Liz Milward led this live discussion on 22 January 2002. Readers are invited to submit additional comments by using our Comments form at the bottom of the page.

Summary of Live Discussion Comments

Participants: Ashley Bush, James R. Connor, Ernest Beutler, John Olynyk, Sharon Moalem, Liz Milward, Maire Percy, George Perry, Maurizio Sampietro, Shai Shoham, Hyman M. Schipper.

Transcript unavailable.


Background Text
By Liz Milward

We are proposing that polymorphisms/mutations in a gene known to be important in hemochromatosis, or 'iron overload' disease, can also contribute to Alzheimer's disease in some patients. We believe this requires attention for several reasons, most importantly because iron overload can be controlled. This means there could be the potential for immediate therapeutic actions to delay AD onset and/or progression in affected people.

Hemochromatosis is a condition resulting from excessive uptake of dietary iron that is subsequently deposited in the liver, heart, pancreas and other organs. Although hemochromatosis has not been traditionally associated with brain iron loading or neurological sequelae, this is now being questioned. Modern iron staining techniques and MRI show brain iron loading in hemochromatotic patients. While in the past, many hemochromatosis sufferers did not survive past their 40s or 50s, most patients now have normal lifespans due to improved treatment. These people may develop brain conditions that would not previously have been seen. Furthermore, even 'subclinical' iron loading might eventually have cumulative consequences as lifespans increase. A new Framingham study suggests high iron stores are far more common in the elderly than has been previously realized.

It has been recognized for many years that iron accumulates in the AD hippocampus and other severely affected brain regions, in association with neuritic plaques and neurofibrillary tangles. While so far the relevance of this iron accumulation to the pathogenesis of AD has been uncertain, there are various feasible mechanisms by which iron excess could contribute to AD. One simple hypothesis is that iron overload superimposes on other AD processes to accelerate brain damage and exacerbate symptoms, possibly by amplifying oxidative damage to neurons. There could also be more direct mechanisms whereby excess iron alters amyloid deposition in vivo.

Hemochromatosis is the most common genetic disease so far identified, with around 1 in 200 people severely affected. Most people with hemochromatosis have mutations in the HFE gene, discovered in 1996 within the major histocompatibility (MHC)/HLA locus on the short arm of chromosome 6 and named systematically by the WHO Human Gene Nomenclature Committee. Most patients with severe hemochromatosis are homozygous for the major C282Y mutation of the HFE gene. This mutation has an allelic frequency of 1-15% in Caucasian populations, being frequent in populations of Celtic ancestry but less common in Mediterranean countries. The second common mutation is H63D (allelic frequency 10-20%), which usually has less severe effects on iron status than C282Y. Many homozygotes and some heterozygotes, particularly C282Y/H63D compound heterozygotes, will develop clinical hemochromatosis with aging. Overall estimates suggest 20-40% of people with European ancestry carry at least one mutant HFE allele. Because these mutations are so common, a significant percentage of AD patients can be expected to carry one or more HFE mutations.

The HFE protein is proposed to interact with the transferrin (Tf) receptor to regulate cellular iron uptake. Mutations leading to abnormal HFE protein can dysregulate uptake, causing cellular overloading. Iron uptake into the brain by transcytosis across the blood-brain barrier and at other sites is likewise carried out by the Tf receptor. We have now reported evidence (Connor et al.) that the HFE protein is found on brain capillaries, choroid plexus and ependymal cells along with Tf receptors. Consequently, iron uptake throughout the brain could be influenced by HFE mutations. Furthermore, we have reported additional evidence that in AD, the HFE protein appears to be induced on neurons, on the cells associated with neuritic plaques and on astrocytes associated with blood vessels.

In addition, three recent studies have now found associations between HFE gene mutations and dementia. In a North American study, men lacking both the apoE4 allele and the two major HFE mutations were less likely to have familial Alzheimer's disease (Moalem et al. 2000). Many symptoms of iron overload manifest earlier and more severely in men than women, possibly because of protection afforded by menstruation and perhaps also estrogen. Carriage of one or more allelic copies of the major C282Y mutation was significantly associated with early onset AD (50-65 yrs) in a set of Australian and Irish men (Milward et al. unpublished observations). In addition, Sampietro and colleagues (2001) have reported that in an Italian sample, where the C282Y mutation is very rare, onset of AD occurred about 5 years earlier in subjects carrying one or more copies of the H63D mutation, independent of gender. In patients under 70 yrs at disease onset, the frequency of the H63D mutation was five times higher than in those over 80 yrs at onset. These studies suggest that not just homozygosity but also heterozygosity for the main HFE mutations may influence AD pathogenesis. If this is true, as many as 25% or more of Americans and others of European descent could be affected.

