Antitumor activity of eosinophils activated by IL-5 and eotaxin against hepatocellular carcinoma.

We examined the antitumor effects of eosinophils to explore the potential of eosinophils as effector cells in tumor cytotoxicity. We expressed eotaxin in hepatocellular carcinoma cells, MH134, and injected them into either normal or IL-5 TG mice intradermally and monitored cell growth. In normal mice, growth of MH134 cells containing the expression plasmid pCXN2-eotaxin was similar to that of vector-transfected MH134 cells for a period of 2 weeks, suggesting that expression of eotaxin does not change the growth rate of tumor cells. In IL-5 TG mice, however, the growth of eotaxin expressing MH134 cells was significantly suppressed. LPS induced eosinophils to produce TNF-alpha to kill MH134 cells in vitro. Intratumor injection of LPS is effective to kill MH134-pCXN2 and MH134-pCXN2-eotaxin only in normal mice. Administration of anti-CD4 or anti-CD8 antibodies suppressed growth of MH134-pCXN2-eotaxin cells compared with control antibodies, suggesting that T cells may interfere with immunity against MH134. Administration of anti-IL-5Ralpha and anti-asialo GM1 antibodies enhanced growth of MH134-pCXN2-eotaxin cells, suggesting involvement of eosinophils and NK cells in suppression of tumor cell growth. Although we cannot exclude the possibility that NK cells participate in tumor cell killing in vivo, the presence of NK markers such as DX5, asialo GM1, Ly49, and CD94, and NKG2D on large numbers of eosinophils activated by eotaxin suggests that eosinophils function in such suppression of tumor cell growth. Furthermore, we showed that anti-NKG2D antibodies could significantly inhibit the LPS-induced cytotoxicity against MH134 by highly enriched fraction of eosinophils.

IL-5-Induced Eosinophils Suppress the Growth of Leishmania amazonensis In Vivo and Kill Promastigotes In Vitro in Response to Either IL-4 or IFN-gamma.

In IL-5 transgenic mice (C3H/HeN-TgN(IL-5)-Imeg), in which 50% of peripheral blood leukocytes are eosinophils, the development of infection by Leishmania amazonensis was clearly suppressed. To determine mechanistically how this protozoan parasite is killed, we performed in vitro killing experiments. Either IL-4 or IFN-gamma effectively stimulated eosinophils to kill Leishmania amazonensis promastigotes, and most of the killing was inhibited by catalase but not by the NO inhibitor L-N5-(1-iminoethyl)-ornithine, suggesting that hydrogen peroxide is responsible for the killing of L. amazonensis by eosinophils. There was no significant degranulation of eosinophils in the culture, because eosinophil peroxidase was not detected in culture supernatants when L. amazonensis promastigotes were killed by activated eosinophils. Such resistance was also observed in BALB/c mice, which are highly susceptible to L. amazonensis. Expression plasmids for IL-4, IL-5, and IFN-gamma were transferred into muscle by electroporation in vivo starting 1 week before infection. Expression plasmid for IL-5 was most effective in slowing the development of infection among three expression plasmids. Expression plasmid for IL-4 was slightly effective and that for IFN-gamma had no effect on the progress of disease. These results suggest that IL-5 gene transfer into muscle by electroporation is useful as a supplementary protection method against L. amazonensis infection.