Search

In a previous study of lung cancer patients and controls from the Stockholm area in central Sweden, we found a significantly decreased frequency of the transferrin (TF) variant C3 in small cell and squamous epithelial lung cancer but not in adenocarcinoma, suggesting a protective effect of TF C3 in small cell and squamous lung cancer. In an attempt to replicate this association we studied TF types in lung cancer patients and controls from two additional Swedish subpopulations, viz. northern Sweden and southwestern Sweden. We were able to confirm the significantly decreased frequency of TF C3, especially in small cell lung cancer, in northern Sweden but not in southwestern Sweden. Thus the eventual protective effect of TF C3 in small cell lung cancer is an open question. We hypothesize that the association between TF C alleles and lung cancer may be secondary and dependent on linkage disequilibrium with allelic variants of newly discovered tumor-associated genes known to map to the same position (3q21) as TF, e.g. NCK and H-RYK.

Although a large number of human serum transferrin (TF) variants have been described, only one RFLP (AvaI) has so far been found. Here we report three new RFLPs (MvaI in intron 5 and exon 7, BbvI in exon 7) and correlations between RFLPs and between RFLPs and serum TF types. There were strong, but not always complete, disequilibria between RFLP and serum protein alleles. Thus, the most common serum TF variant, C1, was heterogeneous and could be subdivided into two common haplotypes, whereas the C2, C3, and DCHI variants were completely or almost completely (C2) homogeneous. There was a total genotypic agreement between the BbvI polymorphism and the presence/absence of the TF C3 variant, and the mutation that creates the BhvI site was found to lead to a G258S amino acid substitution.

In a study of transferrin receptor (TFR) polymorphism in different ethnic groups using PCR and restriction cleavage we found a new Hin6I polymorphism in intron 7 and confirmed a tentative BanI polymorphism in exon 4 reported by Evans and Kemp [Gene 1997;199:123–131]. In all ethnic groups there was a complete and highly significant (p < 10–10) linkage disequilibrium where all BanI 1 alleles were linked to Hin6I 1 alleles. Furthermore in the European populations, but not in the Chinese, there was a close correlation between the three BanI-Hin6I haplotypes and the alleles of a previously described three-allelic RsaI polymorphism in the TFR gene studied by Southern blotting. There were distinct ethnic differences in TFR allele and haplotype frequencies. Thus the Saamis were significantly different from the other European ethnic groups, and the Lithuanians had a significantly increased frequency of the BanI 2-Hin6I 1 haplotype, suggesting that this marker may be informative in tracing prehistoric migrations and admixture by Baltic peoples. The new TFR polymorphisms and haplotypes may also be useful markers in studies of interactions with the transferrin and hemochromatosis genes, the genetic influence on body iron stores and disease associations.

In hepatocellular carcinoma (HCC) iron has been implicated as a risk factor primarily in patients with hereditary haemochromatosis (HH) and cirrhosis. The wild-type HH (HFE) protein complexes with the transferrin receptor (TFR), and two HFE mutations (Cys282Tyr and His63Asp) have been found to increase the affinity of the TFR for transferrin resulting in an increased cellular uptake of iron. In previous studies we found an interaction between HFE and TFR genotypes in multiple myeloma and breast and colorectal carcinomas. In the present investigation we have studied HFE and TFR genotypes in 54 Swedish patients with HCC, using DNA from archival samples of paraffin wax blocks. The same HFE-TFR interaction as in the previously studied neoplastic disorders was found. Individuals carrying the HFE282Tyr allele (homo- and heterozygotes) in combination with homozygosity for the TFR Ser allele showed an increased risk for HCC (OR = 3.5; 95% confidence interval, CI = 1.3–9.3), which was further increased in HFE Tyr homozygotes and compound (Tyr/Asp) heterozygotes in combination with TFR 142Ser homozygosity (OR = 17.2; 95% CI = 1.8–168.9). The presence of liver cirrhosis could only be assessed in part of the patient material. In patients with verified liver cirrhosis the risk figures were substantially increased: for HFE 282 Tyr carriers in combination with TFR 142Ser/Ser OR = 7.2; 95% CI = 2.0–25.5 and for HFE 282Tyr homozygotes and compound heterozygotes in combination with TFR 142Ser homozygosity, OR = 62.8; 95% CI = 6.1–642.5.

In contrast to CuZn superoxide dismutase (SOD), only a very limited number of mutations have been described in MnSOD. One interesting example is a polymorphism (Ala-9Val) in the mitochondrial targeting sequence of this radical-scavenging enzyme. We have studied the Ala-9Val polymorphism in various ethnic groups by means of the oligonucleotide ligation assay. There were significant variations in this unique polymorphism between three different language groups: Baltic (Lithuanians), Finnic (Finns and Saamis) and Germanic (Swedes). The Ala frequency in an Asiatic population (Chinese) was significantly lower than in most European populations. This polymorphism may affect the mitochondrial targeting rate of MnSOD which may result in mitochondrial damage with implication in various late-onset neurological diseases.

A number of genes are involved in iron metabolism, including the transferrin receptor (TFR) and haemochromatosis (HFE) genes. In previous investigations an increased risk for neoplastic disease has been observed in individuals homo- and heterozygous for hereditary haemochromatosis. The HFE wild-type gene product complexes with the transferrin receptor (TF) and two different HFE mutations (Cys282Tyr and His63Asp) have been found to increase the affinity of TFR for TF and increase cellular iron uptake. In a recent study we found no associations for HFE and TFR separately, but an interaction between HFE and TFR genotypes in multiple myeloma. Individuals carrying the HFE Tyr282 allele (homo- and heterozygotes) in combination with homozygosity for the TFR Ser142 allele had an increased risk. In the present study the same association was found in breast and colorectal cancer. The odds ratio for all three neoplasms combined was 2.0 (95% CI 1.0-3.8). The risk for neoplastic disease was further increased (OR 7.7, 95% CI = 1.0-59.9) when the analysis was restricted to HFE Tyr homozygotes and compound heterozygotes in combination with TFR Ser homozygosity. Thus, an interaction between HFE and TFR alleles may increase the risk for different neoplastic disorders.

Recent studies have shown that hereditary hemochromatosis (HH) is likely to be caused by homozygosity for a Cys282Tyr mutation in the HFE gene located 4.5 Mb telomeric to HLA-A. Population studies of this polymorphism are facilitated by the fact that the Cys282Tyr mutation creates a Rsal restriction site. We have studied the codon 282 (Cys/Tyr) polymorphism in different ethnic groups. In agreement with previous observations the Tyr allele appeared to be rare or absent in Asiatic (Indian, Chinese) populations. The highest allele frequency (7.5%) was found in Swedes. Saamis (2%) and Mordvinians (1.8%) had significantly lower frequencies of the Tyr allele. Comparisons with allele frequencies based on prevalence estimates of HH showed some disagreements with the RFLP data, particularly in Finns. The newly described HFE marker provides a new approach to the screening of HH as well as studies of the relationship between the HFE Tyr allele and different disorders including cancer.