The Composition of Cigarette Smoke: A Chronology of the Studies of Four Polycyclic Aromatic Hydrocarbons

Abstract Among the polycyclic aromatic hydrocarbons (PAHs), a major class of identified cigarette mainstream smoke (MSS) components, are several shown to be tumorigenic in laboratory animals and suspect as possible tumorigens to humans. To date, nearly 540 PAHs have been completely or partially identified in tobacco smoke [Rodgman and Perfetti (1)]. A detailed chronology is presented of studies on four much discussed PAHs identified in tobacco smoke, namely, benz[a]anthracene (B[a]A), its 7,12-dimethyl derivative (DMB[a]A), dibenz[a, h]anthracene (DB[a, h]A), and benzo[a]pyrene (B[a]P). Of the four, DMB[a]A, DB[a, h]A, and B[a]P are considered to be potently tumorigenic on mouse skin painting and subcutaneous injection. Opinions on the tumorigenicity of B[a]A to mouse skin vary. DMB[a]A is frequently used in tumorigenicity studies as an initiator. Examination of the number of tobacco smoke-related citations listed for these four PAHs reveals the enormous effort devoted since the early 1950s to B[a]P vs. the other three. An annotated chronology from 1886 to date describes the tobacco smoke-related research pertinent to these four PAHs, their discovery, isolation and/or identification, quantitation, and contribution to the observed biological activity of MSS or cigarette smoke condensate (CSC). Much of the major literature on these four PAHs in tobacco smoke is presented in order to permit the reader to decide whether the current evidence is sufficient to classify them as a health risk to smokers. There has certainly been a tremendous effort by researchers to learn about these PAHs over the past several decades. Each of these PAHs when tested individually has been shown to possess the following biological properties: 1) Mutagenicity in certain bacterial situations, 2) tumorigenicity in certain animal species, to varying degrees under various administration modes, and 3) a threshold limit below which no tumorigenesis occurs. For more than five decades, it has been known that some of the PAHs, when co-administered in pairs of a potent tumorigen plus a non-tumorigen or weak tumorigen, show inhibitory effects on the tumorigenicity of the most potent, e.g., B[a]A plus DB[a, h]A; B[a]A plus B[a]P; anthracene plus DB[a, h]A. Over the period studied, some regulatory agencies considered these tobacco smoke PAHs to be serious health concerns, others did not. With respect to cigarette MSS, certainly the ‘danger is in the dose’ for any MSS component tested singularly to be tumorigenic. But is the level of any of these MSS PAHs high enough to be of concern to smokers? The information herein presented indicates that over the last five decades the following has occurred: 1) The per cigarette yields of these four PAHs have decreased substantially, 2) compared to CSC or Federal Trade Commission (FTC) ‘tar’, their per cigarette yields have also decreased to a point that they may be below any significance biologically, and 3) the specific tumorigenicity in mouse skin-painting studies of the CSC has decreased. These are the three criteria originally proposed to define the ‘less hazardous’ cigarette. Actually, criterion 1) was first directed only at B[a]P. Previous studies highlighted the concern that some regulatory bodies had in attempting to understand why lung cancer and other forms of cancer seemed more prevalent in smokers. But cigarette smoking alone could not reconcile the evidence. Social, ethnic, environmental, and economic factors are also very important in understanding the entire biological effect. In fact, the level of B[a]P in CSC could only explain about 2% of its specific tumorigenicity observed in skin-painted mice and the combination of the levels of all the known tumorigenic PAHs in CSC could only explain about 3% of its tumorigenicity. Despite an 18-month study in the late 1950s, the search for a ‘supercarcinogen’ in MSS and CSC to explain the observed biological effects was unsuccessful. In addition, the exceptional study on MSS PAHs by United States Department of Agriculture (USDA) personnel in the 1970s indicated no ‘supercarcinogen’ was present. Only recently has the concept of complex mixtures in relation to the understanding of the complexity of carcinogenesis taken hold. Perhaps the reason why MSS is less tumorigenic than expected in humans is because of the presence of other MSS components that inhibit or prevent tumorigenesis. For example, it is well known that MSS contains numerous anticarcinogens present in quantities significantly greater than those of the PAHs of concern. When one reviews the history of these four PAHs in MSS or CSC it is clear that many unanswered questions remain.