The mutagenicity of indoor and outdoor airborne particulate matter (APM) has been demonstrated by previous in vitro studies (Alink et al., 1983; Van Houdt et al., 1984, 1986, 1987). The aim of the present thesis was to contribute to a better understanding of the mode of action of AIM in the pathogenesis of lung cancer by expanding the scope to the in vivo initiating (Chapter 3 and 4), tumor promoting (Chapter 5 and 6) and carcinogenic (Chapter 7) potential of APM.The aromatic DNA adduct level in white blood cells was used as a biomarker for measuring the exposure of man to residential wood combustion particulate matter under conditions which most likely reflect the pattern of use of open fire places in The Netherlands (Chapter 3). Although indoor air mutagenicity and benzo(a)pyrene (BaP) and pyrene levels were increased, no combustionrelated increase in the amount of adducts was observed. This indicates that these levels of aromatic compounds are probably not an important source of genotoxic damage to DNA.In Chapter 4 experiments are described in which outdoor as well as indoor AIM extracts were tested for in vivo genotoxicity in the DNA-repair host-mediated assay. Although the APM extracts were genotoxic in vitro, high doses of APM induced no genotoxic activity in mice. Because polycyclic aromatic hydrocarbons (PAHs) and nitroPAHs are known to contribute to the mutagenicity of APM, BaP and 2-nitrofluorene (NF) were used as positive controls. High concentrations of BaP caused a moderate effect in mice whereas NF was not genotoxic at all in vivo. It was concluded that the DNA-repair host-mediated assay is (relative) insensitive for PAHs and nitroPAHs. In addition, deactivation of APM in vivo might be responsible for the lack of genotoxicity observed.Because it is generally accepted that carcinogenesis is a multi-step process we also studied the tumor promoting potential of APM. Inhibition of gap-junctional intercellular communication (GJIC) was used for measuring tumor promoting effects (Chapter 5). We showed that outdoor as well as indoor APM strongly inhibited GJIC in V79 Chinese hamster lung fibroblasts. Mutagenicity data, obtained with Salmonella typhimurium TA 98, and the data on inhibition of GJIC were correlated, and the concentrations at which a positive effect was found, were in the same range. V79 cells lack the capacity to metabolize foreign compounds. Thus the effects on communication were caused by unmetabolized compounds. We also showed that communication was strongly inhibited in primary cultures of rat alveolar type II cells, a target cell for inhaled particles. Type II cells do have the capacity to metabolize foreign compounds, but it is unknown whether parent compounds and/or their metabolites inhibited IC.Because vitamin A is an important regulator of normal epithelial differentiation and proliferation and a disturbance in its kinetics might influence the process of carcinogenesis, interactions of APM with the kinetics of vitamin A were studied (Chapter 6). A single high dose of APM extract increased plasma retinol levels in rats whereas lung vitamin A levels were depleted. Such a depletion might indicate an increased susceptibility for the development of lung tumors.Finally, the carcinogenicity of outdoor APM in the newborn mouse bioassay was studied (Chapter 7). The APM samples were taken under conditions of increased air pollution. Because man is exposed to such high pollution levels only for a limited amount of time, the dose levels of APM applied in this animal model were an overestimation compared to the actual cumulative dose man is exposed to. BaP, used as a positive control, significantly increased the lung tumor incidence. On the contrary no evidence was found for a carcinogenic response to APM.In a recent assessment of the state of the art in chemical carcinogenesis by Gori (1992) it is mentioned that initiation may not solely be of a mutagenic nature, but may more reflect a shift of the normal behavior of cells towards an adaptation to stressors adverse to cell and organism homeostasis, mediated by epigenetic events. It is further stated in that study that "initiation may be a trivial step in carcinogenesis, given that all cells in an organism are exposed since embryonic development to multiple initiator stresses, either physiologically generated of from the myriad of naturally ocurring external encounters." The idea of Gori implicates also that the initiating properties of chemicals might be of less importance then generally is assumed. All this would plead for a more important role of tumor promotion in the process of carcinogenesis. For cigarette smoking this indeed seems to be the case. For people who stop smoking the risk freezes at the time of quitting without further advantages, suggesting a promotional role of smoking. Upon smoking cessation the effects of nongenotoxic promoting agents abruptly stop. As a consequence preneoplastic lesions remain static or regress, whereas in the continuing smoker they progress (Doll and Peto, 1978; Weisburger, 1990). Furthermore, from epidemiological data of cigarette smoking and lung cancer it can be concluded that a threshold exists at measurable doses, again pleading for an important role of tumor promotion (Gori and Mantel, 1991).As in the studies described in this thesis no evidence is found for an in vivo initiating or carcinogenic potential of APM, it might be that analogous with cigarette smoke also for APM the tumor promoting potential might be of more importance than the initiating potential. Therefore it is concluded that exposure to APM alone probably does not pose an important cancer risk to man. However, in combination with other factors a contribution of APM to the development of cancer can not be excluded.