Gated pore mutation affects ferritin iron in vivo

Abstract

Ferritins, spherical protein cages, concentrate iron and oxidants as ferric oxy biominerals. Eight pores, formed by junctions of three subunits, have crucial roles in iron removal from ferritin minerals based on the coincidence of pore unfolding (X‐ray crystallography, CD spectroscopy in mutant and wild type proteins) and changes in iron removal rates in frog model ferritins. Sequence similarity between human and frog ferritins is 65%, but quaternary structure and pore gates are the same. To examine the effects of pore mutations in human cell lines we made the analogous mutation in human H ferritin: L138P (L134P in horse/frog numbering). We observed, as in the frog mutant, there was no effect on cage assembly (gel filtration), and the initial rate of Fe removal from the mineral was faster (4.9‐fold, P<0.01), with 50% Fe removed from L138P in 2.2 min vs. 46.6 min from the wild type (WT). To study the pores in vivo, HeLa cells were transiently transfected with plasmids encoding H‐WT or H‐L138P ferritin, with comparable increases in ferritin protein (8 fold). When the retention of Fe in ferritin (cpm/ng ferritin protein) was compared, using cells incubated two hours with 59Fe‐transferrin, H‐L138P had only 25% (P<0.01) as much 59Fe as WT; the background used was 59Fe in endogenous ferritin (mock transfection). The data on ferritin pores in cultured cells, together with a study identifying a ferritin binding peptide with effects similar to L134 mutation in vitro, (Liu et al, JBC, 282:31821), suggest that cytoplasmic factors regulate recovery of iron from ferritin in vivo by pore folding/unfolding. Part support (NIH Grant DK20251).