Abstract
Apoptotic eosinophils (EOs), atypically, release cytotoxic specific granule proteins that promote tissue damage in eosinophilic inflammatory states such as asthma. PMN myleloperoxidase strongly promotes PMN apoptosis by generating HOCl. We tested whether EPO plays a similarly pivotal role in EO apoptosis. In vivo, bromide (Br−), nitrite (NO2−), and thiocyanate (SCN−) compete for oxidation by EPO and H2O2 yielding, respectively, HOBr, NO2°, and HOSCN. We have shown that SCN- is the strongly preferred substrate for EPO in vivo and that HOSCN, in striking contrast to HOBr, NO2° and HOCl, is a weak, sulfhydryl-specific oxidant. Because we recently showed that HOSCN is a uniquely potent oxidant activator of endothelial cell NF-kB (Blood 107:558 2006), a powerful antagonist of apoptosis, we hypothesized that endogenously generated HOSCN would inhibit EO apoptosis. Blood EOs were isolated from mildly atopic donors by immunomagnetic separation to > 95% purity. EOs were cultured in RPMI + 10% FCS without added IL-5 and assayed for cell viability by Annexin V and Propidium Iodide (PI) staining and flow cytometry. Apoptosis was confirmed using an immunoassay of cytoplasmic histone-associated DNA fragments, caspase 3 activation and morphology. In every time course examined, EOs were first annexin+/PI-, then later annexin+/PI+. Therefore data are here reported as viability normalized to a control of EOS with 1 ng/ml IL-5, the remainder being comprised of early and late apoptotic cells. EOs cultured 2 days with 1 mM SCN− were 69% viable, a 77% relative increase (n=9, p < 0.0001) over those cultured with nothing (i.e., Cl- only), 1 mM Br-, or 1 mM NO2-, all of which had the same viability (~39%). When 0.5 nM PMA was added to activate the respiratory burst, viability with SCN− after 2 days was 63%, Cl- 5%, Br- 2%, and NO2- 14%. Surprisingly, viability with PMA and SCN− was 20% higher than that with Cl- without PMA (p < 0.05), suggesting that HOSCN not only fails to promote apoptosis but instead engenders an anti-apoptotic tone. Addition of the EPO inhibitor azide (1 mM) abrogated the protective effect of SCN− with PMA. Moreover, SCN- failed to protect EPO-devoid monocytes and lymphocytes from both spontaneous and PMA-induced apoptosis. EOs activated with the physiologic agonists C5a (33 nM) and PAF (5 μM) exhibited the same protective effect of SCN− and increased viability in activated vs. unactivated EOs. EOs treated 2 hours with and without 1mM H2O2 before adding an agonist anti-Fas antibody (1 μg/ml) had viabilities ~30% higher with SCN− than with the other halides at day 1 and at day 2, late-apoptotic cells were 42% that in Cl-. BAY 11-7085 (10μM), an inhibitor of NF-kB activation, caused rapid EO apoptosis (70% at day 1) but in this setting SCN− was not protective with or without PMA activation. Unlike PMN, Western blots for IkB-alpha showed no degradation with PMA irrespective of halide. We conclude that HOSCN generated endogenously in EOs by the EPO/H2O2/ SCN− system plays a previously unsuspected role to maintain both constitutive and agonist-stimulated EO survival. HOSCN antagonizes EO apoptosis through a mechanism that may require constitutive, but not inducible, activation of NF-kB. Because serum SCN− levels vary widely (10–300 μM) and are dietarily determined, oral supplementation with this inexpensive and innocuous pseudohalide may mitigate tissue injury in eosinophilic inflammatory states by inhibiting EO apoptosis in infiltrated organs.
Protective Effect of Diethyldithiocarbamate Pretreatment on Chromium(VI)-Induced Cytotoxicity and Lipid Peroxidation in Primary Cultures of Rat Hepatocytes.
