Mutagenicity

Of the available reverse mutation studies in Salmonella typhimurium, 10/13 reported that perylene was at least weakly mutagenic in the presence of metabolic activation in one or more strains (Carver et al., 1986, 1985; Sakai et al., 1985; De Flora et al., 1984; Lofroth et al., 1984; De Flora, 1981; Ho et al., 1981; Florin et al., 1980; Lavoie et al., 1979; Anderson and Styles, 1978) (see Table 4A). In general, perylene induced borderline or low reverse mutation rates in comparison to other mutagens (e.g., other PAHs) (Hera and Pueyo, 1988; Carver et al., 1986; De Flora et al., 1984; Lofroth et al., 1984; De Flora, 1981; Ho et al., 1981; Florin et al., 1980; Anderson and Styles, 1978). Three studies indicated that perylene was not mutagenic with metabolic activation (Hera and Pueyo, 1988; Anderson et al., 1987; Salamone et al., 1979). Two of these studies [Anderson et al. (1987) and Hera and Pueyo (1988)] used low concentrations of S9 (2–3%), which may have been inadequate to promote metabolic transformation; however, Salamone et al. (1979) reported negative results with 10% S9. Perylene also induced forward mutations in S. typhimurium strains in the presence of metabolic activation (Hera and Pueyo, 1988; Thilly et al., 1983; Penman et al., 1980; Kaden et al., 1979). In contrast to reverse mutations, the forward mutagenic potential of perylene in the 8-azaguanine resistance assay was similar or more potent compared to other PAHs (e.g., benzo[a]pyrene [BaP]) (Thilly et al., 1983; Penman et al., 1980; Kaden et al., 1979). In the absence of metabolic activation, perylene did not induce reverse mutations in S. typhimurium in four of six assays (Anderson et al., 1987; Sakai et al., 1985; Lofroth et al., 1984; Salamone et al., 1979) and was weakly mutagenic in one or more strains in two of six assays (De Flora, 1981; Florin et al., 1980). Perylene did not induce forward mutations in S. typhimurium in the absence of metabolic activation (Penman et al., 1980).

Perylene was not mutagenic in metabolically competent human cell lines (Durant et al., 1996; Crespi and Thilly, 1984) or in human cell lines with exogenous metabolic activation (Penman et al., 1980).

Clastogenicity

Evidence from in vitro studies in mammalian cells regarding clastogenicity is mostly negative. Sister chromatid exchanges (SCEs) were not observed in Chinese hamster V79 cells in the absence of metabolic activation (Popescu et al., 1977). Similarly, chromosomal aberrations (CAs), specifically chromosomal G bands, were not observed during metaphase in human peripheral lymphocytes or Syrian hamster embryonic (SHE) cells in the absence of metabolic activation (Popescu et al., 1977; DiPaolo and Popescu, 1974). However, the frequency of CAs was increased in Chinese hamster V79 cells without metabolic activation, including dicentric chromosomes, chromatid gaps, and chromatid interchanges (Popescu et al., 1977). None of the in vitro clastogenicity studies were conducted in the presence of metabolic activation. In a host-mediated assay, SCEs were not observed in Chinese hamster V79 cells implanted into the peritoneal cavity of mice prior to perylene injection (Sirianni and Huang, 1978).

DNA Damage and Repair

Findings were generally negative in in vitro assays of DNA damage and repair. In Escherichia coli, perylene did not induce DNA damage or repair without metabolic activation; with activation, findings were negative or equivocal/borderline (Mersch-Sundermann et al., 1992; von der Hude et al., 1988; De Flora et al., 1984). In mammalian cells, perylene did not induce DNA synthesis in Sprague Dawley rat primary hepatocytes or unscheduled DNA synthesis in SHE cells in the absence of metabolic activation; assays were not conducted in the presence of metabolic activation (Zhao and Ramos, 1995; Casto et al., 1976).

Perylene did not form DNA adducts in cultured human peripheral blood lymphocytes (Gupta et al., 1988). Similarly, no DNA adducts were identified in mouse skin following repeated topical application of perylene to shaved skin of mice (four exposures over 54 hours) (Reddy et al., 1984).

