Iron Metabolism in the Bone Marrow as Seen by Electron Microscopy: A Critical Review

High resolution electron microscopy has made possible the visualization of transport and storage iron in the form of ferritin, both in dispersed form and in aggregates and in the form of "iron micelles" in mitochondria. Hemosiderin was found to consist either of pure ferritin in crystalline clusters or, more frequently, of ferritin associated with other substances, including a lipid component in the form of myelinic figures and PAS positive material.

In the following paragraphs we have summarized the new morphologic findings and what appears to us the most likely interpretation in the light of known biochemical and isotopic studies. Alternative interpretations have been discussed in the body of the paper.

Electron microscopy has established the erythroblastic island as a morphologic and functional unit of the bone marrow. A central reticular "nurse cell" appears to impart nutrients to surrounding rows of erythroblasts by the process of rhopheocytosis. Transfer of ferritin by this process is probably a passive phenomenon, since the amount transferred parallels the amount of iron present in the central reticular cell. Ferritin is increased both in the reticular cell and in erythroblasts in hemochromatosis. It is absent in iron deficiency, although rhopheocytosis remains prominent. Normally all erythroblasts (proerythroblasts and normoblasts) and reticulocytes contain ferritin. Only the larger aggregates can be visualized by the Prussian blue reaction in sideroblasts and siderocytes. Ferritin generally disappears when reticulocytes mature, even in hemochromatosis and infections, two conditions in which there is an excess of ferritin in erythroblasts. Interestingly, the increase in infections is entirely in form of dispersed ferritin and cannot be visualized by the Prussian blue reaction; i.e., sideroblasts are absent, in contrast to hemochromatosis where they are normal or increased.

It appears most likely that ferritin disappears from normal maturing reticulocytes because it is utilized for hemoglobin formation. It persists in mature red cells in Cooley’s anemia, hypersideremia, hypochromic anemia and lead poisoning where hemoglobin formation is disturbed.

The origin of the ferritin in the nurse cells and the extent to which ferritin rather than siderophilin contributes to hemoglobin synthesis are unsolved problems. Isotopic studies indicate that almost all of the iron used for hemoglobin synthesis is derived from siderophilin and hemoglobin synthesis can proceed without any visible ferritin, as in iron deficiency anemia. These facts must be reconciled with the electron microscopic observations which suggest that normally some iron reutilization within the marrow proceeds by way of erythrophagocytosis, fragmentation, intracellular hemolysis of red cells, formation of ferritin and ropheocytosis. Iron derived from erythrophagocytosis elsewhere in the body probably reaches the marrow bound to siderophilin. Such iron can be incorporated into ferritin of reticular cells as may be seen in hyperferremia and following injection of iron compounds. The process of rhopheocytosis would then lead to utilization of at least part of this ferritin iron for hemoglobin synthesis.

In certain pathologic states, accumulation of ferritin and related visible dispersed or conglomerated iron micelles may point to the sites where hemoglobin synthesis or iron transport is blocked. In Cooley’s anemia and the hypersideremic hypochromic (non-thalassemic) anemias, iron accumulates in the mitochondria, which are known to be involved in hemoglobin synthesis. In lead poisoning, the mitochondria are markedly abnormal, and probably correspond to the areas of punctate basophilia. However, the iron accumulates in other areas of the cell, suggesting a different type of block.