The Relative Toxicity of Substituted Phenols Reported in Cigarette Mainstream Smoke

Carr J. Smith*,1, Thomas A. Perfetti*, Michael J. Morton*, Alan Rodgman{dagger},2, Rajni Garg{ddagger}, Cynthia D. Selassie{ddagger} and Corwin Hansch{ddagger}

* R&D, Bowman Gray Technical Center, RJRT Company, Winston-Salem, North Carolina 27102–1487; {dagger} Fundamental Research, R&D, RJRT Company, Winston-Salem, North Carolina 27102–1487; and {ddagger} Department of Chemistry, Pomona College, Claremont, California

Received February 25, 2002; accepted May 31, 2002


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cigarette mainstream smoke (MS) contains a number of structurally diverse substituted phenols. Recent quantitative structure activity relationship (QSAR) studies on phenols show that substituted phenols with electron-releasing groups can form potentially toxic phenoxyl-free radicals. In contrast, substituted phenols with electron-withdrawing groups do not form phenoxyl-free radicals but exert their toxicity primarily through lipophilicity. The chemical structures of 253 different substituted phenols reported in MS have been described in sufficient detail to allow identification of the individual compounds. From a laterally validated equation based on published data on the toxic effects of phenols on cultured cells, the relative toxicity, on a molar basis, of the 253 MS phenols has been determined. Based on this scheme, the most toxic phenols in MS include, in descending order of toxicity, 2-(dimethylamino)-phenol, 2-ethyl-6-methyl-1,4-benzenediol, 2-methoxy-1,4-benzenediol, and 4-ethyl-2-methoxy-6-methylphenol. The least toxic phenols include, in ascending order of toxicity, 4-hydroxybenzoic acid and 3-hydroxybenzenepropanoic acid. In the human exposure situation, the toxicity of MS phenols is a complex interaction, with contributions made by the following factors: toxicity per mole; MS concentration; synergistic, additive or antagonistic interactions with other MS components; host susceptibility; metabolism; and individual smoking behavior and inhalation patterns. In the absence of data to the contrary, reduction in the number and concentration of toxic MS smoke components may be considered to be advantageous. Studies of this type can play an important role in identifying MS components for reduction or removal.

Key Words: free radicals; toxic smoke components; lung disease.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Free radicals have been implicated in the pathology of aging (Scarfiotti et al., 1997Go), cancer (Pryor, 1997Go), cardiovascular disease (Mehta and Mehta, 1999Go), and occupational and environmental lung diseases (Vallyathan and Shi, 1997Go). When compared with nonsmokers, cigarette smokers may be exposed to an increased burden of free radicals from at least 3 sources. First, the vapor (Shinagawa et al., 1998Go) and particulate phases of the cigarette smoke aerosol display free-radical activity (Blakley et al., 2001Go). Second, elevated levels of pulmonary macrophages and neutrophils are frequently found in the lungs of cigarette smokers. These cells can generate activated oxygen species (Rennard and Daughton, 1993Go). In addition, the peripheral white blood cell count is frequently elevated in smokers (McKarns et al., 1995Go; Taylor, 1987Go), which can increase systemic free-radical exposure. Third, many studies have reported that smokers' intake of dietary antioxidants is lower than that seen in nonsmokers (Midgette et al., 1993Go; Whichelow et al., 1988Go). In this study, primary emphasis was placed on the phenoxyl radicals found in the vapor and particulate phases of the complex cigarette smoke aerosol.

QSAR equations are developed and used to describe correlations between chemical structure and biological activity (Hansch and Leo, 1995Go). The biological activity of interest in this study was cytotoxicity via formation of phenoxyl-free radicals and nonspecific cytotoxicity due to lipophilicity. The data gathered from the biological-activity results are used to develop a regression equation that can be used to predict the biological activities of many more congeners structurally related to the original series. The particular chemical mechanism by which the congeneric series of compounds exerts its biological effects is not directly relevant to the predictive power of the resultant QSAR equation, although mechanistic information can frequently be garnered by comparing QSAR equations utilizing different molecular parameters (Hansch and Leo, 1995Go).

A survey of the international literature reveals that 380 different phenols have been reported in cigarette mainstream smoke (MS) (Ishiguro and Sugawara, 1980Go; Rodgman, personal communication). Of the initial 380 reported phenolic compounds, the chemical structures of 253 of these are described in sufficient detail to allow identification of the individual compounds. While the remaining 127 compounds are indeed phenols, their structures are not precisely defined because of uncertainties in the position and nature of substituents. The purpose of the current study is to use QSAR to rank the 253 phenolic compounds reported in cigarette MS by their predicted toxicity. Understanding the relative toxicity of these compounds may prove useful in identifying target compounds within the leaf or smoke for removal or inhibition.

