Technical Report 122

Relevance of inflammation

As evidenced by numerous experimental studies, pulmonary inflammation has to be regarded as a key driver in the cascade of pathogenic events following chronic inhalation of PSP including the neoplastic lung effects observed in rats. However, it remains unclear why only rats respond with the development of neoplasms of the lung but other animal species chronically exposed to PSPs do not. In fact, lung tumours were never found in rats when pulmonary inflammation was absent (Levy 1995, Oberdörster 1997; ILSI 2000; Greim et al, 2001; Kolling et al, 2011). In addition, a direct relationship between chronic inflammation and carcinogenesis in exposed rats has been established (Oberdörster, 1997; Kolling et al, 2011). Using a qualitative scoring system for specific non-neoplastic endpoints of lung toxicity such as inflammation, epithelial hyperplasia and fibrosis, a correlation between tumour frequencies and increased scores could be established (Kolling et al, 2011). Based hereupon it is hypothesised that the mechanism of tumour development presumably involves the severe, persistent inflammatory reaction which induces marked cell proliferation which has been demonstrated for several so-called inert particles (Driscoll, 1996). Increased epithelial cell proliferation may influence the number and survival of transformed cells, thus increasing the chances for lung tumour development (Oberdörster, 1995). According to Driscoll et al (1995, 1996) such a persistent inflammatory reaction can lead to an elevated influx of reactive oxygen species releasing activated neutrophils into the lung, which in turn can induce mutagenic effects in epithelial lung cells. It is speculated that the greater sensitivity of the rat lung with regard to oxidative stress and subsequent epithelial cell responses is due to a more pro-inflammatory environment compared to other species. This hypothesis, that lung tumours in rats following chronic exposure to PSP are induced by such an indirect mechanism, is supported by results of various other experiments. Inflammatory cells and activated alveolar macrophages, which are found in large numbers in animals exposed to PSP, can release ROS and other mediators of inflammation which in turn can then cause DNA damage by a secondary mechanism (Driscoll et al, 1997; Jackson et al, 1989; Weitzman and Gordon 1990; Oberdörster 1997). The following mechanistic findings were experimentally established:

1. Activated inflammatory cells (macrophages, granulocytes) from the rat lung cause DNA damage at the HPRT locus in vitro (Driscoll 1996)
2. The amount of DNA damage is directly dependent on the level of activation and on the capacity for generating ROS (Driscoll et al, 1997)

3. Exposure to concentrations which do not induce inflammatory reactions in the rat lung causes neither DNA damage nor lung tumour development (Driscoll, 1996; Dricoll et al, 1997)

Based on the outcome of a workshop on the toxicity of fibres and particles, it was concluded that the tumourigenesis of PSP involves a mechanism of secondary genotoxicity at doses that induce inflammation (Greim et al, 2001). Schins and Knaapen (2007) defined this secondary genotoxic effect as a pathway of genetic damage resulting from the oxidative DNA attack by reactive oxygen/nitrogen species, generated during particle-elicited inflammation. Measurable biological endpoints had been mutations in the HPRT gene of isolated alveolar cells in ex vivo assays (Drisoll et al, 1997), changes in markers of cellular injury and/or inflammation in bronchoalveolar lavage fluid (BALF), expression of mRNA for chemokines and detection of oxidative DNA damage (Johnston et al, 2000).

The hypothesis of secondary genotoxicity is based on findings that various PSP are carcinogenic in rat lungs, irrespective of their chemical composition, but obviously only after prolonged high exposures that are associated with persistent inflammation and lung overload (Miller, 2000; Greim et al, 2001; Borm et al, 2004). Importantly it has to be stressed that these substances do not present a carcinogenic hazard in rats at doses/concentrations not producing concurrently severe lung inflammation reactions. Based hereupon it can be concluded that the observed lung tumours in rats are not due to a substance specific toxicity but the result of generic particle toxicity in a highly sensitive species. Exposure to concentrations of inert particles which do not induce inflammatory reactions in the rat lung causes neither DNA damage nor lung tumour development (Driscoll et al, 1996) indicating that this mechanism has a threshold.

In the lung, neutrophilic granulocytes are considered to be a source of ROS/RNS. A significant increase in the level of DNA damage in the epithelial cells of the rat lungs exposed to non-genotoxic particles was found only if the number of neutrophils had increased to 40-50% of the total cell population (Driscoll et al, 1997). Other studies performed to detect any increases in the rate of proliferation of pulmonary epithelial cells after inhalative exposure of rats to an inert particle revealed an early onset of increased cell proliferation of pulmonary epithelial cells, assayed by means of bromodeoxyuridine (BrdU) indices distinguishing between cells labelled in the S-phase (BrdU positive) and unlabelled cells (BrdU negative). In all cases, exposed animals exhibited an epithelial cell labelling index significantly above control indicating increased proliferation of pulmonary epithelial cells (IPA 1999 cited by MAK 1999).

In conclusion, particle induced pulmonary inflammation is a key driver in the hypothesised cascade of PSP induced events leading to non-neoplastic as well as neoplastic changes in the rat lung. Thus, they are not driven by a substance specific toxicity but due to generic particle responses for which thresholds can be established. In a dose-dependent manner, the level of pro-inflammatory cellular responses (e.g. release or ROS/RNS) is increased and antioxidant defences are weakened. This in turn increases oxidative stress and the susceptibility of airway epithelium towards both, genetic damage in alveolar cells as well as advancing cell proliferation and tissue remodelling. One major conclusion to be drawn from the synopsis of existing data is that the tumourigenesis of PSP in rats apparently involves mechanisms of secondary genotoxicity only at doses that induce also inflammatory pulmonary responses in vivo. This is of significant importance for hazard and risk assessments, because such secondary mechanisms involve thresholds to be considered for the derivation of DNELs in risk characterisation as well as for risk management decisions.