Fe(III) and Fe (II) mostly. The transition between these two states is among the major causes for the oxidative stress (OS) in a living body. You can see the exact mechanisms in the book of Halliwell, but also in any book or review paper on any aspect of the OS issue. Not only Fe, but any metal ion that can change its valency reversibly (such as Cu for example) can prompt OS. To prevent the spontaneous oxidation being prompted by the free iron ions, they travel within the body coordinatively bonded with Fe-bonding transport proteins. If latter fail or missing, the Fe ions induce massive peroxidation of bioactive molecules (DNA< RNA< proteins, lipids, etc) leading to cellular damage, tissues oxidative injuries and finally - to pathologies.
When iron is not bonded to the transporter - the labile bivalent iron reacts with hydrogen peroxide and is oxidated to Fe(III) via the Fenton`s reaction leading to the composition of extremely reactive hydroxyl radical which can damage almost every molecule in the cell.
@Maria Thomas, there didn't seem to be any a few years ago, at least none that were specific to iron. I used expression levels of certain iron-dependent transcripts as a surrogate measurement for the state of intracellular iron. The stability of transferrin receptor, dmt1, ferritin heavy and light chain mRNA transcripts are all well characterized in response to fluctuations in iron. Thus by measuring their changes, one can assess the changes in intracellular iron concentration. Check out this paper if interested:
Article An Optimal Method of Iron Starvation of the Obligate Intrace...
Thank you, Christopher! So, the assessment of dmt1, ferritin and perhaps hepcidin, Hamp gene, and Alas1 expression could serve as an "iron"-relevant gene signature