Human exposure to phenols is ubiquitous. These compounds are found in tea, fruits, vegetables (Ho et al., 1994Go; Huang et al., 1994Go), and cigarette smoke (Rodgman et al., 2000Go). Because of their volatility, many phenols are found in both the vapor and particulate phases of MS (George, 1968Go; George and Keith, 1967Go; Guerin et al., 1992Go). Phenols are also widely used in industrial processes, consumer products, and pharmaceuticals (Rice-Evans and Packer, 1998Go). The toxicology of these compounds is complex, because phenols display a variety of disparate biological activities. A database of QSAR equations (Biobyte Corp., 2002) contains well over 100 examples for various types of phenol toxicity (Hansch et al., 2002Go). In contrast, other phenols function as free-radical scavengers, including {alpha}-tocopherol (vitamin E), the anti-oxidant butylated hydroxytoluene (BHT), and some of the polyphenols in foods and teas.

Many of the same phenols found in tobacco smoke are also present in materials used in foodstuffs. The following are several examples of phenolic compounds found in both foodstuffs (Baltes, 1977Go; Flament, 1991Go; Weidemann and Mohr, 1970Go) and tobacco smoke (Wynder and Hoffmann, 1967Go): 2-methoxy-4–1(2-propenyl)phenol [eugenol] and 2-methoxy-4–1(1-propenyl)phenol [isoeugenol], components of the much used spices clove and cinnamon; 4-(2-propenyl)phenol occurs in aniseed oil; 4-hydroxy-3-methoxybenzaldehyde [vanillin] and 4-hydroxy-3-methoxybenzoic acid [vanillic acid] are vanilla components; 2-methyl-5-(1-methylethyl)-[carvacrol] and 5-methyl-2-(1-methylethyl)phenol [thymol] are found in numerous natural essential oils; 3,4-dihydroxybenzoic acid [protocatechuic acid] occurs in wheat; and caffeic acid is a roasted coffee bean component. Phenol itself is permitted by the FDA to be included at a 0.5% level in various medications to cauterize mouth ulcers (Anbesol®). Different cigarettes might also vary compositionally in the aerosol concentration of phenoxyl radical scavenging compounds. The final contribution of the phenolics to overall phenoxyl radical toxicity will be determined by the complex interaction of their MS concentration, degree of absorption, extent of radical scavenging, metabolic fate, biological half-life, distribution between the particulate and vapor phases vis-à-vis deposition pattern in the lung, and host susceptibility factors. However, in the absence of mechanistic information to the contrary, reducing the number and concentration of phenoxyl radical-forming MS compounds can be considered to be advantageous.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cytotoxicity evaluations.
The following section describes the determination of the concentrations of various catechols (IC50) that induce 50% inhibition of growth of L1210 murine leukemia cells. L1210 cells were maintained in asynchronous logarithmic growth at 37°C in RPMI medium with L-glutamine supplemented with 10% (v/v) FBS. All stock solutions and dilutions were made in unsupplemented RPMI medium.

Cell cultures were seeded at 2–5 x 104 cells/ml in duplicate for each inhibitor concentration in a 96-well microtiter plate (180 ml/well). The test compounds (20 ml) were then added to the cell cultures in 1:10 dilution to achieve the desired concentrations. Each inhibitor was tested at a minimum of 8 concentrations. After 48 h of continuous chemical exposure, the cells were counted by using the CyQuant GR assay kit from Molecular Probes. For this purpose, the media were removed from the plates that were then frozen at –80°C for a minimum of 1 h. The cells were thawed at 37°C and 200 ml of CyQuant GR dye/cell lysis buffer were added to each well. The plates were incubated for 5 min at 37°C and their fluorescence was measured with a Cytofluor II multiwell fluorescence plate reader. The excitation maximum was 485 nm and the emission maximum was 530 nm. From the data, a dose-response curve was drawn and the IC50 determined. The CyQuant GR assay measures the ability of CyQuant GR dye to bind to the cellular nucleic acids of viable cells. Cytotoxicities are expressed as the concentration of the catechol (IC50) that causes a 50% reduction in fluorescence as compared with the controls (Jones et al., 2001Go).

Molecular Parameters Used
Hammett electronic parameter sigma plus ({sigma}+).
A minus sign (–) preceding the numerical value for {sigma}+ indicates that reactivity is favored by electron-releasing substituents, thereby increasing electron density in the benzene ring. Release of electrons into the benzene ring enhances the formation of free radicals by facilitating abstraction of H • from the phenolic hydroxyl group. Conversely, positive coefficient values of {sigma}+ indicate that substituents inhibit radical formation by decreasing electron density in the benzene ring and decreasing the tendency for abstraction of the H • .

In the case of the substituted phenols containing a propanoic acid moiety, this structural group was treated as a methyl group because of electronic behavioral similarity. The measured {sigma}+ value for the methylated compound was used to represent the {sigma}+ value of the analogous propanoic compound.

