Lack of Plasma Protein Hemopexin Results in Increased Duodenal Iron Uptake

Abstract

Purpose: The body concentration of iron is regulated by a fine equilibrium between absorption and losses of iron. Iron can be absorbed from diet as inorganic iron or as heme. Hemopexin is an acute phase protein that limits iron access to microorganisms. Moreover, it is the plasma protein with the highest binding affinity for heme and thus it mediates heme-iron recycling. Considering its involvement in iron homeostasis, it was postulated that hemopexin may play a role in the physiological absorption of inorganic iron.

Methods and results: Hemopexin-null mice showed elevated iron deposits in enterocytes, associated with higher duodenal H-Ferritin levels and a significant increase in duodenal expression and activity of heme oxygenase. The expression of heme-iron and inorganic iron transporters was normal. The rate of iron absorption was assessed by measuring the amount of (57)Fe retained in tissues from hemopexin-null and wild-type animals after administration of an oral dose of (57)FeSO4 or of (57)Fe-labelled heme. Higher iron retention in the duodenum of hemopexin-null mice was observed as compared with normal mice. Conversely, iron transfer from enterocytes to liver and bone marrow was unaffected in hemopexin-null mice.

Conclusions: The increased iron level in hemopexin-null duodenum can be accounted for by an increased iron uptake by enterocytes and storage in ferritins. These data indicate that the lack of hemopexin under physiological conditions leads to an enhanced duodenal iron uptake thus providing new insights to our understanding of body iron homeostasis.

Figure 1. Hx deficiency results in increased iron deposits in the duodenum.

(A) Total iron content in the duodenum of 2 month-old wild-type and Hx-null mice measured by ICP-MS. Values are expressed as μg iron/g wet tissue. Data represent mean ± SEM; n=10; * = P<0.05. (B) Duodenal sections of a wild-type mouse (i, ii) and an Hx-null mouse (iii, iv) at 2 months of age stained with DAB enhanced Perls’ reaction. Note the more intense staining in the duodenum of the Hx-null mouse. Bar i, iii = 500µm; bar ii, iv = 100µm.

Figure 2. Increased H-Ft expression in Hx-null mice duodenum.

Representative Western blots of H-Ft (A) and L-Ft (C) expression in the duodenum of wild-type and Hx-null mice. Band intensities were measured by densitometry and normalized to actin expression. Densitometry data represent mean ± SEM; n=3 for each genotype. (B) Duodenal sections of a wild-type mouse (i) and an Hx-null mouse (ii) processed by immunohistochemistry with an anti-H-Ft antibody. The H-Ft-positive signal is increased in the duodenum of the Hx-null animal. Bar= 100µm.

Figure 3. Hx deficiency does not affect the expression of duodenal iron transporters.

(A) qRT-PCR analysis of DcytB, DMT1, Fpn1, TfR1 and Heph expression in the duodenum of wild-type and Hx-null mice. These assays do not discriminate between the different DMT1 and Fpn1 isoforms. The results of specific qRT-PCR assays for DMT1-IRE and DMT1-noIRE expression and for Fpn1A and Fpn1B expression are shown in (B) and (C), respectively. (D) qRT-PCR analysis of Hepc expression in the liver of wild-type and Hx-null mice. In A-D, transcript abundance, normalized to 18S RNA expression, is expressed as a fold increase over a calibrator sample. Data represent mean ± SEM, n= 6 for each genotype. (E) Representative Western blots of DMT1, Fpn1 and TfR1 expression in the duodenum of wild-type and Hx-null mice. Band intensities were measured by densitometry and normalized to actin expression. Densitometry data represent mean ± SEM; n=3 for each genotype. Results shown are representative of 3 independent experiments.

Figure 4. Hx deficiency results in enhanced heme catabolism in the duodenum.