Previous attempts to treat AD with antioxidants or iron chelation have been controversial and the outcomes ambiguous. Patients carrying hemochromatosis mutations may respond better to therapeutic interventions directed at minimizing iron-induced oxidative injury. Alternatively, these patients may be more resistant to iron chelation or anti-oxidant therapy because of an increased iron burden. Different predispositions to iron accumulation need to be taken into account in implementing future studies. The particular urgency to exploring and defining the relationship between AD and hemochromatosis is that life-style modifications may substantially affect the onset of AD.


Connor JR, Milward EA, Moalem S, Sampietro M, Boyer P, Percy ME, Vergani C, Scott RJ, Chorney M. Is hemochromatosis a risk factor for Alzheimer's disease? Journal of Alzheimer Disease. 2001 Oct. 3(5):471-477. Abstract

Moalem S, Percy ME, Andrews DF, Kruck TP, Wong S, Dalton AJ, Mehta P, Fedor B, Warren AC. Are hereditary hemochromatosis mutations involved in Alzheimer disease? Am J Med Genet. 2000 Jul 3;93(1):58-66. Abstract

Sampietro M, Caputo L, Casatta A, Meregalli M, Pellagatti A, Tagliabue J, Annoni G, Vergani C. The hemochromatosis gene affects the age of onset of sporadic Alzheimer's disease. Neurobiol Aging. 2001 Jul-Aug;22(4):563-8. Abstract

Comment by Maire Percy and Theo Kruck
Posted 22 January 2002

Re: Paragraph 7 of discussion paper "Previous attempts to treat AD with antioxidants or iron chelation have been controversial and the outcome ambiguous."

With due respect for the discussant's opinion, we wish to draw attention to two promising clinical trials with chelation therapy or antioxidants. It is suggested that the controversial issue is not so much whether or not these two studies are reliable, but why such treatments actually have worked.

The analysis of longitudinal neurocognitive and neurobehavioural data in Alzheimer disease (AD) clinical trials is difficult due to missing data, floor and ceiling effects in components of the test battery, and non-linearity in the rate of decline in neurocognitive and/or neurobehavioural functions (1). Furthermore, there currently is much debate as to the role that excesses of metals such as aluminum, iron, copper and zinc might play in the development and /or progression of AD. Nevertheless, two clinical trials have demonstrated effectiveness of treatment with chelation or treatment with antioxidants in slowing the progression of AD (2,3).

Because focal deposits of Al 3+ with or without co-deposits of Fe3+ have been observed in the AD brain, and a correlation has been established between the increased risk for developing sporadic AD (ADand increased levels of Al3+ in drinking water, a clinical trial was undertaken to see if treatment with the iron and aluminum chelator desferrioxamine mesylate (DFO) would slow down the progression of AD (2). This study involved 50 patients with AD; 25 were treated with 125 mg of DFO twice a day for 5 days a week over a 24 month period by muscular injection. It was demonstrated that the rate of mental deterioration in the treated patients was reduced in excess of 50% over the 24 month trial period. This clinical trial demonstrated the effectiveness of DFO in altering beneficially the course of AD. In our opinion, there is nothing ambiguous or controversial about the outcome of this particular study. What is controversial, however, is why DFO had a beneficial effect. Not known is if it reduced body and/or brain levels of aluminum, or iron, or both.

Because of mounting evidence that excessive oxidation might be involved in the development or progression of AD, a clinical trial was undertaken to see if selegiline, alpha-tocopheral or both might slow down the progression of the disease in patients with moderately severe impairment from AD (3). Both compounds, in fact, were found to have beneficial effects. The outcome of this second is not controversial; it is promising. But again, what is controversial is why such treatments have had an effect. For example, is it the anti-oxidant property of alpha-tocopheral or its anti-coagulant effect that is important in slowing down the progression of AD? -Maire Percy and Theo Kruck, Surrey Place Centre and the University of Toronto

Future research should attempt to identify the molecular targets of these compounds and the pathological processes that they are affecting to maximize treatment effectiveness and tolerance.