Octanol-water partition coefficients (P).
Lipophilicity, as measured by the base-10 logarithm of the octanol-water partition coefficient (P) and denoted as log P, correlates with a number of biological activities including in vitro mutagenicity (Debnath et al., 1994Go) and carcinogenicity in rodents (Franke et al., 2001Go). Lipophilic compounds can cross biological barriers that contain lipid, e.g., cell or microsomal membranes and skin (Tayar et al., 1991Go). In addition, lipid-partitioning properties can govern the ability of a molecule to enter into the largely hydrophobic active sites of biomacromolecules, such as metabolizing enzymes. Hence, log P influences metabolic fate, intrinsic biological activity, and the biological transport properties of chemicals (Hansch et al., 1995Go). Whereas the log P value is often a predictor of toxicity, the range of log P values associated with maximum toxicity is influenced both by the chemical structure of the compound and by the particular type of toxicity.

The measured partition coefficients of a subset of the studied phenols were determined by the shake-flask method (Hansch and Leo, 1995Go). Calculated log P values were determined by the method of Leo (1993). All values represent the uncharged forms of the phenol. The values of {sigma}+ are also experimentally determined (Hansch et al., 1995Go).

In some cases (as seen in the "Adjusted Log P" values in Table 3Go), the substituted phenol contained a carboxylic acid side chain, thereby resulting in a charged species at neutral pH. The calculated log P values for these charged species were determined by subtracting 4.0 from the calculated log P values of the neutral forms (Hansch and Leo, 1995Go).


View this table:
[in this window]
[in a new window]
 
TABLE 3 Phenols Ordered by Estimated Toxicity
 
Method for rank-ordering the 253 substituted phenols.
A total of 380 substituted phenols have been reported in cigarette smoke. However, many of those have not been sufficiently characterized to allow their exact identification. Of the 380, 253 substituted phenols were characterized structurally in sufficient detail for initial consideration. Those 253 compounds are listed in Table 1Go, together with their experimentally determined {sigma}+ values and their calculated and measured log P values. Since Table 1Go shows a clear correspondence between calculated and measured log P values (i.e., r2 = 0.97, n = 89), and the list of calculated log P values is more complete, the calculated log P values were used in the rank-ordering process.


View this table:
[in this window]
[in a new window]
 
TABLE 1 List of Phenols Considered
 
Of the 253 substituted phenols shown in Table 1Go, 51 have missing {sigma}+ values. Since their toxicity could not be estimated accurately, those 51 compounds were omitted from consideration in the ranking process. An additional 41 substituted phenols had positive {sigma}+ values (Table 2Go). These substituted phenols do not readily form a free-radical species and are not likely to be of concern, because they only exert toxic effects via lipophilicity (Hansch and Zhang, 1995Go; Selassie et al., 1998Go, 1999Go). These 41 substituted phenols were also omitted from consideration in the ranking process.


View this table:
[in this window]
[in a new window]
 
TABLE 2 List of Phenols with Positive {sigma}+
 
Preliminary results with 8 catechols indicate that the substituted catechols represent a special case of the substituted phenols by exerting their toxicity by a different mechanism, possibly related to their metabolism to benzoquinones (Snyder and Andrews, 1996Go). The estimated toxicity of the catechols is determined by utilizing a regression equation that is based on the cytotoxicity evaluation of a small set of catechols. The IC50 values were determined in a cultured L1210 leukemia cell line after 48 h. The QSAR model is as follows:

((Eq. 1))
where n = 8, r2 = 0.882, s = 0.094, and q2 = 0.807. C is the IC50 and is defined as the concentration of substituted catechols that inhibits growth by 50%. The Hammett electronic parameter is {sigma}+, n is the number of substituted catechols used in the regression, r is the correlation coefficient, and s is the standard deviation around the regression equation. A measure of the goodness of fit of the data is provided by q2, which approaches the value of r2 as the quality of the fit improves.

QSAR support for a different mechanism is provided by comparing the cytotoxicity of the catechols versus the phenols against rapidly dividing L1210 leukemia cells. In this in vitro system, the QSAR equation for the catechols has a {sigma}+ coefficient of –0.41, in contrast with a comparable value of –1.35 for the substituted phenols. The substituted catechols tend to form free radicals; however, they do so more easily than do the substituted phenols, i.e., they need less help from the substituents. There are 37 substituted catechols listed in Table 1Go. Four of the 37 also had missing values for {sigma}+ and were previously eliminated. Therefore, consideration of the mechanistic uniqueness of the catechols resulted in using QSAR Equation 1Go rather than that employed for the phenols.