(A) HO activity in the duodenum of wild-type and Hx-null mice. Data represent mean ± SEM; n= 8 for each genotype. * = P<0.05. (B) Representative Western blot of HO-1 expression in the duodenum of wild-type and Hx-null mice. Band intensities were measured by densitometry and normalized to actin expression. Densitometry data represent mean ± SEM; n=3 for each genotype. (C) Sections of the duodenum of a wild-type mouse (i, iv, vii) and an Hx-null mouse (ii, v, viii) stained with an antibody to HO-1. Enlarged details of sections i, ii, iii are shown in iv, v, vi respectively The HO-1-positive signal was more intense in the Hx-null mouse than in the wild-type control (arrows) Sections on the right (iii, vi, ix) represent negative controls in which the primary antibody was omitted. Bar i, ii, iii = 100µm; bar iv, v, vi = 57 µm; bar vii, viii, ix = 20 µm.

Figure 5. Hx deficiency does not affect the expression of duodenal heme transporters.

(A) qRT-PCR analysis of PCFT/HCP1, FLVCR1, ABCG2 and HRG-1 expression in the duodenum of wild-type and Hx-null mice. Transcript abundance, normalized to 18S RNA expression, is expressed as a fold increase over a calibrator sample. Data represent mean ± SEM, n= 6 for each genotype. (B) Representative Western blot of PCFT/HCP1 expression in the duodenum of wild-type and Hx-null mice. Band intensities were measured by densitometry and normalized to actin expression. Densitometry data represent mean ± SEM; n=3 for each genotype. Results shown are representative of 3 independent experiments.

Figure 6. Hx deficiency results in enhanced iron uptake in the duodenum cells.

(A) ⁵⁷Fe retention in the duodenal mucosa of wild-type and Hx-null mice measured by ICP-MS 30, 60, 90, 135 and 180 minutes after oral administration of a solution containing 20 mmol/L ⁵⁷FeSO₄. Control mice were administered vehicle solution and represented the “0” time point of the experiment. Values are expressed as μg ⁵⁷Fe/ g tissue. Data represent mean ± SEM; n=6 for each experimental point; *** = P<0.001 (comparing control mice with the corresponding group of ⁵⁷FeSO₄-administered mice), # # = P<0.01 (comparing the two genotypes). (B) ⁵⁷Fe retention in the duodenal mucosa of wild-type and Hx-null mice measured by ICP-MS 30, 60, 90 and 135 minutes after oral administration of a solution containing 20 mmol/L ⁵⁷Fe labelled heme (⁵⁷Fe-heme). Control mice were administered vehicle solution and represented the “0” time point of the experiment. Values are expressed as μg ⁵⁷Fe/ g tissue. Data represent mean ± SEM; n=6 for each experimental point; *** = P<0.001 (comparing control mice with the corresponding group of ⁵⁷Fe-heme-administered mice), # = P<0.05 (comparing the two genotypes).

Figure 7. Hx deficiency does not affect iron transfer from the duodenum to other tissues.

(A) ⁵⁷Fe retention in the liver of wild-type and Hx-null mice measured by ICP-MS 30, 60, 90, 135 and 180 minutes after oral administration of a solution containing 20 mmol/L ⁵⁷FeSO₄. Control mice were administered vehicle solution and represented the “0” time point of the experiment. Values are expressed as μg ⁵⁷Fe/ g tissue. Data represent mean ± SEM; n=6 for each experimental point; * = P<0.05, ** = P<0.01 (comparing control mice with the corresponding group of ⁵⁷FeSO₄-administered mice). (B) ⁵⁷Fe retention in the bone marrow of wild-type and Hx-null mice measured by ICP-MS 48 hours after oral administration of a solution containing 20 mmol/L ⁵⁷FeSO₄. Control mice were administered vehicle solution and represented the “0” time point of the experiment. Values are expressed as μg ⁵⁷Fe/ g tissue. Data represent mean ± SEM; n=10 for each experimental point; * = P<0.05. (C) ⁵⁷Fe retention in the kidney of wild-type and Hx-null mice measured by ICP-MS 30, 60, and 90 minutes after oral administration of a solution containing 20 mmol/L ⁵⁷FeSO₄. Control mice were administered vehicle solution and represented the “0” time point of the experiment. Values are expressed as μg ⁵⁷Fe/ g tissue. Data represent mean ± SEM; n=6 for each experimental point.