Experiment
Disorders of iron metabolism are among the most common acquired and constitutive diseases.
Iron homeostasis is maintained through the intestinal absorption of a limited amount of iron to compensate for daily losses, and the recycling of heme iron following phagocytosis of senescent red blood cells by macrophages.
In the current study a specific linkage approach using recombinant congenic strains (RCSs) of the C3H/DiSnA - C57BL/10ScSnA/Dem series, HcB, was developed to identify genes that modify iron homeostasis. The B6 strain exhibits statistically significant lower tissue iron than C3H, in the HcB/Dem strains the genes responsible for the low tissue iron level are present on the genetic background of the high tissue iron strain C3H/DiSnA.
A total of 225 F2 mice were generated six (C3H/DiSnA x HcB15/Dem)F1 x (C3H/DiSnA x HcB15/Dem)F1 mating pairs. One hundred F2 mice were generated from 7 (C3H/HeNrj x C57BL/6) F1 x (C3H/HeNrj x C57BL/6) F1 mice. Iron contents of tissue (liver and spleen) have been reported to differ between male and female mice; thus only male mice that displayed the highest tissue iron content were selected for subsequent QTL analysis.
Serum iron and transferrin levels were measured on an Olympus AU400 automatic analyzer.
Tissue iron content was determined by acid digestion of tissue samples followed by iron quantification using an IL test. Tail genomic DNA was extracted using a QIAamp DNA purification kit. Differential segments were genotyped using a PCR-based method in the F2 HcB-15/Dem x C3H/DiSnA hybrid mice using 36 microsatellite markers. The C3H/HeNrj x C57BL/6 F2 hybrids were genotyped solely for the two markers (D9Mit184 and D1Mit234) that displayed the highest linkage in the C3H/DiSnA x HcB15 cross.
Genotypes were compared with phenotypes to identify QTLs by linkage analysis. Statistical analysis revealed strong linkage of the spleen iron content, and to a lesser extent the liver iron content, with marker D9Mit184 on Chromosome 9.
QTL Splic1 (spleen iron content 1) mapping with marker D9Mit184 had an LRS=70.3, LOD=15.5 with a p value of 0.00001. The C3H/DiSnA allele was associated with higher spleen iron contents and accounted for 29% of trait variance. [Fig 1A].
QTL Lvic6 (liver iron content 6) also mapped with marker D9Mit184. It had an LRS=36.4, LOD=7.9 with a p value of 0.00001. The C3H/DiSnA allele was associated with higher liver iron content and accounted for 15% of the trait variance. The best fitting inheritance model was the dominance of the B10 allele over the C3H/DiSnA allele for both the spleen and the liver QTL.
In previous studies a missense N374S mutation has been identified in Mon1a that was assumed to be responsible for the low macrophage iron loading in the C57BL/10J mice. The same mutation was found here in the C57BL/6 donor strain in the HcB15 RCS strain suggesting that a founder event had occurred in the C57BL/6 lineage.
A third QTL was identified close to marker D1Mit234. Splic2 (spleen iron content 2) was identified with a LRS=22.6, LOD score of 5.0 and a p value of 0.00002. The locus was attributed with 8% of the trait variance [Fig 2A]. The C3H/DiSnA allele was dominant over the B10 allele. D1Mit234 is 30 kb downstream from Slc40a1, a transmembrane iron exporter in macrophages, making it a possible candidate gene for Splic2.
An interaction was detected between D9Mit184 and D1Mit234, with an LRS value for the interacting factor of 36.8, exceeding the permutation derived threshold of 35.8, indicative of highly significant evidence for interaction (p<0.001). The lowest spleen iron appeared to be attributable mainly to B10 homozygosity at both loci.
Macrophage iron content is modulated by the amount of Slc40a1 present at the cell surface. To determine whether Slc40a1 expression depended on the chromosome 9 and chromosome 1 genotypes, cultures were established from bone marrow-derived macrophages isolated from D9Mit184BB mice to be CC, BB, or BC at D1Mit234. Slc40a1 was detected predominantly in cytosolic vesicles, but there appeared to be more Slc40a1 on the plasma membrane in D1Mit234BB macrophages than in D1Mit234CC macrophages. To evaluate the subsequent effect of Slc40a1 cell surface expression on intracellular iron retention among the three genotypes, the intracellular ferritin content was measured as an indirect measurement of iron content.
Interestingly, the D1Mit234BB macrophages, which display more cell surface Slc40a1, had significantly lower ferritin content (Fig. 4b). Taken together, these results suggest that the gain-of-function effect of the N374S Mon1a mutation on Slc40a1 is restricted to animals that are homozygous for the B10 allele at the D1Mit234 locus. This locus is located 30 kb downstream of Slc40a1, suggesting that a functional epistatic interaction between Mon1a and Slc40a1 may occur.
Previously a nonsense R435X mutation in the Cp gene was identified in the C3H/DiSnA strain that markedly increased liver iron but not spleen iron. Whether this specific background might modulate the effect of the N374S Mon1a mutation on liver iron was tested.
QTL analysis was performed of 100 F2 hybrids between C3H/HeNrj mice, which are free of the R435X Cp mutation, and C57BL/6 mice, which carry N374S Mon1a (Table 1). Results mapped a third QTL, Splic3 (spleen iron content 3) that in this genetic background mapped with marker D9Mit184 with a LOD score 8.3, p value <0.00001, and accounted for 32% of trait variance; but no longer affected liver iron load (LOD score = 1.5; Fig. 5b), confirming that the lowering effect of the N374S Mon1a mutation on liver iron is dependent upon the absence of expression of Cp in the liver.
Interestingly, the low spleen iron level of D9Mit184BB (C3H/HeNrj x C57BL/6) F2 hybrids (Fig. 5a) was again attributable mainly to mice homozygous for the B6 background at D1Mit234 (Fig. 5c). This demonstrated that in contrast to liver iron, the epistatic interactions between D9Mit184 and D1Mit234 loci in modulating spleen iron load were independent of the nonsense Cp mutation.
The study highlighted the existence of genetic interactions between Cp, Mon1a, and the Slc40a1 locus in iron metabolism, suggesting that epistasis may be a crucial determinant of the variable biological and clinical presentations in iron disorders.
Analysis Tools