• Alzheimer's disease
• iron
• hemochromatosis
• transferrin

Alzheimer's disease is a neurodegenerative disease characterised pathologically by the presence of neurofibrillary tangles, senile plaques, and selective loss of neurons. Numerous hypotheses have been suggested for the aetiology and pathogenesis of Alzheimer's disease and one that has gained considerable attention is the disruption of the brain iron metabolism in Alzheimer's disease that could lead to an oxidative stress and neuronal damage.¹ An increased iron deposition has been found in the Alzheimer's disease brain, especially in the regions containing more senile plaques and neurofibrillary tangles.^1,2 Tissue iron can promote oxidative damage through an increase of free radical formation that can lead to subsequent oxidative stress.¹ Among genetic risk factors associated with Alzheimer's disease, the APOE genotype is the major genetic risk factor for sporadic and familial late onset disease. Recently, two genetic risk factors involved in iron metabolism have been associated with an increased risk for Alzheimer's disease. The first one is the allele C2 of the transferrin (Tf) gene, an iron transporting protein detected in senile plaques.^3,4 In another study⁵ performed on a small group of patients, mutations in the haemochromatosis associated gene (HFE) were overrepresented in Alzheimer's disease compared with controls. We postulated that if these genetic defects in iron metabolism were indeed involved in the pathogenesis of Alzheimer's disease they should be detected in independent populations. Thus, in the present work we investigated whether the C2 allele of the Tf gene or the two common HFE mutations were involved in the pathogenesis or were a disease modifying factor in our Alzheimer's disease population.

We included in this study 108 patients with Alzheimer's disease (80 woman) recruited from both community (n=37) and clinic (n=71) sources. The control sample consisted of 110 unrelated subjects (68 woman) recruited from the community (n=44) and clinic sources (n=66). All control subjects underwent a complete neurological and neuropsychological examination to exclude medical illness and cognitive impairment. All patients were fully evaluated and met the conventional NINCDS-ADRA criteria for probable Alzheimer's disease. After informed consent a blood sample was obtained from patients and controls.

The Tf polymorphism (codon 570) was determined after polymerase chain reaction (PCR) amplification using the mismatched sense primer 5′-GCTGTGCCTT GATGGTACC AGGTAA-3′ and antisense primer 5′-GGA CGCA AGCTTCCTTATCT-3′ as described.³ Polymorphic exon 15 was amplified from genomic DNA using described conditions.³ The 110 bp product was digested with BstEII, separated in a 6% polyacrylamide gel, and stained with silver nitrate. After digestion the C1 allele was converted to a 89 bp fragment while the C2 allele remained 110 bp long. We also genotyped the two common mutations (H63D and C282Y) involved in hereditary haemochromatosis.⁵ APOE genotyping was performed through PCR amplification and HhaI restriction enzyme digestion. Allelic and genotypic distributions were analysed by the χ² test with the SPSS (version 10.0) statistical package.

Mean age for patients and controls was 78.8 (range 61 to 93) and 73.6 years (range 45 to 92) respectively. Both populations were in Hardy-Weinberg equilibrium for all the polymorphisms. The HFE mutation frequency in the control group was consistent with the frequency of the Spanish population. The frequency distribution of the Tf C2 allele, and C282Y and H63D genotypes among patients with Alzheimer's disease and controls is given in table 1. We did not find associations between Tf C2 allele, H63D, and C282Y mutation frequencies and Alzheimer's disease. Stratification for sex yielded a trend toward an increase in the H63D mutation frequency among male patients with Alzheimer's disease (53.6%) compared with male controls (33.3%, p=0.09). Stratification for age or APOE status did not yield any significant difference. As expected APOE ε4 was increased in the group of patients (47.2% at least one ε4 allele) compared with controls (11.8%, p<0.0001).

In this study we did not find any significant association between the Tf C2 allele or the two common mutations in the HFE gene (H63D and C282Y) and Alzheimer's disease. However, this is by contrast with several studies that have indicated that there is a disruption of brain iron metabolism in Alzheimer's disease.^1,2 In neuropathological studies iron has been found to be increased in the brain in Alzheimer's disease, especially in regions containing abundant neurofibrillary tangles and senile plaques such as the hippocampus and amygdala.² In particular, selective accumulation of iron has been found within the neurofibrillary tangles and senile plaques in the Alzheimer's disease brain.^1,2 It is of interest that iron is specifically localised to lesions of Alzheimer's disease and not the glial cells surrounding senile plaques, which contain abundant iron binding proteins.¹ Thus, the accumulation of iron in the Alzheimer's disease brain and the increasing reports implicating oxidative stress, lead us to hypothesise that genetic factors involved in iron metabolism, such as the C2 allele of Tf gene and HFE mutations, could act as a risk factor for the disease. In fact, the C2 allele of the transferrin gene has been associated with an increased risk for Alzheimer's disease in some studies.^3,4 Furthermore, the two mutations of the HFE gene involved in hereditary haemochromatosis, have also been associated with an increased risk for other diseases, such as dilated cardiomyopathy, myocardial infarction, and type 2 diabetes, which are common complications of iron overload. There is only one study⁵ assessing HFE mutations in Alzheimer's disease. In this study, which was performed in 26 patients with familial Alzheimer's disease, HFE mutations were overrepresented in the group of patients compared with controls.

However, our study is the first assessing HFE mutations in Alzheimer's disease using a large sample. Based on our results neither the C2 allele of the Tf gene nor the HFE mutations were associated with an increased risk for Alzheimer's disease. Thus, the effect of the C2 allele of the Tf gene seems to be lower than previously reported. However, our study can not address the influence of these genetic factors on iron deposition. Resolving this point deserves further studies evaluating iron quantification in vivo using MRI or at neuropathological examination.