipowicz PJ. (2005) Smoke composition and predicting relationships for international commercial cigarettes smoked with three machine-smoking conditions. Regul. Toxicol. Pharmacol. 41:185–227.[CrossRef][Web of Science][Medline]
DeMarini DM. (2004) Genotoxicity of tobacco smoke and tobacco smoke condensate: A review. Mutat. Res. 567:447–474.[CrossRef][Web of Science][Medline]
Diana JN. (1990) The Reference Cigarette(The University of Kentucky Printing Services, Lexington, KY).
DIN. (2004) DIN-Fachbericht 133: Toxicological Evaluation of Additives for Tobacco Products—A Guide Bestellnummer: 97615. Beuth Verlag GmbH, pp. 1–21.
Dixon M and Borgerding MF. (2006) Recent advances in the application and understanding of alternative smoking regimes. Recent Adv. Tob. Sci. 32:3–84.
Dube MF and Green CR. (1982) Methods of collection of smoke for analytical purposes. Recent Adv. Tob. Sci. 8:42–102.
Foy JWD, Bombick BR, Bombick DW, Doolittle DJ, Mosberg AT, Swauger JE. (2004) A comparison of in vitro toxicities of cigarette smoke condensate from Eclipse cigarettes and four commercially available ultra low-"tar" cigarettes. Food Chem. Toxicol. 42:237–243.[CrossRef][Web of Science][Medline]
Gori GB. (1976) Chemistry and In Vitro Bioassays of Smoke from Experimental Filter Cigarettes DHEW Publication No. (NIH) 76-1076. Smoking and Health Program, U. S. Department of Health, Education, and Welfare, U. S. Government Printing Office., Washington, DC.
Health Canada. (1999) Determination of "Tar", Nicotine and Carbon Monoxide in Mainstream Tobacco Smoke Official method T-115, pp. 1–7. Available at: http://www.hc-sc.gc.ca/hl-vs/alt_formats/hecs-sesc/pdf/tobac-tabac/legislation/reg/indust/method/main-principal/nicotine/nicotine_e.pdf. Accessed on August 16, 2006.
Health Canada. (2004a) Bacterial Reverse Mutation Assay for Mainstream Tobacco Smoke Official method T-501, Version 1.1, pp. 1–19. Available at: http://www.hc-sc.gc.ca/hl-vs/tobac-tabac/legislation/reg/indust/propose/tox_e.html. Accessed on August 16, 2006.
Health Canada. (2004b) Neutral Red Uptake Assay for Mainstream Tobacco Smoke Official method T-502, Version 1.1, pp. 1–18. Available at: http://www.hc-sc.gc.ca/hl-vs/tobac-tabac/legislation/reg/indust/propose/tox_e.html. Accessed on August 16, 2006.
ISO 3402. (1999) Tobacco and Tobacco Products—Atmosphere for Conditioning and Testing 4th ed (International Organization for Standardization, Geneva, Switzerland) pp. 1–4.
ISO 4387. (2000) Cigarettes—Determination of Total and Nicotine-Free Dry Particulate Matter Using a Routine Analytical Smoking Machine 3rd ed (International Organization for Standardization, Geneva, Switzerland) pp. 1–17.
Johannsen FR and Levinskas GJ. (1986) Relationships between toxicity and structure of Aliphatic Nitriles. Fundam. Appl. Toxicol. 7:690–697.[CrossRef][Web of Science][Medline]
Kensler CJ. (1969) Letter and Attachments Dated February 5, 1969, to T. J. Grady Bureau of Deceptive Practices, Federal Trade Commission. Available at: http://legacy.library.ucsf.edu/tid/dgy71e00. Accessed on August 16, 2006.
Lauterbach JH. (2000) A critical assessment of recent work on the application of gas/particle partitioning theories to cigarette smoke. Beitr. Tabakforsch. Int. 19:65–83.
Lauterbach JH. (2004) Which Ames is which? Lessons from some recent studies of mainstream smoke condensate mutagenicity using the Ames assay. Program Booklet and Abstracts 58th Tobacco Science Research Conference, Vol. 58, p. 60 (Abstract).
Nall JF. (1966) Complexed cyanide in collected cigarette smoke. Program Booklet and Abstracts 20th Tobacco Chemists' Research Conference, Vol. 20, pp. 25–26 (Abstract).
OECD SIDS. (1994) 2-Hydroxypropanenitrile UNEP Publications. Available at: http://www.inchem.org/documents/sids/sids/HYDROXYPROPANE.pdf. Accessed June 29, 2006.
Putnam KP, Bombick DW, Doolittle DJ. (2002) Evaluation of eight in vitro assays for assessing the cytotoxicity of cigarette smoke condensate. Toxicol. In Vitro 16:599–607.[CrossRef][Web of Science][Medline]
Putnam KP, Bombick DW, James EF, Doolittle DJ. (2001) The cytotoxicity of smoke condensates from cigarettes in the US market compared to Kentucky reference cigarettes. Toxicol. Sci. 60:305.
Rickert WS, Trivedi A, Wright W. (2002) Effect of smoking condition and method of collection on TA98 and TA100 response to crude smoke condensate (CSC) from control cigarettes (Kentucky reference 1R4F, 1R5F and a Canadian flue-cured monitor). (CORESTA Congress, New Orleans) Abstract ST 7 (Abstract).
Rickert WS, Trivedi A, Wright W. Relative cytotoxicity of tobacco smoke vapor and particulate phases when cigarettes are smoked under ISO and Canadian intense conditions. Program Booklet and Abstracts 57th Tobacco Science Research Conference, Vol. 57, p. 27 (Abstract).
Roemer E, Stabbert R, Rustemeier K, Veltel DJ, Meisgen TJ, Reininghaus W, Carchman RA, Gaworski CL, Podraza KF. (2004) Chemical composition, cytotoxicity and mutagenicity of smoke from US commercial and reference cigarettes smoked under two sets of machine smoking conditions. Toxicology 195:31–52.[CrossRef][Web of Science][Medline]
Roemer E, Tewes FJ, Meisgen TJ, Veltel DJ, Carmines EL. (2002) Evaluation of the potential effects of ingredients added to cigarettes. Part 3: In vitro genotoxicity and cytotoxicity. Food Chem. Toxicol. 40:105–111.[CrossRef][Web of Science][Medline]
Sakuma H, Kusama M, Shimojima N. (1978) Gas chromatographic analysis of the p-nitrophenylhydrazones of low boiling carbonyl compounds in cigarette smoke. Tob. Sci. 22:156–158.
Steele RH, Payne VM, Fulp CW, Rees DC, Lee CK, Doolittle DJ. (1995) A comparison of the mutagenicity of mainstream cigarette smoke condensates from a representative sample of the US cigarette market with a Kentucky reference cigarette (K1R4F). Mutat. Res. 342:179–190.[CrossRef][Web of Science][Medline]
Tewes FJ, Meisgen TJ, Veltel DJ, Roemer E, Patskan G. (2003) Toxicological evaluation of an electrically heated cigarette. Part 3: Genotoxicity and cytotoxicity of mainstream smoke. J. Appl. Toxicol. 23:341–348.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||