Hemin uptake and release by neurons and glia

Abstract

Hemin accumulates in intracerebral hematomas and may contribute to cell injury in adjacent tissue. Despite its relevance to hemorrhagic CNS insults, very little is known about hemin trafficking by neural cells. In the present study, hemin uptake and release were quantified in primary murine cortical cultures, and the effect of the hemin-binding compound deferoxamine (DFO) was assessed. Net uptake of (55)Fe-hemin was similar in mixed neuron-glia, neuron, and glia cultures, but was 2.6-3.6-fold greater in microglia cultures. After washout, 40-60% of the isotope signal was released by mixed neuron-glia cultures into albumin-containing medium within 24 h. Inhibiting hemin breakdown with tin protoporphyrin IX (SnPPIX) had minimal effect, while release of the fluorescent hemin analog zinc mesoporphyrin was quantitatively similar to that of (55)Fe-hemin. Isotope was released most rapidly by neurons (52.2 ± 7.2% at 2 h), compared with glia (15.6 ± 1.3%) and microglia (17.6 ± 0.54%). DFO did not alter (55)Fe-hemin uptake, but significantly increased its release. Mixed cultures treated with 10 μM hemin for 24 h sustained widespread neuronal loss that was attenuated by DFO. Concomitant treatment with SnPPIX had no effect on either enhancement of isotope release by DFO or neuroprotection. These results suggest that in the presence of a physiologic albumin concentration, hemin uptake by neural cells is followed by considerable extracellular release. Enhancement of this release by DFO may contribute to its protective effect against hemin toxicity.

Figure 1

Time course of hemin release from neuron-glia cultures, and effect of HO inhibition. A) Percentage ⁵⁵Fe signal in cell lysate and culture medium at indicated time points after ⁵⁵Fe-hemin loading; B) Effect of indicated concentrations (μM) of the HO inhibitor tin protoporphyrin IX (Sn) on percentage isotope release into the culture medium at 4 and 24 hours after ⁵⁵Fe-hemin loading; C) Percentage of zinc mesoporphyrin fluorescence signal in cell lysate and culture medium at indicated time points after loading. All values represent mean ± S.E.M. (*P < 0.05 vs. mean value in corresponding control cultures (CTRL) lacking tin protoporphyrin IX, Bonferroni multiple comparisons test, n = 6 per condition).

Figure 2

Comparison of ⁵⁵Fe-hemin uptake and release in neurons, glia, and microglia. A) ⁵⁵Fe-hemin signal (± S.E.M.) in cell lysates of indicated culture types after 2 hour treatment with 5 μM ⁵⁵Fe-hemin. B–D) % ⁵⁵Fe signal in cell lysate and culture medium at indicated time points after ⁵⁵Fe-hemin loading. ***P < 0.001 vs. mean value in microglia cultures, n = 8–10/condition).

Figure 3

Deferoxamine (DFO) increases release of cell hemin. A) Uptake: bars represent mean cell ⁵⁵Fe signal in mixed cortical cultures after two hour incubation with 5 μM ⁵⁵Fe-hemin alone (CTRL) or with 100 μM DFO or 2,2′-dipyridyl (DP); Release: mean medium ⁵⁵Fe signal at 24 hours after ⁵⁵Fe-hemin washout followed by incubation in medium containing 0.67 mg/ml bovine serum albumin alone (CTRL) or with 100 μM DFO or DP. B) Medium ⁵⁵Fe signal at 24 hours after ⁵⁵Fe-hemin washout followed by incubation in BSA-free medium alone (CTRL) or with indicated concentrations of DFO. C) As in A, testing the effect of concomitant treatment with the HO inhibitor tin protoporphyrin IX (Sn, 60 μM) on enhancement of ⁵⁵Fe-hemin release by DFO, in medium containing BSA. **P < 0.01, ***P < 0.001 vs. mean value in corresponding CTRL condition.

Figure 4

Deferoxamine attenuates hemin neurotoxicity. A) Bars represent mean LDH release (± S.E.M., n = 10–12/condition) in cultures treated for 24 hours with 10 μM hemin or with 10 μM hemin plus 100 μM DFO, 60 μM tin protoporphyrin IX (Sn) or both DFO and tin protoporphyrin IX. B) Cultures (n= 6/condition) were treated with indicated concentrations of hemin for 2 hours without DFO; after hemin washout, cultures were placed into medium containing or lacking 100 μM DFO. *P < 0.05, **P < 0.01 vs. corresponding condition treated with hemin without DFO. LDH values are normalized to those in sister cultures treated with NMDA for 24 hours (=100), which kills all neurons in these cultures. The LDH signal in controls exposed to washes only was subtracted from all values to calculate the signal due to hemin toxicity.