A QSAR equation was developed (Selassie et al., 1998Go, 1999Go) to estimate the free radical- and hydrophobicity-induced growth inhibitory activity of phenolic compounds. Whereas this particular QSAR equation was used in the ranking process herein, the general concept of radical mediated phenoxyl toxicity has been laterally validated in a number of test systems (Hansch et al., 2000Go). Developmental toxicity endpoints were determined by using the Somite, tail-defect, and tail-defect tests with rat embryos plus co-cultured hepatocytes and WB rat liver epithelial cells (Hansch et al., 2000Go; Kavlock, 1990Go). The estimated toxicity is given by the regression equation:

((Eq. 2))
where n = 51, r2 = 0.895, s = 0.227, and q2 = 0.882. C has been previously defined. Equation 3Go was formulated for the same set of phenols, but calculated OH hemolytic bond dissociation energies (BDE) were used in lieu of {sigma}+ values:

((Eq. 3))
where n = 52, r2 = 0.920, s = 0.202, and q2 = 0.909. The BDE values were based on B3 LYP/6–31G**//AMI energies and were obtained using Jaguar 3.0 and Spartan 5.0 (Selassie et al., 1999Go). They are defined as the reaction energy (in kcal/mol) for the following reaction:

Equation 3Go strongly indicates that a radical mediated reaction is responsible for the cytotoxicity. QSAR Equation 2Go, using {sigma}+ (r2 = 0.895), is not quite as good as QSAR Equation 3Go (r2 = 0.920), because it was necessary to estimate {sigma}+ values for a number of complex phenols. In addition, many other QSAR studies have shown that radical-mediated reactions of phenols are well correlated by {sigma}+ (Hansch and Gao, 1997Go). By use of Equation 2Go, the remaining 162 phenolic compounds are listed in order of decreasing estimated toxicity in Table 3Go.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The free radical and hydrophobicity-related toxicity of 203 MS substituted phenols are considered. Of the 203, 41 have positive coefficients of {sigma}+ and are considered to be less toxic than the 162 substituted phenols with negative coefficients. As a result of this process of elimination, the 162 substituted phenols, of which 33 are catechols, are rank-ordered by toxicity in Table 3Go.

Predicted toxicity values for the ranked substituted phenols in Table 3Go were calculated from the QSAR equations described in the Materials and Methods section and represent the combined effects of the tendency to form phenoxyl- or catechol-free radicals, and in the case of the non-catechols, additional toxicity due to lipophilicity. On a logarithmic scale, the predicted toxicity values for the 162 toxic phenols ranged from a high value of 5.90 to a lower value of 2.90. Therefore, 2-(dimethylamino)-phenol (the most toxic of the 162 ranked) is predicted to be 1000-fold more toxic per mole than the least toxic compound ranked, 4-hydroxybenzoic acid. The greater impact of the {sigma}+ term in the ranking equation is illustrated by the calculated log P values for the 162 toxic phenols ranging from 6.75 to –3.97, an approximate range of 10 log units, in lipophilicity. Also, on a logarithmic scale, the coefficients of the {sigma}+ values for the ranked compounds ranged from 0 to –1.87.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
From laterally validated equations based on published data on the toxic effects of phenols on cultured cells, the relative cytotoxicity, on a molar basis, of 253 MS phenols has been determined. Based on this scheme, the most cytotoxic phenols in MS include in descending order of toxicity: 2-(dimethylamino)-phenol, 2-ethyl-6-methyl-1,4-benzenediol, 2-methoxy-1,4-benzenediol and 4-ethyl-2-methoxy-6-methylphenol. The least toxic phenols include, in ascending order of cytotoxicity, 4-hydroxybenzoic acid, 3-hydroxyphenylpropanoic acid, and 3-hydroxybenzenepropanoic acid.

The toxicity of substituted phenols has been correlated with {sigma}+ values (Hansch et al., 2000Go; Selassie et al., 1998Go). A review of the literature (Hansch and Gao, 1997Go) revealed that in 25 QSAR equations describing the formation of free radicals by abstracting H • from the phenolic hydroxyl group, 23 equations were correlated with {sigma}+. This observation implies that following H • abstraction from the phenol hydroxyl group, the remaining phenoxyl radical has some electron-deficient character. Such a species would be stabilized by electron-donating substituents such as amino, hydroxy, methoxy, and methyl groups and presumably would possess an increased lifetime to facilitate biological/cellular damage. Substituted phenols bearing electron-donating substituents would be predicted to be more toxic. In contrast, with electron-withdrawing substituted phenols, non-specific toxicity as modulated by hydrophobicity appears to predominate.