Cell Transformation

Perylene did not induce cell transformation in SHE cells in the absence of metabolic activation (Popescu et al., 1977; Casto et al., 1973). In an assay specifically designed to evaluate tumor initiation and promotion potentials of compounds, perylene induced transformation foci in metabolically competent Bhas 42 cells (derived from BALB/c mouse 3T3 cells) in both initiation and promotion stages (Asada et al., 2005).

Immunotoxicity

Humoral immunity in mice was not suppressed following subcutaneous (s.c.) injection of perylene for 1 or 14 days (see Table 4B), as measured by antibody responses to an antigen (sheep red blood cells [sRBCs]) in a plaque-forming cell (PFC) assay (White et al., 1985). The study also evaluated spleen weight, spleen cellularity, and thymus weight, and observed no statistically significant changes. A statistically significant increase in body weight was reported following a 14-day exposure to perylene. This study compared the response of perylene to the robust immunosuppressive response of BaP administered at the same dose level (40 mg/kg-day). It is unclear whether perylene would have elicited an immunosuppressive response if given at higher doses, for a longer duration, or via a different route of exposure. No additional in vivo studies evaluating immunotoxicity were identified. In an in vitro assay in human skin keratinocytes, perylene induced dose-dependent inflammatory cytokine secretion (interleukin-1α, interleukin-6) (Bahri et al., 2010).

Table 4B

Other Supporting Studies of Perylene.

Cancer

Available skin painting studies (see Table 4B), while limited by design flaws (lack of negative controls, potentially inadequate dose levels, small group sizes) and/or limited reporting (abstract only), do not indicate that perylene is a complete dermal carcinogen or cocarcinogen. No skin tumors were observed in 16 mice following exposure to 0.15% perylene twice weekly in a decalin vehicle for up to 82 weeks; in contrast, the same protocol with 0.14% BaP induced skin tumors (Horton and Christian, 1974). Similarly, no skin tumors were observed in 20 mice exposed to 0.3% perylene in benzene every 4 days for up to 24 weeks, while exposure to 0.3% BaP induced skin tumors using the same protocol (Finzi et al., 1968). In a study only available as an abstract, dermal application of 1% perylene (vehicle not reported) 3 times weekly for 1 year did not increase the number of skin tumors, compared to vehicle control (incidence data not reported) (Anderson and Anderson, 1987).

Skin tumor initiation-promotion studies do not indicate that perylene is a strong tumor initiator or promotor. Perylene was not a mouse skin tumor initiator following single exposure (800 μg in benzene) or 10 repeated exposures (total dose 1 g in acetone) with 12-O-tetradecanoylphorbol-13 acetate (TPA) as a promotor (El-Bayoumy et al., 1982; Van Duuren et al., 1970). In both studies, the positive control group (BaP or 7,12-dimethylbenz[a]anthracene) showed expected tumor-initiating activity. In a study only available as an abstract, perylene was not a tumor promotor when applied as a 1% solution (vehicle not reported) 3 times weekly for 1 year following a single initiating dose of 300 μg BaP (Anderson and Anderson, 1987).

Two dermal studies also evaluated carcinogenic potential of perylene in the presence of other chemicals. Horton and Christian (1974) evaluated cocarcinogenic potential of 0.15% perylene twice weekly in a decalin/dodecane vehicle (50:50 ratio). Dodecane had previously been shown to be a skin cocarcinogen with benz[a]anthracene, but is not a skin carcinogen when administered alone (Horton and Christian, 1974). Animals coexposed to perylene did not have significantly increased skin papillomas (1/15) compared with animals exposed only to decalin/dodecane (2/13). In contrast, BaP and dodecane showed evidence of cocarcinogenicity following a similar protocol (Horton and Christian, 1974). In a study by Finzi et al. (1968), perylene coexposure decreased the number of skin tumors induced by BaP (13/40) compared to BaP alone (36/40); however, statistical analysis of the data was not performed. The study authors proposed that perylene prevented BaP-induced tumor formation via competitive binding to epidermal proteins.

In a study only available as an abstract, intraperitoneal (i.p.) injections of perylene at doses of 200–1,000 mg/kg 3 times/week for 8 weeks did not induce lung tumors in Strain A mice, a sensitive model for detecting lung tumor induction, following a 16-week postexposure observation period (Anderson and Anderson, 1987). Due to the limitations of these available injection and dermal cancer studies, it is not possible to thoroughly assess perylene’s carcinogenic potential following oral and inhalation exposures.