Cell Biology and Metabolism Branch, National Institute of Child Health and Human Development, Bethesda, Maryland 20892, USA. trou@helix.nih.gov
Accumulations of iron are often detected in the brains of people suffering from neurodegenerative diseases. But it is often not known whether such accumulations contribute directly to disease progression. The identification of the genes mutated in two such disorders suggests that errors in iron metabolism do indeed have a key role.
For many years it has been well accepted that accumulations of iron in organs such as the liver and heart can cause disease. In disorders such as genetic hemochromatosis and thalassemia, hepatic iron overload causes cirrhosis; cardiac iron overload leads to heart failure1. Accumulations of iron are also frequently observed in areas of the brain that degenerate in disorders such as Parkinson and Alzheimer diseases. But the importance of iron accumulation in the progression of most neurodegenerative diseases has been unclear2, in part because all aging humans normally accumulate iron in brain regions such as the substantia nigra. On pages 345 and 350 of this issue, Curtis and colleagues3 and Zhou and co-workers4 provide some answers. They describe how they used genetics and positional cloning to identify the genes defective in two neurodegenerative diseases that are characterized by profound accumulations of iron in the brain.
Iron and neurodegeneration. A model of iron damage to axons. Ferritin is an iron-storage protein. Ferritin subunits are synthesized in the neuronal cell body, and mature, assembled heteropolymers are found within axons6. Some of the ferritin in distal axons and presynaptic terminals are degraded within lysosomes, potentially releasing ferrous iron into a region of the neuron in which proteins, such as components of neurofilaments, are vulnerable to iron-binding and oxidative damage. Alternatively, ferritin heteropolymers that contain abnormal subunits spontaneously release free iron into the axon or synapse.
Bob Crimi
Curtis et al.3 first describe a previously unrecognized, adult-onset neurodegenerative disease that affects the basal ganglia and is associated with iron accumulation. They then identify a dominantly inherited gene that is mutated in the affected individuals, who are from northern England. All of the patients have the same mutation—insertion of an adenine within the portion of the gene that encodes the carboxy terminus of the L chain of ferritin, an iron-storage protein. The mutation leads to synthesis of a unique 22-amino-acid carboxy terminus.
In the brains of these patients, the globus pallidus shows abundant spherical inclusions that contain ferritin. Throughout the white matter of the brain, axonal swellings are immunoreactive for neurofilaments, ubiquitin and tau—a characteristic of neurodegenerative diseases. Interestingly, however, serum ferritin levels are remarkably low, and the pancreas, liver and heart appear to function normally. So, although the abnormal ferritin is almost certainly ubiquitously expressed, it would seem that only neurons develop significant pathology. The authors propose that this disorder should be referred to as 'neuroferritinopathy'.
On the basis of the crystal structure of ferritin, Curtis et al.3 suggest that the altered carboxy terminus may affect the protein's function and stability. Ferritin is a heteropolymeric iron-storage protein, composed of H and L subunits that assemble to form a hollow sphere in which ferric iron precipitates are sequestered5. The carboxy terminus of the aberrant L chain might interfere with stable polymer formation, perhaps allowing inappropriate release of iron from iron-laden ferritin.
Interestingly, the simultaneous overexpression of the ferritin H and L chains is implicated in the progression of a newly described neurodegenerative disease in mice that lack iron-regulatory protein-2 (IRP2)6. Here, axonal degeneration is the earliest pathological event identified in neurons that overexpress ferritin. Immunohistochemical staining of wildtype and Irp2-/- mice revealed ferritin throughout the length of axons, hinting that ferritin is normally transported from its site of synthesis at cell bodies to synapses. In Irp2-/- mice, however, the levels of ferric iron in axons are markedly increased, and much of this iron is probably sequestered within the overexpressed ferritin.
There could be many reasons why neurons degenerate in the newly described human neuroferritinopathy and in Irp2-/- mice. But it is attractive to consider models that could be applied to both diseases. It is difficult to ascertain where the initial neuronal damage occurs in humans3, but the studies of Irp2-/- mice6 point to the axon. The transport of ferritin down the axon might therefore be the key to the pathology associated with the overexpression of both ferritin subunits in Irp2-/- mice, and with the production of the abnormal ferritin L chain in humans with neuroferritinopathy.
Neurons are highly polarized, with proteins being synthesized in the cell body and then transported over potentially long distances to synapses. Ferritin probably sequesters iron in the cell body, but transport of ferritin down the axon may allow net transport of iron to synapses. Normally, the half-life of ferritin appears to be determined by its degradation in lysosomes7; in the Irp2-/- mice, the degradation of overexpressed ferritin would lead to the increased release of free iron. This could occur within lysosomes present in distal axons8. The positively charged iron atoms could then bind to axonal proteins, including negatively charged components of neurofilaments, causing oxidation and loss of function. In people with neuroferritinopathy, perhaps the ferritin heteropolymers containing the aberrant L chain are inherently unstable; here, the release of iron might also occur spontaneously as ferritin is transported down the axon. Thus, iron-dependent oxidative damage to the axon may be an early event that is common to both diseases. The highly polarized nature of the neuron, together with axonal trafficking of iron-laden ferritin, may explain why significant pathology is seen only in the nervous system.
Meanwhile, Zhou et al.4 have identified the defective gene that causes one of the neurodegenerative diseases in which iron accumulation is most dramatic — an autosomal, recessively inherited disease known until recently as Hallervorden−Spatz disease. Adolescents and young adults with this disease develop a progressive, disabling movement disorder, characterized by spasmodic and uncontrollable movements of the trunk and limbs and by distorted body positions. In autopsied brains, accumulations of iron in the globus pallidus and pars reticulata of the substantia nigra are profound.
Zhou et al. detected the underlying mutations in a gene that encodes pantothenate kinase. This enzyme is essential in coenzyme A biosynthesis, and catalyses the phosphorylation of pantothenate (vitamin B5) and related substrates. The product of this reaction, 4'-phosphopantothenate, is then converted to 4'-phosphopanthetheine in a reaction that consumes cysteine. Humans have four genes encoding pantothenate kinases, and the one identified here, PANK2, appears to be expressed specifically in the brain, potentially explaining the pattern of pathology in specific regions of the central nervous system.
Moreover, given that cysteine is consumed in the conversion of 4'-phosphopantothenate, an absence of functional PANK2 might also explain the previously observed accumulation of cysteine in the degenerating brain areas of patients with Hallervorden−Spatz disease9. In addition, the well-known iron-chelating properties of cysteine might account for the observed regional iron accumulations, and cysteine-bound iron may promote iron-dependent oxidative damage in these regions. So, although pantothenate kinase-2 is not directly involved in iron metabolism, its absence may contribute to secondary iron accumulation, the consequence of which might still be important in the disease process.
The best news to come from the identification of this disease gene is that the mutations are loss-of-function mutations, so delivery of 4'-phosphopantothenate or coenzyme A might prevent neurodegeneration. The generation of Pank2−null mice will allow treatment strategies to be evaluated.
There is still much to learn about the basic biology of this disease, which Zhou et al.4 suggest should be called 'pantothenate-kinase-associated neurodegeneration', or 'PKAN'. Likewise, the mechanisms underlying the neuroferritinopathy described by Curtis et al.3 remain to be determined. However, these discoveries may lead to new insights into these neurodegenerative diseases and to the design of specific treatments—and, not least, to respect for the importance of the regulation of iron metabolism in cells that must survive and function over the lifespan of human beings.