As reported in Table 3Go, four of the substituted phenols with relatively high estimated toxicities contain methoxy groups at the 2- and 6-positions. The four compounds are 2,6-dimethoxy-4-ethylphenol, 2,6-dimethoxy-4-methylphenol, 2,6-dimethoxy-4-ethenylphenol, and 2,6-dimethoxyphenol. A recent measurement of the toxicity of 2,6-dimethoxyphenol against L1210 cells suggests that the predicted value shown in Table 3Go overestimates the toxicity of this compound. The predicted log 1/C value from Table 3Go is 5.62, while the experimental log 1/C value recently measured by Selassie et al. (in press) is 3.86. While the mechanistic reason for this overestimation of toxicity is unknown, this result implies that the toxicity of the 2,6-dimethoxy-4-ethylphenol, 2,6-dimethoxy-4-methylphenol, and 2,6-dimethoxy-4-ethenylphenol may also be overestimated in Table 3Go.

The QSAR predicted toxicities of the substituted phenols reported in cigarette smoke span a very wide range. From a target-setting perspective, removal or inhibition of these 162 toxic substituted phenols would be desirable if the resultant whole smoke aerosol, condensate, or vapor phase did not inadvertently display increased toxicity due to an unexpected inhibitory interaction with some other toxic MS constituent. The more complicated question would be raised if the free-radical activity of the substituted phenols were completely inhibited, but none of the standard tests for the toxicity of cigarette smoke was altered. In this case, if it were technologically feasible, it would still be prudent to inhibit the potentially carcinogenic chemicals in the complex mixture because of the inability of current toxicological tests to model adequately the human in vivo situation. The limitations of current in vivo models are illustrated by the inability of animal inhalation exposures to induce pulmonary carcinoma with cigarette smoke (Henry and Kouri, 1984Go, 1986Go; Huber, 1989Go).

Table 3Go contains one of the interesting observations in this study. Previous animal studies have reported that 4-methoxyphenol can function as an inhibitor of B[a]P carcinogenesis (Asakawa et al., 1994Go; Boutwell and Bosch, 1959Go; Wattenberg, 1981Go; Wattenberg et al., 1980Go). However, an examination of its negative {sigma}+ coefficient of –0.78 indicates that it can form a free radical and is thus likely to be a biologically reactive compound. There are several examples of biologically active compounds that possess some inherent toxicity, but function to reduce the tumorigenicity of a complex mixture. For example, low molecular weight polycyclic aromatic hydrocarbons (PAHs) inhibit the tumorigenicity of higher molecular weight PAHs in the mouse skin-painting model, presumably by tying up liver enzymes required for the metabolic activation of the more tumorigenic higher molecular weight PAHs (Rodgman et al., 2000Go).

The significance of the toxicology results regarding tobacco smoke phenols is unclear, despite extensive experimentation (Rodgman et al., 2001Go). The toxicological results on tobacco smoke phenols date from the mid-1950s, when the promoting effect of specific phenols on PAH tumorigenicity in the mouse skin-painting assay was reported (Boutwell and Bosch, 1959Go; Boutwell et al., 1955Go, 1956Go). The results, showing tumor-promoting potential of phenols on initiated mouse skin, were met with enthusiasm as an explanation of the tumor-promoting potency of cigarette MS condensate, despite minor contributions from PAHs toward tumorigenicity in the same rodent system (Hoffmann and Wynder, 1963aGo,bGo; Wynder et al., 1958Go).

The observation that mixtures of substituted phenols promoted PAH tumorigenicity in mouse skin was offset by reports on the effects of phenols within the complex mixture of CSC. First, the almost complete removal of volatile phenols from MS by selective filtration (Hoffmann and Wynder, 1963aGo,bGo; Laurene et al., 1963Go) displayed no significant effect on the tumorigenicity or tumor-promoting ability of the resulting MS CSC in the mouse skin-painting assay (Wynder and Hoffmann, 1967Go, 1968Go, 1969Go). Second, inhibition by phenol of the specific tumorigenicity of B[a]P in the mouse-skin painting assay was reported (Van Duuren, 1980Go; Van Duuren et al., 1971Go, 1973Go). Third, several other phenols, e.g., 4-methoxyphenol (Spears, 1963Go) and {alpha}-tocopherol (Risner, 1996Go, 1997Go; Rodgman and Cook, 1960Go), known to be effective anticarcinogens vs. several potent tumorigenic PAHs such as B[a]P were isolated from cigarette MS (cf. Rodgman, 1991Go, 1992Go, 1994Go).

In the absence of data to the contrary, reduction in the biological activity of toxic MS phenols may be considered to be advantageous. Achieving meaningful reductions in the MS concentration of substituted phenols poses a significant technical challenge as the major tobacco leaf precursors are cellulose, pectins, starch, lignin, and chlorogenic acid (Bell et al., 1966Go; Schlotzhauer et al., 1982Go). QSAR studies of this type, in conjunction with toxicology test results, can play an important role in identifying MS components for inhibition, reduction, or removal.