1. Thomas GR, Berg JD, Sano M, Thal L. 2000. Analysis of longitudinal data in an Alzheimer disease clinical trial. Stat Med 19:1433-40.

2. Crapper-McLachlan DR, Dalton AJ, Kruck TPA, Bell MY, Smith W, Kalow W, Andrews DF. 1991. Intramuscular desferrioxamine in patients with Alzheimer's disease. Lancet 337:1304-1308.

3. Sano M, Ernesto C, Thomas RG, Klauber MR, Schafer K, Grundman M, Woodbury P, Growdon J, Cotman CW, Pfeiffer E, Schneider LS, Thal LJ.1997.A controlled trial of selegiline, alpha-tocopherol, or both as treatment for Alzheimer's disease. The Alzheimer's Disease Cooperative Study. N Engl J Med. 1997 Apr 24;336(17):1216-22.

Comment by Hyman M. Schipper
Posted 29 January 2002

1) Although the epidemiologic data linking heterozygosity for the hfe mutations with earlier expression of Alzheimer disease (AD) are intriguing, I am having some difficulty reconciling certain facts related to the proposed mechanism of deranged brain iron deposition in patients harbouring these mutations: There exists ample evidence that the excessive sequestration of iron in the brains of Alzheimer and Parkinson (PD) patients occurs via non-transferrin pathways. Indeed, there appears to be no paucity of brain iron accumulation in atransferrinemic rodents (Connor et al.). It is believed that the normal hfe protein attenuates the affinity of the transferrin receptor for diferric transferrin (Jazwinska, 1998). In hemochromatosis, there is less inhibition of transferrin receptor/transferrin interaction due to mutation of the hfe protein, resulting in excessive tissue iron loading. Thus, if pathological brain iron deposition in AD and PD is independent of the transferrin receptor, how do hfe mutations affect the disposition of brain iron and influence disease progression in these neurodegenerative conditions?

2) In normal human brain, immunoreactive hfe protein is detectable in choroid plexus, ependymal cells and endothelial cells, but not in neurons (Connor et al. J Alz Dis 3: 471-477, 2001). In AD brain, there appears to be additional expression of hfe protein in neurons, astrocytes and senile plaques (ibid.). If, as mentioned above, the hfe protein attenuates the affinity of the transferrin receptor for diferric transferrin, shouldn't the up-regulation of the hfe protein (normal or mutant) be construed as an attempt by the brain to suppress iron deposition in AD rather than be interpreted as contributory to the disease process?

3) Is the age of onset of Parkinson's disease, a condition characterized by excessive iron deposition in the substantia nigra, also accelerated in patients bearing the C282Y or H63D mutations? Is the hfe protein expressed in the human nigra and, if so, in which cell types?

4) Moalem et al cite literature implicating elevated levels of circulating lead and aluminum in individuals bearing hfe mutations. Given the longstanding debate concerning the role of brain aluminum deposition in AD, it may be worthwhile to determine whether brain aluminum levels in AD patients correlate with hfe status.

5) Moalem et al describe some interesting gender differences in the relationship of hfe status to the development of familial AD. Several authors have argued that menses-related iron losses may be protective in women homozygous or heterozygous for mutant hfe. Is there any evidence that sex hormones modulate tissue iron stores by directly affecting the expression of the hfe protein?


  1. I have read with interest the comment on HFE mutations and AD. Our group also postulated that a genetic defect in iron metabolism could be associated with a greater risk for AD. Therefore, we analyzed the two common HFE mutations and the Transferrin C2 allele in 108 AD patients and 110 controls. Against our hypothesis we did not find differences between AD and controls. This will be published this year in J Neurol Neruosurg and Psychiatry. In my opinion the next step would be to investigate the abnormalities in brain iron metabolism associated with these genetic polymorphisms.

Make a Comment

To make a comment you must login or register.


Webinar Citations

  1. Hemochromatosis as a Factor in AD

Other Citations

  1. Liz Milward

External Citations

  1. Connor et al.
  2. Moalem et al. 2000
  3. Sampietro and colleagues (2001)
  4. Abstract
  5. Abstract
  6. Jazwinska, 1998)

Further Reading


  1. . Apolipoprotein E-epsilon4 genotype predicts a poor outcome in survivors of traumatic brain injury. Neurology. 1999 Jan 15;52(2):244-8. PubMed.