Alternatives to Laboratory Animals

Clive Hollands, former Director of the Scottish Society for the Prevention of Vivisection, Secretary of the Committee for the Reform of Animal Experimentation, former member of the Animal Procedures Committee, and member of the Farm Animal Welfare Council, died in November 1996 at the age of 67. His many contributions to human and animal welfare are reviewed by some of his many friends and colleagues.

α-Hexachlorocyclohexane (α-HCH) was examined for cytotoxicity, genotoxicity and cytochrome P450 induction in primary cultures of mouse, rat and human hepatocytes and in three hepatoma cell lines (Hepa 1c1c7, FaO and Hep G2, from mouse, rat and man, respectively). The cell lines were much more sensitive to the cytotoxicity of the classical inducers phenobarbital and 3-methylcholanthrene than that of α-HCH, whereas no cytotoxicity was observed in primary hepatocytes. Exposure for 24 hours to 0.32mM α-HCH produced a modest, but statistically significant, frequency of DNA breaks, as measured by the alkaline elution assay, in the mouse Hepa 1c1c7 cell line, and the human Hep G2 cell line, but not in the rat FaO cell line. In the Hep G2 cell line, the amount of DNA fragmentation was found to increase with the length of exposure. Compared with the results of previous observations on primary cultures, with regard to species specificity, only the human cell line gave a concordant positive response. Monooxygenase activity induction in primary hepatocytes, despite rather high initial levels of 7-ethoxycoumarin-O-deethylase activity, was low with the classical inducers phenobarbital and 3-methylcholanthrene. α-HCH caused no induction of monooxygenase. The rat FaO and human Hep G2 cell lines were sensitive to α-HCH, but only after long exposure. The results of this study support the hypothesis that α-HCH might act as a weak genotoxic agent in humans, but they also suggest caution in the extrapolation to the in vivo situation of the observations made in established cell lines.

Although primary cultures of human thyroid cells are used for endocrinological and toxicological studies, until now no attention has been paid toward verifying whether the hormonal conditions to which the gland was exposed in vivo prior to surgery could influence in vitro responses. Our findings suggest that the hormonal situation in vivo cannot be used as a predictive indicator of triiodothyronine and thyroxine release and/or S-phase frequency in vitro, either with or without the addition of bovine thyrotropin.

The effects of four cosmetic surfactants on interleukin (IL)-1α and IL-8 release from human keratinocytes were studied to investigate the feasibility of using these effects for the prediction of the irritation potential of chemicals. After exposure of cells to surfactants, the amounts of IL-1α and IL-8 released into culture medium were measured by ELISA. Cytotoxicity was evaluated by using the neutral red uptake (NRU) cytotoxicity assay. Cytokine release was increased 7–15 times by sodium lauryl sulphate (SLS), laurtrimonium chloride, cocamidopropyl betaine (CPB) and Oleth-5 at cytotoxic concentrations. IL-8 release was increased 3–4 times by SLS, CPB and Oleth-5 at subcytotoxic concentrations. After exposure to SLS, IL-1α was released within 1 hour, suggesting that IL-1α release is associated with membrane damage, whereas IL-8 release continued for 24 hours, suggesting that IL-8 was produced within the cells. Cytotoxicity tests and IL-8 release assays were also performed on seven other surfactants. The results show that moderate irritants CPB and PEG-4 dioleate, which have weak cytotoxic effects, significantly increased IL-8 release from human keratinocytes. It is suggested that measurement of IL-8 release is useful for predicting the irritation potential of chemicals which cannot be detected by using the NRU cytotoxicity assay.

This study describes an alternative adjuvant for the immunisation of chicken. The lipopeptide, Pam₃Cys-Ser-(Lys)₄ (PCSL), has been demonstrated to be a very potent adjuvant, by immunisation with the hapten methyl phosphonic acid paraaminophenyl 1,2,2-trimethylpropyl diester, with coronavirus, with rotavirus and with the Escherichia coli K99 pilus antigen. The antibody titres in chickens that received PCSL were compared with those in animals that received Freund's complete adjuvant/Freund's incomplete adjuvant (FCA/FIA). Antigen-dependent differences in antibody titres could be shown. The optimal dosage for PCSL was 250μg per injection. Whereas injection of FCA/FIA resulted in chronic inflammatory alterations, mostly accompanied with granulomatous reactions, no long-term tissue damage could be found with PCSL. When PCSL was used as the adjuvant, the total immunoglobulin Y (IgY) levels in chicken sera were constant during the immunisation period (20.2mg/ml, SD ± 1.3), and were significantly different (p < 0.0001) from those in chickens given FCA as adjuvant (26.6mg7ml, SD ± 4.9). IgY concentrations in the egg yolks showed no significant differences in both groups (PCSL — 13.7mg/ml, SD ± 3.7; FCA — 13.5mg/ml, SD ± 4.2), and were significantly lower (p < 0.0001) than the IgY serum levels.