    NOTES
 
1 To whom correspondence should be addressed. Fax: (336) 741-5197. E-mail: smithc4{at}rjrt.com. Back

2 Present address: 2828 Birchwood Dr., Winston-Salem, NC 27103. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Asakawa, E., Hirose, M., Hagiwara, A., Takahashi, S., and Ito, N. (1994). Carcinogenicity of 4-methoxyphenol and 4-methylcatechol in F344 rats. Int. J. Cancer 56, 146–152.[ISI][Medline]

Baltes, W. (1977). Effect of roasting on the composition of coffee. Colloq. Sci. Int. Café 8, 85–96.

Bell, J. H., Saunders, A. O., and Spears, A. W. (1966) The contribution of tobacco constituents to phenol yield of cigarettes. Tob. Sci. 10, 138–142.

BioByte Corporation (2002). 201 West 4th Street, Suite 204, Claremont, CA 91711.

Blakley, R. L., Henry, D. D., and Smith, C. J. (2001). Lack of correlation between cigarette mainstream smoke particulate-phase radicals and hydroquinone yield. Food Chem. Toxicol. 39, 401–406.[ISI][Medline]

Boutwell, R. K., and Bosch, D. K. (1959). The tumor-promoting action of phenols and related compounds for mouse skin. Cancer Res. 19, 413–424.[ISI]

Boutwell, R. K., Rusch, H. P., and Booth, B. (1956). Tumor production by phenol and related compounds. Proc. Am. Assoc. Cancer Res. 2, 96.

Boutwell, R. K., Rusch, H. P., and Bosch, D. K. (1955). The action of phenols and related compounds in tumor formation. Proc. Am. Assoc. Cancer Res. 2, 6–7.

Debnath, A. K., Shusterman, A. J., deCompadre, R. L., and Hansch, C. (1994). International Commission for Protection Against Environmental Mutagens and Carcinogens. Importance of the hydrophobic interaction in the mutagenicity of organic compounds. Mutat. Res. 305, 63–72.[ISI][Medline]

Flament, I. (1991). Coffee, cocoa, and tea. Food Sci. Technol. 44, 617–669.

Franke, R., Gruska, A., Giuliani, A., and Benigni, R. (2001). Prediction of rodent carcinogenicity of aromatic amines: A quantitative structure-activity relationship model. Carcinogenesis 22, 1561–1571.[Abstract/Free Full Text]

George, T. W. (1968). Selective removal of components of tobacco smoke by filtration. In Toward a Less Harmful Cigarette. (E. L. Wynder and D. Hoffmann, Eds.). National Cancer Institute Monograph 28, 237–248.

George, T. W., and Keith, C. H. (1967). The selective filtration of tobacco smoke. In Tobacco and Tobacco Smoke: Studies in Experimental Carcinogenesis. (E. L. Wynder and D. Hoffmann, Eds.), pp. 577–622 (see 601–607). Academic Press, New York.

Guerin, M. R., Jenkins, R. A., and Tomkins, B. A. (1992). Mainstream and sidestream cigarette smoke. In The Chemistry of Environmental Tobacco Smoke: Composition and Measurement. (M. Eisenberg, Ed.). Lewis Publishers, Chelsea, MI.

Hansch, C., and Gao, H. (1997). Comparative QSAR: Radical reactions of benzene derivatives in chemistry and biology. Chem. Rev. 97, 2995–3060.[ISI][Medline]

Hansch, C., Hoekman, D., Leo, A., Weininger, D., and Selassie, C. D. (2002). Chem-Bioinformatics: Comparative QSAR at the interface between chemistry and biology. Chem. Rev. 102, 783–812.[ISI][Medline]

Hansch, C., and Leo, A. (1995). Exploring QSAR. Fundamentals and Applications in Chemistry and Biology (S.R. Heller, Ed.). ACS Professional Reference Book, Washington, DC.

Hansch, C., Leo, A., and Hoekman, D. (1995). Exploring QSAR. Hydrophobic, Electronic, and Steric Constants. (S. R. Heller, Ed.). ACS Professional Reference Book, Washington, DC.

Hansch, C., McKarns, S. C., Smith, C. J., and Doolittle, D. J. (2000). Comparative QSAR evidence for a free-radical mechanism of phenol-induced toxicity. Chem. Biol. Interact 127, 61–72.[ISI][Medline]

Hansch, C., and Zhang, L. (1995). Comparative QSAR: Radical toxicity and scavenging: Two different sides of the same coin. SAR QSAR Environ. Res. 4, 73–82.[Medline]

Henry, C. J., and Kouri, R. E. (1984). Chronic Exposure of Mice to Cigarette Smoke. Final Report On "Smoke Inhalation in Mice." Field, Rich, and Associates, New York.

Henry, C. J., and Kouri, R. E. (1986). Chronic inhalation studies in mice: II. Effects of long-term exposure to 2R1 cigarette smoke on (C57BL/Cum x C3H/AnfCum)F1 mice. J. Natl. Cancer Inst. 77, 203–212.[ISI][Medline]

Ho, C.-T., Osawa, T., Huang, M.-T., and Rosen, R. T., eds. (1994). Food Phytochemicals for Cancer Prevention, Vol. II. American Chemical Society, Washington, DC.

Hoffmann, D., and Wynder, E. L. (1963a). Die Filtration von Phenolen aus Cigarettenrauch. Beit. Tabakforsch. 2, 51–66.

Hoffmann, D. and Wynder, E. L. (1963b). Filtration of phenols from cigarette smoke. J. Natl. Cancer Inst. 30, 67–84.[ISI][Medline]

Huang, M.-T., Osawa, T., Ho, C.-T., and Rosen, R. T., Eds. (1994). Food Phytochemicals for Cancer Prevention, Vol. I. American Chemical Society, Washington, DC.

Huber, G. L. (1989). Physical, chemical, and biologic properties of tobacco, cigarette smoke, and other tobacco products. Seminars in Resp. Med. 10, 297–332.

Ishiguro, S., and Sugawara, S. (1980). The chemistry of tobacco smoke. Sci. Papers, Cent. Res. Inst. Japan Monopoly Corp. 298, 1.

Jones, L. J., Gray, M., Yue, S. T., Haugland, R. P., and Singer, V. L. (2001). Sensitive determination of cell number using the CyQuant cell proliferation assay. J. Immunol. Methods 254, 85–98.[ISI][Medline]

Kavlock, R. J. (1990). Structure-activity relationships in the developmental toxicity of substituted phenols: in vivo effects. Teratology 41, 43–59.[ISI][Medline]

Laurene, A. H., Young, G. W., and Lyerly, L. A. (1963). Factors which affect the phenol content of cigarette smoke. RDR, 1963, No. 58 (November 13). See Bates number 500962130 –2163 at www.rjrtdocs.com.

Leo, A. J. (1993). Calculating log P oct from structures. Chem. Rev. 93, 1281–1306.[ISI]

McKarns, S. C., Smith, C. J., Payne, V. M., and Doolittle, D. J. (1995). Blood parameters associated with atherogenic and thrombogenic risk in smokers and nonsmokers with similar life-styles. Mod. Pathol. 8, 434–440.[ISI][Medline]

Mehta, J. L., and Mehta, J. (1999). Antioxidants and vitamins in your cardiac patient: Are they helpful? Cardiol. Rev. 7, 56–61.[Medline]

Midgette, A. S., Baron, J. A., and Rohan, T. E. (1993). Do cigarette smokers have diets that increase their risks of coronary heart disease and cancer? Am. J. Epidemiol. 137, 521–529.[Abstract]

Pryor, W. A. (1997). Cigarette-smoke radicals and the role of free radicals in chemical carcinogenicity. Environ. Health Perspect. 105(Suppl. 4), 875–882.[ISI][Medline]

Rennard, S. I., and Daughton, D. M. (1993). Smoking reduction and biological markers of response. Monaldi Arch Chest Dis. 48, 580–582.[Medline]

Rice-Evans, C. A., and Packer, L., eds. (1998). Flavonoids in Health and Disease. Marcel Dekker, New York.

Risner, C. H. (1996). The determination of +{alpha}-tocopherol in tobacco and mainstream tobacco smoke. 50th Tob. Chem. Res. Conf., Richmond, VA: Paper No. 46.

Risner, C. H. (1997). The determination of +{alpha}-tocopherol in tobacco and mainstream tobacco smoke. Tob. Sci. 41, 53–61.

Rodgman, A. (1991). A comparison of the chemical and physical properties of cigarette mainstream smoke (MS), cigarette sidestream smoke (SS), and environmental tobacco smoke (ETS). Document submitted to the Environmental Protection Agency, December, 1991. Revised version submitted, June, 1992. See Bates number 510649075–9111, accessed at www.rjrtdocs.com.

Rodgman, A. (1992). Environmental tobacco smoke. Regul. Toxicol. Pharmacol. 16, 223–244.[ISI][Medline]

Rodgman, A. (1994). The chemical composition of environmental tobacco smoke: Some comments on the Occupational Safety and Health Administration's notice on "Indoor Air Quality." Document submitted to the Occupational Safety and Health Administration, 5 August, 1994, pp. xiii + 172. See Bates number 515923456 –3645, accessed at www.rjrtdocs.com.

Rodgman, A. (2001). Personal communication regarding the identification of 383 different substituted phenols in cigarette smoke.

Rodgman, A., and Cook, L. C. (1960). The composition of cigarette smoke. IV. {alpha}- Tocopherol. Tob. Sci. 4, 7–8.

Rodgman, A., Smith, C. J., and Perfetti, T. A. (2000). The composition of cigarette smoke: A retrospective, with emphasis on polycyclic components. Hum. Exp. Toxicol. 19, 573–595.[ISI][Medline]

Scarfiotti, C., Fabris, F., Cestaro, B., and Giuliani, A. (1997). Free radicals, clinical atherosclerosis, ageing, and related dysmetabolic pathologies: Pathological and aspects. Eur. J. Cancer Prev. 6(Suppl. 1), S31–36.[ISI][Medline]

Schlotzhauer, W. S., Martin, R. M., Snook, M. E., and Williamson, R. E. (1982). Pyrolytic studies on the contribution of tobacco leaf constituents to the formation of smoke catechols. J. Agric. Food Chem. 30, 372–374.[ISI]

Selassie, C. D., DeSoyza, T. V., Rosario, M., Gao, H., and Hansch, C. (1998). Phenol toxicity to leukemia cells: A radical process? Chem. Biol. Interact 113, 175–190.[ISI][Medline]

Selassie, C. D., Shusterman, A. J., Kapur, S., Verma, R. P., Zhang, L., and Hansch, C. (1999). On the toxicity of phenols to fast growing cells. A QSAR model for a radical-based toxicity. J. Chem. Soc., Perkin Transaction 2, 2729.

Selassie, C. D., Verma R. P., Kapur S., Shusterman A. J., and Hansch, C. QSAR for the action of 2-alkyl or 2,6-dialkyl, 4-X-phenols: The nature of the radical reaction. J. Chem. Soc., Perkin Transaction (in press).

Shinagawa, K., Tokimoto, T., and Shirane, K. (1998). Spin trapping of nitric oxide in aqueous solutions of cigarette smoke. Biochem. Biophys. Res. Commun. 253, 99–103.[ISI][Medline]

Snyder, R., and Andrews, L. S. (1996). Toxic effects of solvents and vapors. In Casarett & Doull's Toxicology, 5th ed. (C. D. Klaassen, Ed.), pp. 737–771. McGraw-Hill, New York, NY.

Spears, A. W. (1963). Quantitative determination of phenol in cigarette smoke. Anal. Chem. 35, 320–322.[ISI]

Tayar, N. E., Tsai, R. S., Testa, B., Carrupt, P. A., Hansch, C., and Leo, A. (1991). Percutaneous penetration of drugs: A quantitative structure-permeability relationship study. J. Pharm. Sci. 80, 744–749.[ISI][Medline]

Taylor, R. G. (1987). Smoking and the leucocyte count. Eur. J. Respir. Dis. 71, 65–68.[ISI][Medline]

Vallyathan, V., and Shi, X. (1997). The role of oxygen free radicals in occupational and environmental lung diseases. Environ. Health Perspect. 105(Suppl. 1), 165–177.[Medline]

Van Duuren, B. L. (1980). Carcinogens, cocarcinogens, and tumor inhibitors in cigarette smoke condensate. In A safe cigarette? Banbury Report 3. (G. B. Gori and F. G. Bock, Eds.), pp. 105–112. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Van Duuren, B. L., Katz, C., and Goldschmidt, B. M. (1973). Cocarcinogenic agents in tobacco carcinogenesis. J. Natl. Cancer Inst. 51, 703–705.[ISI][Medline]

Van Duuren, B. L., Sivak, A., Katz, C., and Melchionne, S. (1971). Cigarette smoke carcinogenesis: Importance of tumor promoters. J. Natl. Cancer Inst. 47, 235–240.[ISI][Medline]

Wattenberg, L. W. (1981). Inhibitors of chemical carcinogens. In Cancer: Achievements, Challenges and Prospects for the 1980`s. (J. J. Burchenal, Ed.), 517–539. Grune and Stratton, New York NY.

Wattenberg, L. W., Coccia, J. B., and Lam, L. K. (1980). Inhibitory effects of phenolic compounds on benzo[a]pyrene-induced neoplasia. Cancer Res. 40, 2820–2823.[Abstract]

Weidemann, H. L., and Mohr, W. (1970). Specificity of roasted coffee aroma. Lebensm. Wiss. Technol. 13, 23–32.

Whichelow, M. J., Golding, J. F., and Treasure F. P. (1988). Comparison of some dietary habits of smokers and non-smokers. Br. J. Addict. 83, 295–304.[ISI][Medline]

Wynder, E. L., and Hoffmann, D. (1967). Tobacco and Tobacco Smoke: Studies in Experimental Carcinogenesis, p. 626. Academic Press, New York.

Wynder, E. L., and Hoffmann, D. (1968). Experimental tobacco carcinogenesis. Science 162, 862–871.[ISI][Medline]

Wynder, E. L., and Hoffmann, D. (1969). A study of tobacco carcinogenesis: X. Tumor promoting activity. Cancer 24, 289–301.[ISI][Medline]

Wynder, E. L., Wright, G. F., and Lam, J. (1958). A study of tobacco carcinogenesis. V. The role of pyrolysis. Proc. Am. Assoc. Cancer Res. 2, 357–358; Cancer 11, 1140–1148.